Special Issue: KDIGO Clinical Practice Guideline for the Care of Kidney Transplant Recipients

Abstract

The 2009 Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guideline on the monitoring, management, and treatment of kidney transplant recipients is intended to assist the practitioner caring for adults and children after kidney transplantation. The guideline development process followed an evidence-based approach, and management recommendations are based on systematic reviews of relevant treatment trials. Critical appraisal of the quality of the evidence and the strength of recommendations followed the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) approach. The guideline makes recommendations for immunosuppression, graft monitoring, as well as prevention and treatment of infection, cardiovascular disease, malignancy, and other complications that are common in kidney transplant recipients, including hematological and bone disorders. Limitations of the evidence, especially on the lack of definitive clinical outcome trials, are discussed and suggestions are provided for future research.

Foreward

Since the first successful kidney transplantation in 1954, there has been an exponential growth in publications dealing with the care of kidney transplant recipients (KTRs). In addition, the science of conducting and interpreting both clinical trials and observational studies has become increasingly controversial and complex. Caring for KTRs requires specialized knowledge in areas as varied as immunology, pharmacology, nephrology, endocrinology and infectious disease. The last two comprehensive clinical practice guidelines on the care of KTRs were published in 2000 by the American Society of Transplantation and the European Best Practices Guidelines Expert Group. Both of these guidelines were based primarily on expert opinion, not rigorous evidence review. For these reasons, the international consortium of kidney guideline developers, Kidney Disease: Improving Global Outcomes (KDIGO), concluded that a new comprehensive evidence-based clinical practice guideline for the care of KTRs was necessary.

It is our hope that this document will serve several useful purposes. Our primary goal is to improve patient care. We hope to accomplish this in the short term by helping clinicians know and better understand the evidence (or lack of evidence) that determines current practice. By making this guideline broadly applicable, our purpose is to also encourage and enable the establishment and development of transplant programs worldwide. Finally, by providing comprehensive evidence-based recommendations, this guideline will also help define areas where evidence is lacking and research is needed. Helping to define a research agenda is an often neglected, but very important function of clinical practice guideline development.

We used the GRADE system to rate the strength of evidence and the strength of recommendations. In all, there were only 4 (2%) recommendations in this guideline for which the overall quality of evidence was graded ‘A,’ whereas 27 (13.6%) were graded ‘B,’ 77 (38.9%) were graded ‘C,’ and 90 (45.5%) were graded ‘D.’ Although there are reasons other than quality of evidence to make a grade 1 or 2 recommendation, in general, there is a correlation between the quality of overall evidence and the strength of the recommendation. Thus, there were 50 (25.3%) recommendations graded ‘1’ and 148 (74.7%) graded ‘2.’ There were 3 (1.5%) recommendations graded ‘1A,’ 16 (8.1%) were ‘1B,’ 18 (9.1%) were ‘1C,’ and 13 (6.6%) were ‘1D.’ There was 1 (0.5%) graded ‘2A,’ 11 (5.6%) were ‘2B,’ 59 (29.8%) were ‘2C,’ and 77 (38.9%) were ‘2D.’ There were 45 (18.5%) statements that were not graded.

Some argue that recommendations should not be made when evidence is weak. However, clinicians still need to make clinical decisions in their daily practice, and they often ask ‘what do the experts do in this setting’? We opted to give guidance, rather than remain silent. These recommendations were often rated with a low strength of recommendation and a low strength of evidence, or were not graded. It is important for the users of this guideline to be cognizant of this (see Disclaimer). In every case these recommendations are meant to be a place for clinicians to start, not stop, their inquiries into specific management questions pertinent to the patients they see in daily practice.

We wish to thank Martin Zeier, Co-Chair, along with all of the Work Group members who volunteered countless hours of their time developing this guideline. We also thank the Evidence Review Team members and staff of the National Kidney Foundation who made this project possible. Finally, we owe a special debt of gratitude to the many KDIGO Board members and individuals who volunteered time reviewing the guideline, and making very helpful suggestions.

Kai-Uwe Eckardt, MD KDIGO Co-Chair

Bertram L. Kasiske, MD KDIGO Co-Chair

Guideline Scope and Intended Users

This guideline describes the prevention and treatment of complications that occur after kidney transplantation. We do not include pretransplant care. Specifically, we do not address issues pertinent to the evaluation and management of candidates for transplantation, or the evaluation and selection of kidney donors.

Although many of the issues that are pertinent to KTRs are also pertinent to recipients of other organ transplants, we intend this guideline to be for KTRs only. We cover only those aspects of care likely to be different for KTRs than for patients in the general population. For example, we deal with the diagnosis and treatment of acute rejection, but not with the diagnosis and treatment of community-acquired pneumonia. We also make recommendations pertinent to the management of immunosuppressive medications and their complications, including infections, malignancies, and cardiovascular disease (CVD).

This guideline ends before the kidney fails, either by death of the recipient with a functioning graft, return to dialysis, or retransplantation. We do not deal with the preparation of KTRs for return to dialysis or retransplantation.

This guideline was written for doctors, nurses, coordinators, pharmacists, and other medical professionals who directly or indirectly care for KTRs. It was not developed for administrative or regulatory personnel per se. For example, no attempts were made to develop clinical performance measures. Similarly, this guideline was not written for patients directly, although carefully crafted explanations of guideline recommendations could potentially provide useful information for patients.

This guideline was written for transplant-care providers throughout the world. As such, it addresses issues that are important to the care of KTRs in both developed and developing countries, but nowhere was the quality of care compromised for utilitarian purposes. Nevertheless, we recognize that, in many parts of the world, treatment of end-stage kidney disease (chronic kidney disease [CKD] stage 5) with dialysis is not feasible, and transplantation can only be offered as a life-saving therapy if it is practical and cost-effective. Therefore, in providing a comprehensive, evidence-based guideline for the care of the KTRs, we were cognizant of the fact that programs in some areas of the world may need to adopt cost-saving measures in order to make transplantation possible.

Section I: Immunosuppression

Introduction

Kidney transplantation is the treatment of choice for CKD stage 5. The risk of death for KTRs is less than half of that for dialysis patients (1). Any differences in patient survival attributable to different immunosuppressive medication regimens are substantially smaller than the survival difference between dialysis and transplantation. Specifically, marginally inferior immunosuppressive medication regimens will result in substantially better patient outcomes than dialysis. Thus, it is better to perform kidney transplantation even with an inferior immunosuppressive regimen, than to avoid transplantation altogether.

Recommendations for immunosuppressive medications are necessarily complex, because combinations of multiple classes of drugs are used and because the choices among different regimens are determined by the tradeoffs between benefits and harm. Typically, a greater degree of immunosuppression may reduce the risk of rejection, but may also increase the risk of infection and cancer. Decision analysis with patient-based utilities may be needed to correctly assess the tradeoffs between benefits and harm, but this has not usually been done.

Rating Guideline Recommendations

Within each recommendation, the strength of recommendation is indicated as Level 1, Level 2, or Not Graded, and the quality of the supporting evidence is shown as A, B, C, or D.

Grade*	Wording
*The additional category ‘Not Graded’ was used, typically, to provide guidance based on common sense or where the topic does not allow adequate application of evidence. The most common examples include recommendations regarding monitoring intervals, counseling, and referral to other clinical specialists. The ungraded recommendations are generally written as simple declarative statements, but are not meant to be interpreted as being stronger recommendations than Level 1 or 2 recommendations.
Level 1	‘We recommend’
Level 2	‘We suggest’

Grade for quality of evidence	Quality of evidence
A	High
B	Moderate
C	Low
D	Very low

Chapter 1: Induction Therapy

• 1.1: 
  We recommend starting a combination of immunosuppressive medications before, or at the time of, kidney transplantation. (1A)
• 1.2: 
  We recommend including induction therapy with a biologic agent as part of the initial immunosuppressive regimen in KTRs. (1A)

  • 1.2.1: 
    We recommend that an IL2-RA be the first-line induction therapy. (1B)
  • 1.2.2: 
    We suggest using a lymphocyte-depleting agent, rather than an IL2-RA, for KTRs at high immunologic risk. (2B)

IL2-RA, interleukin 2 receptor antagonist; KTRs, kidney transplant recipients.

Background

Except perhaps for transplantation between identical twins, all kidney transplant recipients (KTRs) need immunosuppressive medications to prevent rejection. Induction therapy is treatment with a biologic agent, either a lymphocyte-depleting agent or an interleukin 2 receptor antagonist (IL2-RA), begun before, at the time of, or immediately after transplantation. The purpose of induction therapy is to deplete or modulate T-cell responses at the time of antigen presentation. Induction therapy is intended to improve the efficacy of immunosuppression by reducing acute rejection, or by allowing a reduction of other components of the regimen, such as calcineurin inhibitors (CNIs) or corticosteroids. Available lymphocyte-depleting agents include antithymocyte globulin (ATG), antilymphocyte globulin (ALG) and monomurab-CD3. Basiliximab and daclizumab, the two IL2-RAs that are currently available in many parts of the world, bind the CD25 antigen (interleukin-2 [IL2] receptor α-chain) at the surface of activated T-lymphocytes and thereby competitively inhibit IL2-mediated lymphocyte activation, a crucial phase in cellular immune response of allograft rejection.

Rationale

• • 
  There is high-quality evidence that the benefits of IL2-RA vs. no IL2-RA (or placebo) outweigh harm in a broad range of KTRs with variable immunological risk and concomitant immunosuppressive medication regimens.
• • 
  There is moderate-quality evidence that a lymphocyte-depleting agent vs. no lymphocyte-depleting agent (or placebo) reduces acute rejection and graft failure in high-immunological-risk patients.
• • 
  There is moderate-quality evidence across a broad range of patients with different immunological risk and concomitant immunosuppressive medication regimens, which shows that (compared to IL2-RA) lymphocyte-depleting agents reduce acute rejection, but increase the risk of infections and malignancies.
• • 
  Economic evaluations for IL2-RA demonstrate lower cost and improved graft survival compared with placebo.
• • 
  Although there are sparse data in KTRs <18 years old, there is no biologically plausible reason why age is an effect modifier of treatment, and the treatment effect of IL2-RA appears to be homogenous across a broad range of patient groups.
• • 
  Induction therapy with a lymphocyte-depleting antibody reduces the incidence of acute rejection compared with IL2-RA, but has not been shown to prolong graft survival.
• • 
  Induction therapy with a lymphocyte-depleting antibody increases the incidence of serious adverse effects.
• • 
  For KTRs ≥18 years old, who are at high risk for acute rejection, the benefits of induction therapy with a lymphocyte-depleting antibody outweigh the harm.

In a large number of long-term, randomized controlled trials (RCTs) in adults, it has been consistently shown that induction therapy with either lymphocyte-depleting agents or IL2-RA reduces acute rejection in patients treated with ‘double therapy’ (calcineurin inhibitor [CNI] and prednisone), or ‘triple therapy’ (CNI, an antiproliferative agent [e.g. mycophenolate or azathioprine], and prednisone). Lymphocyte-depleting antibody induction also reduces the risk of graft failure while, in more recent studies, IL2-RA reduced the risk of death-censored graft failure, but not overall graft loss. Oral maintenance therapy may not produce immediate effects on the immune response when it is most needed, that is at the time of transplantation and antigen presentation. Pharmacokinetic and pharmacodynamic properties of oral maintenance agents may delay their full effect on immune cells.

The efficacy and safety of IL2-RA (compared to placebo or no treatment) have been confirmed in the most recent Cochrane review of 30 RCTs and 4670 patients followed to 3 years (2). In this review, IL2-RA consistently reduced the risk of acute rejection (e.g. for biopsy-proven acute rejection: 14 RCTs, 3861 patients, relative risk [RR] 0.77, 064–0.92) and graft loss (censored for death: 16 RCTs, n = 2973 patients, RR = 0.74, 0.55–0.99). IL2-RA did not affect all-cause mortality (24 RCTs, n = 4468, RR 0.73, 0.50–1.07), malignancy (14 RCTs, n = 3363, RR 0.70, 0.38–1.29) or cytomegalovirus (CMV) infection (17 RCTs, n = 3767, RR 0.90, 0.76–1.06), although all point estimates favor IL2-RA (all outcomes are at 1 year). The use of IL2-RA has also been found to be cost-effective compared to placebo (3).

The evidence for safety and efficacy of lymphocyte-depleting antibodies is more limited than that for IL2-RA. A meta-analysis of seven RCTs (N = 794) comparing lymphocyte-depleting agents with placebo or no treatment reported a reduction in graft failure (RR 0.66, 0.45–0.96) (4). In an individual patient meta-analysis of five of these same trials (N = 628), the reduction in graft loss at 2 years was greater in patients with high panel-reactive antibody (PRA) levels (RR 0.12, 0.03–0.44), compared to the reduction in risk for patients without high PRA (RR 0.74, 0.50–1.09) (5).

Since publication of these meta-analyses, there have been other trials comparing lymphocyte-depleting agents with placebo or no depleting agent. In a single-center RCT, sensitized patients were randomized to induction with ATG or no induction. Patients treated with ATG had a reduction in acute rejection and improvement in graft survival (6). In a three-arm RCT, the incidence of biopsy-proven acute rejection at 6 months was highest in deceased-donor KTRs receiving tacrolimus, azathioprine and prednisone without induction (25.4%, N = 185) compared to a group receiving tacrolimus, azathioprine, prednisone and ATG (15.1%, N = 184) and a group receiving cyclosporine A (CsA), azathioprine, prednisone and ATG (21.2%, N = 186) (7). However, CMV infection occurred in 16%, 24% and 28% of the patients in these groups, respectively (p = 0.012). Similarly, leukopenia, thrombocytopenia, fever and serum sickness were all more common in the two groups receiving antithymocyte induction (7). There is high-quality evidence for a net benefit of IL2-RA compared to placebo for some patient outcomes (graft survival) but not all (all-cause mortality); and high-quality evidence of a net benefit to prevent acute rejection (see Evidence Profile and accompanying evidence in Supporting Tables 1–4 at http://www3.interscience.wiley.com/journal/118499698/toc).

There have been a number of RCTs comparing IL2-RA with lymphocyte-depleting agents. Most of these trials have been small and of low quality. A meta-analysis of nine RCTs (N = 778) found no difference in clinical acute rejection at 6 months (2). There were no differences in graft survival or patient survival (2). Since this meta-analysis, there have been other RCTs. The largest (N = 278), and arguably highest-quality, RCT compared ATG with daclizumab in deceased-donor KTRs selected to be high-risk for delayed graft function (DGF) and/or acute rejection (8). This RCT found no difference in the primary composite end-point, but the ATG induction group had fewer biopsy-proven acute rejections and more overall infections compared to the daclilzumab group (8). In an updated Cochrane review, the risk of acute rejection was higher with IL2-RA compared with lymphocyte-depleting agents (nine RCTs, n = 1166, RR 1.27, 1.00–1.61), but the risk of graft loss (12 RCTs, n = 1430, RR 1.10, 0.73–1.65), and mortality was not significantly different (13 RCTs, n = 1670, RR 1.28, 0.74–2.20). Compared with lymphocyte-depleting agents, the risk of CMV infection (13 RCTs, n = 1480, RR 0.69, 0.49–0.97), and malignancy (six RCTs, n = 840, RR 0.23, 0.06–0.93) is lower with IL2-RA. Thus, there is moderate-quality evidence for trade-offs between IL2-RA and depleting antibodies; depleting antibodies are superior to prevent acute rejection, but there is uncertainty whether this corresponds to improved graft outcomes. Depleting antibodies are associated with more infections (see Evidence Profile and accompanying evidence in Supporting Tables 5–7).

There have been few head-to-head comparisons of different lymphocyte-depleting agents. Thus, it is unclear whether any one of these agents is superior to any other. In meta-analyses, there do not appear to be obvious differences in the effects of different lymphocyte-depleting agents on acute rejection or graft survival.

Alemtuzumab (Campath 1H) is a humanized anti-CD52 monoclonal antibody that depletes lymphocytes. In the United States, it has been approved by the Food and Drug Administration (FDA) for use in patients with B-cell lymphomas. There have been a few small RCTs examining the use of alemtuzumab as an induction agent in KTRs. All of these RCTs lack statistical power to examine the effects of alemtuzumab on patient survival, graft survival or acute rejection. In many of the RCTs, there were differences between the comparator groups other than alemtuzumab, making it difficult to discern the effects of alemtuzumab alone. For example, in a single-center RCT, 65 deceased-donor KTRs received alemtuzumab induction with delayed tacrolimus monotherapy and were compared to 66 KTRs treated with no induction, mycophenolate mofetil (MMF) and corticosteroids. At 12 months, the rate of biopsy-proven acute rejection was 20% vs. 32% in the two groups, respectively (p = 0.09) (9). In 21 high-immunological-risk KTRs randomized to alemtuzumab plus tacrolimus vs. four doses of ATG (plus tacrolimus, MMF and steroids), there were two vs. three acute rejections, respectively (10). Among 20 patients randomized to alemtuzumab plus low-dose CsA vs. 10 patients on CsA plus azathioprine and prednisone, there were biopsy-proven acute rejections in 25% vs. 20%, respectively (11). Ninety deceased-donor KTRs were randomly allocated to ATG, alemtuzumab or daclizumab induction, with those receiving alemtuzumab also receiving a lower tacrolimus target, MMF 500 mg twice daily and no maintenance prednisone, while those in the other two groups received MMF 1000 mg twice daily and prednisone. After 2 years of follow-up, acute rejections occurred in 20%, 23% and 23% in the three groups, respectively, but there was borderline worse death-censored graft survival in the alemtuzumab group (p = 0.05), and more chronic allograft nephropathy (CAN) (p = 0.008) (12,13). Altogether, these small studies fail to clearly demonstrate that the benefits outweigh the harm of alemtuzumab induction in KTRs.

For KTRs treated with an IL2-RA, the reduction in the incidence of acute rejection and graft loss, without an increase in major adverse effects, makes the balance of benefits vs. harm favorable in most patients. However, it is possible that in some KTRs at low risk for acute rejection and graft loss, the benefits of induction with IL2-RA may be too small to outweigh even minor adverse effects (especially cost in developing countries) and so, in this setting, not administering IL2A is reasonable.

In contrast to IL2-RA, induction therapy with lymphocyte-depleting antibodies increases the incidence of serious adverse effects. For KTRs treated with lymphocyte-depleting antibodies, a reduction in the incidence of acute rejections must be balanced against an increase in major infections. This balance may favor the use of depleting agents in some, but not all, patients. Logic would suggest that the chances of a favorable balance between benefits and harm could be maximized by limiting the use of lymphocyte-depleting agents to patients at increased risk for acute rejection.

In an individual patient, meta-analysis of five RCTs comparing lymphocyte-depleting antibody induction with no induction (or placebo), the reduction in graft failure was greater in patients with a high PRA (5). Unfortunately, there are few, if any, studies comparing the relative effectiveness of lymphocyte-depleting agents vs. IL2-RA in subgroups of patients at increased immunological risk. Nevertheless, observational data can be used to quantify the risk for acute rejection and graft failure, and thereby define patients who are most likely to benefit from lymphocyte-depleting agents compared to an IL2-RA.

Risk factors for acute rejection include (Table 1):

Table 1.  Risk of acute rejection in multivariate analyses

Patient characteristic	Study characteristics
Country of study	United States (14)	Spain (15)	North America (16)	Portugal (17)	Netherlands (18)	Norway (19)	UK (20)	Norway (21)
BMI, body mass index; CMV, cytomegalovirus; DGF, delayed graft function; HLA, human leukocyte antigen; NA, not applicable, for example deceased-donor cause of death or cold ischemia time in studies including only living donors. ^aDefined in multivariate analysis by hazards ratios (Cox analysis) or odds ratios (logistic regression): ↔ Indicates not significantly associated with acute rejection (p < 0.05); may have been eliminated in univariate analysis. ↑ and ↓ indicate 10–20% more or less risk of acute rejection, respectively. ↑↑ and ↓↓ indicate 20–30% more or less risk of acute rejection, respectively. ↑↑↑ and ↓↓↓ indicate >30% more or less risk of acute rejection, respectively. ^bUnclear if tested in multivariable analysis. ^cInfection and clinical symptoms or signs of disease. ^dNot defined. ^eCMV pp65 antigen in leukocytes.
Number analyzed (N)	27 377	3365	2779 children	866	790	739	518	451
Percent that used living donors (%)	33%	0%	100%	1.4%	0%	100%	0%	33%
Transplant years included	97–99	90, 94, 98	87–97	85–99	83–96	94–04	91–99	94–97
	Acute rejection risk^a
Deceased (vs. living donor)				NA		NA	NA
Younger recipient age	↑	↑↑↑	↔	↑↑↑		↑↑↑	↔	↔
	per 10 years	<60 y	<2 years	<45 years		<50 years
Older donor age
		≥60 years				≥65 years		per 10 years
Recipient female (vs. male)	↑	↔				↔	↔	↔
Deceased donor cause of death
Cerebral vascular death (vs. other cause)						NA
Trauma (vs. nontrauma)						NA
Recipient ethnicity US black (vs. white)	↑↑		↑↑↑
Recipient Hispanic (vs. non-Hispanic)		NA		NA		NA	NA	NA
Recipient diabetes (vs. no diabetes)	↑						↔^b
HLA mismatches
Any number of ABDR (vs. 0)
Any number of AB (vs. 0)
Any number of DR (vs. 0)
Per each ABDR mismatch 4–6 ABDR (vs. 3–1)
Panel reactive antibody status				NA			↔
>0% (vs. 0%)								↑↑↑
>15% (vs. ≤15%)		↑
>50% (vs. ≤50%)					↑↑↑
Cold ischemia time
>24 h (vs. <24 h)						NA
Per hour						NA
DGF (vs. none)		↑↑↑		↑↑↑	↑↑↑		↑↑↑
CMV disease (vs. none)								↑↑↑^c
CMV infection (vs. none)		↑↑^d						↑↑↑^e
Recipient size
BMI ≥35 kg/m^2
Body weight
Prior transplantation		↔			↔		↔	↔

• • 
  The number of human leukocyte antigen (HLA) mismatches (A)
• • 
  Younger recipient age (B)
• • 
  Older donor age (B)
• • 
  African-American ethnicity (in the United States) (B)
• • 
  PRA >0% (B)
• • 
  Presence of a donor-specific antibody (B)
• • 
  Blood group incompatibility (B)
• • 
  Delayed onset of graft function (B)
• • 
  Cold ischemia time >24 hours (C)

where A is the universal agreement, B is the majority agreement and C is the single study.

Retrospective observational studies have identified a number of risk factors for acute rejection after kidney transplantation (Table 1). Younger recipients are at substantially higher risk than older recipients, although there is no clear age threshold for the risk of acute rejection. Younger recipients may also be better able to tolerate serious adverse effects of additional immunosuppressive medication, making it compelling to treat younger recipients with lymphocyte-depleting antibody than IL2-RA. Kidneys from older donors may impart increased risk for acute rejection to the recipient, but a distinct age threshold has not been clearly defined.

The number of HLA mismatches between the recipient and donor is associated with the risk of acute rejection, but few studies have agreed on the number or type of mismatches (Class 1 [AB] or Class 2 [DR]) that increase the risk for acute rejection. In the United States, African-American ethnicity has been linked to an increased risk of acute rejection. For deceased-donor recipients, the duration of cold ischemia, for example longer than 24 hours, has been associated with acute rejection. DGF has also been associated with acute rejection, although by the time it is apparent that graft function is delayed, it is likely too late to decide whether or not to use a lymphocyte-depleting agent or an IL2-RA. However, induction with a lymphocyte-depleting agent could be used when there is an increased risk for DGF, such as in cases with expanded criteria donation or prolonged cold ischemia time. Finally, the presence of antibodies to a broad panel of potential recipients has been associated with an increased risk of acute rejection.

Chapter 2: Initial Maintenance Immunosuppressive Medications

• 2.1: 
  We recommend using a combination of immunosuppressive medications as maintenance therapy including a CNI and an antiproliferative agent, with or without corticosteroids. (1B)
• 2.2: 
  We suggest that tacrolimus be the first-line CNI used. (2A)

  • 2.2.1: 
    We suggest that tacrolimus or CsA be started before or at the time of transplantation, rather than delayed until the onset of graft function. (2D tacrolimus; 2B CsA)
• 2.3: 
  We suggest that mycophenolate be the first-line antiproliferative agent. (2B)
• 2.4: 
  We suggest that, in patients who are at low immunological risk and who receive induction therapy, corticosteroids could be discontinued during the first week after transplantation. (2B)
• 2.5: 
  We recommend that if mTORi are used, they should not be started until graft function is established and surgical wounds are healed. (1B)

CNI, calcineurin inhibitor; CsA, cyclosporine A; mTORi, mammalian target of rapamycin inhibitor(s).

Background

Maintenance immunosuppressive medication is a long-term treatment to prevent acute rejection and deterioration of graft function. Treatment is started before or at the time of transplantation, and the initial medication may or may not be used with induction therapy. Agents are used in combination to achieve sufficient immunosuppression, while minimizing the toxicity associated with individual agents. Since the risk for acute rejection is highest in the first 3 months after transplantation, higher doses are used during this period, and then reduced thereafter in stable patients to minimize toxicity. In these guidelines, antiproliferative agents refer specifically to azathioprine or mycophenolate (either MMF or enteric-coated mycophenolate sodium [EC-MPS]).

Corticosteroids have traditionally been a mainstay of maintenance immunosuppression in KTRs. However, adverse effects of corticosteroids have led to attempts to find maintenance immunosuppression regimens that do not include corticosteroids. Terminology has often been confusing, but ‘steroid avoidance’ is used here to refer to protocols that call for the initial use of corticosteroids, which are then withdrawn sometime during the first week after transplantation. In contrast, ‘steroid-free’ protocols do not routinely use corticosteroids as initial or maintenance immunosuppression. ‘Steroid withdrawal’ refers to protocols that discontinue corticosteroids after the first week posttransplant. Similar definitions have been applied to the use of CNIs.

Rationale

• • 
  Used in combination and at reduced doses, drugs that have different mechanisms of action may achieve additive efficacy with limited toxicity.
• • 
  The earlier that therapeutic blood levels of a CNI can be attained, the more effective the CNI will be in preventing acute rejection.
• • 
  There is no reason to delay the initiation of a CNI, and no evidence that delaying the CNI prevents or ameliorates DGF.
• • 
  Compared to CsA, tacrolimus reduces the risk of acute rejection and improves graft survival during the first year of transplantation.
• • 
  Low-dose tacrolimus minimizes the risk of new-onset diabetes after transplantation (NODAT) compared to higher doses of tacrolimus.
• • 
  Compared with placebo and azathioprine, mycophenolate reduces the risk of acute rejection; there is some evidence that mycophenolate improves long-term graft survival compared with azathioprine.
• • 
  Avoiding the use of maintenance corticosteroids beyond the first week after kidney transplantation reduces adverse effects without affecting graft survival.
• • 
  Mammalian target of rapamycin inhibitors (mTORi) have not been shown to improve patient outcomes when used either as replacement for antiproliferative agents or CNIs, or as add-on therapy, and they have important short- and long-term adverse effects.

Calcineurin Inhibitors

Timing of initiation

In theory, the earlier that therapeutic blood levels of a CNI can be attained, the more effective the CNI is likely to be in preventing acute rejection. However, there are also theoretical reasons that the early use of CNIs might increase the incidence and severity of DGF. As a result, RCTs have compared early vs. delayed CNI initiation after transplantation. In three RCTs (N = 338), there was no difference in the incidence of DGF with early vs. delayed CsA initiation. In five RCTs (N = 620), there were no differences in acute rejection, graft failure or kidney function in early vs. delayed CsA initiation. Altogether, these RCTs suggest that there is no reason to delay the initiation of CsA. There are no similar studies using tacrolimus, but it is suggested that, with a regimen including induction and reduced-dose tacrolimus, the risk for early CNI nephrotoxicity is minimized and optimal prevention of acute rejection can be achieved. There is moderate-quality evidence that, in CsA-containing regimens, there is no net benefit or harm of early vs. delayed CsA; the evidence is of low quality for CNIs in general, because of a lack of data for tacrolimus-containing regimens (see Evidence Profile and accompanying evidence in Supporting Tables 11–13 at http://www3.interscience.wiley.com/journal/118499698/toc).

Tacrolimus vs. cyclosporine

A meta-analysis of RCTs reported reduced acute rejection and better graft survival with tacrolimus compared to CsA (22). For every 100 patients treated for the first year with tacrolimus rather than CsA, 12 would be prevented from having acute rejection, two would be prevented from having graft failure, but five would develop NODAT. The RCTs in the meta-analysis combined studies of patients receiving the original CsA preparation and cyclosporine A microemulsion (CsA-ME). This study also showed that lower tacrolimus were associated with higher relative risk of graft loss, while higher levels of tacrolimus were associated with an increased risk for NODAT.

Randomized controlled trials comparing tacrolimus with CsA-ME using concomitant azathioprine and corticosteroids, but no induction, have shown reduced acute rejection with tacrolimus; for example, 22% vs. 42% at 12 months, respectively (p < 0.001) (23). The difference in acute rejection between the two CNIs could no longer be observed with concomitant induction and MMF instead of azathioprine; for example 4% vs. 6%, for tacrolimus vs. CsA-ME, respectively (24) or 7% vs. 10% at 6 months, respectively (25) when C₂ monitoring of CsA was also employed. Furthermore, there is evidence that subclinical rejection (acute rejection changes in protocol biopsy not indicated by a change in kidney function) is more effectively prevented by tacrolimus and MMF compared to CsA and MMF; 15% vs. 39% (p < 0.05) (26).

