Kidney dysfunction in the recipients of liver transplants

Authors

  • Alan Wilkinson,

    Corresponding author
    1. Department of Medicine, Kidney and Pancreas Transplant Program, David Geffen School of Medicine at UCLA, Los Angeles, CA
    • Department of Medicine, Kidney and Pancreas Transplant Program, David Geffen School of Medicine at UCLA, 200 UCLA Medical Plaza, Suite 365, Los Angeles, CA 90095-1693
    Search for more papers by this author
    • Telephone: 310-794-1781; FAX: 310-794-9718

  • Phuong-Thu Pham

    1. Department of Medicine, Kidney and Pancreas Transplant Program, David Geffen School of Medicine at UCLA, Los Angeles, CA
    Search for more papers by this author

Abstract

Key Points

  • 1Pretransplant kidney function is an important predictor of posttransplant kidney function.
  • 2Chronic kidney disease is present in 20% of liver transplant recipients by 5 years.
  • 3Kidney function is impacted by pretransplant management of the consequences of the hepatorenal syndrome.