A very large multicenter RCT in de novo KTRs (n = 1645; the Symphony study) showed superior graft function, better prevention of acute rejection (12.3%) and superior graft survival (96.4%) at 12 months with daclizumab induction and low-dose tacrolimus (C₀ 3–7 ng/mL). The comparator groups included low-dose CsA and low-dose sirolimus, both with daclizumab induction and standard-dose CsA without induction. All patients received MMF (2 g/day) and corticosteroids (27).

There is no uniform definition of NODAT used in the literature. Therefore, the reported incidences of NODAT vary to a great extent. Studies reporting a difference between tacrolimus and CsA in the incidence of NODAT, impaired glucose tolerance, or the use of antidiabetic treatment, favor CsA; for example 17% vs. 9% (p < 0.01; tacrolimus vs. CsA) (25). Others have found lower incidences and no significant difference (24,28). One reason for the variation in findings may be differences in the use of corticosteroids as maintenance medication and treatment of acute rejection. Indeed, use of a steroid-free regimen has been associated with a lower incidence of NODAT (29).

Overall, there is moderate-quality evidence for a net benefit of tacrolimus vs. CsA (see Evidence Profile and accompanying evidence in Supporting Tables 8–10). There is no clear evidence of differences in terms of patient mortality, incidence of malignancy, infection, delayed onset of graft function or blood pressure. There is evidence that cholesterol, low-density lipoprotein cholesterol (LDL-C) (but not high-density lipoprotein cholesterol [HDL-C]), acute rejection and graft loss are higher with CsA vs. tacrolimus. However, there is also evidence that NODAT is more common with tacrolimus than CsA, so that there is clear trade-off in the different patient-relevant outcomes with these two CNIs.

Dosing of CNI

Dosing of CNI is important, but is a relatively under-researched area. There are few trials that compare the effects of different doses or target levels of the same drugs in which baseline immunosuppression is kept constant across both arms. Indirect comparisons and case series have shown that high doses might increase adverse events and low doses might increase acute rejection. Standard-dose tacrolimus may be defined as it is recommended by the producer (Astellas Pharma, Tokyo, Japan); the dose achieving 12-h trough levels (C₀) of 10 (5–15) ng/mL. A low-dose tacrolimus has recently been introduced in the Symphony study and was defined as C₀ of 5 (3–7) ng/mL (27). Standard-dose CsA may be defined as the dose achieving C₀ of 200 (150–300) ng/mL (30) or C₂ 1400–1800 ng/mL early and 800–1200 ng/mL later after transplantation (25). Low-dose CsA has been used in some recent clinical studies (27,30) and was defined as achieving C₀ of 75 (50–100) ng/mL.

Mycophenolate Mofetil

Randomized controlled trials have shown that MMF (2 or 3 g, but not 1 g daily) is significantly better in preventing acute rejection than placebo. This was seen in studies using steroids as concomitant medication and either tacrolimus or CsA (31,32). For example, acute rejection at 6 months was reduced from 55% with placebo to 30% and 26% with MMF 2 and 3 g daily doses (31). There were 5–7% improvements of graft survival at 12 months with MMF, but the studies were not powered to evaluate this difference. There were no significant differences in patient survival, graft function, malignancy, NODAT, infection rates or gastrointestinal adverse events such as diarrhea, although there might be evidence that higher doses of MMF cause more diarrhea than lower doses of MMF. More bone marrow suppression was seen with MMF compared to placebo. Overall, there is moderate-quality evidence of a net benefit of MMF over placebo to prevent acute rejection, but low-quality evidence for all graft and patient outcomes overall (see Evidence Profile and accompanying evidence in Supporting Tables 14–15).

Randomized controlled trials comparing outcomes between MMF vs. azathioprine have shown some important inconsistencies. In a recent meta-analysis of 19 trials and 3143 patients, MMF was associated with less acute rejection (RR 0.62, 95% confidence interval [CI] 0.55–0.87) and improved graft survival (RR 0.76, 0.59–0.98) (33). However, there were no differences in patient survival or kidney function (33). There were also no differences in major adverse effects (e.g. infections, CMV, leucopenia, anemia and malignancies) between MMF and azathioprine, but diarrhea was more common with MMF (RR 1.57; 95% CI 1.33–28.6) (33). In several RCTs, MMF reduced the incidence of acute rejection at 6 months; for example from 36% with azathioprine (100–150 mg/day) to 20% with MMF (2 g/day) using CsA and steroids as concomitant medication (34) and from 38% to 20% with the addition of concomitant induction (35). Also, a reduction in acute rejection from 29% to 7% was seen with concomitant tacrolimus, steroids and induction in using MMF 2 g, but not 1 g (36). Conversely, another study showed a smaller reduction in acute rejection at 6 months from 23% with azathioprine (100–150 mg/day) to 18% with MMF (2 g/day), a difference that was not statistically significantly (37). These patients were also treated with CsA-ME and steroids. However, using the same concomitant medication, including CsA-ME, other investigators found a significant reduction of acute rejection at 12 months from 27% with azathioprine to 17% with MMF 2 g (38). In a third arm of this latter study, patients received MMF from day 0 to day 90 and thereafter azathioprine, and the acute rejection rate was the same, 17%, as for those receiving MMF for the entire study period of 12 months. Thus, high-quality evidence finds a net benefit of MMF over azathioprine to prevent acute rejection, but moderate-quality evidence exists for patient-level outcomes. Because of the substantially increased cost of MMF compared with azathioprine and increased side effects compared with azathioprine, there is no clear net benefit, but a decision based upon trade-offs is required (see Evidence Profile and accompanying evidence in Supporting Tables 16–18).

Analyses of observational registry data have shown either a small 4% improvement in graft survival with MMF vs. azathioprine (39) or, more recently, no improvement in graft survival (40). However, for a number of reasons, the results of retrospective analyses of observational registry data need to be interpreted cautiously (41).

MMF Compared to EC-MPS

One RCT compared MMF 2 g daily vs. EC-MPS 1.44 g daily with CsA-ME, steroids, with or without induction (42). There were no significant differences in acute rejection (24% vs. 23%), patient or graft survival or rates of malignancy or infection. There was no difference in rates of gastrointestinal disorders (80% vs. 81%) despite the fact that the potential reduction of gastrointestinal adverse events has been the incentive for the development of EC-MPS. Another study (43) tested the crossover between the two formulations and also found no differences in any of the outcome parameters. A summary of the RCTs on MMF vs. EC-MPS is available in Supporting Tables 25–26.

Steroid avoidance or withdrawal

The rationale for minimizing corticosteroid exposure is compelling and provided by well-established risks of osteoporosis, avascular necrosis, cataracts, weight gain, diabetes, hypertension and dyslipidemia. Such risk is not constant, and varies with comorbidities such as preexisting metabolic syndrome and age. On the other hand, corticosteroids have been the mainstay of immunosuppression for KTRs for decades, and trial data evaluating minimization of steroid exposure are sparse compared to the large number of trials that have included steroids in the regimens being evaluated. In addition, many of the adverse effects attributed to corticosteroids were observed with high doses. Whether or not low doses (e.g. 5 mg prednisone per day) that are commonly used for long-term maintenance immunosuppression are associated with major adverse effects is less clear.

Randomized controlled trials have shown that the withdrawal of corticosteroids from maintenance immunosuppressive medication regimens, when carried out weeks to months after transplantation, is associated with a high risk of acute rejection (44,45). More recent studies have examined whether steroid avoidance (discontinuing corticosteroids within the first week after transplantation) can be done safely. These studies have generally shown higher rates of acute rejection, but lower rates of long-term adverse effects (12,29,46–48). Unfortunately, these trials have had design limitations that make the interpretation of their results difficult.

Overall, there is moderate-quality evidence for trade-offs between steroid avoidance or withdrawal compared to steroid maintenance, with a higher rate of steroid-sensitive acute rejections but avoidance of steroid-related adverse effects (see Evidence Profile and accompanying evidence in Supporting Tables 19–21).

Mammalian target of rapamycin inhibitor(s)

Regimens using the mTORi sirolimus and everolimus have been compared to a number of different regimens in clinical trials in KTRs, for example as replacement for azathioprine, MMF or CNIs, and in combination with CNIs (both at high and low dose). The use of mTORi in the setting of chronic allograft injury (CAI) is described in Chapter 7. mTORi have a number of adverse effects that limit their use, including dyslipidemia and bone marrow suppression (49–56). Although they have been compared with many other regimens in RCTs, in none of these RCTs was there an improvement in graft or patient survival.

mTORi as replacement for antiproliferative agents

In a meta-analysis of 11 RCTs with 3966 KTRs evaluating mTORi as replacement for azathioprine or MMF, there were no differences in graft or patient survival (57). mTORi appear to reduce the risk of acute rejection (RR 0.84, 95% CI 0.71–0.99; p = 0.04), but graft function and LDL-C outcomes were generally better with azathioprine or MMF (57).

mTORi as replacement for CNIs

In a meta-analysis of eight RCTs with 750 patients evaluating mTORi as replacement for CNIs, there were no differences in acute rejection, CAN, graft survival or patient survival (57). mTORi were associated with higher glomerular filtration rate (GFR), but also with increased risk of bone marrow suppression and dyslipidemia (49,57).

mTORi in combination with CNIs

The combined use of mTORi and CNIs should be avoided, because these agents potentiate nephrotoxicity, particularly when used in the early post-transplant period (57). When used as long-term maintenance, mTORi have been used in two different regimens in combination with CNIs. Eight RCTs involving 1360 patients have evaluated low-dose mTORi and standard-dose CNI compared with standard-dose mTORi and low-dose CNI (57). Overall, the low-dose, CNI-standard dose mTORi regimen is associated with a 30% increased risk of rejection with no difference in graft survival. An additional 10 RCTs involving 3175 patients have evaluated the effects of high- vs. low-dose mTORi in combination with fixed-dose CNI, showing less rejection but lower GFR with higher-dose therapy, but no improvement in patient outcomes.

Moderate-quality evidence for sirolimus finds net harm without improved graft or patient survival; CNI toxicity is potentiated when used in combination with sirolimus (see Evidence Profile and accompanying evidence in Supporting Tables 22–24).

Research Recommendations

• • 
  A long-term RCT that has adequate statistical power to detect differences in acute rejection and major adverse events is needed to determine whether the benefits of steroid avoidance outweigh the harm.

Chapter 3: Long-Term Maintenance Immunosuppressive Medications

• 3.1: 
  We suggest using the lowest planned doses of maintenance immunosuppressive medications by 2–4 months after transplantation, if there has been no acute rejection. (2C)
• 3.2: 
  We suggest that CNIs be continued rather than withdrawn. (2B)
• 3.2: 
  If prednisone is being used beyond the first week after transplantation, we suggest prednisone be continued rather than withdrawn. (2C)

CNI, calcineurin inhibitor.

Background

Using high doses of immunosuppressive medications early after transplantation when the risk of acute rejection is highest, but then reducing doses later when the risk of acute rejection is lower, has been used empirically as the mainstay of long-term immunosuppressive medication management since the advent of kidney transplantation. However, there are no randomized trials testing this therapeutic strategy.

Rationale

• • 
  If low-dose CNI was not implemented at the time of transplantation, CNI dose reduction >2–4 months after transplantation may reduce toxicity yet prevent acute rejection.
• • 
  RCTs show that CNI withdrawal leads to increased acute rejection, without altering graft survival.
• • 
  RCTs show that steroid withdrawal more than 3 months after transplantation increases the risk of acute rejection.
• • 
  Different immunosuppressive medications have different toxicity profiles and patients vary in their susceptibility to adverse effects.

CNI dose reduction

Although there are no RCTs comparing dose reduction with maintaining initial high doses and target levels, this dose reduction strategy has been successfully adopted in most RCTs. The assumption is that the immune system gradually adapts to the foreign antigens in the graft, and that the need for immunosuppression is thereby reduced. There is great individual variation, and some patients with a high risk for immunological complications (acute and chronic rejection) may need to continue on higher doses of immunosuppression compared to the majority of patients.

A range of trial designs have directly and indirectly compared the effects of different CNI dose, usually as measured by different target levels. In RCTs in which CNI has been combined with mTORi (eight RCTs, 1178 patients), as either low-dose mTORi with standard CNI or higher mTORi and lower CNI, standard-dose CNI was associated with lower rates of acute rejection (RR 0.67) but lower GFR (9 mL/min/1.73 m²). Such trials are clearly confounded, but do suggest that variable CNI exposure leads to competing benefits and harm. Graft function may be improved by minimizing CNI, leading to reduced CAI, but may be worsened if acute rejection occurs.

The strongest evidence comes from RCTs that have directly compared low vs. high CNI doses (four RCTs, 1256 patients). In these trials, there were no differences in outcomes (including graft survival) except for GFR, which favored low CNI in two of the four studies. Low-quality evidence suggests no net benefit or harm of low- vs. standard-dose CNI (see Evidence Profile and accompanying evidence in Supporting Tables 27–29 at http://www3.interscience.wiley.com/journal/118499698/toc).

Using indirect comparisons of trials of different CNI doses, the risk of diabetes and graft loss was reduced with lower doses. However, there are sparse data on the relative effects of specific CNI target values from head-to-head trials, apart from the broad category of high vs. low.

Low-dose CNI maintenance

The notion of complete CNI withdrawal, after the peak period for immunologically mediated complications (3 months) is attractive, considering the long-term complications of CNI exposure. However, RCTs of complete CNI withdrawal show that, although some small benefit in graft function results, the risk of acute rejection is significantly increased without a clear benefit on improved graft survival (eight RCTs, 1891 patients). As described above, CNI toxicity can be minimized by administering low-dose CNI, while ensuring sufficient immunosuppression is provided. Moderate-quality evidence shows a net harm to CNI withdrawal (see Evidence Profile and accompanying evidence in Supporting Tables 30–32).

Steroid withdrawal

Long-term steroid administration may lead to hypertension, NODAT, osteoporosis, fractures and dyslipidemia, all of which may affect graft survival. However, long-term steroid administration prevents acute rejection and immunologically mediated graft loss. In six RCTs of 1519 KTRs, steroid withdrawal led to increased acute rejection, without a clear benefit for improved patient or graft outcomes, except for a reduction in total cholesterol levels in the steroid-withdrawal group. Low-quality evidence suggests net harm of steroid withdrawal (see Evidence Profile in Supporting Table 33).

Individual tailoring of immunosuppressive medication to the patient's risk profile

Although tailoring immunosuppressive therapies to the individual patient's risk profile (both risk for acute rejection and risk for adverse effects) is considered standard practice, there are few studies that suggest how this should be done. There are some data on the relative incidence and severity of adverse effects, collected in clinical trials and observational studies (Table 2). However, standard definitions have not been used to define adverse effects of immunosuppressive medications. Data collection has generally relied on spontaneous investigator reporting, which can lead to serious under-reporting. For these and other reasons, the quality of data on adverse drug effects is very low.

Table 2.  Toxicity profiles of immunosuppressive medications

Adverse effect	Steroids	CsA	Tac	mTORi	MMF	AZA
AZA, azathioprine; CsA, cyclosporine A; GFR, glomerular filtration rate; MMF, mycophenolate mofetil; mTORi, mammalian target of rapamycin inhibitor(s); Tac, tacrolimus. ↑ indicates a mild-moderate adverse effect on the complication. ↑↑ indicates a moderate-severe adverse effect on the complication. (↑) indicates a possible, but less certain adverse effect on the complication.
New-onset diabetes mellitus	↑	↑	↑↑	↑
Dyslipidemias
Hypertension	↑↑	↑↑	↑
Osteopenia			(↑)
Anemia and leucopenia				↑	↑	↑
Delayed wound healing
Diarrhea, nausea/vomiting			↑		↑↑
Proteinuria
Decreased GFR		↑	↑

Withdrawal of a specific drug in an individual patient with an adverse drug effect may or may not result in clinical improvement. Nevertheless, drug withdrawal or substitution is a logical course of action if the benefits (reducing symptoms) appear to outweigh the harm (acute rejection).

• • 
  NODAT may be caused or exacerbated by corticosteroids, tacrolimus, mTORi and, to a lesser extent, by CsA. In patients with impaired glucose tolerance or NODAT, steroid reduction or withdrawal may be beneficial. If this is not sufficient, a switch from tacrolimus to CsA-ME may be considered.
• • 
  Dyslipidemia may be caused or exacerbated by corticosteroids, CsA and especially by mTORi. Patients with significant dyslipidemia before or after transplantation should probably avoid mTORi.
• • 
  Hypertension may be caused or exacerbated by corticosteroids, CsA and, to a lesser extent, tacrolimus. In patients, who are not normotensive after transplantation, despite adequate antihypertensive treatment, reduction or withdrawal of steroid or CNI may be beneficial.
• • 
  Osteopenia may be caused or exacerbated by corticosteroids, and possibly CsA and tacrolimus. Steroid reduction or withdrawal may be helpful.
• • 
  Bone marrow suppression may be caused or exacerbated by MMF, azathioprine and mTORi. Monitoring of the mycophenolic acid (MPA) area under the concentration–time curve (AUC), and probably reduction of the dose of MMF or azathioprine, are the first suggested actions in case of anemia or leucopenia.
• • 
  Delayed wound healing may be caused or exacerbated by mTORi. Patients who have delayed wound healing on an mTORi may benefit from switching the mTORi to a CNI.
• • 
  Diarrhea, nausea and vomiting may be caused or exacerbated by MMF and tacrolimus. Monitoring MPA, AUC and tacrolimus C₀ levels may help to reduce these complications. However, it is important to rule out treatable, underlying causes other than the immunosuppressive medication. In a recent study, about half of the patients were cured by treatment of an infection (58). Only after ruling out other underlying causes should reducing the MMF, or changing MMF to azathioprine, be considered.
• • 
  Proteinuria may be caused or exacerbated by mTORi. Consider avoiding an mTORi in a patient with persistent urinary protein excretion of more than 500–1000 mg/day.
• • 
  Decreased kidney function may be caused or exacerbated by CsA and tacrolimus. See Chapter 7 regarding treatment of chronic CNI nephrotoxicity.

Chapter 4: Strategies to Reduce Drug Costs

• 4.1: 
  If drug costs block access to transplantation, a strategy to minimize drug costs is appropriate, even if use of inferior drugs is necessary to obtain the improved survival and quality of life benefits of transplantation compared with dialysis. (Not Graded)

  • 4.1.1: 
    We suggest strategies that may reduce drug costs include:

    • • 
      limiting use of a biologic agent for induction to patients who are high-risk for acute rejection (2C);
    • • 
      using ketoconazole to minimize CNI dose (2D);
    • • 
      using a nondihydropyridine CCB to minimize CNI dose (2C);
    • • 
      using azathioprine rather than mycophenolate (2B);
    • • 
      using adequately tested bioequivalent generic drugs (2C);
    • • 
      using prednisone long-term. (2C)
• 4.2: 
  Do not use generic compounds that have not been certified by an independent regulatory agency to meet each of the following criteria when compared to the reference compound (Not Graded):

  • • 
    contains the same active ingredient;
  • • 
    is identical in strength, dosage form, and route of administration;
  • • 
    has the same use indications;
  • • 
    is bioequivalent in appropriate bioavailability studies;
  • • 
    meets the same batch requirements for identity, strength, purity and quality;
  • • 
    is manufactured under strict standards.
• 4.3: 
  It is important that the patient, and the clinician responsible for the patient's care, be made aware of any change in a prescribed immunosuppressive drug, including a change to a generic drug. (Not Graded)
• 4.4: 
  After switching to a generic medication that is monitored using blood levels, obtain levels and adjust the dose as often as necessary until a stable therapeutic target is achieved. (Not Graded)

CCB, calcium-channel blocker; CNI, calcineurin inhibitor.

Background

A number of cost-saving strategies may offer access to transplantation when the cost of immunosuppressive medication is otherwise prohibitive. The use of generic medications can substantially reduce cost. A generic immunosuppressive medication is a medication that is manufactured and distributed without patent protection, but is structurally identical to the brand-name medication. However, manufacturing, distribution and quality control may differ among pharmaceutical companies. Regulatory authorities generally do not require that the efficacy and safety of generic medications be tested in RCTs. Manufacturers of generic drugs must only prove that their preparation is bioequivalent to the existing drug in order to gain regulatory approval.

However, generic drugs approved by the US FDA have met rigid standards. To gain FDA approval (http://www.fda.gov/cder/ogd; last accessed March 30, 2009), a generic drug must:

• • 
  contain the same active ingredients as the brand drug (inactive ingredients may vary);
• • 
  be identical in strength, dosage form and route of administration;
• • 
  have the same use indications;
• • 
  be bioequivalent;
• • 
  meet the same batch requirements for identity, strength, purity and quality;
• • 
  be manufactured under the same strict standards of the FDA's good manufacturing practice regulations.

Similarly, the European Agency for the Evaluation of Medicinal Products, also known as the European Medicinal Agency (http://www.emea.europa.eu/htms/human/raguidelines/datagenerics/biosimilars.htm; last accessed March 30, 2009) defines a generic medicinal product as a medicinal product that has:

• • 
  the same qualitative and quantitative composition in active substances as the reference product;
• • 
  the same pharmaceutical form as the reference medicinal product;
• • 
  bioequivalence with the reference medicinal product demonstrated by appropriate bioavailability studies.

Tacrolimus, CsA, mTORi, MMF, and azathioprine are all available as generics (loosely defined) in many countries around the world. However, the efficacy and the safety of these generics may not always be firmly established by local regulatory authorities charged with approving these agents.

Rationale

• • 
  Lack of dialysis facilities may make kidney transplantation the only life-saving therapy available for some patients with CKD stage 5.
• • 
  Kidney transplantation is the therapy of choice to treat CKD stage 5, since overall costs are lower, and outcomes and quality of life are better compared to dialysis.
• • 
  Cost savings that do not compromise patient safety are beneficial.
• • 
  Use of cytochrome P-450 inhibitors, such as ketoconazole and diltiazem, allow therapeutic blood levels of CsA to be achieved at a lower dose, thereby reducing cost.
• • 
  Azathioprine can be used to achieve most of the efficacy and safety of MMF, but at a much lower cost.
• • 
  An adequately tested bioequivalent generic formulation can lower cost without compromising safety and efficacy of the originally patented formulation.

Chronic maintenance dialysis is not available for many patients in a number of developing countries in Asia, Africa, and South America (59). Patients living in remote areas may not have access to dialysis. Kidney transplantation, especially preemptive transplantation (before the need for chronic dialysis), may be the only viable option for long-term renal replacement therapy in many areas of the world. Transplantation is the most cost-effective form of renal replacement therapy, and offers a superior quality of life compared to dialysis (60). For all of these reasons, there is a growing demand for kidney transplantation in the developing world, and it is imperative that kidney transplantation be affordable. Even where immunosuppressive drugs are available, their high cost may preclude their use if adequate health insurance coverage is not available (61).

Calcineurin inhibitors currently form the backbone of immunosuppressive regimens, but their cost imposes a long-term financial burden on patients in developing countries. Forced discontinuation of CsA due to cost increases the risk of acute rejection and may result in poor long-term outcomes (62).

Calcineurin inhibitors and mTORi (sirolimus and everolimus) are metabolized through the hepatic cytochrome P-450 microsomal oxidase enzyme system. Commonly used drugs such as the antifungal ketoconazole and the nondihydropyridine calcium-channel blocker (CCB) diltiazem are known inhibitors of this enzyme system and increase blood levels of these immunosuppressive drugs. This, in turn, reduces the dose necessary to maintain therapeutic blood levels (63,64).

A number of studies (Table 3) have shown that ketoconazole, when used in a dose of 50–200 mg/day, allows substantial reduction in the daily dose of CsA, tacrolimus and sirolimus, while maintaining therapeutic blood levels (65–76). In a RCT (69), 51 patients received 100 mg/day of ketoconazole along with CsA and 49 served as controls. The dose reduction was highest at 1 month (76.5%) and was maintained at 10 years (64.6%). The cost of CsA decreased by 73% at 1 year, 69% at 5 years and 63% at 10 years in the intervention group, while the decrease in cost was 13% and 20% in the control group at 1 and 10 years, respectively.

Table 3.  CNI cost reduction from the concomitant use of ketoconazole

Study	CNI	Keto (N)	Control (N)	Mean follow-up (months)	Ketoconazole (mg/day)^a	Estimated cost reduction (%)
CNI, calcineurin inhibitor; CsA, cyclosporine A; Keto, ketoconazole; Tac, tacrolimus. ^aFixed total once daily dose, or mean ± standard deviation. ^bRCT
First (66)^b	CsA	24	28	15	200	73
Butman (66A)	CsA	15		11	400	72
Keogh (68)^b	CsA	23	20	25	200	80
Sobh (69)^b	CsA	51	49	53	82.8	73
Carbajal (71)	CsA	14	17	29	54 ± 17	60
El-Dahshan (73)^b	Tac	35	35	24	100	53
Soltero (73A)	Tac	11	–	15	87	78

In another study (73), 70 patients on a tacrolimus-based immunosuppression regimen were randomly allocated to receive ketoconazole (n = 35) or no ketoconazole (controls, n = 35). The tacrolimus dose reduction was 58.7% at 6 months and 53.8% at 2 years, leading to cost reduction of 56.9% and 52.2%, respectively. None of the studies has reported any adverse effect of this approach on graft function.

Ketoconazole requires an acidic milieu in the stomach for its absorption; hence, concomitant use of agents that inhibit gastric acid secretion should be avoided.

In comparison to ketoconazole, the dose reduction achieved with diltiazem is modest (67,77). Hence, some would suggest that a nondihydropyridine CCB, such as diltiazem, be used only in situations where ketoconazole is contraindicated. On the other hand, if patients discontinue ketoconazole abruptly, the levels of immunosuppressive drugs may drop precipitously and result in acute rejection. A precipitous drop is less likely with nondihydropyridine CCBs, and the risk of acute rejection may therefore be less. In addition, most KTRs have hypertension that requires treatment, and nondihydropyridine CCBs may serve the dual purpose of treating hypertension and reducing cost. The choice between ketoconazole and a CCB should be adapted to the patient's situation and preference.

The use of 2-h CsA concentration (C₂) monitoring for adjusting drug dose is not suitable for patients receiving ketoconazole or diltiazem. Metabolic inhibitors interfere with the disposal—but not the absorption—of CsA or tacrolimus, and therefore flatten the AUC. In this situation, the CsA AUC correlates better with C₀ than C₂. Dose adjustments based on C₂ levels may lead to CsA toxicity (78). Trough concentration monitoring therefore should be used to adjust drug dosage.

Although MMF is considered the preferred antimetabolite for KTRs, the Mycophenolate Steroid Sparing follow-up study showed that azathioprine-treated patients experienced similar long-term outcomes compared to those receiving MMF after a median 5.4 years (37). CsA-ME was the CNI used in this study. The length of hospital stay, incidence of acute rejections, and the likelihood of return to dialysis were also similar in the two groups. In a cost-minimization analysis, MMF was found to be 15 times more expensive than azathioprine. This study (and the lack of large differences in outcomes in other studies comparing MMF with azathioprine) suggests that it may be acceptable to use azathioprine in place of MMF when cost is an important consideration.

A number of generic formulations of CsA, tacrolimus, mTORi and MMF are now available around the world. Generic formulations vary from country to country. Most countries require evidence of bioequivalence in only a small number of patients before marketing is permitted. In many countries, however, generic formulations have been available for over 10 years and their efficacy has been established in real-life situations. Head-to-head data comparing efficacy and toxicity are generally not available for most generics (79–81). Caution should therefore be exercised in choosing a generic formulation for use in KTRs. Ideally, a generic formulation should be used only after its safety and efficacy have been established in KTRs.

Chapter 5: Monitoring Immunosuppressive Medications

• 5.1: 
  We recommend measuring CNI blood levels (1B), and suggest measuring at least:

  • • 
    every other day during the immediate post-operative period until target levels are reached (2C);
  • • 
    whenever there is a change in medication or patient status that may affect blood levels (2C);
  • • 
    whenever there is a decline in kidney function that may indicate nephrotoxicity or rejection. (2C)
  • 5.1.1: 
    We suggest monitoring CsA using 12-h trough (C₀), 2-h post-dose (C₂) or abbreviated AUC. (2D)
  • 5.1.2: 
    We suggest monitoring tacrolimus using 12-h trough (C₀). (2C)
• 5.2: 
  We suggest monitoring MMF levels. (2D)
• 5.3: 
  We suggest monitoring mTORi levels. (2C)

AUC, area under concentration–time curve; CNI, calcineurin inhibitor; CsA, cyclosporine A; MMF, mycophenolate mofetil; mTORi, mammalian target of rapamycin inhibitor(s).

Background

Cyclosporine A has a narrow therapeutic window and variable absorption characteristics, even with the microemulsion formulation (CsA-ME). Therefore, the CsA dosage must be individualized to find a balance between high levels that may be toxic and low levels that may be insufficient to prevent rejection. Variability in absorption is greatest during the first 4 h after dosing, and during the first few weeks after transplantation. There are no RCTs comparing monitoring with no monitoring; however, the fact that different target levels influence efficacy and toxicity is strongly suggestive that monitoring is beneficial (82).

The C₀ is the measured concentration after the dosing interval (e.g. 12 h after dosing if the dosing interval is every 12 h), C₂ the concentration 2 h after dosing and AUC_0–4 is the AUC during the first 4 h after dosing. Fewer data are available to guide blood-level monitoring of tacrolimus compared to CsA. MPA is the active metabolite of MMF and the molecule generally used for monitoring of MMF. The half-lives of mTORi are greater than 48 h, making anything but monitoring of C₀ unlikely to be useful. There are no clinical methods for monitoring corticosteroid blood levels.

There continues to be widespread interest in pharmcodynamic assays for monitoring immunosuppressive medication and adjusting dosing accordingly. However, there are insufficient data demonstrating the efficacy of pharmacodynamic monitoring.

Rationale

CsA monitoring

Cyclosporine A absorption may increase substantially during the first 1–2 weeks after transplantation. In KTRs, absorption stabilizes by approximately the end of the first month. Common factors that might change CsA blood levels are the use of other drugs affecting cytochrome P450 3A4 (CYP3A4) and/or P-glycoprotein, diet and intestinal motility. There are no studies comparing one schedule of monitoring vs. another; however, tailoring the monitoring schedule to the expected absorption variability is a reasonable, empirical approach. There are no data to suggest whether monitoring blood levels in stable patients beyond the first few weeks after transplantation is beneficial.

There are few RCTs to define optimal target blood levels. Target levels should generally reflect the overall immunosuppressive medication regimen, and therefore target levels may vary accordingly. For example, it may be prudent to use lower early posttransplant target blood levels when an induction antibody is used. In any case, blood-level monitoring with predetermined targets can be effectively used to balance the risk for rejection with the risk for toxicity.

Cyclosporine A C₀ has often been used for therapeutic drug monitoring, but C₀ does not correlate closely with AUC_0–4. Blood levels at 2 h after drug administration (C₂), instead of at 12 h (C₀ if the dosing interval is 12 h), have been used to monitor CsA therapy with the CsA-ME formulation. Although C₂ levels appear to correlate more closely with AUC_{0 − 4}, no differences have been observed in two RCTs between the incidence of acute rejection, graft loss or adverse events whether patients were monitored by AUC_0–4 or C₂ or C₀ levels (83). Overall, a very low strength of evidence suggests uncertain trade-offs between using C₀ or C₂ (see Evidence Profile and accompanying evidence in Supporting Tables 34–36 at http://www3.interscience.wiley.com/journal/118499698/toc); therefore, either C₀ or C₂ blood levels are acceptable.

Tacrolimus monitoring

There have been fewer studies with blood-level monitoring for tacrolimus than for CsA. However, available evidence suggests that the benefits and harm of therapeutic drug monitoring for these two CNIs are similar. Tacrolimus C₀ is correlated with the AUC of tacrolimus (generally r > 0.8) (84,85). This relationship appears to be better during the first few months after transplant than later; however, there is high inter- and intrapatient variability. As is the case for CsA, there are no studies comparing one schedule of monitoring tacrolimus vs. another; however, tailoring the monitoring schedule to the expected absorption variability is a reasonable, empirical approach. Target levels for tacrolimus should reflect the patient's overall immunosuppressive drug regimen and risk for rejection, with higher targets early after transplantation, and lower targets later.

MMF monitoring

The AUC is widely regarded as the best measure of overall drug exposure of MPA. Pharmacokinetic studies have demonstrated poor correlation of C₀ with the full AUC (86). The inability of single-point sampling strategies, particularly those in the early postdose period, to effectively predict the AUC has resulted in a number of studies investigating the use of limited sampling strategies. These strategies use a number of sampling points, usually between 2 and 4 h, to predict the AUC (87).

Mycophenolate mofetil has conventionally been administered at a fixed dose without routinely monitoring MPA blood levels. Therapeutic drug monitoring during MMF therapy remains controversial. Available studies have serious limitations and report conflicting results. Early after transplantation, MPA AUC might be correlated with a lower risk of acute rejection than C₀, but this is supported by only a single RCT (88). There are two RCTs showing that targeting different MPA AUC resulted in different rates of acute rejection (89,90). Several observational studies have also shown that MPA AUC early after transplantation correlates with acute rejection (91–93). Most studies showed little correlation between MPA pharmacokinetic parameters and adverse effects (89–93). In addition, there is an important intrapatient variability of MPA pharmacokinetics and an increasing number of different drug combinations, which may affect MPA bioavailability. The proposed therapeutic window of the MPA AUC_0–12 (30–60 μg·h/mL) is restricted to the early posttransplant period and when MMF is used in combination with CsA. In general, MPA C₀ 1.0–3.5 mg/L correlates with MPA AUC_0–12 (30–60 μg·h/mL) in patients treated with CsA. A summary of the RCTs about MPA monitoring is provided in Supporting Table 37.

mTORi monitoring

The pharmacokinetics of mTORi sirolimus and everolimus differ substantially (94). Although the time to peak concentration is similar between the two mTORi, the half-life of sirolimus is about 60 h in adults (10–24 in children), while that of everolimus is 28–35 h (95,96). In general, C₀ correlates well with AUC_0–12 (95,97). Therefore, C₀ is probably adequate for monitoring mTORi levels. There are limited observational data suggesting that mTORi C₀ correlate with adverse effects (98). There are no RCTs demonstrating that monitoring mTORi C₀ reduces acute rejection or adverse effects.

Research Recommendations

• • 
  RCTs with adequate statistical power are needed to determine the cost-effectiveness of therapeutic drug monitoring for all immunosuppressive agents with measurable blood levels.

Chapter 6: Treatment of Acute Rejection

• 6.1: 
  We recommend biopsy before treating acute rejection, unless the biopsy will substantially delay treatment. (1C)
• 6.2: 
  We suggest treating subclinical and borderline acute rejection. (2D)
• 6.3: 
  We recommend corticosteroids for the initial treatment of acute cellular rejection. (1D)

  • 6.3.1: 
    We suggest adding or restoring maintenance prednisone in patients not on steroids who have a rejection episode. (2D)
  • 6.3.2: 
    We suggest using lymphocyte-depleting antibodies or OKT3 for acute cellular rejections that do not respond to corticosteroids, and for recurrent acute cellular rejections. (2C)
• 6.4: 
  We suggest treating antibody-mediated acute rejection with one or more of the following alternatives, with or without corticosteroids (2C):

  • • 
    plasma exchange;
  • • 
    intravenous immunoglobulin;
  • • 
    anti-CD20 antibody;
  • • 
    lymphocyte-depleting antibody.
• 6.5: 
  For patients who have a rejection episode, we suggest adding mycophenolate if the patient is not receiving mycophenolate or azathioprine, or switching azathioprine to mycophenolate. (2D)

OKT3, muromonab (anti–T-cell antibody).

Background

An acute rejection episode is the consequence of an immune response of the host to destroy the graft. It is of cellular (lymphocyte) and/or humoral (circulating antibody) origin. An acute rejection is clinically suspected in patients experiencing an increase in serum creatinine, after the exclusion of other causes of graft dysfunction (generally with biopsy). We know from the early days of transplantation, before there were effective antirejection treatments, that untreated acute rejection inevitably results in graft destruction. Therefore, it is strongly recommended that acute rejection episodes be treated, unless the treatment is expected to be life-threatening or to cause harm severe enough to preclude treatment.

Acute rejection is characterized by a decline in kidney function accompanied by well-established diagnostic features on kidney allograft biopsy. Subclinical acute rejection is defined by the presence of histological changes specific for acute rejection on screening or protocol biopsy, in the absence of clinical symptoms or signs. Acute cellular rejections are acute T-cell–mediated rejections and respond to treatment with corticosteroids. Borderline acute rejection is defined by histopathological changes that are only ‘suspicious for acute rejection’ according to the Banff classification schema (99). A rejection episode is said to be unresponsive to treatment when graft function does not return to baseline after the last dose of treatment.

An antibody-mediated rejection is defined by histological changes caused by a circulating, anti-HLA, donor-specific antibody. The following criteria are generally used to determine whether an acute rejection is caused by a donor-specific antibody:

• i) 
  staining of peritubular capillaries with C4d (fourth complement fraction);
• ii) 
  the presence of a circulating, anti-HLA, donor-specific antibody and
• iii) 
  histological changes consistent with an antibody-mediated rejection including (but not limited to) the presence of polymorphonuclear cells in peritubular capillaries.

Rationale

• • 
  Several causes of decreased kidney function can only be distinguished from acute rejection by biopsy.
• • 
  Treatment of decreased kidney allograft function that is not caused by acute rejection with additional immunosuppressive medication may be harmful.
• • 
  Treating subclinical acute rejection discovered on protocol biopsy may improve graft survival.
• • 
  Most acute cellular rejection responds to treatment with corticosteroids.
• • 
  Treating acute cellular rejection that is unresponsive to corticosteroids or recurs with an anti–T-cell antibody may prolong graft survival.
• • 
  Increasing the amount of immunosuppressive medication after an acute cellular rejection may help prevent further rejection.
• • 
  Treating borderline rejection may prolong graft survival.
• • 
  A number of measures may be effective in treating antibody-mediated rejections, including plasma exchange, intravenous immunuoglobulin, anti-CD20 antibody and anti–T-cell antibodies.

Although there are no RCTs to establish that obtaining a biopsy improves outcomes of suspected acute rejection, there are alternative diagnoses that might mimic an acute rejection episode. BK polyomavirus (BKV) nephropathy would generally be treated differently than acute rejection, for example with a reduction in immunosuppressive medication. Therefore, logic dictates that, whenever possible, biopsy confirmation should be obtained to avoid inappropriate treatment.

Some centers use protocol biopsies to detect and treat subclinical acute rejection. In a RCT, the detection and treatment of subclinical acute rejection in patients (N = 72) on CsA, MMF and corticosteroids resulted in better graft function (100,101). However, in a larger (N = 218) multicenter RCT in patients on tacrolimus, MMF and corticosteroids, protocol biopsies and treatment of subclinical acute rejection were not beneficial (102). Finally, in a single-center RCT of 102 recipients of living-donor kidneys (treated with CsA [N = 96] or tacrolimus [N = 6], MMF [N = 55] or azathioprine [N = 47] and corticosteroids) protocol biopsies and treatment of subclinical acute rejection resulted in improved graft function (103). Uncontrolled data suggest that, when the incidence of clinical acute rejection is low, the number of patients with subclinical acute rejection may be too small to warrant the inconvenience and cost of protocol biopsies (104).

Corticosteroid therapy is the most commonly used, first-line treatment for acute cellular rejection episodes. Although most patients respond to corticosteroids, the dose and duration of treatment has not been well defined by RCTs. Treatment starting with intravenous solumedrol 250–500 mg daily for 3 days is a common practice.

Treatment of acute cellular rejection with an anti–T-cell antibody (muromonab [OKT3], ATG or ALG) is more effective in restoring kidney function and preventing graft loss than treatment with corticosteroids (105). The systematic review concluded that treatment with an antibody is associated with more adverse effects, but whether the overall benefits of antibody treatment vs. corticosteroids outweigh harm is uncertain (105). There are no RCTs examining whether anti–T-cell antibodies vs. corticosteroids should be the initial treatment of Banff IIA or IIB (vascular) rejection. A low strength of evidence suggests no net benefits or harm between antibodies or steroids alone (see Evidence Profile in Supporting Table 39 at http://www3.interscience.wiley.com/journal/118499698/toc).

Studies suggest that steroid-resistant or recurrent T-cell–mediated rejection responds to treatment with polyclonal or monoclonal anti–T-cell antibodies (105). It is also possible that the addition of MMF to the postrejection maintenance immunosuppressive medication regimen, or replacement of azathioprine with MMF, will help to prevent subsequent acute rejection. A RCT (N = 221) compared MMF to azathioprine in the treatment of first acute rejection (106). Patients receiving MMF had fewer subsequent rejections, and among the 130 who completed the trial, at 3 years graft survival was better in the MMF group (106). A summary of the RCTs on replacement of azathioprine by MMF in the setting of rejection is provided in Supporting Tables 40–41.

Whether or not to treat borderline acute rejection is controversial. There are no RCTs addressing whether treatment of borderline acute rejection prolongs graft survival, and whether overall benefits outweigh harm.

If function does not return to baseline, or if there is a new decline in function after successful treatment of an acute rejection, a biopsy should be considered to rule out additional rejection, BKV nephropathy and other causes of graft dysfunction.

Anti–T-cell antibodies (OKT3, ATG, ALG) can be used when corticosteroids have failed to reverse rejection or for treatment of a recurrent rejection. In such circumstances, benefits generally outweigh harm. However, there is inadequate evidence from RCTs to conclusively establish the best treatment for steroid-resistant or recurrent acute cellular rejection (see Evidence Profile in Supporting Table 38). Most studies comparing OKT3 to ATG or ALG did not have adequate statistical power to show a difference in efficacy. However, in one RCT, ATG was better tolerated than OKT3 (107). When a steroid-resistant rejection or a recurrent rejection does not respond to a lymphocyte-depleting antibody or OKT3, a new biopsy should be considered to rule out alternative causes of graft dysfunction.

Therapeutic strategies that include combinations of plasma exchange to remove donor-specific antibody, and/or intravenous immunoglobulins and anti-CD20⁺ monoclonal antibody (rituximab) to suppress donor-specific antibody production have been used to successfully treat acute humoral rejection. However, the optimal protocol to treat acute humoral rejection remains to be determined. Indeed, there are no RCTs with adequate statistical power to compare the safety and efficacy of these different therapeutic strategies. In a RCT in 20 children, rituximab was associated with better function and improved postrejection biopsy scores compared to treatment with anti–T-cell antibody and/or corticosteroids (108). Clearly, additional studies to define the optimal treatment of acute humoral rejection are needed.

Research Recommendations

Additional RCTs are needed to determine:

• • 
  whether treating borderline acute rejection improves outcomes;
• • 
  when protocol biopsies and treatment of subclinical acute rejection are cost-effective;
• • 
  the optimal treatment for antibody-mediated acute rejection.

Chapter 7: Treatment of Chronic Allograft Injury

• 7.1: 
  We recommend kidney allograft biopsy for all patients with declining kidney function of unclear cause, to detect potentially reversible causes. (1C)
• 7.2: 
  For patients with CAI and histological evidence of CNI toxicity, we suggest reducing, withdrawing, or replacing the CNI. (2C)

  • 7.2.1: 
    For patients with CAI, eGFR >40 mL/min/1.73 m², and urine total protein excretion <500 mg/g creatinine (or equivalent proteinuria by other measures), we suggest replacing the CNI with a mTORi. (2D)

CAI, chronic allograft injury; CNI, calcineurin inhibitor; CsA, cyclosporine A; eGFR, estimated glomerular filtration rate; mTORi, mammalian target of rapamycin inhibitor(s).

Background

Historically, KTRs with gradually declining kidney allograft function associated with interstitial fibrosis and tubular atrophy (IF/TA) have been said to have ‘chronic rejection,’ or ‘chronic allograft nephropathy.’ However, these diagnoses are nonspecific and the Banff 2005 workshop suggested using ‘chronic allograft injury’ to avoid the misconception that the pathophysiology and treatment of this entity are understood (109). Causes of CAI include hypertension, CNI toxicity, chronic antibody-mediated rejection and others. Overall, death causes up to 50% of graft failures. However, of those who return to dialysis or require retransplantation, the most common cause is CAI, followed by acute rejection and recurrent primary kidney disease (110,111). Moderate to severe CAI is present in about one quarter of KTRs at 1 year after transplant, and in about 90% by 10 years (112–114). CAI is a diagnosis of exclusion characterized by the progressive reduction in graft function not due to recurrence of disease or other recognized causes. Histologically, CAI is defined by IF/TA (109,114). Other features may include subclinical rejection, transplant glomerulopathy or transplant vasculopathy.

Rationale

Graft function 6–12 months after kidney transplantation is an outcome reported in most RCTs of immunosuppressive medications. These are described in the relevant sections of these guidelines. Similarly, the use of other medications (antihypertensive agents, lipid-lowering agents, antiproteinuric agents) to prevent CAI or prevent the progression of CAI are also discussed in other sections of these guidelines.

Some causes of CAI may be reversible. Patients found to have acute rejection, BKV nephropathy or recurrent kidney disease, for example, may respond to appropriate treatments. Therefore, it is important that patients suspected of having CAI undergo biopsy, if possible. Most commonly, when there are no reversible causes of graft dysfunction, the biopsy will show IF/TA with or without other features consistent with CAI. In other words, the diagnosis of CAI is a diagnosis of exclusion. The roles of CNI toxicity, chronic antibody-mediated rejection and other immune and nonimmune mechanisms of injury are unclear. The treatment of CAI has been controversial (115).

CNI withdrawal and/or replacement

Although there are a large number of uncontrolled studies describing the effects of withdrawing CNIs in KTRs with CAI (116), there are only two RCTs. In both RCTs, the CNI was replaced with an alternative immunosuppressive agent. In the ‘Creeping Creatinine’ study of 143 KTRs, MMF was substituted for CsA, and outcomes were reported at 12 months (117). There were no differences in mortality, graft loss, acute rejection, infection or blood pressure between the two groups. Those randomized to MMF had a small improvement in their creatinine clearance (+5.0 mL/min [+0.8 mL/s] vs. –0.7 mL/min [–0.01 mL/s]) at 12 months, but creatinine clearance was not measured in 20%, and the long-term importance of this outcome is uncertain. The ‘Chronic Renal Allograft Failure’ study replaced CsA with tacrolimus in 186 KTRs (2:1 randomization) with moderate CKD. Baseline creatinine was 220 μmol/L and outcomes were reported at 5 years (118). There was no difference in death, graft loss, acute rejection, treatment discontinuations, NODAT, hypertension, infections or cancer between the two arms. However, incident cardiac events favored tacrolimus. Over 5 years, serum creatinine increased in the CsA group by about 60 μmol/L compared with the tacrolimus group. Overall, the quality of evidence evaluating the effects of replacing a CNI in patients with CAI is low, and there is uncertainty regarding benefit–harm trade-offs (see Evidence Profile and accompanying evidence in Supporting Tables 42–44 at http://www3.interscience.wiley.com/journal/118499698/toc).

CNI replacement with mTORi

No RCTs have examined whether switching KTRs with established CAI from a CNI to an mTORi is beneficial. However, a RCT randomly allocated 830 KTRs with estimated glomerular filtration rate (eGFR) ≥20 mL/min/1.73 m² to continuation of CNI (N = 275) vs. converting to sirolimus (N = 555) (119). Patients were stratified into two groups based on eGFR 20–40 mL/min/1.73 m² (N = 87) and eGFR >40 mL/min/1.73 m² (N = 743). The Data Monitoring and Safety Board stopped the trial for patients with eGFR 20–40 mL/min/1.73 m² when the primary safety end point (acute rejection, graft failure or death at 12 months) occurred in 8 of 48 of sirolimus vs. 0 of 25 CNI patients (p = 0.045). In the stratum eGFR >40 mL/min/1.73 m², the primary end point (change in eGFR baseline to 12 months) was not different in the two groups, but there was more proteinuria in the sirolimus group (119). Thus, this post hoc subgroup analysis suggested that converting patients with eGFR 20–40 mL/min/1.73 m² from CNI to sirolimus may be harmful, and that converting patients with eGFR >40 mL/min/1.73 m² may not be beneficial. However, the patients in this trial were not selected to have CAI per se, and it is possible that patients with CAI, preserved kidney function and low levels of proteinuria may still benefit from conversion. Additional study is needed.

Section II: Graft Monitoring and Infections

Rating Guideline Recommendations

Within each recommendation, the strength of recommendation is indicated as Level 1, Level 2, or Not Graded, and the quality of the supporting evidence is shown as A, B, C, or D.

Grade*	Wording
*The additional category ‘Not Graded’ was used, typically, to provide guidance based on common sense or where the topic does not allow adequate application of evidence. The most common examples include recommendations regarding monitoring intervals, counseling, and referral to other clinical specialists. The ungraded recommendations are generally written as simple declarative statements, but are not meant to be interpreted as being stronger recommendations than Level 1 or 2 recommendations.
Level 1	‘We recommend’
Level 2	‘We suggest’

Grade for quality of evidence	Quality of evidence
A	High
B	Moderate
C	Low
D	Very low

Chapter 8: Monitoring Kidney Allograft Function

• 8.1: 
  We suggest measuring urine volume (2C):

  • • 
    every 1–2 hours for at least 24 hours after transplantation (2D);
  • • 
    daily until graft function is stable. (2D)
• 8.2: 
  We suggest measuring urine protein excretion, (2C) at least:

  • • 
    once in the first month to determine a baseline (2D);
  • • 
    every 3 months during the first year (2D);
  • • 
    annually, thereafter. (2D)
• 8.3: 
  We recommend measuring serum creatinine, (1B) at least:

  • • 
    daily for 7 days or until hospital discharge, whichever occurs sooner (2C);
  • • 
    two to three times per week for weeks 2–4 (2C);
  • • 
    weekly for months 2 and 3 (2C);
  • • 
    every 2 weeks for months 4–6 (2C);
  • • 
    monthly for months 7–12 (2C);
  • • 
    every 2–3 months, thereafter. (2C)
  • 8.3.1: 
    We suggest estimating GFR whenever serum creatinine is measured, (2D) using:

    • • 
      one of several formulas validated for adults (2C); or
    • • 
      the Schwartz formula for children and adolescents. (2C)
• 8.4: 
  We suggest including a kidney allograft ultrasound examination as part of the assessment of kidney allograft dysfunction. (2C)

GFR, glomerular filtration rate.

Background

Some tests need to be performed routinely to detect abnormalities that may lead to treatment or prevention of complications that are common in KTRs (Table 4). The frequency of screening is based on the incidence of the complication being screened, because there are no other data to determine the best interval for screening. Serum creatinine is easily measured and readily available in most laboratories. Screening tests for urine protein excretion include dipstick tests for total protein or albumin, as well as randomly collected ‘spot’ urine to measure protein-to-creatinine or albumin-to-creatinine ratios.

Table 4.  Routine screening after kidney transplantation

Screening test	Screening intervals by time after transplantation
	1 week	1 month	2–3 months	4–6 months	7–12 months	>12 months
BKV, BK polyoma virus; EBV, Epstein-Barr virus; NAT, nucleic acid testing. ^aSerum creatinine. ^bUrine total protein and/or urine albumin. ^cComplete blood count including white blood count, hemoglobin and platelet counts. ^dScreen for diabetes with fasting blood glucose, glucose tolerance test, or HbA1c level. ^eLipid profile includes fasting cholesterol, LDL-C, HDL-C, and triglycerides. ^fScreen for tobacco use. ^gScreen for BKV using plasma NAT. ^hScreen for EBV using plasma NAT in patients with no antibody to EBV at transplant.
Creatinine^a	Daily	2–3 per week	Weekly	Every 2 weeks	Monthly	Every 2–3 months
Urine protein^b	…………..Once……………..		…………………Every 3 months………………………………..			Annually
Complete blood count^c	Daily	2–3 per week	Weekly	……………….Monthly……………….		Annually
Diabetes^d	…………..Weekly…………….		…………………..Every 3 months………………………………			Annually
Lipid profile^e	–	–	Once	–	–	Annually
Tobacco use^f	Prior to discharge					Annually
BKV NAT^g	…………………………..Monthly…………………….			………….Every 3 months…………..		–
EBV NAT (seronegative)^h	Once	………………..Monthly………………		………….Every 3 months…………..
Blood pressure, pulse, height, body weight	………………………………………………………………Each clinic visit………………………………………………………….

Rationale

• • 
  Detecting kidney allograft dysfunction as soon as possible will allow timely diagnosis and treatment that may improve outcomes.
• • 
  Urine output that is inappropriately low, or inappropriately high, is an indication of possible graft dysfunction.
• • 
  Serum creatinine and urine protein measurements are readily available and are useful for detecting acute and chronic allograft dysfunction.
• • 
  Ultrasound is relatively inexpensive and reasonably accurate for diagnosing treatable causes of kidney allograft dysfunction.

Urine volume

Urine volume is an easily-measured parameter of early kidney allograft function (120). The recovery of kidney function, measured as a decrease in serum creatinine and blood urea nitrogen, is generally preceded by an increase in urine volume (120). Rarely, excessive urine volume may indicate the presence of a saline diuresis or a water diuresis caused by tubular damage. In addition to its role in assessing early allograft dysfunction, measuring the urine volume is an important part of overall fluid and electrolyte management.

Urine protein excretion

Proteinuria is an early and sensitive marker of kidney damage in CKD (121). Many causes of proteinuria are potentially reversible with appropriate treatment (Table 5) (122), and detection of proteinuria can therefore improve graft outcomes (113,122–132). Patients with proteinuria generally have lower kidney function compared to patients without proteinuria (122,129). Proteinuria is also associated with mortality and CVD events in KTRs (130–132).

Table 5.  Some causes of proteinuria after kidney transplantation

CAI, chronic allograft injury; FSGS, focal segmental glomerulosclerosis; HUS, hemolytic-uremic syndrome; IgA, Immunoglobulin A.

Persistent disease in the native kidneys

Allograft rejection and drug toxicity

Acute rejection

Thrombotic microangiopathy

CAI

Transplant glomerulopathy

De novo and recurrent glomerular diseases

Minimal change disease

FSGS

IgA glomerulonephritis

Membranous glomerulonephritis

Membranoproliferative glomerulonephritis

Postinfectious glomerulonephritis

Thrombotic thrombocytopenic purpura

HUS

Vasculitis

Diabetic nephropathy

Systemic lupus erythematosus

Amyloidosis

Light- and heavy-chain deposition diseases

Proteinuria includes albuminuria as well as other proteins. The urinary excretion rate for albumin and total protein can be estimated from the ratio of albumin or total protein to creatinine concentration in a casual urine specimen (133–136). Creatinine excretion is higher in men than in women. Therefore, the values in the general population and cut-off values for abnormalities in urine albumin-to-creatinine ratio are lower for men than women (137,138 (Table 6). For details, see Kidney Disease Outcomes Quality Initiative (KDOQI) Guidelines for Chronic Kidney Disease, Part 5, Assessment of Proteinuria (http://www.kidney.org/professionals/kdoqi/guidelines_ckd/p5_lab_g5.htm; last accessed March 30, 2009).

Table 6.  Definitions of proteinuria and albuminuria

	Urine collection method	Normal	Microalbuminuria	Albuminiuria or clinical proteinuria
NA, not applicable. Reference values for urinary protein and albumin excretion in pediatric patients (139,140). To convert metric units to SI units, see Conversion Factors, p. Six. Modified with permission (141).
Total protein	24-h excretion	<300 mg/day (adults)	NA	≥300 mg/day (adults)
		<4 mg/m^2/h (children)		≥4 mg/m^2/h (children)
	Dipstick	<30 mg/dL (adults and children)	NA	≥30 mg/dL (adults and children)
	Spot protein-to-creatinine ratio	<200 mg/g (adults)	NA	≥200 mg/g (adults)
		<0.2 mg/mg (children 2 years or older)
		<0.5 mg/mg (<6–24 months old)
Albumin	24-h excretion	<30 mg/day	30–300 mg/day	>300 mg/day
	Albumin dipstick	<3 mg/dL	≥3 mg/dL	NA
	Spot albumin-to-creatinine ratio	<17 mg/g (men)	17–250 mg/g (men)	>250 mg/g (men)
		<25 mg/g (women)	25–355 mg/g (women)	>355 mg/g (women)
		<30 mg/g (children)

Serum creatinine

Causes of kidney allograft dysfunction that require rapid intervention for treatment to be effective include acute rejection, obstruction, urine leak, vascular compromise and some recurrent diseases, for example focal segmental glomerulosclerosis (FSGS). These causes are more common in the first few days to weeks after kidney transplantation than in subsequent months to years. Therefore, it is important to closely monitor kidney function early after transplantation.

Measurement of the serum creatinine concentration is a simple, inexpensive and universally available method for estimating GFR, and it is reliable for detecting acute changes of kidney function (142,143). The level of serum creatinine at year 1 after transplantation is a risk factor for subsequent outcomes, and may help guide care, for example the frequency of visits (144,145).

A gradual increase in serum creatinine after the first year may be due to acute rejection, but more often is caused by CAI, recurrence of the original kidney disease, or de novo kidney disease. Unfortunately, serum creatinine is less reliable for detecting chronic changes (over months to years) in kidney function.

As is true in the general population, measurement of GFR with inulin, iothalamate, iohexol or other suitable markers of GFR, either with urinary or plasma clearance techniques, provides the most accurate measure of allograft function in KTRs. Although these tests are appropriate for clinical use, the Work Group did not recommend their use in routine clinical practice due to cost, low patient acceptance, and lack of availability outside of academic medical centers. Measurement of cystatin C has also been used to monitor kidney function. The advantage of cystatin C is its independence from body weight. However, at present, there is a paucity of validation studies for cystatin C estimates of GFR in KTRs (146–148).

Formulas to estimate GFR have been tested in KTRs, but no formula has been consistently shown to be superior to any other formula (149–156). It is unlikely that these formulas will improve the ability of serum creatinine to estimate acute changes in kidney function since, in most formulas, the only component of the formula that changes significantly is serum creatinine. It is similarly unclear whether formulas improve the ability of serum creatinine to measure chronic changes in kidney transplant function, especially when serum creatinine may change due to changes in muscle mass due to an improved nutritional status after kidney transplantation (157–159).

Kidney allograft ultrasound examination

Many of the most common causes of allograft dysfunction, other than rejection, can be diagnosed by ultrasound. These include arterial occlusion, venous thrombosis, urinary obstruction, a urine leak (large fluid collection), compressing perinephric hematoma and arteriovenous fistula from a kidney biopsy (160–163). Ultrasound is also useful in guiding a kidney allograft biopsy, so it is often obtained at the time of biopsy. In the kidney allograft, mild to moderate calyceal distension can be normal, so a baseline ultrasound examination when kidney function is normal may be useful to compare to subsequent ultrasound examinations for allograft dysfunction.

Chapter 9: Kidney Allograft Biopsy

• 9.1: 
  We recommend kidney allograft biopsy when there is a persistent, unexplained increase in serum creatinine. (1C)
• 9.2: 
  We suggest kidney allograft biopsy when serum creatinine has not returned to baseline after treatment of acute rejection. (2D)
• 9.3: 
  We suggest kidney allograft biopsy every 7–10 days during delayed function. (2C)
• 9.4: 
  We suggest kidney allograft biopsy if expected kidney function is not achieved within the first 1–2 months after transplantation. (2D)
• 9.5: 
  We suggest kidney allograft biopsy when there is:

  • • 
    new onset of proteinuria (2C);
  • • 
    unexplained proteinuria ≥3.0 g/g creatinine or ≥3.0 g per 24 hours. (2C)

Background

Kidney allograft biopsies are performed for specific clinical indications, or as part of a surveillance program (or protocol). An ‘indicated biopsy’ is one that is prompted by a change in the patient's clinical condition and/or laboratory parameters. A ‘protocol biopsy’ is one obtained at predefined intervals after transplantation, regardless of kidney function. In both cases, the biopsy is obtained to find histological changes prompting treatment to improve outcomes. DGF is graft function low enough to require dialysis in the first week after kidney transplantation, or lack of improvement in pretransplant kidney function.

New-onset proteinuria (defined in Table 6) may indicate treatable causes of graft dysfunction, including acute rejection and thrombotic microangiopathy. In patients who already have proteinuria, an increase exceeding a threshold usually defined as ‘nephrotic range’ proteinuria, for example ≥3.0 g/g creatinine or ≥3.0 g/24 h, may indicate treatable causes of graft dysfunction.

Rationale

• • 
  Increased serum creatinine that is not explained by dehydration, urinary obstruction, high CNI levels or other apparent causes is most likely due to an intragraft parenchymal process, such as acute rejection, CAI, drug toxicity, recurrent or de novo kidney disease or BKV nephropathy.
• • 
  The optimal diagnosis and treatment of intragraft parenchymal causes of allograft dysfunction require an adequate biopsy.
• • 
  In patients with DGF, change in serum creatinine is not useful for ruling out acute rejection, and protocol biopsies are needed to rule out acute rejection.
• • 
  Proteinuria, or a substantial increase in proteinuria, may indicate a potentially treatable cause of graft dysfunction.

Biopsies for an increase in serum creatinine

Although serum creatinine has many limitations for estimating GFR (see Chapter 8), an unexplained rise in serum creatinine is generally indicative of a decline in GFR. Some fluctuation in creatinine can result from normal laboratory or physiological variability. Hence, only a persistent increase that is outside this normal, but poorly defined, range is clinically relevant. A 25–50% increase over baseline is often arbitrarily used in studies. At least one study suggested that a persistent 30% rise in serum creatinine was an excellent predictor of subsequent graft failure (144,145). The Acute Kidney Injury Network (164) has proposed a definition and classification scheme for evaluating acute kidney injury (Table 7).

Table 7.  Diagnostic criteria for acute kidney injury

BMI, body mass index. Adapted with permission (164).
Criteria	An abrupt (within 48 h) reduction in kidney function currently defined as an absolute increase in serum creatinine of ≥0.3 mg/dL (≥26.4 μmol/L), a percentage increase in serum creatinine of ≥50% (1.5-fold from baseline), or a reduction in urine output (documented oliguria of less than 0.5 mL/kg/h for more than 6 h).
Notes	The above criteria include both an absolute and a percentage change in creatinine to accommodate variations related to age, gender and BMI, and to reduce the need for a baseline creatinine but do require at least two creatinine values within 48 h.

Causes of acute, reversible declines in GFR should be ruled out, including dehydration, urinary obstruction or acute CNI toxicity (by demonstrating high blood levels), before a biopsy is performed. If there are no apparent causes of a decline in GFR, then an allograft biopsy is generally warranted to detect the nature of potentially treatable causes of kidney injury, including rejection, infections like BKV nephropathy, recurrent or de novo kidney disease or infiltration with posttransplant lymphoproliferative disease (PTLD). Since any of these conditions can develop in the setting of preexisting graft pathology, additional biopsies may be required when an abrupt change in the rate of progression is observed.

Biopsies can determine both the type and severity of immunologic damage (109). Different types of acute rejection may require different treatment approaches. For example, acute cellular rejection is usually treated with steroid pulses, but acute antibody-mediated rejection may prompt the use of specific treatments in addition to steroids.

Biopsies for a lack of improvement in graft function

When acute rejection does not respond to first-line treatment with steroids, additional treatment (e.g. with a lymphocyte-depleting antibody) may be successful (105,165). Alternatively, a failure of function to return to baseline could be due to a new pathological process, such as coexistent acute tubular necrosis, drug toxicity or BKV nephropathy, that would require a different treatment approach. Therefore, a biopsy is indicated to determine the correct treatment.

Patients should always be assessed for their suitability for biopsy before undertaking the procedure. Biopsies may be hazardous in those with a bleeding diathesis, or in the presence of large fluid collections or infection.

Biopsies for DGF

Observational studies have shown that the incidence of acute rejection during DGF is higher than in patients without DGF (166–168). Kidney function cannot be used as an indication for biopsy to diagnose superimposed acute rejection while the patients are already being treated with dialysis due to DGF, or when the serum creatinine does not fall from pretransplant values. It is therefore prudent to obtain periodic biopsies of the kidney during DGF to diagnose acute rejection. There are few data to determine when and how often biopsies during DGF should be obtained. However, studies in which biopsies have been obtained every 7–10 days, while patients are receiving dialysis for DGF, have shown that acute rejection can be present for the first time on the second, third or even fourth biopsy (167).

In centers that have a very low overall incidence of acute rejection, the incidence of acute rejection during DGF could also be low enough to obviate the need for biopsies during DGF. A biopsy may no longer be needed when there are signs that DGF is resolving, for example when urine output is increasing rapidly or serum creatinine is declining.

Protocol biopsies

Acute rejection, CAI and CNI toxicity can occur in the absence of a measurable decline in kidney function. Several studies have shown that protocol biopsies can detect clinically inapparent (subclinical) acute rejection, CAI and CNI nephrotoxicity. The reported prevalence of subclinical rejection (Banff grade 1A or higher) varies from 13% to 25% at 1–2 weeks, 11–43% at 1–2 months, 3–31% at 2–3 months and 4–50% at 1 year (169–175).

Data from observational studies indirectly suggest that detecting and treating subclinical acute rejection with protocol biopsies may be beneficial. Subclinical rejection is associated with CAI (170,173,176,177) and reduced graft survival (176–179).

In another study, subclinical acute rejection in 14-day protocol biopsies was associated with poorer 10-year graft survival (179). Graft survival rates with subclinical rejection, borderline subclinical rejection or no rejection were 88%, 99% and 98% at 1 year (p < 0.05), and 62%, 94% and 96% at 10 years (p < 0.05), respectively. In a pediatric study, subclinical rejection was associated with progressive CAI, reduced creatinine clearance and shorter graft survival (177).

Treatment of subclinical rejection may improve outcomes. In a RCT, 72 patients were randomly allocated to undergo protocol biopsies and treatment of subclinical rejection at 1, 2, 3, 6 and 12 months (biopsy group), or protocol biopsies without treatment at 6 and 12 months only (control group) (100). Patients in the biopsy arm of the study had a significant decrease in acute rejection episodes, a reduced 6-month chronic tubulointerstitial score and a lower 2-year serum creatinine. Interstitial fibrosis was less in those treated for subclinical rejection (100). In another trial, 52 living-donor KTRs were randomized to undergo protocol biopsies and 50 controls had only indicated biopsies (103). At 1 and 3 months, protocol biopsies revealed borderline changes in 11.5% and 14% patients, acute rejection in 17% and 12% and CAI in 4% and 10%, respectively. The incidence of clinically evident acute rejection episodes was similar in the two groups, but the biopsy group had lower serum creatinine at 6 months (p = 0.0003) and 1 year (p < 0.0001).

Baseline immunosuppression is likely important in determining the incidence of subclinical rejection and thereby the benefit of protocol biopsies. Tacrolimus- and MMF-treated patients generally have a lower rate of acute rejection than patients treated with CsA and azathioprine, and tacrolimus is associated with a reduced incidence of subclinical rejection (104,113,176,180,181), lower acute Banff scores (182,183) and 1-year serum creatinine (181). In a RCT, 121 patients were randomly allocated to biopsies at 0, 1, 2, 3 and 6 months, and 119 to biopsies at 0 and 6 months (102). At 6 months, 35% of the biopsy arm and 20.5% of the control arm patients had interstitial fibrosis and tubular atrophy (ci + ct) scores ≥2 (p = 0.07). Of note, the frequency of clinical acute rejection episodes was only 10% in the biopsy arm and 7% in the control arm (p > 0.05). The prevalence of subclinical rejection in the biopsy arm was 4.6%. Creatinine clearance at 6 months was not different (p > 0.05) in the two groups. Use of protocol biopsies, therefore, for diagnosis of subclinical rejection may not be appropriate in tacrolimus- and MMF-treated patients.

Other conditions that can be detected on protocol biopsies include CNI toxicity, recurrent disease, transplant glomerulopathy, CAI and BKV nephropathy. However, it is unclear whether the detection of these conditions by protocol biopsy improves outcomes.

The safety of biopsies has been documented in several series (180,184). The reported risk of major complications from protocol biopsy, including substantial bleeding, macroscopic hematuria with ureteric obstruction, peritonitis or graft loss, is approximately 1% (185–187). The reported incidence of graft loss from protocol biopsy is 0.03%. Protocol biopsies can be done safely as an outpatient procedure. Data collected on 1705 protocol kidney transplant biopsies at one center showed that all of the complications became evident in the first 4 h after the biopsy (188).

Protocol biopsies, however, may be expensive. The Mayo Clinic reported that protocol biopsies cost US$ 3000 per biopsy, and it cost US$ 114 000 to detect one case of acute subclinical rejection (104). Therefore, decisions on whether or not to perform protocol biopsies should take these and other factors, including patient preferences, into account. Altogether, based on very-low-quality evidence, the benefit of performing protocol biopsies in CsA/azathioprine-treated patients without induction therapy may outweigh the harm (see Evidence Profile and accompanying evidence in Supporting Tables 45–47 at http://www3.interscience.wiley.com/journal/118499698/toc).

Research Recommendations

• • 
  RCTs are needed to determine when the benefits of protocol biopsies outweigh harm.

Chapter 10: Recurrent Kidney Disease

• 10.1: 
  We suggest screening KTRs with primary kidney disease caused by FSGS for proteinuria (2C) at least:

  • • 
    daily for 1 week (2D);
  • • 
    weekly for 4 weeks (2D);
  • • 
    every 3 months, for the first year (2D);
  • • 
    every year, thereafter. (2D)
• 10.2: 
  We suggest screening KTRs with potentially treatable recurrence of primary kidney disease from IgA nephropathy, MPGN, anti-GBM disease, or ANCA-associated vasculitis for microhematuria, (2C) at least:

  • • 
    once in the first month to determine a baseline (2D);
  • • 
    every 3 months during the first year (2D);
  • • 
    annually, thereafter. (2D)
• 10.3: 
  During episodes of graft dysfunction in patients with primary HUS, we suggest screening for thrombotic microangiopathy (e.g. with platelet count, peripheral smear for blood cell morphology, plasma haptoglobin, and serum lactate dehydrogenase). (2D)
• 10.4: 
  When screening suggests possible treatable recurrent disease, we suggest obtaining an allograft biopsy. (2C)
• 10.5: 
  Treatment of recurrent kidney disease:

  • 10.5.1: 
    We suggest plasma exchange if a biopsy shows minimal change disease or FSGS in those with primary FSGS as their primary kidney disease. (2D)
  • 10.5.2: 
    We suggest high-dose corticosteroids and cyclophosphamide in patients with recurrent ANCA-associated vasculitis or anti-GBM disease. (2D)
  • 10.5.3: 
    We suggest using an ACE-I or an ARB for patients with recurrent glomerulonephritis and proteinuria. (2C)
  • 10.5.4: 
    For KTRs with primary hyperoxaluria, we suggest appropriate measures to prevent oxalate deposition until plasma and urine oxalate levels are normal (2C), including:

    • • 
      pyridoxine (2C);
    • • 
      high calcium and low oxalate diet (2C);
    • • 
      increased oral fluid intake to enhance urinary dilution of oxalate (2C);
    • • 
      potassium or sodium citrate to alkalinize the urine (2C);
    • • 
      orthophosphate (2C);
    • • 
      magnesium oxide (2C);
    • • 
      intensive hemodialysis to remove oxalate. (2C)

ACE-I, angiotensin-converting enzyme inhibitor; ANCA, antineutrophil cytoplasmic autoantibody; ARB, angiotensin II receptor blocker; FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; HUS, hemolytic-uremic syndrome; IgA, immunoglobulin A; KTRs, kidney transplant recipients; MPGN, membranoproliferative glomerulonephritis.

Background

The primary kidney disease is generally documented by pretransplant biopsy of the native kidney, or of a previous kidney transplant. Recurrence of the primary kidney disease is usually established when there is biopsy-documented involvement of the kidney allograft with the primary kidney disease.

Rationale

• • 
  Some recurrent kidney diseases cause allograft failure.
• • 
  Treatment of some recurrent kidney diseases may prevent, or delay, the onset of graft failure.
• • 
  Screening for treatable recurrent kidney disease may result in early diagnosis and treatment that may be beneficial.

Recurrence of primary kidney diseases is an important cause of morbidity and graft loss following kidney transplantation, in both adults and children. In a study of 1505 cases with both native kidney and kidney allograft biopsies documenting recurrent glomerular disease, graft loss due to recurrent glomerulonephritis was the third most frequent cause for graft failure 10 years after kidney transplantation (110). Recurrence may present as increased serum creatinine (reduced GFR), new-onset or increased proteinuria and/or hematuria. The impact of recurrence varies according to the primary kidney disease. Not all diseases recur with equal frequency. The risk of recurrence is particularly increased in FSGS, immunoglobulin A (IgA) nephropathy, membranoproliferative glomerulonephritis (MPGN), hemolytic-uremic syndrome (HUS), oxalosis and Fabry's disease and, to a lesser extent, with lupus nephritis, anti-glomerular basement membrane (GBM) disease and vasculitis (189). Also, the timing of recurrence and manner of presentation vary for different diseases. FSGS, HUS and oxalosis may recur in the first few days to weeks after transplantation, whereas the timing is variable in the others (127).

In a majority of instances, proteinuria and/or reduced GFR provide the initial basis for suspecting disease recurrence. Since these parameters are periodically assessed in KTRs as part of their routine monitoring, a separate strategy for detection of disease recurrence is not warranted.

The modality of screening for some of these diseases, however, may vary from the usual posttransplant monitoring if timely detection is not achieved by the routine posttransplant monitoring strategies (Table 8). For example, FSGS can recur early; hence, screening for FSGS recurrence requires early and frequent monitoring for proteinuria. HUS recurrence requires looking for evidence of microangiopathic hemolysis. Screening for recurrent IgA nephropathy, MPGN, anti-GBM disease and vasculitis require examination of urinary sediment to detect microhematuria and/or presence of casts in addition to screening for proteinuria. It is appropriate to perform dipstick testing for proteinuria followed by quantitation using spot protein creatinine ratio or timed urine collection. Depending on the primary disease, biopsy evaluation may require immunofluorescence and electron microscopy in addition to light microscopy to confirm recurrence and to rule out other causes of proteinuria, hematuria or graft dysfunction (190).

Table 8.  Screening for recurrent diseases

Disease	Screening (in addition to serum creatinine)	Minimum screening frequency	Diagnostic tests (in addition to kidney biopsy)	Potential treatment
ANCA, antineutrophil cytoplasmic antibody; FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; HUS, hemolytic-uremic syndrome; IgA, immunoglobulin A; LDH, lactate dehydrogenase; MPGN, membranoproliferative glomerulonephritis.
FSGS	Proteinuria	Daily for 1 week, weekly for 4 weeks, every 3 months for 1 year, then annually		Plasmapheresis
IgA nephropathy	Proteinuria, microhematuria	Once in the first month, every 3 months in the first year, then annually
MPGN	Proteinuria, microhematuria		Serum complement levels
Anti-GBM disease	Proteinuria, microhematuria		Anti-GBM antibodies	Plasmapheresis
Pauci-immune vasculitis	Proteinuria, microhematuria		ANCA	Cyclophosphamide and corticosteroids
HUS	Proteinuria, platelet count	During episodes of graft dysfunction	Platelet count, peripheral blood smear, LDH	Plasmapheresis

There is also weak evidence (uncontrolled case studies and case reports) that disease-specific treatment may be beneficial for some recurrent diseases.

Idiopathic FSGS

Idiopathic, or primary, FSGS is characterized by typical sclerosis in a segment of glomerular tuft, along with foot-process fusion on electron microscopy. Sclerosis may not be evident in early recurrence, and light microscopy may show normal glomerular architecture. Recurrence is suspected when a patient with a documented primary FSGS in the native kidneys or a prior kidney allograft develops proteinuria and/or increase in serum creatinine, typically soon after transplantation (127).

Idiopathic FSGS recurs in 20–50% of KTRs (up to 80% if it has recurred in a prior kidney transplant) (191). It is important to distinguish idiopathic from secondary causes of FSGS that generally do not recur. Recurrence of familial FSGS has also been documented, if the donor is an obligate carrier (191). Putative risk factors for recurrence include age of onset of FSGS in native kidneys between 6 and 15 years (192), rapid course of the original disease (e.g. less than 3 years from diagnosis to CKD stage 5), diffuse mesangial proliferation on histology and non-African American ethnicity. The strongest risk factor is recurrence in a previous transplant.

The demonstration of increase in the albumin permeability of isolated rat glomeruli by sera from patients with a recurrent FSGS offers the possibility of more accurate prediction of the risk of recurrent disease (193). However, this assay is still experimental.

Idiopathic FSGS can recur at any time after transplantation, but recurrence is more common early after transplantation. Recurrent disease presents with proteinuria, which is usually heavy. About 80% of cases recur in the first 4 weeks (193). Proteinuria screening therefore needs to be more frequent in the early posttransplant period in those with CKD stage 5 due to FSGS, especially those with risk factors for recurrence. The exact frequency has not been worked out. Interpretation of proteinuria, especially in the early posttransplant period, requires knowledge of pretransplant proteinuria. Although proteinuria from the native kidneys declines after transplantation (194), the time taken for its disappearance is variable. Posttransplant proteinuria therefore should be interpreted in light of the pretransplant values.

There have been no RCTs of therapy for recurrent idiopathic FSGS. However, there have been individual cases, and uncontrolled series, reporting that patients with recurrent idiopathic FSGS may have a substantial reduction in urine protein excretion after plasma exchange (195,196). This probably occurs by removing circulating factors that alter glomerular permeability to protein. Predictors of response to plasma exchange include early initiation of treatment after recurrence, and possibly an early recurrence of disease (196). Unfortunately, proteinuria may recur after treatment, and may require additional plasma exchange, or even periodic, ongoing treatments. The presumption is that reducing protein excretion with plasma exchange will help preserve allograft function, but no studies have examined this.

It is unclear how many plasma-exchange treatments are required to reduce protein excretion, but one review found a median of nine treatments before there was a remission in proteinuria (195). In small case series, prophylactic plasma exchange has been reported, but the data are not convincing that this is effective in preventing recurrent FSGS (197,198).

High-dose CsA may induce remission of proteinuria. In one series, 14 of 17 children entered lasting remission (199). The rationale behind maintaining a high CsA blood level is to overcome the effect of high serum cholesterol often seen in patients with recurrent FSGS (lipoproteins bind CsA and reduce free CsA levels). High-dose CsA may be combined with plasmapheresis. A study concluded that plasmapheresis alone was not sufficient to induce remission except when combined with high-dose CsA (200).

For patients who do not respond to plasma exchange, or for patients who have non-nephrotic proteinuria, a reduction in proteinuria with an angiotensin-converting enzyme inhibitor (ACE-I) and/or an angiotensin II receptor blocker (ARB) may be beneficial.

IgA nephropathy

IgA nephropathy is the most common type of glomerulonephritis worldwide and is the primary cause of CKD stage 5 in 20% of KTRs in many parts of the world. Recurrent IgA nephropathy is common after transplantation. Reported incidence of recurrence varies from 13% to 53% according to differences in duration of follow-up and biopsy policy of different transplant centers, with the highest rates in centers that perform routine protocol biopsies (201). Latent IgA deposits in the donor kidney (identified on preimplantation biopsies) are responsible for ‘recurrence’ in some cases transplanted for kidney failure due to IgA nephropathy in areas with high disease prevalence (202). Single-nucleotide polymorphisms in the interleukin-10 and TNF-alpha genes have been shown to predict recurrence risk (203,204). The estimated 10-year incidence of graft loss due to recurrence was 9.7% (CI = 4.7–19.5%) (110). Recurrence risk in retransplants is increased if the first graft was lost due to recurrent IgA nephropathy in less than 10 years (205). There is no effective therapy for preventing recurrent IgA nephropathy. ACE-Is and ARBs have been shown to reduce proteinuria and possibly preserve kidney function in recurrent IgA nephropathy (206). In a study of 116 KTRs with IgA nephropathy, use of ATG as induction therapy was associated with a reduction in recurrence risk from 41% to 9% when compared to IL2 receptor antagonists (207).

Membranoproliferative glomerulonephritis

Secondary causes of MPGN, such as hepatitis C, should be ruled out. The histological recurrence rate in idiopathic type I MPGN is 20–30% and exceeds 80% in type 2 disease (192). Manifestations include microhematuria, proteinuria and deterioration of kidney function. Risk factors for recurrence include severity of histological lesions in native kidneys, HLA-B8DR3, living related donors and previous graft loss from recurrence (208,209). There are reports of response to long-term cyclophosphamide (210), plasmapheresis (211–213) and CsA (214).

Hemolytic-uremic syndrome

Hemolytic-uremic syndrome is defined histopathologically by intimal cell proliferation, and thickening and necrosis of the wall, thrombi and narrowed lumens of glomerular, arteriolar or interlobular artery. The severity can range from endothelial swelling to complete cortical necrosis. It manifests clinically with microangiopathic hemolytic anemia and rapid worsening of kidney function with or without involvement of other organs. HUS is often classified as diarrhea-associated (D)− HUS (atypical) and D+ HUS (typical).

Hemolytic-uremic syndrome recurs commonly in adults and in children in whom the original kidney disease was D− variant. The overall recurrence risk is less than 10% in the pediatric population; D+ HUS usually does not recur, while idiopathic D− or familial HUS may recur in 21–28% of children (215). Recurrence occurs in about 80–100% of patients with factor H or factor I mutation, while patients with a mutation in membrane cofactor protein do not have recurrence (216,217). The risk is higher in adults, with 33–56% (218–220) showing clinical manifestations and an additional 16–20% of patients demonstrating clinically silent recurrence. Recurrence is particularly frequent in adults with autosomal recessive or dominant HUS (215). Recurrence develops within 4 weeks in most cases. Most patients show microangiopathic anemia, thrombocytopenia and kidney dysfunction, whereas others present with rapidly progressive graft dysfunction without showing the classic hematologic manifestations. Platelet count should be performed during episodes of graft dysfunction in KTRs with HUS as the original cause of CKD stage 5. In those with falling counts, additional tests such as examination of peripheral blood smear to look for fragmented cells (schistocytes), haptoglobin and lactate dehydrogenase estimation to document hemolysis are warranted. Long-term graft survival is lower (about 30%) in those with recurrence.

Treatment strategies have included plasmapheresis, intravenous immunoglobulin and rituximab. Aggressive plasmapheresis using fresh frozen plasma (40–80 mL/kg per session) increases the levels of deficient factors and has provided encouraging results, even in those with factors H and I mutations (221–223). As factor H is synthesized in the liver, combined liver and kidney transplantation (together with preoperative and intraoperative plasmapheresis using fresh frozen plasma and low-molecular-weight heparin) could reduce the risk of recurrence (222,224–226). Intravenous immunoglobulin and rituximab have been reported to rescue recurrent HUS resistant to multiple courses of plasma exchanges (227,228). There is no evidence that avoidance of CNI, mTORi and OKT3 (that may themselves cause thrombotic microangiopathy) will reduce the recurrence risk.

ANCA-associated vasculitis and anti-GBM disease

Both antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis and anti-GBM disease may present with rapidly progressive CKD and crescentic glomerulonephritis. Recurrence rates are low if the disease is quiescent at the time of transplant. In an analysis of pooled data from 127 patients with ANCA-associated vasculitis, 17% of patients had recurrence, with kidney manifestation in 57.1%. Kidney dysfunction occurred in 33% of those with recurrence (229). More recent studies (230) report lower (7%) recurrence rates, most beyond the first posttransplant year with no direct or indirect impact on allograft function. ANCA-associated vasculitis relapses in the kidney allograft usually manifest as pauci-immune necrotizing glomerulonephritis, but graft function can also be affected by acute arteritis, ureteral stenosis and obstructive uropathy due to granulomatous vasculitis.

Pretransplantation disease course, disease subtype, ANCA type or titer, time of transplantation or donor type does not predict recurrence. Kidney ANCA-associated vasculitis generally responds well to high-dose prednisolone and cyclophosphamide (231–233). Other treatment modalities that have been tried include MMF, plasmapheresis with or without intravenous immunoglobulin and rituximab (234–240).

Histological evidence of anti-GBM disease can be found in biopsies in 15–50% of cases. Clinical recurrence is rare and consisted of isolated case reports only (201,241). Graft failure due to recurrence is rare (110). The incidence of recurrence may be higher in those with circulating anti-GBM antibody at the time of transplantation. Treatment of clinically active anti-GBM disease may include pulse steroids, cyclophosphamide and plasma exchange, particularly if there is potentially life-threatening pulmonary involvement (241).

Primary hyperoxaluria

Primary hyperoxaluria is caused by deficiency of hepatic peroxisomal alanine:glyoxylate aminotransferase, leading to increased synthesis and urinary excretion of oxalate, recurrent calcium oxalate urolithiasis, irreversible nephrocalcinosis and eventually CKD. In CKD, insoluble oxalates accumulate throughout the body, especially in bone and arteries. Because the enzyme defect in primary hyperoxaluria is not corrected by isolated kidney transplantation, oxalate overproduction persists, leading to recurrence of calcium oxalate deposits in over 90% of transplanted kidneys, and eventually leading to graft loss (242), unless the enzyme is replaced through a simultaneous liver transplant (243). The total body oxalate burden is very high in CKD stage 5 patients, and the urinary oxalate excretion increases greatly as soon as graft function is established. Plasma and urine oxalate levels may remain high for some period of time even in patients undergoing simultaneous kidney and liver transplantation. High urinary oxalate concentration promotes precipitation of calcium oxalate crystals first in the distal tubules, leading to graft dysfunction. This secondarily results in deposition in the parenchyma of the graft, leading to allograft failure. This risk is obviously increased further in those with primary nonfunction of the graft. Transplant protocols designed to minimize complications of recurrent disease include early posttransplant urinary dilution through aggressive fluid administration, and early and frequent dialysis in those with DGF.

Although isolated kidney transplantation is not recommended in primary hyperoxaluria, it is sometimes carried out in developing countries where liver transplantation is not available. Primary hyperoxaluria recurs invariably in those who receive kidney transplant alone and leads to graft loss. Patients with the Gly170Arg mutation are pyridoxine-sensitive, and should be given high-dose pyridoxine if they receive kidney transplant alone (244).

The disease is sometimes diagnosed for the first time after kidney transplantation when oxalate deposits are detected on biopsy in patients with graft dysfunction. Whenever possible, these patients should be referred to specialized centers for liver transplantation. In the immediate postoperative phase, extra dialysis sessions may be necessary to control oxalate blood levels until the liver is completely working (245).

Specific measures designed to increase oxalate excretion and reduce production help in minimization of recurrence, and should be in place for all patients during the first months or years after kidney or combined liver–kidney transplantation (246). These include maintenance of urine output >3.0–3.5 L/day, and the use of alkaline citrate, neutral phosphate and magnesium oxide. Severe dietary oxalate restriction is of limited benefit (247), but intake of nutrients extremely rich in oxalate and ascorbic acid, a precursor of oxalate, should be discouraged. Pharmacological doses of pyridoxine may reduce hyperoxaluria in some patients, especially in those with a Gly170Arg mutation (244). Pyridoxine responsiveness can be assessed by observation of >30% reduction in urinary oxalate excretion to 10 mg/kg/day dose of pyridoxine (248) in patient's sibs with less severe kidney disease if it was not done at the predialysis stage. Urinary alkalinization with citrate reduces the risk of urinary calcium oxalate supersaturation by forming a soluble complex with calcium, which reduces the likelihood of binding and precipitation with other substances, such as oxalate (249). The dosage is 0.1–0.15 g/kg body weight of a sodium or sodium/potassium citrate preparation. The adequacy of therapy and patient compliance can be verified by measuring urinary pH and citrate excretion. Orthophosphate (20–60 mg/day), along with pyridoxine, has also been shown to reduce urinary calcium oxalate crystallization (250).

Fabry disease

Fabry disease is a rare, X-linked inherited disease characterized by a deficiency of alpha-galactosidase A (alpha-Gal-A), resulting in progressive systemic accumulation of glycosphingolipids. Transplantation is the treatment of choice for most patients with CKD stage 5 due to Fabry disease (251). Although patients with Fabry disease may have histological recurrence of the disease in the allograft, how often recurrence causes graft failure is not clear. In a recent US Organ Procurement and Transplantation Network registry study, 197 KTRs with Fabry disease had 74% 5-year graft survival, compared to 64% in KTRs with other kidney diseases (252). Two formulations of recombinant human alpha-Gal A are currently available: agalsidase alpha (Replagal, Transkaryotic Therapies, Cambridge, MA) and agalsidase (Fabrazyme, Genzyme, Cambridge, MA). In non-KTRs, treatment with recombinant human alpha-Gal A has been shown to reduce the rate of decline in kidney function. However, it is unclear whether treatment improves graft survival, or reduces other complications of Fabry disease in KTRs. Treatment appears to be safe in KTRs (253,254); however it is very expensive, and whether it is cost-effective for improving KTR outcomes is not known.

Chapter 11: Preventing, Detecting, and Treating Nonadherence

• 11.1: 
  Consider providing all KTRs and family members with education, prevention, and treatment measures to minimize nonadherence to immunosuppressive medications. (Not Graded)
• 11.2: 
  Consider providing KTRs at increased risk for nonadherence with increased levels of screening for nonadherence. (Not Graded)

KTRs, kidney transplant recipients.

Background

Adherence is ‘the extent to which the patient's behavior matches the agreed-upon prescriber's recommendations’ (255). At a recent consensus conference, this definition was modified to take into account the threshold of the effect of nonadherence on the therapeutic outcome. We have adopted this definition of nonadherence as ‘deviation from the prescribed medication regimen sufficient to adversely influence the regimen's intended effect’ (255). Nonadherence encompasses primary (at initiation) and secondary (subsequent) nonadherence, partial and/or total nonadherence, as well as the timing of medication use (256–260).

Rationale

• • 
  Nonadherence is associated with a high risk of acute rejection and allograft loss.
• • 
  Nonadherence may occur early and/or late after transplantation.
• • 
  The transition from pediatric to adult nephrology care may be a time when nonadherence is particularly common.
• • 
  Measures can be taken to reduce nonadherence and thereby improve outcomes.

Nonadherence is common in the first months after kidney transplantation and increases by duration of follow-up. The level of adherence affects clinical outcomes, and is associated with early and late allograft rejection, which reduces graft function and graft survival (261–263). Graft loss is sevenfold more likely in nonadherent compared to adherent individuals (264). In another study, nonadherence (missed appointments, fluctuating drug concentration) accounted for over a half of kidney allograft loss (265).

Nonadherence is multidimensional (255), although we have focused primarily on adherence with immunosuppressive medication use. Additional areas of nonadherence include prescribed diet; exercise; tobacco, alcohol and drug use; self-monitoring of vital signs, for example blood pressure, body weight and clinical appointments.

Satisfactory adherence to medication use is achieved when the gaps between dosing history and the prescribed regimen have no effect on therapeutic outcome. This pharmacoadherence definition emphasizes therapeutic outcome in contrast to specific medication intake or drug level. Measurement of outcome and drug levels is commonly used in the transplant population. Measurable parameters of pharmacoadherence are acceptance (whether the patient accepts the recommended treatment), execution (how well the patient executes the recommended regimen), and discontinuation (when the patient stops taking the medication) (264,266). Measurement of adherence can be by direct observation that medication was consumed, indirect measures that medication had been consumed or self-reporting (Table 9). Indirect measures include serum drug levels, biological markers, electronic monitoring, pill count and refill/prescription records. Since there is no perfect measure of adherence, consideration should be given to use more than one approach to measure adherence (267–270).

Table 9.  Assessment of medication adherence (255,260)

Modified with permission (260).

Self-reporting medication use by patient

Collateral reporting of medication use by relatives, friends or caretakers

Patient diaries

Questionnaires

Laboratory tests (drug and metabolite levels)

Medical record review, outcomes

Prescription refills

Monitored pill counts

Electronic monitoring devices

In organ transplant recipients, the average nonadherence rate was highest for diet (25 cases per 100 people per year), immunosuppressive medication (22.6 cases per 100 people per year), monitoring vital signs (20.9 cases per 100 people per year) and exercise (19.1 cases per 100 people per year) (264). Among KTRs, nonadherence with immunosuppressive medications was highest (35.6 cases per 100 people per year). Nonadherence to long-term medication is as high as 50% in developed countries and even higher rates have been reported in developing countries (264). Meta-analysis showed that the odds of having a good outcome is 2.9 times higher if the patient is adherent (271).

Risk factors for nonadherence include long duration of treatment (with decline in rates of adherence over time), poor communication and lack of social support (Table 10). Risk factors for nonadherence can be categorized into four interrelated areas: patient/environment, caregiver, disease and medication. The patient/environment is central and interrelates with the other three categories. The primary patient–medication factors are side effects, regimen complexity, costs and poor access. Negative beliefs in medication and lack of medication knowledge have a moderate impact. Patient–caregiver factors include poor communication and poor aftercare/discharge planning (272–274). Patient–disease factors are primarily poor disease knowledge and insights, disease duration and comorbid psychiatric disease. A meta-analysis of 164 studies in the nonpsychiatric literature reported risk factors for adherence, including: age (adolescents less adherent), sex (girls more adherent than boys among pediatric patients), education level (positively associated with adherence in chronic disease) and socioeconomic status (positively correlated with adherence in adults) (275–277).

Table 10.  Risk factors for medication nonadherence (255,260)

Modified with permission (260).

Nonadherence behavior prior to transplantation

Psychiatric illness

Personality disorders

Poor social support

Substance abuse and other high-risk behavior

Adolescence

High education level

Time since transplantation (higher earlier)

Lack of adequate follow-up with transplant specialists

Inadequate pretransplant education

Multiple adverse effects from medications

Complex medication regimens

A team approach consisting of education, monitoring, recognition and intervention is essential to secure the benefit of transplantation. A combination of educational, behavioral and social support interventions provides the best results (Table 11) (271,278). Simplified drug regimens, pillboxes to organize medications, individualized instructions (particularly for travelers and night-shift workers), combining medication administration with daily routine activities and electronic devices can contribute to improve adherence.

Table 11.  A summary of interventions aimed at improving medication adherence

Modified with permission (260).

Education and medical intervention

Ensure that patients know their medications by name, dosage and reason for prescription; reinforce these points during every clinic visit.

Inform patients about the adverse effects of drugs.

Provide written instructions for each change in medication dose or frequency.

Reduce the number and frequency of medications. Where possible, medications should be given either once or, at most, twice daily.

Ensure the patients understand that they need to continue taking immunosuppressive agents even if the transplanted organ is functioning well.

Teach patients that chronic rejection is insidious in onset, hard to diagnose in its early stages and often not reversible once established.

Attempt to treat adverse effects by means other than dose reduction.

Inquire about problems during every clinic visit, and address specific patient concerns.

Monitor compliance with laboratory work, clinic visit and prescription refills.

Behavioral and psychosocial approaches

Provide positive support to encourage adherent behaviors during preparation for transplant.

Encourage patient to demonstrate a track record of medication adherence and knowledge.

Encourage individual team members to develop rapport with patient.

Identify and involve a backup support system (family or friends).

Treat depression, anxiety or other psychological issues.

Elicit a personal promise of adherence (e.g. a written contract).

Use a nonjudgmental approach to the discussion of adherence.

Address social problems such as insurance changes or difficulties at school or work.

Tailor interventions for nonadherence to its root cause.

Integrate taking medication into the daily routine.

Consider reminders such as digital alarms or alerts.

Provide ongoing education, discussion and easily accessible counseling.

Simply forgetting to take their pills is one of the most common reasons that patients give for missing doses of their medication (268). Patients should be counseled about various possibilities to integrate their medication administration into their daily routine. Pillboxes may be helpful for complex regimens consisting of multiple drugs with multiple daily dose-administration schedules. Electronic compliance devices, including alarms, are also available for improving medication adherence. The disease-management assistance system is a device that delivers a programmed voice message reminder at set times and has been applied in patients on antiretroviral therapy (279). Finally, an online pager or mobile phone system may improve adherence to medication regimens (280). However, except for the observation method, which can be onerous, all measures have significant disadvantages, primarily related to their lack of accuracy. Because there is no perfect measure of nonadherence, consideration should be given to use more than one approach to measure adherence. The overall approach to measure adherence requires individualization.

The number of prescribed medications and the dosing frequency has an effect on adherence rates (280,281). When a regimen is extremely complex, forgetfulness becomes a contributing factor to nonadherence (282). The complexity of a medication regimen is inversely proportional to the rate of adherence, with an increasing number of prescribed medications favoring nonadherence (283). Medications requiring twice-daily administration have resulted in greater adherence than those administered more than twice daily (284). The simplification of therapy strategies includes immunosuppressive as well as nonimmunosuppressive medications (e.g. antihypertensives). In addition, steroid- or CNI-sparing protocols should be considered for the benefit of reduction of number of drugs, and reduction of adverse events. Involving a clinical pharmacist may be helpful to provide comprehensive patient education regarding benefits and adherence effects of their medications. A significantly greater proportion of patients were adherent with their immunosuppressive medications at 1 year after transplant when a pharmacist was involved (284,285).

Behavioral change strategies have been applied in the clinical setting. Behavior modifications have been incorporated in six adherence-improvement RCTs in KTRs (286,287). The methods included behavioral contacting, education, skills training, feedback and reinforcement. These data indicated that such behavioral intervention is a very individualized process and adherence motivation needs to be patient-specific and updated continuously. Using the medication event-monitoring system to monitor monthly azathioprine adherence during a 6-month period in KTRs demonstrated a significant correlation with adherence and rejection-free survival in the first 6 months after transplantation (288).

Research Recommendations

• • 
  Additional prospective cohort studies are needed to establish the best measures of adherence and the association between adherence and outcomes.
• • 
  RCTs are needed to test interventions to improve adherence in KTRs.

Chapter 12: Vaccination

• 12.1: 
  We recommend giving all KTRs approved, inactivated vaccines, according to recommended schedules for the general population, except for HBV vaccination. (1D)

  • 12.1.1: 
    We suggest HBV vaccination (ideally prior to transplantation) and HBsAb titers 6–12 weeks after completing the vaccination series. (2D)

    • 12.1.1.1: 
      We suggest annual HBsAb titers. (2D)
    • 12.1.1.2: 
      We suggest revaccination if the antibody titer falls below 10 mIU/mL. (2D)
• 12.2: 
  We suggest avoiding live vaccines in KTRs. (2C)
• 12.3: 
  We suggest avoiding vaccinations, except influenza vaccination, in the first 6 months following kidney transplantation. (2C)

  • 12.3.1: 
    We suggest resuming immunizations once patients are receiving minimal maintenance doses of immunosuppressive medications. (2C)
  • 12.3.2: 
    We recommend giving all KTRs, who are at least 1-month post-transplant, influenza vaccination prior to the onset of the annual influenza season, regardless of status of immunosuppression. (1C)
• 12.4: 
  We suggest giving the following vaccines to KTRs who, due to age, direct exposure, residence or travel to endemic areas, or other epidemiological risk factors are at increased risk for the specific diseases:

  • • 
    rabies, (2D)
  • • 
    tick-borne meningoencephalitis, (2D)
  • • 
    Japanese B encephalitis—inactivated, (2D)
  • • 
    Meningococcus, (2D)
  • • 
    Pneumococcus, (2D)
  • • 
    Salmonella typhi—inactivated. (2D)
  • 12.4.1: 
    Consult an infectious disease specialist, a travel clinic or public health official for guidance on whether specific cases warrant these vaccinations. (Not Graded)

KTRs, kidney transplant recipients; HBsAb, antibody to hepatitis B surface antigen; HBV, hepatitis B virus.

Background

Recommended vaccinations are those approved and suggested by local and national health authorities for their constituent populations. These may vary by country of origin and geographic location. The efficacy of hepatitis B vaccination is determined by the prevention of hepatitis B infection, which is indirectly measured by the development of antibody to hepatitis B surface antigen (HBsAb) titers >10 mIU/mL. Individuals who are at increased risk include those with direct exposure, or residence in or travel to an endemic geographic area. In the case of meningococcal infection, patients who have undergone splenectomy are at increased risk.

Rationale

• • 
  The harm of different infections, and thereby the potential benefits of vaccinations, vary by geographic region.
• • 
  Little or no harm has been described with the use of licensed, inactivated vaccines in KTRs.
• • 
  Most vaccines produce an antibody response, albeit diminished, in immunocompromised individuals, including KTRs.
• • 
  The potential benefits outweigh the harm of immunization with inactivated vaccines in KTRs.
• • 
  Serious infection can result from live vaccines in immunocompromised patients, including KTRs.
• • 
  In the absence of adequate safety data to the contrary, it should be assumed that the harm of live vaccines outweigh their benefits in KTRs.
• • 
  Vaccinations are most likely to be effective when immunosuppression is lowest, when KTRs are receiving the lowest possible doses of immunosuppressive medication.
• • 
  Influenza vaccination needs to be provided on an annual basis in advance of the onset of the annual influenza season. Even while KTRs are receiving high levels of immunosuppression, the benefits of timely vaccination outweigh the risks of delaying vaccination.
• • 
  Some KTRs are at increased risk to develop disease attributable to one or more (rare) pathogens based upon direct exposure from residence in, or travel to, endemic areas. Although limited efficacy data are available for these inactivated vaccines to rare pathogens, potential benefits likely outweigh harm.

Inactivated vaccines

The American Society for Transplantation's Guidelines for the Prevention and Management of Infectious Complications of Solid Organ Transplantation provides guidance on immunizations relevant to their patient populations (289). While these recommendations may be appropriate for North America, they may not apply to KTRs worldwide.

Although only a limited number of studies evaluating the safety and efficacy of inactivated vaccines have been performed in solid-organ transplant recipients in general, and in KTRs in particular, available evidence suggests that inactivated vaccines are safe. There is no evidence that vaccinations lead to an increased risk of rejection.

Unfortunately, data on the efficacy of individual inactivated vaccines are limited. In general, existing data suggest that the response to vaccination in KTRs is diminished compared to immunization prior to transplantation. Accordingly, the optimal timing for immunizing KTRs is prior to transplantation. However, this is not always possible and, in some cases, repeated vaccinations after transplantation are necessary. A number of studies have been performed in organ transplant recipients that demonstrate immunogenicity of several inactivated vaccines after solid-organ transplantation. Influenza vaccination is among the most thoroughly evaluated in organ transplant recipients. Although response to influenza vaccination may vary among KTRs and from year to year, 30–100% of immunized KTRs will achieve protective hemagglutination-inhibiting serum antibody titers. Of note, the efficacy of influenza vaccination appears to be superior in pediatric compared to adult KTRs (290). Data are also available supporting the use of the 23-valent polysaccharide pneumococcal vaccine for KTRs >2 years of age. In contrast, hepatitis B vaccine has significantly diminished immunogenicity in organ transplant recipients compared to organ transplant candidates (291). Specific data regarding the immunogenicity of most of the remaining inactivated vaccinations are not available for solid-organ transplant recipients. Although data are lacking, most experts agree that the benefits outweigh the risks of immunization with inactivated vaccines (289).

There are sufficient data in KTRs indicating that the risk of vaccination with inactivated vaccines is minimal. The risk of infection, on the other hand, is higher in KTRs than in the general population. Therefore, vaccination with inactivated vaccines is warranted (Table 12).

Table 12.  Recommended vaccines after kidney transplantation

*For travel, occupational or other specific risk, and edemic regions. Consider providing booster polysaccharide pneumococcal vaccination every 3–5 years.

Diphtheria—pertussis–tetanus

Haemophilus influenza B

Hepatitis A*

Hepatitis B

Pneumovax

Inactivated polio

Influenza types A and B (administer annually)

Meningococcus: administer if recipient is in high risk

Typhoid Vi

Live vaccines

The currently licensed live vaccines use either attenuated viral strains that have been manipulated to reduce their virulence while attempting to maintain their immunogenicity, or, as in the case of Bacillus Calmette-Guérin (BCG), substitute a related bacterium that is thought to be less pathogenic, but still able to provide cross-reacting immunity to the target pathogen. While data are limited, significant concern exists for the use of live vaccines in immunocompromised patients. To date, only a limited number of studies have evaluated the use of live viral vaccines in organ transplant recipients (292). The high incidence of infections in KTRs is ample cause for concern that live vaccinations may cause infection in KTRs. While limited published experience is available describing the use of some live viral vaccines in organ transplant recipients (292), the limited number and small sample sizes included in these studies raise concerns about both the safety and efficacy of these vaccines in KTRs. Accordingly, most experts agree that, in general, the risks outweigh the potential benefits of using live vaccines in KTRs (293).

A number of live vaccinations licensed for use in the general population are contraindicated in KTRs (Table 13).

Table 13.  Contraindicated vaccinations after transplantation

BCG, Bacillus Calmette-Guérin.

Varicella zoster

BCG

Smallpox

Intranasal influenza

Live oral typhoid Ty21a and other newer vaccines

Measles (except during an outbreak)

Mumps

Rubella

Oral polio

Live Japanese B encephalitis vaccine

Yellow fever

Vaccination timing

The reduced antibody response to different vaccines in KTRs is most likely due to immunosuppressive medication. Although there are no RCTs, it is reasonable to assume that giving vaccines when the amount of immunosuppressive medications patients are receiving is lowest is most likely to maximize the response to the vaccine (289)

Immunosuppressive medication amounts are usually highest in the first few months after transplantation, when the risk of acute rejection is also the greatest. Some time during the first 6–12 months, the amount of immunosuppressive medication is generally reduced to the lowest maintenance levels, if there is no acute rejection, and this is likely to be the best time for vaccination. This time of minimal maintenance immunosuppressive medication, and optimal time for vaccination, may be different in patients treated for acute rejection.

Influenza infection is a potentially important cause of morbidity and mortality in KTRs. The use of influenza vaccination has been demonstrated to be safe and generally effective in organ transplant recipients, including KTRs (294,295). In particular, it is worth noting that there is no proven association between the use of influenza vaccination in organ transplant recipients and the development of rejection. Accordingly, annual use of influenza vaccination is recommended for both KTRs and their household contacts. Because acquisition of influenza will occur during annual seasonal epidemics, it may not be possible to delay giving this vaccine until the patient is out far enough from transplant or on low levels of immunosuppression. Given that this is an inactivated viral vaccine, the major consequence of using this too early is that the immunization will not work. Given the potential benefit of providing the vaccine, it is recommended to give this vaccine prior to the onset of the annual influenza season, as long as the recipient is at least 1-month posttransplant. This timing is chosen as the vaccine is least likely to work during the first month after transplant, especially if the KTR has received induction therapy.

Hepatitis B revaccination

The need for hepatitis B vaccination booster is controversial and the practice varies from country to country. Patients with impaired immune function tend to have lower peak HBsAb levels compared to immunocompetent individuals. There are few data on durability of immunologic memory in immunocompromised hosts. However, there have been reports of clinically significant infection due to hepatitis B virus (HBV) in previously immunized dialysis patients in whom production of HBsAb was no longer measurable (296).

Serial measurements of HBsAb levels to inform the use of a booster dose of hepatitis B vaccine has been recommended for dialysis patients by the US Advisory Committee on Immunization Practices (296). In addition, the European Consensus Group on Hepatitis B immunity has expanded this recommendation to include patients with impaired immune function (297). Immunological memory wanes faster in immunocompromised renal transplant recipients. A level above 10 mIU/mL is generally taken to be protective, but transplant recipients with titers less than 100 mIU/mL tend to lose them rapidly. The potential for low anti-HBs levels to mask significant infection (indicated by hepatitis B surface antigen (HBsAg)) and the rapid decline led a European Consensus Group to suggest booster vaccination at titers below 100 mIU/mL. Although there is no clear evidence to support this recommendation, given the relative risk–benefit ratio of hepatitis B vaccine, it seems prudent to assess annually the need for a booster dose of this immunization.

Additional vaccines

Kidney transplant recipients may be at increased risk for vaccine-preventable pathogens through residence or travel to endemic areas, or due to inadvertent exposure. Recommendations for individuals traveling to certain geographic locations frequently include receipt of one or more immunizations against these pathogens. These recommendations would logically apply to KTRs, as long as the recommended vaccinations are inactivated, for example salmonella typhi Vi polysaccharide vaccine, or meningococcal vaccine. Consultation with an infectious disease specialist, travel clinic or public health official is recommended to clarify appropriate use of vaccinations for scenarios where travel or exposure may warrant use of these additional vaccinations.

Although efficacy data may not be available in KTRs, inactivated vaccines are generally safe. In contrast, some immunizations typically recommended for travelers are available only as live-attenuated vaccines. The use of these vaccines cannot be recommended, as neither safety nor efficacy data are available in this patient population.

Research Recommendations

Studies are needed to determine:

• • 
  the optimal timing of immunization in KTRs;
• • 
  the durability of immunologic response in KTRs vaccinated before and after transplantation.

Chapter 13: Viral Diseases

• 13.1: 
  BK POLYOMA VIRUS

  • 13.1.1: 
    We suggest screening all KTRs for BKV with quantitative plasma NAT (2C) at least:

    • • 
      monthly for the first 3–6 months after transplantation (2D);
    • • 
      then every 3 months until the end of the first post-transplant year (2D);
    • • 
      whenever there is an unexplained rise in serum creatinine (2D); and
    • • 
      after treatment for acute rejection. (2D)
  • 13.1.2: 
    We suggest reducing immunosuppressive medications when BKV plasma NAT is persistently greater than 10 000 copies/mL (10⁷ copies/L). (2D)

BKV, BK polyoma virus; KTRs, kidney transplant recipients; NAT, nucleic acid testing.

Background

BK polyoma virus (BKV) is a member of the polyoma family of viruses. BKV can cause nephropathy, which is diagnosed by kidney biopsy. Reduction of immunosuppression is defined as a decrease in the amount and intensity of immunosuppressive medication. Nucleic acid testing (NAT) is defined as one or more molecular methods used to identify the presence of DNA or RNA (e.g. polymerase chain reaction).

Rationale

• • 
  The use of NAT to detect BKV in plasma provides a sensitive method for identifying BKV infection and determining KTRs who are at increased risk for BKV nephropathy.
• • 
  Early identification of BKV infection may allow measures to be taken that may prevent BKV nephropathy.
• • 
  When NAT is not available, microscopic evaluation of urine for the presence of decoy cells is an acceptable, albeit nonspecific, alternative screening method for BKV disease and the risk for BKV nephropathy.
• • 
  Fifty percent of patients who develop BK viremia do so by 3 months after kidney transplantation.
• • 
  Ninety-five percent of BKV nephropathy occurs in the first 2 years after kidney transplantation.
• • 
  BKV plasma NAT >10 000 copies/mL (10⁷ copies/L) has a high positive predictive value for BKV nephropathy.
• • 
  Reduction of immunosuppressive medication may result in reduced BKV load and decreased risk of BKV nephropathy.
• • 
  Histologic evidence of BKV nephropathy may be present in the absence of elevated serum creatinine.
• • 
  Reduction in maintenance immunosuppressive medication is the best treatment for BKV nephropathy.

Whether to screen KTRs with NAT of plasma or urine has been controversial. A negative urine NAT for BKV has almost a 100% negative predictive value (298). By testing urine, one can avoid performing BKV testing of blood on those patients with negative urine studies. Based on this, some experts recommend screening of urine as the definitive site for BKV surveillance (298). However, the presence of a positive NAT for BKV in urine, in the absence of an elevated BKV load in the plasma, is not associated with an increased risk for BKV disease (298). Hence, the use of urine screening requires performance of NAT on the blood of those patients whose level of BK viruria exceeds established thresholds. This requires patients to return to the clinic for the additional test. Accordingly, it is suggested that NAT be performed on plasma, and not the urine of KTRs.

When NAT is not available, microscopic evaluation of the urine for the presence of decoy cells is an acceptable, albeit nonspecific, alternative screening method for BKV disease and the risk for BKV nephropathy. A negative screening test rules out BKV nephropathy in most cases (high negative predictive value). However, a positive screening test has a very low positive predictive value for BKV nephropathy (298,299). Thus, many patients with urine decoy cells will not develop BKV nephropathy. It may be inappropriate to change therapy in such patients based on the presence of urine decoy cells alone.

Emerging data suggest that BKV nephropathy can be prevented if immunosuppressive medications are reduced in patients with BKV detected by a high viral load in plasma (determined by NAT) (300).

Timing of BKV NAT

The presence of BKV can be identified prior to the onset of clinical symptoms at a time when only subclinical infection is present, or in association with clinically apparent BKV nephropathy. Evidence to date suggests that the presence of BK viremia precedes BKV nephropathy by a median of 8 weeks. Approximately, 50% of patients who will develop BK viremia will do so by 3 months after transplant (298).

Most BKV nephropathy occurs in the first 2 years after transplant with only 5% of cases occurring between 2 and 5 years after transplant (298). Accordingly, the timing and frequency of testing in recommended screening algorithms should reflect these data and balance the cost of screening with the potential to prevent BKV nephropathy. The proposed screening algorithm is most intense early after kidney transplantation, with decreasing frequency as patients are out longer from the transplant. Although we have not recommended screening beyond the first year after transplant, an international consensus conference suggested continued annual screening for patients between 2 and 5 years after kidney transplantation (298). Centers with higher frequency of BKV might follow this approach. Screening for the presence of BKV should also be performed for patients with unexplained rises in serum creatinine, as this may be attributable to BKV nephropathy. Finally, screening should be considered for those patients who have undergone a major increase in immunosuppressive medication, as they may be at risk of developing BKV nephropathy.

Rising BKV load

There is increased risk of BKV nephropathy associated with a rising BKV load in plasma (298,299). Although plasma NAT assays for BKV lack standardization, a threshold plasma BKV level of >10 000 copies/mL (10⁷ copies/L) is associated with a 93% specificity for the presence of BKV nephropathy. In the absence of evidence of clinical disease, KTRs with BKV levels in excess of this threshold are considered to be at risk of progression to BKV nephropathy (298,299). Histologic evidence of early BKV nephropathy may be present prior to detection of elevated serum creatinine (298).

The risk of BKV nephropathy appears to be correlated with the intensity of immunosuppression, and reduction of immunosuppression can result in a decrease in BKV load and a concomitant reduction of risk of development of BKV nephropathy (301). A RCT reported that withdrawal of the antimetabolite resulted in clearance of viremia without progression to BKV nephropathy (300). Although some would use antiviral therapy (including cidofovir, leflunomide and/or ciprofloxacin) as treatment, to date there are no definitive data confirming the effectiveness of these agents for either treatment or prevention of BKV nephropathy (298,299).

Some centers may choose different treatment strategies for patients with elevated BKV loads in the absence of any histologic changes, compared to patients with findings of BKV nephropathy in the absence of serum creatinine elevation. The international consensus group recommended performance of kidney biopsy for these patients (298). When a kidney biopsy is obtained, it should be evaluated for the presence of BKV using the cross-reacting antibody for simian virus 40. However, other experts have not recommended the performance of a kidney biopsy for asymptomatic patients with an elevated BKV load (300).

Treating biopsy-proven BKV nephropathy

The treatment of BKV nephropathy is unsatisfactory. Although there are some centers that would use antiviral therapy (including cidofovir, leflunomide and/or ciprofloxacin) as treatment, to date there are no definitive data confirming their effectiveness. However, reduction of immunosuppression does appear to have some impact on BKV nephropathy, though variable rates of graft loss attributable to BKV nephropathy have been reported even when reduction of immunosuppression has been employed (Table 14). A common practice of immunosuppressive dose reduction is withdrawal of antimetabolite (azathioprine or MMF) and reduction in CNI dosage by 50%. An algorithm for the treatment of BKV nephropathy through modification of baseline immunosuppression has been proposed (298). Switching from the antimetabolite MMF or EC-MPS to leflunomide (an immunosuppressive agent with antiviral activity) has been associated with declining BKV load in blood and improving histology (302), although convincing evidence of the efficacy of this, or other antiviral agents, is lacking.

Table 14.  Treatment of BKV nephropathy by modification of maintenance immunosuppression

Switching	Decreasing	Discontinuing
BKV, BK polyoma virus; CsA, cyclosporine A; MMF, mycophenolate mofetil. B-III, ‘moderate evidence to support a recommendation for use’ based on ‘evidence from opinions of respected authorities, based on clinical experience, descriptive studies or reports of expert committees.’ Likely equivalent to a 2D recommendation. C-III, ‘poor evidence to support a recommendation’ based on ‘evidence from opinions of respected authorities, based on clinical experience, descriptive studies or reports of expert committees.’ Likely equivalent to a 2D recommendation. Reprinted with permission (298).
Tacrolimus→CsA (trough levels	Tacrolimus (trough levels	Tacrolimus or MMF (maintain
100–150 ng/mL) (B-III)	< 6 ng/mL) (B-III)	or switch to dual-drug therapy):
MMF→azathioprine (dosing ≤100 mg/day) (B-III)	MMF dosing ≤1 g/day (B-III)	CsA/prednisone (B-III)
Tacrolimus→sirolimus (trough levels <6 ng/mL)	CsA (trough levels 100–150 ng/mL)	Tacrolimus/prednisone (B-III)
(C-III)	(B-III)
MMF→sirolimus (trough levels <6 ng/mL) (C-III)		Sirolimus/prednisone (C-III)
MMF→leflunomide (C-III)		MMF/prednisone (C-III)

Research Recommendations

Studies are needed to determine:

• • 
  the most cost-effective strategies for screening for BKV in different populations;
• • 
  the efficacy of altering immunosuppressive medication regimens and of antiviral agents in the prevention and treatment of BKV nephropathy.

• 13.2: 
  CYTOMEGALOVIRUS

  • 13.2.1: 
    CMV prophylaxis: We recommend that KTRs (except when donor and recipient both have negative CMV serologies) receive chemoprophylaxis for CMV infection with oral ganciclovir or valganciclovir for at least 3 months after transplantation, (1B) and for 6 weeks after treatment with a T-cell–depleting antibody. (1C)
  • 13.2.2: 
    In patients with CMV disease, we suggest weekly monitoring of CMV by NAT or pp65 antigenemia. (2D)
  • 13.2.3: 
    CMV treatment:

    • 13.2.3.1: 
      We recommend that all patients with serious (including most patients with tissue invasive) CMV disease be treated with intravenous ganciclovir. (1D)
    • 13.2.3.2: 
      We recommend that CMV disease in adult KTRs that is not serious (e.g. episodes that are associated with mild clinical symptoms) be treated with either intravenous ganciclovir or oral valganciclovir. (1D)
    • 13.2.3.3: 
      We recommend that all CMV disease in pediatric KTRs be treated with intravenous ganciclovir. (1D)
    • 13.2.3.4: 
      We suggest continuing therapy until CMV is no longer detectable by plasma NAT or pp65 antigenemia. (2D)
  • 13.2.4: 
    We suggest reducing immunosuppressive medication in life-threatening CMV disease, and CMV disease that persists in the face of treatment, until CMV disease has resolved. (2D)

    • 13.2.4.1: 
      We suggest monitoring graft function closely during CMV disease. (2D)

CMV, cytomegalovirus; KTRs, kidney transplant recipients; NAT, nucleic acid testing.

Background

Cytomegalovirus disease is defined by the presence of clinical signs and symptoms attributable to CMV infection, and the presence of CMV in plasma by NAT or pp65 antigenemia. CMV disease may manifest as a nonspecific febrile syndrome (e.g. fever, leukopenia and atypical lymphocytosis) or tissue-invasive infections (e.g. hepatitis, pneumonitis and enteritis). Tissue-invasive CMV disease is defined as CMV disease and CMV detected in tissue with histology, NAT or culture. Serologically, negative CMV is defined by the absence of CMV immunoglobulin G (IgG) and immunoglobulin M. Serologically positive for CMV is defined as being CMV IgG-positive. Interpretation of CMV serologies may be confounded by the presence of passive antibody that may have been acquired from a blood or body-fluid contamination. Chemoprophylaxis is defined as the use of an antimicrobial agent in the absence of evidence of active infection, to prevent the acquisition of infection and the development of disease.

Rationale

• • 
  CMV disease is an important cause of morbidity and mortality.
• • 
  There are strategies for preventing CMV infection and disease that result in marked improvements in outcomes.
• • 
  Risk for CMV after transplantation is strongly dependent on donor (D) and recipient (R) serology, with patients who are D+/R−, D+/R+ or D−/R+ at risk for developing CMV infection and disease, and D+/R− at highest risk for severe CMV disease.
• • 
  The incidence of CMV disease in D−/R− is <5%.
• • 
  Chemoprophylaxis with ganciclovir or valganciclovir for at least 3 months after transplantation reduces CMV infection and disease in high-risk patients.
• • 
  Chemoprophylaxis is associated with improved graft survival compared to preemptive antiviral therapy initiated in response to increased CMV load.
• • 
  The use of a T-cell–depleting antibody is a risk factor for CMV disease.
• • 
  Chemoprophylaxis with ganciclovir for patients receiving a T-cell–depleting antibody protects against the development of CMV disease.
• • 
  A detectable CMV load at the end of antiviral therapy is associated with an increased risk of disease recurrence.
• • 
  CMV infection is associated with acute rejection.

Preventing CMV

Cytomegalovirus is a frequent and important cause of clinical disease in KTRs. In the absence of antiviral prophylaxis, symptomatic CMV disease can be seen in approximately 8% of KTRs (303), although older estimates placed it at 10–60% of KTRs (304). In addition to directly attributable morbidity, CMV may also have an immunomodulatory effect, and active CMV disease has been associated with infectious complications as well as acute rejection and CAI (305). Accordingly, strategies that can prevent CMV infection and disease should lead to improved outcomes following kidney transplantation.

Randomized controlled trials have demonstrated that the incidence of CMV disease can be reduced by prophylaxis and preemptive therapies in solid-organ transplant recipients (306–308). In trials of KTRs alone, there is low-quality evidence, largely due to sparse data, that prophylaxis results in less acute rejection and CMV infection, with no clear evidence of increased adverse events (see Evidence Profile and accompanying evidence in Supporting Tables 48–49 at http://www3.interscience.wiley.com/journal/118499698/toc). However, there is high-quality evidence from a large systematic review that CMV prophylaxis in solid-organ transplant recipients (307) significantly reduces all-cause mortality, CMV disease mortality, CMV disease, but not acute rejection or graft loss. In most of these trials, the majority of organ recipients received kidneys. Thus, the Work Group concluded that overall there is moderate-quality evidence to support this recommendation. Observational data suggest that D+/R− KTRs are at the highest risk of developing severe CMV disease compared to all other KTRs (306). Studies in this high-risk population have shown that antiviral chemoprophylaxis reduces the incidence of CMV disease by about 60% (306). The use of antiviral chemoprophylaxis has also been shown to reduce the incidence of CMV-associated mortality, all-cause mortality, as well as clinically important disease due to opportunistic infections (306). Chemoprophylaxis has also been shown to be effective in KTRs at moderate risk for CMV disease (e.g. CMV D+/R+, or D−/R+).

In contrast to the situation for antiviral chemoprophylaxis, the number of studies evaluating the efficacy of viral load monitoring to inform preemptive therapy in high-risk patients is limited (308). While results of these studies are encouraging, they have only demonstrated a reduction in CMV disease, and this strategy has not yet been shown to reduce CMV-related mortality (306). At the present time, the use of viral load monitoring to prompt preemptive therapy is not recommended for these high-risk KTRs (307). The basis for this concern is both a lack of data in CMV D+/R− KTRs, the implications of a failure to comply with the preemptive monitoring approach (an important potential limitation of this strategy) and the relative safety and efficacy of universal chemoprophylaxis in high-risk organ transplant recipients.

The use of CMV viral load monitoring to inform preemptive antiviral treatment with ganciclovir in patients at moderate risk for developing CMV disease has been shown to be effective (308) and has several potential advantages compared to the use of universal chemoprophylaxis. Primary among these is limiting exposure to antiviral agents only to those KTRs who have demonstrated evidence of subclinical CMV infection. Based upon this, a consensus has existed to limit this approach to patients at moderate (but not high) risk for CMV disease (305,307). However, a recently published RCT comparing oral ganciclovir prophylaxis to CMV surveillance monitoring to inform preemptive ganciclovir therapy demonstrated an advantage in long-term graft survival in those KTRs randomized to received ganciclovir chemoprophylaxis (309). Accordingly, while many experts have previously felt that both strategies (universal chemoprophylaxis or viral load monitoring to inform pre-emptive antiviral therapy) were acceptable for the prevention of CMV disease in this population (305,308), if confirmed, the newer data may provide evidence that all KTRs at risk for the development of CMV should receive chemoprophylaxis and not a preemptive therapy approach. Some experts recommend the use of viral load monitoring to inform preemptive antiviral treatment in this cohort of KTRs at moderate risk for developing CMV disease.

A number of observational studies have shown that the incidence of CMV disease is very low (<5%) in CMV seronegative recipients of CMV seronegative donors (D−/R−) (307). Although there are no cost–benefit studies in this low-risk population, the very low incidence of CMV disease makes it very unlikely that the benefits of preventive strategies outweigh their harm. The latter include adverse effects of medication and costs.

There is strong evidence linking the use of antibody treatment of rejection with increased risk of CMV infection and disease. The use of these agents results in activation of CMV from latency to active infection.

Chemoprophylaxis

A variety of potential antiviral agents have been evaluated. RCTs demonstrated that ganciclovir, valganciclovir, acyclovir and valacyclovir were each effective in the preventing CMV infection and disease (307). However, head-to-head comparisons demonstrated that ganciclovir was more effective than acyclovir in preventing both CMV infection and CMV disease. Oral valganciclovir was as effective as intravenous ganciclovir in the prevention of both CMV infection and disease. Oral and intravenous ganciclovir yielded similar results. The use of acyclovir and valacyclovir should be restricted to situations where ganciclovir/valganciclovir cannot be used.

Most recent RCTs evaluating oral antiviral agents for the prevention of CMV disease have treated patients for 3 months after transplantation (307). A recent meta-analysis did not find a difference in treatment efficacy for patients receiving less or more than 6 weeks of therapy. The impetus behind prolonged treatment is an increasing recognition of late CMV disease. A RCT evaluating 3 vs. 6 months is currently being conducted.

Three studies have evaluated prophylaxis or CMV disease in KTRs treated for acute rejection. Two studies evaluating ganciclovir in patients receiving antilymphocyte antibody therapy demonstrated a reduction in CMV disease (310). A third study evaluated the use of intravenous immunoglobulin followed by acyclovir prophylaxis in patients receiving OKT3 (311). This latter study failed to demonstrate a protective effect against CMV compared with no therapy. Accordingly, the use of intravenous ganciclovir or oral valganciclovir has been recommended for CMV prophylaxis during antilymphocyte antibody therapy (305). The use of oral ganciclovir should be avoided for patients with high-level CMV viremia (305). The use of acyclovir or famciclovir is not recommended, given the absence of data supporting the efficacy of these agents. It is also suggested that CMV serologies be repeated for patients CMV-seronegative prior to transplant, who require antibody therapy as treatment for rejection to decide their current risk status.

CMV treatment

The presence of CMV in plasma, detected by NAT or pp65 antigenemia, at the end of treatment is a major predictor of recurrent CMV disease (305). Recent evidence suggests that the use of oral valganciclovir was effective in the treatment of CMV disease (312). Although the results of this study are encouraging, the determination of what level of disease is appropriate for oral therapy in the ambulatory setting vs. treatment with intravenous ganciclovir (at least initially) remains unclear. At this point, most experts would be willing to use oral therapy to treat adult KTRs with mild CMV disease. A consensus does not exist as to which patients with tissue-invasive disease might be candidates for oral therapy. Clearly, patients with more severe disease, including those with life-threatening disease should be hospitalized and treated with intravenous ganciclovir.

It is worth noting that similar data are not available for pediatric KTRs or other children undergoing solid-organ transplantation. Accordingly, while the use of oral valganciclovir may be appropriate for some adult KTRs experiencing mild to moderate CMV disease, all pediatric KTRs should receive intravenous ganciclovir for the treatment of CMV disease. Further, concern also exists with regards to the use of oral valganciclovir in patients in whom there are questions regarding adequate absorption of this medication.

CMV viral load testing

While resolution of clinical signs and symptoms are critical in the management of CMV disease, measurement of the CMV viral load provides additional useful information. The use of viral load monitoring identifies both virologic response (guiding duration of therapy) as well as the possible presence of antiviral resistance. The presence of detectable CMV load at the end of therapy is associated with an increased rate of recurrent disease (313). The time to clearance of CMV in plasma as measured by NAT may be prolonged compared to pp65, and may be associated with an increase risk of recurrent CMV disease (314).

Immunosuppression and graft function monitoring during CMV disease

The reduction of immunosuppression used as part of the treatment of CMV disease places patients at some risk for the development of rejection. The presence of CMV infection and disease has been associated with the development of rejection independent of reduction of immunosuppression. Accordingly, careful monitoring of kidney allograft function is warranted during treatment of CMV disease to guide the use of immunosuppression.

Research Recommendations

Randomized controlled trials are needed to determine:

• • 
  the benefits and harm of CMV chemoprophylaxis vs. preemptive antiviral therapy informed by CMV viral load monitoring;
• • 
  the optimal duration of antiviral chemoprophylaxis.

• 13.3: 
  EPSTEIN-BARR VIRUS AND POST-TRANSPLANT LYMPHOPROLIFERATIVE DISEASE

  • 13.3.1: 
    We suggest monitoring high-risk (donor EBV seropositive/recipient seronegative) KTRs for EBV by NAT (2C):

    • • 
      once in the first week after transplantation (2D);
    • • 
      then at least monthly for the first 3–6 months after transplantation (2D);
    • • 
      then every 3 months until the end of the first post-transplant year (2D); and
    • • 
      additionally after treatment for acute rejection. (2D)
  • 13.3.2: 
    We suggest that EBV-seronegative patients with an increasing EBV load have immunosuppressive medication reduced. (2D)
  • 13.3.3: 
    We recommend that patients with EBV disease, including PTLD, have a reduction or cessation of immunosuppressive medication. (1C)

EBV, Epstein-Barr virus; KTRs, kidney transplant recipients; NAT, nucleic acid testing; PTLD, post-transplant lymphoproliferative disease.

Background

Epstein-Barr virus (EBV) disease is defined by signs and symptoms of active viral infection and increased EBV load. The EBV viral load is defined as the amount of viral genome that is detectable in the peripheral blood by NAT. PTLD are clinical syndromes associated with EBV and lymphoproliferation, which range from self-limited, polyclonal proliferation to malignancies containing clonal chromosomal abnormalities (315). The World Health Organization (WHO) has developed a histological classification for PTLD (323).

Rationale

• • 
  There is a 10- to 50-fold increased risk for EBV disease (including PTLD) in EBV-seronegative compared to EBV-seropositive KTRs.
• • 
  The EBV viral load measurement is sensitive, but not specific, for EBV disease and PTLD, particularly in previously seronegative KTRs.
• • 
  The EBV viral load becomes positive before the development of EBV disease.
• • 
  Early identification of primary infection and viral load monitoring allows therapeutic interventions to prevent progression to EBV disease.
• • 
  Reducing immunosuppressive medication may prevent EBV disease and PTLD.
• • 
  Reducing immunosuppressive medication is an effective treatment for many patients with EBV disease and PTLD.
• • 
  EBV viral load is detectable and elevated in many patients experiencing EBV disease, including PTLD, but can also be elevated in asymptomatic patients.
• • 
  The presence of EBV-negative PTLD has been reported, and these lesions may behave differently than EBV-positive PTLD lesions.

Primary EBV (human herpes virus 4) infection is associated with an increased incidence of PTLD in KTRs. An EBV-negative KTR from an EBV-positive donor is at increased risk for developing PTLD (316,317). A newly detectable or rising EBV load often precedes EBV disease and PTLD (318). Identification of seronegative patients with a rising EBV load offers the opportunity to preemptively intervene and potentially prevent progression to EBV disease including PTLD (319). While this has been observed most frequently in pediatric KTRs, there is no reason to assume that EBV-seronegative adult KTRs who receive a kidney from a EBV seropositive recipient are not also at increased risk of developing EBV disease, and likely to benefit from EBV load monitoring.

Primary EBV infection in EBV-seronegative organ transplant recipients occurs most frequently in the first 3–6 months following organ transplantation (320). This is most likely due to the fact that the source of the EBV infection is attributable to either the donor organ or blood products received by the patient at or near the time of transplant. Serial measurement of EBV loads in previously seronegative patients allows the identification of onset of infection (318). Continued observation of EBV loads in newly infected patients identifies those patients with rapidly rising viral loads who are likely to be at greatest risk of progressing to EBV disease. Because the most likely sources of EBV infection in KTRs are either passenger leukocytes from the donor allograft or blood products exposure (which are more likely at or near the time of transplantation), the likelihood that they will develop primary EBV infection is reduced with time after transplantation. Accordingly, EBV load monitoring should be performed most frequently during the first 3–6 months after transplant. Because the risk of developing EBV infection after this time period is diminished, but not eliminated, continued surveillance of EBV load is recommended, albeit at less frequent intervals.

EBV-seronegative patients with an increasing EBV viral load

The development of primary EBV infection after kidney transplantation is associated with a marked increased risk for the development of EBV disease and PTLD (316,317). High EBV loads have been found at the time of diagnosis of PTLD. Because the EBV load becomes positive 4–16 weeks prior to development of PTLD (318), the presence of a rising EBV load identifies patients in whom intervention may prevent PTLD.

The potential role of antiviral therapy as a preemptive response to a rising viral load is controversial. Children undergoing liver transplantation had a reduction in the risk for EBV PTLD with reduced immunosuppressive medication (tacrolimus) without concomitant use of antiviral therapy (321). In contrast, evidence is lacking for the efficacy of preemptive antiviral therapy (e.g. acyclovir, ganciclovir) in response to an elevated or rising EBV load in the absence of reduction of immunosuppression.

EBV disease diagnosis

Epstein-Barr virus disease can present with varied manifestations, including nonspecific febrile illness, gastroenteritis, hepatitis and other manifestations that may be attributable to CMV or other pathogens. Although biopsy to detect the presence of EBV infection within affected tissue is the most definitive way to confirm the diagnosis of EBV disease, histological confirmation may not be feasible for patients with some nonspecific clinical syndromes that may not localize to specific tissue (e.g. febrile syndromes). Because the EBV viral load is detectable and elevated in the vast majority of KTRs with EBV disease, including PTLD, the combination of the presence of a compatible clinical syndrome in association with a high EBV load provides a sensitive and specific approach to the diagnosis of EBV disease 322). However, it is still necessary to be cautious in considering this diagnosis, as many patients may have asymptomatic elevations of EBV load. Accordingly, such patients may be misdiagnosed as having EBV disease, if they develop intercurrent infections due to an alternative pathogen at a time that they are having an asymptomatic elevation in their EBV load. In such patients, a tissue diagnosis may be the only method of confirming the presence or absence of EBV disease.

EBV-associated PTLD

The term PTLD describes a broad category of EBV-related diseases that have distinct histological appearances (Table 15) (323). The approach to the management of PTLD can vary according to the PTLD disease classification. Furthermore, EBV-negative PTLD lesions have been reported. These lesions may behave differently then EBV-positive lesions and may warrant alternative therapeutic options. In addition, lesions with a characteristic clinical appearance on physical examination or imaging studies may be due to alternative pathogens (e.g. pulmonary nodules attributable to fungal pathogens). Because of all these concerns, it is imperative that suspected PTLD lesions be biopsied and undergo histolopathologic evaluation by a pathologist experienced with the diagnosis of PTLD (315).

Table 15.  Categories of PTLD

PTLD, posttransplant lymphoproliferative disease; T/NK, T-cell/natural killer cell. Modified with permission (323).
1: Early lesions	Reactive plasmacytic hyperplasia
	Infectious mononuclueosis-like
2: PTLD—polymorphic	Polyclonal (rare) Monoclonal
3: PTLD—monomorphic (classify according to lymphoma classification)	B-cell lymphomas
	Diffuse large B-cell lymphoma (immunoblastic, centroblastic, anaplastic)
	Burkitt/Burkitt-like lymphoma
	Plasma cell myeloma
	T-cell lymphomas
	Peripheral T-cell lymphoma, not otherwise categorized
	Other types (hepatosplenic, gamma-delta, T/NK)
4: Other types (rare)	Hodgkin's disease-like lesions (associated with methotrexate therapy)
	Plasmacytoma-like lesions

Observational studies have suggested KTRs with EBV disease are at high risk of developing PTLD (324). Observational studies have also shown that mortality from EBV-associated PTLD is over 50% (325,326). The presence of immunosuppression is major risk factor for the development of EBV disease, including PTLD, in KTRs (317,327). In most cases, the progression of clinical symptoms is a consequence of the inability to mount an adequate EBV-specific cytotoxic T-cell response, because of the immunosuppressive medications. It is therefore logical to assume that reduction of immunosuppression may result in resolution of EBV disease. As many as two thirds of patients presenting with EBV-associated PTLD will respond to reduction or withdrawal of immunosuppressive medication (315,328). This is less likely to be the case for patients presenting more than 1 year after transplantation, or with EBV-associated lymphoma. In these cases, there is an increased tendency for the lesions to behave in a truly malignant fashion. However, because some patients presenting late after transplant with biopsy evidence of lymphoma have responded to reduction of immunosuppression, this strategy may still be considered even in these patients, though expectations of efficacy will be reduced.

Epstein-Barr virus disease and PTLD are important causes of morbidity and mortality following kidney transplantation. Rates of PTLD are higher in pediatric KTRs and those patients who are EBV-seronegative prior to transplant who experience primary infection after transplant. While EBV disease and PTLD may be more common among pediatric KTRs, adult EBV-seronegative recipients of kidneys from an EBV-seropositive donor are also felt to be at increased risk for the development of these complications. Because of the complexity of this disease and its management, involvement of infectious diseases specialists, oncologists and transplant physicians in a team approach will likely maximize therapeutic outcomes.

• 13.4: 
  HERPES SIMPLEX VIRUS 1, 2 AND VARICELLA ZOSTER VIRUS

  • 13.4.1: 
    We recommend that KTRs who develop a superficial HSV 1, 2 infection be treated (1B) with an appropriate oral antiviral agent (e.g. acyclovir, valacyclovir, or famciclovir) until all lesions have resolved. (1D)
  • 13.4.2: 
    We recommend that KTRs with systemic HSV 1, 2 infection be treated (1B) with intravenous acyclovir and a reduction in immunosuppressive medication. (1D)

    • 13.4.2.1: 
      We recommend that intravenous acyclovir continue until the patient has a clinical response, (1B) then switch to an appropriate oral antiviral agent (e.g. acyclovir, valacyclovir, or famciclovir) to complete a total treatment duration of 14–21 days. (2D)
  • 13.4.3: 
    We suggest using a prophylactic antiviral agent for KTRs experiencing frequent recurrences of HSV 1,2 infection. (2D)
  • 13.4.4: 
    We recommend that primary VZV infection (chicken pox) in KTRs be treated (1C) with either intravenous or oral acyclovir or valacyclovir; and a temporary reduction in amount of immunosuppressive medication. (2D)

    • 13.4.4.1: 
      We recommend that treatment be continued at least until all lesions have scabbed. (1D)
  • 13.4.5: 
    We recommend that uncomplicated herpes zoster (shingles) be treated (1B) with oral acyclovir or valacyclovir (1B), at least until all lesions have scabbed. (1D)
  • 13.4.6: 
    We recommend that disseminated or invasive herpes zoster be treated (1B) with intravenous acyclovir and a temporary reduction in the amount of immunosuppressive medication (1C), at least until all lesions have scabbed. (1D)
  • 13.4.7: 
    We recommend that prevention of primary varicella zoster be instituted in varicella-susceptible patients after exposure to individuals with active varicella zoster infection (1D):

    • • 
      varicella zoster immunoglobulin (or intravenous immunoglobulin) within 96 hours of exposure (1D);
    • • 
      if immunoglobulin is not available or more than 96 h have passed, a 7-day course of oral acyclovir begun 7–10 days after varicella exposure. (2D)

HSV, herpes simplex virus; KTRs, kidney transplant recipients; VZV, varicella zoster virus.

Background

Superficial herpes simplex virus (HSV) infection is defined as disease limited to the skin or mucosal surfaces without evidence of dissemination to visceral organs.

Systemic HSV infection is defined by disease involving visceral organs.

Primary varicella zoster virus (VZV) infection is infection in a patient who is immunologically naive to VZV. In general, primary VZV presents as ‘chickenpox,’ which most frequently manifests as multiple crops of cutaneous lesions that evolve from macular, papular, vesicular and pustular stages. The lesions tend to erupt over the entire body and will be in different stages. Disseminated VZV can develop in immunocompromised individuals with involvement of the lungs, liver, central nervous system and other visceral organs.

Uncomplicated herpes zoster (shingles) is defined as the presence of cutaneous zoster limited to no more than three dermatomes.

Disseminated or invasive herpes zoster is defined as the presence of cutaneous zoster in more than three dermatomes, and/or evidence of organ system involvement.

The definition of a clinically significant exposure to an individual with active VZV infection varies by whether the infected individual presents with varicella (chickenpox) or zoster (shingles). Varicella may be spread to a susceptible individual by either airborne exposure or direct contact with a lesion. In contrast, an infectious exposure to someone with zoster requires direct contact with a lesion. Accordingly, a significant exposure to varicella is defined by face-to-face contact with someone with chickenpox, while a significant exposure to someone with zoster requires direct contact with a lesion. The minimum duration of airborne exposure necessary to allow transmission is not known. In general, most experts consider the minimum to be somewhere in the range of 5–60 min.

Rationale

• • 
  Superficial HSV infections are typically self-limited in immunocompetent patients, but immunosuppressive medication in KTRs increases the risk for invasive and disseminated HSV infection; treatment of superficial HSV infections with oral acyclovir or valacyclovir is safe and effective.
• • 
  Systemic HSV infections represent a potentially life-threatening complication to immunosuppressed KTRs. Intensive treatment of systemic HSV infection with intravenous acyclovir and a reduction in the amount of immunosuppressive medication is warranted to prevent progression and further dissemination of HSV.
• • 
  Primary VZV infection is potentially life-threatening to KTRs. Treatment with intravenous acyclovir is safe and effective.
• • 
  Herpes zoster infection is potentially life-threatening to KTRs. Treatment with oral acyclovir or valacyclovir is safe and effective.
• • 
  Disseminated or invasive herpes zoster is life-threatening to KTRs. Treatment with intravenous acyclovir and a temporary reduction in the amount of immunosuppressive medication is safe and effective.
• • 
  The use of varicella zoster immunoglobulin or commercial intravenous immunoglobulin products within 96 h of exposure to VZV prevents or modifies varicella in susceptible individuals.
• • 
  Oral acyclovir begun within 7–10 days after varicella exposure and continued for 7 days appears to be a reasonable alternative to immunoglobulin to prevent or modify primary varicella in susceptible individuals (329,330).

Superficial HSV infection

Serologic evidence of HSV1 and HSV2 is common in the general population. Although periodic reactivation of HSV1 and HSV2 infection occurs, these episodes tend to be self-limited in immunocompetent individuals. However, episodes of invasive or disseminated HSV may occur in KTRs receiving immunosuppressive medications, and indeed the incidence of invasive HSV is higher in KTRs than in the general population (331,332).

The highest incidence of HSV reactivation occurs early after transplantation, with the greatest risk occurring during the first month following transplantation (333). While presentation later after transplant is associated with a lower risk of dissemination, treatment of superficial infection with oral acyclovir, valacyclovir or famciclovir is still recommended, given the safety and efficacy of these medications (333). To prevent dissemination, it seems prudent to continue treatment until there are no new, active lesions.

Systemic HSV infection

In contrast to superficial HSV infection, systemic HSV infection involving the lungs, liver, central nervous system or other visceral organs represents a potentially life-threatening complication. Because systemic HSV is life-threatening, hospitalization and treatment with intravenous acyclovir is warranted (333). If possible, immunosuppressive medications should be reduced or withdrawn until the infection has resolved.

Intravenous acyclovir should be continued until there is demonstrative evidence of clinical improvement as measured by resolution of fever, hypoxia and signs or symptoms of hepatitis. Once the patient has reached this level of improvement, completion of therapy may be carried out using oral acyclovir or valacyclovir.

Primary varicella zoster infection

Varicella zoster infection can be life-threatening in KTRs (334,335). Although some centers have begun to institute the use of oral acyclovir in the outpatient setting for KTRs, there is little evidence to confirm the safety and efficacy of this approach. Careful selection of patients with assurance of close clinical follow-up is necessary if oral acyclovir is to be used in these patients.

Uncomplicated herpes zoster

Although herpes zoster can be seen in immunocompetent patients, the presence of immunosuppression is associated with an increased risk for the development of both uncomplicated and complicated herpes zoster infection. Patients with only skin disease, but who have lesions involving more than three dermatomes, are considered to have disseminated cutaneous zoster. Similarly, patients with visceral involvement in addition to skin disease are considered to have disseminated zoster.

Uncomplicated zoster is a clinical syndrome characterized by cutaneous clustering of vesicular lesions in a dermatomal distribution of one or more adjacent sensory nerves. An important complication of herpes zoster in immunocompetent adults is the potential development of postherpetic neuralgia. RCTs in healthy adults have demonstrated that the use of acyclovir, valacyclovir or famciclovir have been associated with more rapid healing of the skin, as well as a decreased incidence of both acute neuritis and postherpetic neuralgia (336,337). In immunocompromised hosts, patients are at risk not only of postherpetic neuralgia but also of severe local dermatomal infection (334). Similarly, immunosuppressed patients are at increased risk for the development of disseminated cutaneous zoster and visceral dissemination. The more severe the level of immunosuppression, the greater the risk of dissemination. Accordingly, prompt initiation of antiviral therapy with close follow-up is warranted for these patients, even if they have only superficial skin infection (333).

Disseminated or invasive herpes zoster

Treatment with intravenous acyclovir and temporary reduction in the amount of immunosuppressive medication is efficacious (333,338). Although specific evidence is not available to guide which immunosuppressive agent should be reduced, it would seem logical, whenever possible, to reduce the dosage of CNIs as well as steroids. In the absence of any evidence of intercurrent rejection, an effort should be made to maintain the reduced level of immunosuppression for a minimum of 3–5 days and until there is evidence of clinical improvement.

Prevention of primary varicella zoster infection

The use of varicella zoster immunoglobulin has been demonstrated to prevent or modify varicella in immunosuppressed individuals exposed to varicella (330,333,339). If varicella zoster immunoglobulin is not available, or if >96 h have passed since the exposure, some experts recommend prophylaxis with a 7-day course of oral acyclovir (80 mg/kg/day administered in four divided doses with a maximum of 800 mg per dose) beginning on day 7–10 after varicella exposure (330,339). The use of varicella vaccine is not recommended as a postexposure prophylactic strategy in KTRs.

• 13.5: 
  HEPATITIS C VIRUS

  • 13.5.1: 
    We suggest that HCV-infected KTRs be treated only when the benefits of treatment clearly outweigh the risk of allograft rejection due to interferon-based therapy (e.g. fibrosing cholestatic hepatitis, life-threatening vasculitis). (2D)[Based on KDIGO Hepatitis C Recommendation 2.1.5.]
  • 13.5.2: 
    We suggest monotherapy with standard interferon for HCV-infected KTRs in whom the benefits of antiviral treatment clearly outweigh the risks. (2D)[Based on KDIGO Hepatitis C Recommendations 2.2.4 and 4.4.2.]
  • 13.5.3: 
    We suggest that all conventional current induction and maintenance immunosuppressive regimens can be used in HCV-infected patients. (2D)[Based on KDIGO Hepatitis C Recommendation 4.3.]
  • 13.5.4: 
    Measure ALT in HCV-infected patients monthly for the first 6 months and every 3–6 months, thereafter. Perform imaging annually to look for cirrhosis and hepatocellular carcinoma. (Not Graded)[Based on KDIGO Hepatitis C Recommendation 4.4.1.] (See Recommendation 19.3.)
  • 13.5.5: 
    Test HCV-infected patients at least every 3–6 months for proteinuria. (Not Graded)[Based on KDIGO Hepatitis C Recommendation 4.4.4.]

    • 13.5.5.1: 
      For patients who develop new-onset proteinuria (either urine protein/creatinine ratio >1 or 24-hour urine protein >1 g on two or more occasions), perform an allograft biopsy with immunofluorescence and electron microscopy. (Not Graded)[Based on KDIGO Hepatitis C Recommendation 4.4.4.]
• 13.5.6: 
  We suggest that patients with HCV-associated glomerulopathy not receive interferon. (2D)[Based on KDIGO Hepatitis C Recommendation 4.4.5.]

ALT, alanine aminotransferase; HCV, hepatitis C virus; KDIGO, Kidney Disease: Improving Global Outcomes; KTRs, kidney transplant recipients.

Background

The Work Group reviewed the KDIGO Hepatitis C Guidelines (340) that were applicable to KTRs, and ultimately agreed with the pertinent guideline statements. Only minor modifications (to guideline statement 4.4.1 in the KDIGO Hepatitis C Guidelines) were made, resulting in recommendation statement 13.5.4. The Transplant Work Group did not conduct a systematic review, but relied on the evidence reviewed by the Hepatitis C Work Group. A brief synopsis of the rationale for the KDIGO Hepatitis C Guidelines that are pertinent to KTRs is presented, with further discussion to the modification of recommendation 13.5.4. Details may be found in the Hepatitis C guidelines.

Rationale

Kidney transplant recipients infected with hepatitis C virus (HCV) have worse patient- and allograft-survival rates than KTRs without HCV infection. In addition, HCV-infected KTRs are at increased risk for several complications, including worsening liver disease, NODAT and glomerulonephritis. Thus, close follow-up of the HCV-infected KTR is prudent.

There are few data to suggest when and how to screen HCV-infected KTRs for posttransplant complications. However, given the higher level of immunosuppression early after transplantation, the Transplant Guideline Work Group determined that liver enzymes should be checked every month for the first 6 months of the posttransplant period, and every 3 months thereafter. The detection of clinically worsening liver enzymes should prompt referral for hepatologic evaluation. Annual liver ultrasound and alpha-fetoprotein level to screen for hepatocellular carcinoma should be considered in patients with cirrhosis on liver biopsy.

Available evidence indicates that all currently available induction and maintenance immunosuppressive agents can be used in KTRs infected with HCV. Although immunosuppression may cause or contribute to complications of HCV in KTRs, there is scant evidence that one type of immunosuppressive agent is more or less likely to be harmful. The exception is tacrolimus, which increases the risk for NODAT, and might be expected to impart at least an additive risk for NODAT to HCV-infected KTRs.

Interferon is effective for viral eradication in HCV-infected patients, especially when combined with ribavirin. However, the administration of interferon after kidney transplantation can be deleterious to the allograft and should generally be avoided in KTRs, unless there is indication of worsening hepatic injury.

Hepatitis C virus infection has also been implicated in the pathogenesis of glomerular disease in both native and transplanted kidneys. Therefore, the Hepatitis C and Transplant Guideline Work Groups concluded that HCV-infected KTRs should be tested for proteinuria every 3–6 months. As recommended for all KTRs, patients who develop new-onset proteinuria (either urine protein/creatinine ratio >1 or 24-hour urine protein greater than 1 g on two or more occasions) should have an allograft biopsy with immunofluorescence and electron microscopy.

Interferon-based therapies may be effective in treating HCV-related glomerulopathy in native kidney disease. However, interferon use in KTRs is associated with an increased risk of rejection. The risk of kidney allograft loss from progressive HCV-associated glomerulopathy compared to that from interferon-induced rejection is unknown. Ribavirin can reduce proteinuria in HCV-associated glomerulopathy, although its impact on kidney function is unknown and it does not lead to viral clearance.

• 13.6: 
  HEPATITIS B VIRUS

  • 13.6.1: 
    We suggest that any currently available induction and maintenance immunosuppressive medication can be used in HBV-infected KTRs. (2D)
  • 13.6.2: 
    We suggest that interferon treatment should generally be avoided in HBV-infected KTRs. (2C)
  • 13.6.3: 
    We suggest that all HBsAg-positive KTRs receive prophylaxis with tenofovir, entecavir, or lamivudine. (2B)

    • 13.6.3.1: 
      Tenofovir or entecavir are preferable to lamivudine, to minimize development of potential drug resistance, unless medication cost requires that lamivudine be used. (Not Graded)
    • 13.6.3.2: 
      During therapy with antivirals, measure HBV DNA and ALT levels every 3 months to monitor efficacy and to detect drug resistance. (Not Graded)
  • 13.6.4: 
    We suggest treatment with adefovir or tenofovir for KTRs with lamivudine resistance (>5 log₁₀ copies/mL rebound of HBV-DNA). (2D)
  • 13.6.5: 
    Screen HBsAg-positive patients with cirrhosis for hepatocellular carcinoma every 12 months with liver ultrasound and alpha feto-protein. (Not Graded) (See Recommendation 19.3.)
  • 13.6.6: 
    We suggest that patients who are negative for HBsAg and have HBsAb titer <10 mIU/mL receive booster vaccination to raise the titer to ≥100 mIU/mL. (2D)

ALT, alanine aminotransferase; HBsAb, antibody to hepatitis B surface antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; KTRs, kidney transplant recipients.

Background

Patients with CKD stage 5 are at increased risk of acquiring HBV infection. Infection can be acquired through infected blood products, or transmission from another infected patient in a dialysis unit. The risk has come down considerably in Western countries following the introduction of universal immunization and strict isolation practices, but remains substantial in developing countries. Screening for HBV infection is done by serologic testing for hepatitis B surface antigen (HBsAg). NAT for the presence of HBV DNA gives a more accurate idea of the infection load. Viral replication is accelerated following introduction of immunosuppression in KTRs. A number of studies have shown that HBV infection increases the risk of mortality, most often due to liver disease and graft failure. Effective antiviral therapy permits inhibition of viral replication and retards development of progressive liver disease, and may lower the risk of liver cancer.

Rationale

• • 
  HBV-infected patients exhibit increased viral replication and are at risk for progressive liver disease after kidney transplantation.
• • 
  HBsAg positivity is an independent risk factor for mortality and graft failure.
• • 
  HBsAg-negative patients are at low risk of increased viral replication and progressive liver disease.
• • 
  Prospective studies have shown that antiviral agents normalize alanine aminotransferase (ALT), and induce clearance of HBV-DNA and hepatitis B E antigen (HBeAg). Antiviral agents are best used as prophylaxis, since KTRs not initiated on antiviral agents at the time of transplantation often develop enhanced viral replication and hepatic dysfunction.
• • 
  ALT activity is lower in KTRs than in the general population, and is unreliable as a marker of liver disease activity by itself. Serial monitoring of HBV DNA is required to assess treatment efficacy. A rise in DNA copy number suggests development of resistance.
• • 
  The newer nucleoside analogues, adefovir and tenofovir are effective for treatment of lamivudine-resistant HBV infection.

Hepatitis B virus infected patients are at risk of exacerbation of the infection, progressive liver disease and development of hepatocellular carcinoma after kidney transplantation. The rate of HBV infection in CKD stage 5 patients as determined by seropositivity for HBsAg varies between 0% and 8% in developed countries (341). The US Centers for Disease Control and Prevention (CDC) estimates that the prevalence of HBsAg-positive patients in the US dialysis population has declined from 7.8% to 0.9%, with an estimated incidence of disease in 2000 of 0.05% (342). This has largely been due to widespread use of universal precautions, screening of the blood supply, the use of erythropoiesis-stimulating agents (ESAs), HBV vaccination and strict implementation of segregation of HBsAg-positive from HBsAg-negative patients during hemodialysis with dedicated machines and staff for each group. The prevalence, however, is much higher (10–20%) in developing countries.

Hepatitis B virus infection in CKD stage 5 patients is usually asymptomatic even in the acute phase, with about 80% of patients progressing to a chronic carrier state (343). Immunosuppression following kidney transplantation leads to increased replication of HBV and results in progressive liver disease. Assessing the natural history of hepatitis B among KTRs is difficult for several reasons (344). Aminotransferase activity is lower in this population, which hampers recognition of HBV-related liver disease (345).

In a meta-analysis (346) of six observational studies (6050 patients), HBsAg positivity was found to be an independent and significant risk factor for mortality (RR 2.49, 95% CI 1.64–3.78) and graft failure (RR 1.44, 95% CI 1.02–2.04). This finding was confirmed in later observational studies. In a study of 286 kidney transplant patients, liver-related death was the most common cause of death in HBV-positive patients (347). A survey from the South Eastern Organ Procurement Foundation demonstrated a detrimental effect of HBV infection on patient survival (p = 0.02) and graft survival (p = 0.05) in 13 287 patients who underwent kidney transplantation between 1977 and 1987 in the United States (348). Patient survival was 62% and 66% at 10 years for HBsAg-positive and -negative KTRs (p = 0.02). The 10-year survival rate of HBsAg-positive KTRs (45%) compares poorly with HCV-infected patients (65%). In patients with biopsy diagnosis of cirrhosis, 10-year survival was 26% (349).

Many studies provided only limited details of virology and did not incorporate liver histology before kidney transplantation, leading to underestimation of the severity of liver disease at the time of transplantation. The only study that carried out serial biopsies found histological deterioration in 85% of HBsAg-positive patients at a mean interval of 66 months. Approximately, 28% showed cirrhosis, whereas no patients had been cirrhotic on baseline biopsy (350). Among those with cirrhosis, hepatocellular carcinoma was found in 23%, suggesting an annual incidence of between 2.5% and 5%. Based on these data, an expert group recommended hepatic imaging every 3 months to detect hepatocellular carcinoma in patients with cirrhosis (351).

The standard practice of screening for HBV infection is testing for HBsAg. The place of routine NAT in these patients is unclear. Some recent studies have shown that a proportion of dialysis patients may exhibit occult HBV infection as detected by NAT in the face of a negative HBsAg (352–360) but not all (361–363). These patients have generally low viral loads and may have mutations that prevent appearance of HBsAg. A large proportion of those with occult infection have antibody to hepatitis B core antigen (HBcAb) and it has been suggested that testing these patients by NAT may be a cost-effective strategy for confirming occult infection. The risk of reactivation of HBV among patients who are HBsAg-negative and HBcAb-positive is low, however (364). Berger et al. (365) found recurrence in 2 of 229 (0.9%) such patients. Savas et al. (366) reported two cases of reactivation and provided a review of 25 previously reported cases. They noted a wide age range of patients experiencing recurrence (22–75 years), a male preponderance, and a posttransplant time of onset between 8 weeks and 15 years. All but one patient had HBsAb titers of less than 100 mIU/mL, leading the authors to suggest that vaccination of such patients may be an effective preventative measure. An expert group recommended routine use of vaccination in such patients to boost the titers above 100 mIU/mL and lamivudine prophylaxis (see section ‘Pharmacotherapy,’ below) during periods of intensified immunosuppression (351).

The primary goals of management are maximal suppression of viral replication, while minimizing development of resistance and prevention of hepatic fibrosis. In view of the poor likelihood of seroconversion to HBsAb, low rates of conversion from HBeAg to anti-HBeAg antibody positivity, and poor reliability of following ALT as a measure of activity, HBV DNA levels need to be followed to assess response to therapy. Serological markers of fibrosis, such as the commercially available Fibrotest panel, have not been evaluated in KTRs with HBV infection. Since the replication is dependent on the overall extent of immunosuppression rather than an individual drug, efforts should be made to minimize the doses of all immunosuppressive drugs without compromising graft outcomes. These include use of the lowest possible dose of steroids. Currently, there is no evidence for the differential effect of any specific immunosuppressive agent on HBV replication.

Pharmacotherapy

There are currently seven medications available for the treatment of hepatitis B: interferon alfa-2b, pegylated interferon alfa 2a, lamivudine, adefovir, tenofovir, telbivudine and entecavir. Interferon therapy for HBV infection in KTRs is associated with high rates of graft loss due to rejection. In a series (367) of 31 HBsAg-positive KTRs treated with recombinant interferon-alpha (three million international units) three times a week for 6 months, long-term ALT normalization was noted in 47% of patients and 13% cleared HBeAg. However, graft loss occurred in five out of 17 patients during therapy and an additional four patients after the completion of therapy. The use of interferon in this setting, therefore, is not recommended (351).

Lamivudine, a cytosine analog that inhibits HBV reverse transcriptase, has been used extensively in KTRs with HBV infection (Table 16). The utility of lamivudine in stabilization of liver function was shown in several observational studies. A meta-analysis (368) that included 14 prospective cohort studies (184 patients) determined the mean overall estimate for ALT normalization, and HBV-DNA and HBeAg clearance at 81% (95% CI 70–92%), 91% (95% CI 86–96%) and 27% (95% CI 16–39%), respectively. The duration of lamivudine therapy was 6–12 months in the majority (11 of 14) of the studies. Later clinical trials (369–375) have shown similar results with lamivudine monotherapy given for 24–69 months. HBeAg and HBV-DNA clearance occurred in 0–25% and 43–78%, respectively. Changes in ALT paralleled those in viremia, and 33–77% of patients maintained normal ALT levels.

Table 16.  Outcomes of clinical trials of lamivudine therapy

Author (year) (ref no)	ALT normalization (%)	HBsAg clearance (%)	HBeAg clearance (%)	HBeAg seroconversion (%)	HBV DNA clearance (%)
ALT, alanine aminotransferase; HBV, hepatitis B virus; HBeAg, hepatitis B E antigen; HBsAg, hepatitis B surface antigen; NA, not available. Reproduced with permission (368).
Rostaing (1997) (376)	4/5 (80)	0	0	NA	6/6 (100)
Goffin (1998) (377)	4/4 (100)	0	0	0/1 (0)	4/4 (100)
Jung (1998) (378)	6/6 (100)	0	1/3 (33)	NA	6/6 (100)
Kletzmayr (2000) (379)	3/3 (100)	0	2/12 (17)	2/12 (17)	15/16 (93)
Tsai (2000) (380)	NA	0	0	NA	7/8 (87.5)
Lewandowska (2000) (381)	17/28 (61)	0	2/26 (8)	NA	10/10 (100)
Antoine (2000) (382)	NA	0	8/12 (67)	NA	9/12 (75)
Mouquet (2000) (383)	8/15 (53)	0	NA	NA	13/15 (87)
Fontaine (2000) (384)	NA	0	6/13 (46)	6/13 (46)	26/26 (100)
Lee (2001) (385)	NA	1/13 (8)	3/8 (37.5)	3/8 (37.5)	10/13 (77)
Han (2001) (386)	6/6 (100)	0	2/3 (67)	NA	6/6 (100)
Chan (2002) (369)	14/14 (100)	0	3/14 (21)	NA	26/26 (100)
Park (2001) (387)	8/10 (80)	0	1/5 (20)	NA	7/10 (70)
Mosconi (2001) (388)	NA	0	NA	NA	4/4 (100)

Timing of initiation

Data on optimal timing of initiation of antiviral therapy are scarce. However, the available data support starting treatment at the time of transplantation in HBsAg-positive patients, irrespective of HBV DNA levels. In a study of 15 patients with normal preoperative ALT (389), seven were started on lamivudine at the time of kidney transplantation. Half of those not treated showed transaminase elevations and HBV viremia in the first year of follow-up, requiring initiation of lamivudine therapy. In contrast, all seven individuals who received lamivudine at the time of transplantation continued to have normal ALT and were negative for HBV DNA throughout the follow-up. In another study of HBsAg-positive KTRs (386), where lamivudine was given prophylactically (HBV DNA negative) or preemptively (HBV DNA positive) to 10 patients or reserved for hepatic dysfunction in 10 patients, 42% in the latter group developed viremia during follow-up, compared to 10% in the former. Six in the reactive group developed hepatic dysfunction compared to none in the prophylactic/preemptive group. In another study (369) where the decision to start lamivudine was based on HBV DNA levels or liver function status, all patients had to be started on lamivudine at a mean time period of 8 months after transplant. More than half the patients were started on treatment because of abnormal ALT.

Duration of therapy

The optimal duration of therapy that ensures long-term remission of viremia and maintenance of normal liver function and minimizes the development of resistance is not known. In a meta-analysis, increased duration of lamivudine therapy was positively associated with frequency of HBeAg loss (r = 0.51, p = 0.04) (Figure 1) (368). Lamivudine discontinuation was attempted by Chan et al. (369) in 12 low-risk patients after stabilization, and was successful in only five (42%).

Figure 1. HBeAg clearance vs. lamivudine duration. HBeAg, hepatitis B E antigen; Lam, lamivudine. Reproduced with permission (368).

At least 24 months of prophylactic treatment has been recommended (390). The optimal treatment and the choice of drugs require further study. Withdrawal of antiviral therapy may be associated with a relapse and increased viral replication, even resulting in liver failure.

Development of resistance is a major clinical problem with long-term lamivudine use. This is usually reflected by a secondary increase in the HBV DNA titers. A commonly used definition is demonstration of >5 log₁₀ copies/mL rebound of HBV DNA. In most, but not all, instances, it is caused by a mutation in the tyrosine–methionine–aspartate–aspartate (YMDD) locus of the HBV DNA polymerase (384). The clinical presentation varies. While some patients show no significant biochemical changes or clinical symptoms, others develop deterioration in liver function (391).

In a study of 29 KTRs (392), resistance was noted in 48% of patients during a mean follow-up period of 69 months; all due to YMDD mutations. Resistance was not related to patient demographics, HBeAg status, seroconversion rates or genotype. About 80% with the YMDD mutation had a hepatitis flare. In the meta-analysis (368), the mean overall estimate for lamivudine resistance was 18% (95% CI 10–37%). An increased duration of lamivudine therapy was positively associated with lamivudine resistance (r = 0.62, p = 0.02). The cumulative probability of developing resistance was approximately 60% in the later studies.

Patients with lamivudine resistance should be treated with adefovir or tenofovir. Limited data are available regarding use of these agents in KTRs. Fontaine et al. (393) gave adefovir to 11 KTRs with lamivudine-resistant HBV infection and found it to be effective in bringing about a reduction in serum HBV DNA, without any significant adverse effects. Entecavir, a guanosine analog, is 30 times more potent than lamivudine in suppressing viral replication. In a multicenter, double-blind RCT comparing entecavir to lamivudine in the general population, entecavir was shown to result in larger reductions in HBV DNA than lamivudine. At a dose of 0.5 mg daily, 83% of patients treated with entecavir had undetectable HBV DNA compared to 58% of those treated with lamivudine (394). In a study (395) that treated eight adefovir- and lamivudine-resistant KTRs with entecavir for 16.5 months, there was a significant decrease in HBV DNA viral load without any significant adverse effects. Data in the non-CKD population shows that, while the risk of resistance to entecavir is low in treatment-naïve patients, it may be as high as 51% at 5 years (396) in lamivudine-resistant cases. In a recent study, tenofovir was shown to be superior to adefovir in achieving remission of HBV viremia and hepatic histologic scores in non-CKD patients. Tenofovir was effective in lamivudine-resistant cases, and did not produce resistance up to 48 months of treatment (397). Of the two agents, tenofovir has a much lower renal toxicity than adefovir, and hence would be the preferred agent in KTRs. It is not known whether substitution of lamivudine with entecavir or tenofovir for prophylaxis will prevent development of resistance.

Research Recommendations

• • 
  The frequency of occult HBV infection in patients with CKD stage 5 should be evaluated in different parts of the world, and its impact on posttransplant outcomes determined.
• • 
  Studies are required to determine whether substitution of lamivudine with entecavir or tenofovir for prophylaxis will prevent development of resistance in KTRs.

• 13.7: 
  HUMAN IMMUNODEFICIENCY VIRUS

  • 13.7.1: 
    If not already done, screen for HIV infection. (Not Graded)
  • 13.7.2: 
    To determine antiretroviral therapy, refer HIV-infected KTRs to an HIV specialist, who should pay special attention to drug–drug interactions and appropriate dosing of medications. (Not Graded)

HIV, human immunodeficiency virus; KTRs, kidney transplant recipients.

Background

Screening for human immunodeficiency virus (HIV) infection is defined as the performance of serologic testing for HIV. A two-step screening is usually performed. In the first step, patients are screened for the presence of antibodies against HIV, usually with an enzyme-linked immunosorbent assay (ELISA). This is an extremely sensitive test. However, it is not specific. Accordingly, those patients who are positive on ELISA are then screened using a Western Blot assay. The presence of a positive Western Blot assay for HIV confirms the diagnosis of HIV infection except in children <18 months of age, where a positive serologic test may be attributable to the presence of passive antibody acquired from the child's mother during the pregnancy. NAT for the presence of HIV DNA or HIV RNA viral load should be performed on children <18 months of age with a positive HIV antibody. Antiretroviral medications are used specifically for the treatment of HIV infection. Drug–drug interactions are pharmacokinetic interactions between separate medications that may result in accumulation or more rapid metabolism of one or both compounds.

Rationale

• • 
  Patients with HIV require specialized care in centers with appropriate expertise.
• • 
  Screening for HIV infection should be carried out on all KTRs (ideally before transplantation) in order to identify those KTRs that will require specialized care.
• • 
  Antiretroviral therapy is necessary to maintain virologic suppression and normal immunologic function in HIV patients undergoing kidney transplantation.
• • 
  The concomitant use of antiretroviral agents and immunosuppressive medications creates the potential for drug–drug interactions that may substantially alter blood levels of drugs and require appropriate monitoring and adjustments in dosing.

Case series have documented successful outcomes of KTRs with HIV (398–400). However, these HIV patients had been carefully selected and adequately treated for HIV at the time of transplantation (400). Although HIV is not an absolute contraindication to kidney transplantation, the presence of HIV has major implications in the management of patients following transplantation. A major issue of concern in the management of HIV patients is the need to be aware of potential drug–drug interactions among antiretroviral therapy and other medications, including immunosuppressants. Care must be taken to identify and select those HIV-infected patients who are most likely to benefit from kidney transplantation without an unacceptably high risk of opportunistic infections.

Evidence from a National Institutes of Health (NIH)—sponsored study of organ transplantation in HIV patients has demonstrated both the effectiveness of transplantation as well as the complexity of management of KTRs with HIV (400). Data accrued from the NIH-sponsored study of organ transplantation in HIV-infected patients has identified specific drug combinations that are associated with drug–drug interactions in these patients (401). Accordingly, attention must be paid and caution must be used in these patients to account for the potential impact of these interactions. Although the data from the NIH study demonstrate the feasibility of transplantation for HIV-infected KTRs, the limited number of HIV patients with CKD stage 5 undergoing kidney transplantation to date suggests the need to continue performing this procedure under research protocols and in selected centers with appropriate expertise. Finally, it is worth noting that review of experience to date suggests that there may be an increased risk for the development of acute cellular rejection in patients with HIV undergoing organ transplantation.

Research Recommendations

• • 
  There is a need to determine the optimal immunosuppression medication regimen, as well as the best antiretroviral regimens, for HIV-infected KTRs.

Chapter 14: Other Infections

• 14.1: 
  URINARY TRACT INFECTION

  • 14.1.1: 
    We suggest that all KTRs receive UTI prophylaxis with daily trimethoprim–sulfamethoxazole for at least 6 months after transplantation. (2B)
  • 14.1.2: 
    For allograft pyelonephritis, we suggest initial hospitalization and treatment with intravenous antibiotics. (2C)

KTRs, kidney transplant recipients; UTI, urinary tract infection.

Background

A urinary tract infection (UTI) is an infection causing signs and symptoms of cystitis or pyelonephritis (including the presence of signs of systemic inflammation), which is documented to be caused by an infectious agent. Kidney allograft pyelonephritis is an infection of the kidney allograft that is usually accompanied by characteristic signs and symptoms of systemic inflammation and a positive urine and/or blood culture. Occasionally, pyelonephritis is diagnosed by allograft biopsy. Antibiotic prophylaxis is the use of an antimicrobial agent (or agents) to prevent the development of a UTI.

Rationale

• • 
  UTI is a frequent and potentially important complication of kidney transplantation.
• • 
  The use of antibiotic prophylaxis can reduce the risk of UTI.
• • 
  Kidney allograft pyelonephritis may be associated with bacteremia, metastatic spread, impaired graft function and even death.
• • 
  KTRs with clinical and laboratory evidence suggestive of kidney allograft pyelonephritis should be hospitalized and treated with intravenous antibiotics.

Observational studies have documented a high incidence of UTI in KTRs (402). Pyelonephritis of the kidney allograft is a common complication in KTRs (402). It may cause graft failure, sepsis and death. The use of antibiotic prophylaxis with trimethoprim–sulfamethoxazole has been demonstrated to decrease the frequency of bacterial infections, including UTI in KTRs (403). The use of trimethoprim–sulfamethoxazole for the first 9 months following kidney transplant was associated with statistically significant decreases in number of any bacterial infection, overall number of UTI and number of noncatheter UTI. There is moderate-quality evidence that the benefit of UTI prophylaxis (primarily preventing infection, but unclear evidence for reducing mortality or preventing graft loss) outweighs the risks (see Evidence Profile and accompanying evidence in Supporting Tables 50–51 at http://www3.interscience.wiley.com/journal/118499698/toc). Based upon this, and several other small studies, prophylactic trimethoprim–sulfamethoxazole for 6–12 months following kidney transplantation is warranted.

Although the use of ciproflaxicin also appeared effective in the prevention of UTI after KTRs, patients treated with this regimen were at risk for, and developed Pneumocystis jirovecii pneumonia (PCP) (see Recommendation 14.2) (404). Accordingly, the use of trimethoprim–sulfamethoxazole is preferred over ciprofloxacin at least during the first 6 months after transplantation.

Although some investigators have recommended indefinite use of trimethoprim–sulfamethoxazole, data are not available demonstrating clinical benefit beyond the first 9 months following kidney transplantation. Evidence suggests that late UTIs tend to be benign, without associated bacteremia, metastatic foci or effect on long-term graft function (405). For this reason, we recommend providing prophylaxis for a minimum of 6 months. For patients who are allergic to trimethoprim–sulfamethoxazole, the recommended alternative agent would be nitrofurantoin. This agent, which is widely recommended as an alternative to trimethoprim/sulfamethoxazole, is chosen over ciprofloxacin (despite demonstrated effectiveness in KTRs) in an effort to limit the likelihood of emergence of antibacterial resistance.

Kidney allograft pyelonephritis may be associated with bacteremia, metastatic spread, impaired graft function and even death. Accordingly, KTRs with clinical and laboratory evidence suggestive of kidney allograft pyelonephritis should be hospitalized and be treated with intravenous antibiotics for at least the initial course of therapy. This is particularly true in early infections (first 4–6 months following kidney transplantation). Recognition of the morbidity and mortality associated with allograft pyelonephritis led to recommendations in the 1980s to treat UTIs with as long as a 6-week course of antimicrobials for early UTI following transplantation. More recently, UTI after kidney transplantation has been associated with considerably lower morbidity and mortality (405). Accordingly, a less-prolonged course may be required, although patients experiencing relapsing infection should be considered for a more prolonged therapeutic course.

Because of the potential for serious complications, KTRs with kidney allograft pyelonephritis should be hospitalized and treated with intravenous antibiotics, at least initially. Although evidence derived from RCTs on the optimal duration of therapy for kidney allograft pyelonephritis are not available, it is anticipated, in the absence of a kidney abscess, that 14 days should be adequate.

• 14.2: 
  PNEUMOCYSTIS JIROVECII PNEUMONIA

  • 14.2.1: 
    We recommend that all KTRs receive PCP prophylaxis with daily trimethoprim–sulfamethoxazole for 3–6 months after transplantation. (1B)
  • 14.2.2: 
    We suggest that all KTRs receive PCP prophylaxis with daily trimethoprim–sulfamethoxazole for at least 6 weeks during and after treatment for acute rejection. (2C)
  • 14.2.3: 
    We recommend that KTRs with PCP diagnosed by bronchial alveolar lavage and/or lung biopsy be treated with high-dose intravenous trimethoprim–sulfamethoxazole, corticosteroids, and a reduction in immunosuppressive medication. (1C)
  • 14.2.4: 
    We recommend treatment with corticosteroids for KTRs with moderate to severe PCP (as defined by PaO₂ <70 mm Hg in room air or an alveolar gradient of >35 mm Hg). (1C)

KTRs, kidney transplant recipients; PaO₂, partial pressure of oxygen in arterial blood; PCP, Pneumocystis jirovecii pneumonia.

Background

Pneumocystis jirovecii (formally known as Pneumocystis carinii) is an opportunistic fungal pathogen known to cause life-threatening pneumonia in immunocompromised patients, including KTRs. P. jirovecii pneumonia (PCP) is defined as the presence of lower respiratory-tract infection due to P. jirovecii. A definitive diagnosis of PCP is made by demonstration of organisms in lung tissue or lower respiratory tract secretions. Because no specific diagnostic pattern exists on any given imaging test, it is imperative that the diagnosis of PCP be confirmed by lung biopsy or bronchoalveolar lavage.

Rationale

• • 
  Infection with P. jirovecii is life-threatening in KTRs.
• • 
  Prophylaxis with trimethoprim–sulfamethoxazole is safe and effective.
• • 
  Although thrice-weekly dosing of trimethoprim–sulfamethoxazole is adequate prophylaxis for PCP, daily dosing also provides prophylaxis for UTI and may be easier for patient adherence.
• • 
  Treatment of PCP with high-dose, intravenous trimethoprim–sulfamethoxazole and reduction of immunosuppressive medications are the treatments of choice for PCP.
• • 
  Based upon data from HIV-infected adults, the use of corticosteroids has been uniformly recommended for all patients experiencing moderate to severe PCP.

PCP prophylaxis

Pneumocystis jirovecii is an important opportunistic pathogen known to cause life threatening PCP in KTRs (406). The most typical time of onset of symptoms of PCP is 6–8 weeks following initiation of immunosuppressive therapy. Although PCP is potentially a life-threatening complication of KTRs, the use of chemoprophylaxis has been shown to be extremely effective in preventing the development of clinical disease attributable to this pathogen. The use of trimethoprim–sulfamethoxazole prophylaxis resulted in a RR of 0.08 (95% CI 0.023–0.036) of developing PCP compared to either a placebo, control or no intervention (403). Treatment also decreased mortality.

There was no difference in efficacy for PCP when trimethoprim–sulfamethoxazole was given daily or three times per week (407). However, in KTRs, the use of daily trimethoprim–sulfamethoxazole may be associated with a decreased risk of bacterial infection (403). Although definitive evidence for the duration of PCP prophylaxis is not available, most experts agree that it should be continued for at least 6 months (and perhaps as long as 1 year) following transplantation (406). Because most KTRs will remain on immunosuppression for the rest of their lives, some experts recommend a more prolonged and perhaps even indefinite use of PCP prophylaxis. Indications for the use of alternative preventive agents include the development of allergic reactions and/or drug-induced neutropenia from trimethoprim–sulfamethoxazole. Potential alternative agents include dapsone, aerosolized pentamidine, atovaquone or the combination of clindamycin and pyrimethamine (Table 17).

Table 17.  Antimicrobial agents for the prevention of PCP in KTRs^a

Agent	Adult dose	Pediatric dose
KTRs, kidney transplant recipients; PCP, Pneumocystis jirovecii pneumonia. ^aExcerpted from (408). ^bThis is first-line therapy. All other agents should be considered second-line therapy. ^cMust screen for glucose 6-phosphate dehydrogenase deficiency prior to using dapsone, as this is a risk factor for development of methemoglobinemia associated with use of dapsone.
Trimethoprim/sulfamethoxazole^b	Single-strength pill (80 mg as trimethoprim) or double-strength pill (160 mg as trimethoprim) daily or three times per week	150 mg/m^2/day as trimethoprim daily or three times per week
Aerosolized pentamidine	300 mg inhaled every 3–4 weeks via Respirgard II™ nebulizer	For children ≥5 years old, 300 mg inhaled monthly via Respirgard II™ nebulizer
Dapsone^c	100 mg/day as a single dose or 50 mg twice a day	Can be administered on a daily or weekly schedule as 2.0 mg/kg/day (maximum total dosage of 100 mg/day) or 4.0 mg/kg/week (maximum total dosage of 200 mg/week) orally. Approximately two thirds of patients intolerant to Trimethoprim/sulfamethoxazole can take dapsone successfully. Studies in adults show dapsone is as effective as atovaquone or aerosolized pentamidine but slightly less effective than Trimethoprim/sulfamethoxazole
Atovaquone	1500 mg/day	Administered with a meal as an oral yellow suspension in single dosage of 30 mg/kg/day for patients 1–3 months and >24 months of age, and 45 mg/kg/day for infants aged 4–24 months

PCP treatment

Prior to the use of trimethoprim–sulfamethoxazole, mortality from PCP in KTRs was very high (409,410). The treatment of PCP includes both the use of intravenous trimethoprim–sulfamethoxazole as well as corticosteroids for KTRs with significant hypoxemia (406). RCTs have demonstrated that the use of corticosteroids in the first 72 hours of PCP in HIV patients resulted in improved outcome, including morbidity, mortality and avoidance of intubation (406). The usual duration of treatment is 2–3 weeks. The use of intravenous pentamidine isethionate should be considered in patients with proven trimethoprim–sulfamethoxazole allergy. Other treatment strategies should be restricted to patients with mild PCP only.

• 14.3: 
  TUBERCULOSIS

  • 14.3.1: 
    We suggest that TB prophylaxis and treatment regimens be the same in KTRs as would be used in the local, general population who require therapy. (2D)
  • 14.3.2: 
    We recommend monitoring CNI and mTORi blood levels in patients receiving rifampin. (1C)

    • 14.3.2.1: 
      Consider substituting rifabutin for rifampin to minimize interactions with CNIs and mTORi. (Not Graded)

CNI, calcineurin inhibitor; KTRs, kidney transplant recipients; mTORi, mammalian target of rapamycin inhibitor(s); TB, tuberculosis.

Rationale

• • 
  KTRs are at increased risk of developing disease due to tuberculosis (TB).
• • 
  KTRs with latent TB, identified by a positive purified protein derivative (PPD) skin test or a history of TB disease without adequate treatment, are at highest risk of developing clinical TB after transplantation and are therefore good candidates for chemoprophylaxis with isoniazid.
• • 
  Treatment of TB in KTRs has been shown to respond to standard antimycobacterial therapy.
• • 
  The use of rifampin is associated with numerous drug–drug interactions through its activation of the CYP3A4 pathway.
• • 
  This interaction can affect drug levels for CNIs as well as mTORi.
• • 
  Rifabutin achieves similar therapeutic efficacy while minimizing the potential for drug–drug interactions.

The incidence of TB among KTRs varies according to geographic locations, with rates of 0.5–1.0% reported in North America, 0.7–5% in Europe and 5–15% in India and Pakistan (411,412). This represents a marked (50- to 100-fold) increase in the frequency of TB compared to the general population. In addition, there is also a marked increase in severity of disease in KTRs with mortality rates 10-fold higher than in immunocompetent individuals with TB.

The most frequent source of TB infections in KTRs is reactivation of quiescent foci of Mycobacterium tuberculosis that persist after initial asymptomatic infection (413). Accordingly, screening and identification of individuals with evidence of prior latent infection with TB should allow treatment prior to development of clinical disease, resulting in improved outcome.

Data from a variety of immunosuppressed populations demonstrate that treatment of latent TB markedly reduces the risk of subsequent progression to clinically active TB (414). A limited number of RCTs have evaluated the benefit of prophylactic treatment with isoniazid for KTRs (415) or organ transplant patients, including KTRs (416,417). Results of these studies suggest a benefit to KTRs, although study size and design limit the strength of these observations. The use of prophylactic isoniazid in patients with a past or current positive PPD skin test, and/or a history of TB without adequate documented treatment, has been previously recommended by the European Best Practice Guidelines for Renal Transplantation (411) and the American Society of Transplantation Guidelines for the Prevention and Management of Infectious Complications of Solid Organ Transplantation (418).

If, according to these guidelines, vaccination with BCG can give a ‘false-positive’ PPD skin test, then some patients may be treated unnecessarily. Most believe that the effect of BCG should not persist for more than 10 years (419). The use of BCG vaccine is especially common in regions where the prevalence of TB is high. In these regions, it is therefore difficult to distinguish PPD skin tests that are positive due to BCG from those that are positive due to prior infection with M. tuberculosis. Accordingly, it is recommended that the history of BCG vaccination should be ignored and that a 9-month course of prophylactic isoniazid should be used (411). It is also possible that dialysis and transplant patients frequently have false-negative PPD skin tests. Accordingly, some experts have recommended use of isoniazid prophylaxis in selected KTRs with a negative PPD skin test. These would include those with history of active TB that was not adequately treated, those with radiographic evidence of previous TB without a history of treatment and those who have received an organ from a donor with a history of a positive PPD skin test (418).

Interferon-gamma release assays such as T-SPOT.TB and QuantiFERON are an alternative to the tuberculin skin test for detecting latent TB infection. Their sensitivity and specificity, however, have not been systematically evaluated in KTRs. Data from CKD stage 5 patients suggest important limitations for detecting latent TB infection which preclude their routine use at present (420–423).

Extensive experience in the treatment of immunosuppressed patients (including transplant recipients) suggests that the response to treatment is the same as in immunocompetent patients. Unfortunately, rifampin is a strong inducer of the microsomal enzymes that metabolize CNIs and mTORi, and it may be difficult to maintain adequate levels of these immunosuppressive drugs to prevent rejection. The use of rifampin has required doses of CNIs to be increased two- to threefold (418). One potential alternative is to substitute rifabutin for rifampin. Rifabutin has activity against M. tuberculosis that is similar to rifampin, but rifabutin is not as strong an inducer of CYP3A4 as rifampin. However, there is little published experience with rifabutin in KTRs.

There are reports of successful treatment of posttransplant TB with rifampin-sparing regimens (415). In this report, rifampin is substituted with a fluoroquinolone along with isoniazid, ethambutol and pyrazinamide for the first 2 months. At this point, the latter two are stopped and fluoroquinolone and isoniazid continued for another 10–12 months. According to the authors, the success rate is 100% (424–426).

Finally, the rate of recovery of drug-resistant TB is increasing. Since both KTRs and their donors may come from diverse geographic locations where the prevalence of drug resistance may vary, all isolates of TB recovered from KTRs should be submitted for susceptibility testing. Modifications in treatment should be made once the results of susceptibility testing become available.

• 14.4: 
  CANDIDA PROPHYLAXIS

  • 14.4.1: 
    We suggest oral and esophageal Candida prophylaxis with oral clotrimazole lozenges, nystatin, or fluconazole for 1–3 months after transplantation, and for 1 month after treatment with an antilymphocyte antibody. (2C)

Rationale

• • 
  KTRs are at increased risk for oral and esophageal infections due to Candida species.
• • 
  The use of oral clotrimazole troches or nystatin provides effective prophylaxis without systemic absorption and hence without concerns for side effects.
• • 
  Although data regarding the duration of prophylaxis are not available for KTRs, prophylaxis should logically be continued until patients are on stable, maintenance immunosuppression, particularly corticosteroids.

Observational studies have reported a high incidence of oral and esophageal Candida infections in KTRs. There are limited data supporting the use of antifungal therapy in KTRs, although it is beneficial in liver transplant recipients (427). The standard immunosuppressive agents typically used in KTRs are associated with an increased risk of developing Candida infections. The most common source for these infections is colonization of the oral mucosa. Accordingly, use of topical antifungal therapies such as clotrimazole troches and nystatin offer the opportunity to eradicate fungal colonization without associated risks that may be present for systemically absorbed antifungal agents. However, a recent report suggested a potential drug–drug interaction between clotrimazole and tacrolimus (428). It is important to note that there are drug–drug interactions between fluconazole and CNIs.

Although data regarding the appropriate duration of prophylaxis for these agents are not available for KTRs, the risk is greatest early after transplantation when patients are receiving their highest levels of immunosuppression, and are more likely to be exposed to antibacterial agents that increase the risk for Candida infections. Accordingly, these agents can likely be discontinued once the patient is on maintenance immunosuppression, particularly when steroid doses are stable and low.

Research Recommendations

• • 
  RCTs are needed to determine the optimal duration and type of prophylaxis for Candida infections in KTRs.

Section III: Cardiovascular Disease

Introduction

The incidence of CVD is high after kidney transplantation (429–434). The annual rate of fatal or nonfatal CVD events is 3.5–5.0% in KTRs, 50-fold higher than in the general population (435). By 36 months after transplantation, nearly 40% of patients have experienced a CVD event (436). Although acute myocardial infarction is common after transplantation, especially in elderly patients and those with diabetes (437) congestive heart failure (CHF) is also a common CVD complication (436). Most of the ‘traditional risk factors’ in the general population, including cigarette smoking, diabetes, hypertension and dyslipidemias, are also risk factors for CVD in KTRs (Table 18). In addition, many KTRs have had CKD for an extended period of time prior to transplantation, and have thereby acquired additional CVD risk by the time they undergo transplantation. For all of these reasons, KTRs should be considered to be at the highest risk for CVD and managed accordingly.

Table 18.  Independent predictors of CVD in KTRs

Predictor	Number of studies (number of analyses)	Total number of subjects (range)	Outcomes	Number statistically significant (p < 0.05)
BMI, body mass index; CAD, coronary artery disease; CeVD, cerebrovascular disease; CHF, congestive heart failure; CVD, cardiovascular disease; PVD, peripheral vascular disease. ^aBased on studies that met criteria for systematic reviews of tobacco use, diabetes and obesity. ^bBased on a systematic review performed for the KDOQI Dyslipidemia guidelines (466). This includes smaller studies than would have been included in a de novo systematic review. In addition, not all associations are independent in multivariable analysis.
			All CVD	1/1
			CAD	1/2
Tobacco use (438–443)	6 (10)	57 027	CeVD	1/2
		(427–27 011)	PVD	0/2
			CHF	1/1
			All-cause mortality	2/2
Diabetes (430,442,444–453)	12 (17)	115 510	All CVD	1/1
		(158–76 481)	CAD	3/3
			CeVD	2/2
			PVD	2/2
			CV mortality	3/3
			All-cause mortality	6/6
Obese/elevated BMI (14,443,454–456)	5 (6)	103 295 (2067–51 927)	CHF	1/1
			CV mortality	1/1
			All-cause mortality	2/4
Hypertension^a (439–441,443,450)	5 (5)	29 259	All CVD	1/1
		(403–27 011)	CeVD	1/1
			CHF	2/2
			All-cause mortality	1/1
Dyslipidemia^b (457–465)	9 (9)	3657	All CVD (combined in	5/9
		(21–1124)	systematic review)	1/4
				2/7
				2/7

Rating Guideline Recommendations

Within each recommendation, the strength of recommendation is indicated as Level 1, Level 2, or Not Graded, and the quality of the supporting evidence is shown as A, B, C, or D.

Grade*	Wording
*The additional category ‘Not Graded’ was used, typically, to provide guidance based on common sense or where the topic does not allow adequate application of evidence. The most common examples include recommendations regarding monitoring intervals, counseling, and referral to other clinical specialists. The ungraded recommendations are generally written as simple declarative statements, but are not meant to be interpreted as being stronger recommendations than Level 1 or 2 recommendations.
Level 1	‘We recommend’
Level 2	‘We suggest’

Grade for quality of evidence	Quality of evidence
A	High
B	Moderate
C	Low
D	Very low

Chapter 15: Diabetes Mellitus

• 15.1: 
  SCREENING FOR NEW-ONSET DIABETES AFTER TRANSPLANTATION

  • 15.1.1: 
    We recommend screening all nondiabetic KTRs with fasting plasma glucose, oral glucose tolerance testing, and/or HbA_1c(1C) at least:

    • • 
      weekly for 4 weeks (2D);
    • • 
      every 3 months for 1 year (2D); and
    • • 
      annually, thereafter. (2D)
  • 15.1.2: 
    We suggest screening for NODAT with fasting glucose, oral glucose tolerance testing, and/or HbA_1c after starting, or substantially increasing the dose, of CNIs, mTORi, or corticosteroids. (2D)

CNI, calcineurin inhibitor; HbA_1c, hemoglobin A_1c; KTRs, kidney transplant recipients; mTORi, mammalian target of rapamycin inhibitor(s); NODAT, new-onset diabetes after transplantation.

Background

Diabetes is defined according to the WHO and American Diabetes Association (ADA) (Table 19).

Table 19.  Criteria for the diagnosis of diabetes

WHO, World Health Organization. *In the absence of unequivocal hyperglycemia, these criteria should be confirmed by repeat testing on a different day. Modified with permission (467).
1.	Fasting plasma glucose ≥126 mg/dL (7.0 mmol/L). Fasting is defined as no caloric intake for at least 8 hours.*
OR
2.	Symptoms of hyperglycemia and a casual plasma glucose ≥200 mg/dL (11.1 mmol/L). Casual is defined as any time of day without regard to time since last meal. The classic symptoms of hyperglycemia include polyuria, polydipsia and unexplained weight loss.
OR
3.	Two-hour plasma glucose ≥200 mg/dL (11.1 mmol/L) during an oral glucose tolerance test. The test should be performed as described by the WHO, using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water.*

New-onset diabetes after transplantation is diabetes defined by the WHO and ADA that develops for the first time after kidney transplantation.

Rationale

• • 
  The chances of reversing or ameliorating NODAT may be improved by early detection and intervention.
• • 
  Early treatment of NODAT may prevent complications of diabetes.
• • 
  The incidence of NODAT is sufficiently high to warrant screening.

Fasting plasma glucose, 2-h glucose tolerance testing (after a 75-g glucose load) and hemoglobin A_1c (HbA_1c) are probably suitable screening tests to detect NODAT in KTRs. The frequency of screening for NODAT is based on the incidence of NODAT at different times after kidney transplantation. The reported incidence varies by the definition of diabetes and the type of immunosuppressive medications used. However, the incidence of NODAT is highest in the first 3 months after transplantation. The cumulative incidence of NODAT by the end of the first year has generally been found to be 10–30% in adults receiving CsA or tacrolimus plus corticosteroids (468–479), and 3–13% in children (480,481). The high incidence of NODAT justifies frequent screening during the first year after transplantation. A number of risk factors increase the incidence of NODAT (Table 20), and patients with one or more of these additional risk factors may benefit from more frequent screening.

Table 20.  Risk factors for NODAT^a

Predictor	No. of subjects (range)	Association (No. of studies p < 0.05)	No association (No. of studies)
BMI, body mass index; CsA, cyclosporine A; CMV, cytomegalovirus; D, transplant donor; HCV, hepatitis C virus; HDL-C, high-density lipoprotein cholesterol; HLA, human leukocyte antigen; nd, no data; NODAT, new-onset diabetes after transplantation; R, transplant recipient. ^aNODAT was variously defined in studies identifying risk factors and having a sample size at least 100. To convert HDL-C mg/dL to mmol/L multiply by 0.02586; to convert triglycerides mg/dL to mmol/L multiply by 0.01129.
Tacrolimus (474–477,479,482–485)	100 418 (386–28 941)	7	2
CsA (479,484)	1066 (528–538)		2
Corticosteroids (477,478,484,486)	2035 (386–589)	2	2
Sirolimus (479,484,487,488)	22 525 (528–21 459)	2	2
Acute rejection (477–479)	1436 (386–528)	3
Obesity/higher BMI (471,472,474,476–479,482,484,485,488)	97 702 (386–28 942)	9	2
African American ethnicity (471,472,474–476,479,482,485,488)	103 383 (528–28 942)	8	1
Hispanic ethnicity (US) (474)	15 787	1
Older age (471,472,474–479,484,485,488)	94 487 (386–28 942)	9	2
Male (471,474,476–479,484,485)	64 090 (386–28 942)		8
HLA mismatch (474,476,478,485)	60 560 (522–28 942)	2	2
Deceased-donor kidney (471,474,476–478,485)	63 024 (386–28 942)	1	5
Hepatitis C (474,477,478,482,485,488)	63 805 (386–21 459)	5	1
HCV risk (D+/R−) (476)	28 942	1
CMV risk (D+/R−) (477)	386		1
Beta-blockers	nd
Thiazide diuretics	nd
History of:
Type 2 diabetes in family (478,484)	1060 (522–538)	1	1
Gestational diabetes	nd
Impaired fasting glucose	nd
Impaired glucose tolerance	nd
HDL-C <40 mg/dL	nd
Triglycerides >150 mg/dL (472)	1811	1

Since tacrolimus, CsA, mTORi and corticosteroids can cause NODAT, it is reasonable to screen for NODAT after starting, or substantially increasing the dose of one of these medications. Treating acute rejection with high-dose corticosteroids, for example, should prompt screening for NODAT.

Tacrolimus and CsA may cause NODAT by directly decreasing insulin secretion of pancreatic beta cells (489–493). Logically, reducing the dose or discontinuing these agents as soon as possible could potentially limit the damage to beta cells, although the clinical evidence is anecdotal (494,495). There is anecdotal evidence from case reports/series that NODAT may be reversed by reducing, replacing or discontinuing CsA, tacrolimus or corticosteroids (494,495). There are few data on the effects of corticosteroid reduction on reversing NODAT once it has occurred. Similarly, few, if any, data are available on whether discontinuing mTORi will reverse NODAT.

The relative effects of different immunosuppressive agents on NODAT are difficult to quantify, because RCTs use different regimens and doses, as well as different definitions of NODAT, all of which make comparisons difficult. Nevertheless, it appears that the risk of NODAT with tacrolimus is greater than with CsA. It is also clear that high doses of corticosteroids used immediately after transplantation, and in the treatment of acute rejection, are risk factors for NODAT. Sirolimus has not been as well studied. Some observational studies have found that sirolimus use was associated with an increased incidence of NODAT (487,496,497). Randomized trials have produced conflicting results (498–502). There is no evidence that azathioprine or MMF causes NODAT.

The risk of NODAT from immunosuppressive medications is no doubt higher in individuals with other risk factors, for example African American or American Hispanic ethnicity, obesity and age. Thus, the choice of immunosuppressive medications could be individualized to the risk for NODAT attributable to other risk factors in each individual patient. In addition, the risk of NODAT should be considered in light of the risk of acute rejection. Indeed, the occurrence of acute rejection and its treatment with corticosteroids is a risk factor for NODAT. Unfortunately, it is difficult to weigh the relative risks of rejection and NODAT in individual patients to determine the best immunosuppressive medication regimen.

By almost any definition, the risk of NODAT is increased by obesity. African American and Hispanic ethnicity are generally defined as self-reported. Since data on African American and Hispanic ethnicity are largely from the United States, it is unclear if ethnicities defined otherwise and in other countries have similar risk for NODAT. Older age is a risk factor that shows a linear relationship with risk, but there is no clear threshold. HCV infection is defined by the presence of antibody to the HCV at the time of transplantation.

A number of other risk factors for diabetes have not been rigorously studied in KTRs, but there is little reason to believe that they would not also be risk factors after transplantation. These risk factors include: family history (type 2 diabetes), gestational diabetes, impaired fasting glucose, impaired glucose tolerance and dyslipidemia (high fasting triglycerides and/or low HDL-C) (503–507).

Data from observational studies have shown that NODAT is associated with worse outcomes, including increased graft failure, mortality and CVD (474). It is possible that some of these associations result from unmeasured risk factors that are common to both NODAT and poor outcomes. However, it is certainly plausible that NODAT directly and indirectly contributes to worse outcomes. Untreated diabetes may increase the risk of metabolic complications, including hyperkalemia, and even ketoacidosis. However, there is no evidence from observational studies to suggest how frequently these complications occur after NODAT.

Research Recommendations

• • 
  Future RCTs of immunosuppressive medication regimens should measure fasting glucose, HbA_1c and/or glucose tolerance tests, and any treatments of diabetes, to determine the effect of the medication regimens on the incidence of NODAT.

• 15.2: 
  MANAGING NODAT OR DIABETES PRESENT AT TRANSPLANTATION

  • 15.2.1: 
    If NODAT develops, consider modifying the immunosuppressive drug regimen to reverse or ameliorate diabetes, after weighing the risk of rejection and other potential adverse effects. (Not Graded)
  • 15.2.2: 
    Consider targeting HbA_1c 7.0–7.5%, and avoid targeting HbA_1c≤6.0%, especially if hypoglycemic reactions are common. (Not Graded)
  • 15.2.3: 
    We suggest that, in patients with diabetes, aspirin (65–100 mg/day) use for the primary prevention of CVD be based on patient preferences and values, balancing the risk for ischemic events to that of bleeding. (2D)

CVD, cardiovascular disease; HbA_1c, hemoglobin A_1c; NODAT, new-onset diabetes after transplantation.

Background

The management of diabetes that is present at the time of transplantation may be complicated by severe autonomic neuropathy and other complications of long-standing diabetes that may make ‘tight’ control of blood glucose difficult to achieve. Therefore, we recommend avoiding intensive therapies targeting HbA_1c levels <6.0%. However, complications of long-standing diabetes that make the management of diabetes difficult are less likely to be present in patients with NODAT, and it is not clear whether NODAT can be safely and effectively managed within a narrow range of low blood glucose and HbA_1c targets.

Rationale

• • 
  The benefits and harm of altering the immunosuppressive medication regimen in response to the development of NODAT are unclear.
• • 
  In the general diabetic population, there is insufficient evidence for or against targeting a specific HbA_1c level to reduce CVD; however, recent data suggest that mortality may be increased in patients with type 2 diabetes by targeting HbA_1c levels that are <6.0%.
• • 
  In KTRs, attempting to reduce HbA_1c levels in order to reduce CVD may result in more complications than in the general diabetic population.
• • 
  Randomized trials in the general population suggest that aspirin prophylaxis may prevent CVD in patients with diabetes.

There are no RCTs testing whether changing to different immunosuppressive medication regimens reverses or ameliorates NODAT. There are uncontrolled (largely anecdotal) reports on the effects of changing immunosuppressive agents once NODAT has developed (494,495). Given the associations of NODAT with CsA, tacrolimus, mTORi and corticosteroids, it is plausible that reducing or eliminating these immunosuppressive medications may reverse or ameliorate NODAT. Changes in immunosuppressive medications that may reverse or ameliorate NODAT include:

• i) 
  reducing the dose of tacrolimus, CsA or corticosteroids;
• ii) 
  discontinuing tacrolimus, CsA or corticosteroids;
• iii) 
  replacing tacrolimus with CsA, MMF or azathioprine;
• iv) 
  replacing CsA with MMF or azathioprine.

We could find no published reports of reducing the dose or discontinuing a mTORi to reverse or ameliorate NODAT.

Optimal glycemic control to prevent microvascular disease complications has been defined in a number of guidelines for the general population. A recent systematic review of these guidelines concluded that the goal for glycemic control should be as low as feasible without incurring undue risk for adverse events (508). These authors concluded that a HbA_1c level <7% is a reasonable goal for many, but not all, patients in the general diabetic population.

While there is evidence in the general diabetic population that strict glycemic control reduces microvascular disease complications, there is less evidence that glycemic control reduces CVD. The United Kingdom Prospective Diabetes Study (UKPDS) and the Diabetes Control and Complications Trial reported nonsignificant trends toward lower CVD with lower HbA_1c levels (509,510). A long-term follow-up of this trial reported that intensive insulin therapy reduced CVD (511). Similarly, in a 10-year follow-up of the UKPDS, there were reduced myocardial infarctions in the sulfonylurea–insulin and metformin intensive-therapy groups (compared to usual care) (512).

Recently, the blood glucose control arm of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) was stopped early, because participants in the intensive-treatment group had experienced increased mortality (513). In ACCORD, 10 251 adults with long-standing (average 10 years) type 2 diabetes, and either heart disease or two or more other risk factors for heart disease, were randomly allocated to target HbA_1c <6.0% vs. standard treatment targeting HbA_1c 7.0–7.9%. Half of the participants in the intensive-treatment group achieved a HbA_1c of <6.4%, and half of the participants in the standard treatment group achieved a HbA_1c of <7.5%. The Data Safety Monitoring Board halted these diabetes control arms of the trial 18 months early, because of a higher mortality rate in the group targeting lower HbA_1c levels. In the intensive-treatment group 257 died, compared with 203 in the standard-treatment group. This was a difference of 54 deaths, or 3 per 1000 participants per year, over an average of almost 4 years of treatment. For both the intensive- and standard-treatment groups in ACCORD, clinicians could use all major classes of diabetes medications available. Extensive analyses did not determine a specific cause for the increased deaths, and there was no evidence that any medication or combination of medications was responsible.

Similarly, the Action in Diabetes and Vascular Disease (ADVANCE) study (514) failed to demonstrate that more intensive glycemic control compared to standard practice reduced CVD events. The ADVANCE study achieved a median HbA_1c of 6.3% in the intensive-management group compared with 7.0% in the standard-intervention group. The results from ACCORD and ADVANCE studies may not apply to patients with type 1 diabetes, patients with recently diagnosed type 2 diabetes or those whose cardiovascular risk is different than the participants studied in ACCORD and ADVANCE. In particular, the results may not apply to patients with CKD or to KTRs. Nevertheless, the results of the ACCORD and ADVANCE trials cast serious doubt on the advisability of targeting low HbA_1c levels to reduce CVD. Additional trials in the general diabetic population may help to determine the optimal strategy for managing diabetes (515).

Kidney transplant recipients with diabetes, especially if the diabetes was the cause of CKD stage 5, often have difficult-to-control diabetes, with advanced autonomic neuropathy causing diabetic gastroparesis and hypoglycemic unawareness. In a RCT comparing intensive glucose control with usual care in 99 KTRs, the incidence of severe hypoglycemia was significantly higher in the intensive glucose-control arm (516). Therefore, it may be more difficult to achieve a HbA_1c level <7.0% without undue risk and burden in many KTRs. In addition, some medications used to treat diabetes may need dose reduction, or should be avoided in patients with reduced kidney function (Table 21).

Table 21.  Pharmacological management of diabetes in KTRs

Class	Drug	Dose adjustment	Drug–drug interactions
CKD, chronic kidney disease; CsA, cyclosporine A; DDP-4, dipeptidyl peptidase-4; GFR, glomerular filtration rate; KTRs, kidney transplant recipients; Scr, serum creatinine. ^aGlibenclamide is the same compound as glyburide (526). ^bGliquidone and glisentide are not currently available in the United States (522). ^cThiazolidinediones may cause fluid retention.
First-generation sulfonylureas	Acetohexamide	Avoid (517)	↑ CsA levels
	Chlorpropamide	↓50% if GFR 50–70 mL/min/1.73 m^2	↑ CsA levels
		Avoid if GFR <50 mL/min/1.73 m^2 (517,518)
	Tolazamide	Avoid	↑ CsA levels
	Tolbutamide	Use with caution (519,520)	↑ CsA levels
Second-generation sulfonylureas	Glipizide	No dose adjustment	↑ CsA levels
	Gliclazide	No dose adjustment	↑ CsA levels
	Glyburide (Glibenclamide)^a	Avoid if GFR <50 mL/min/1.73 m^2 (521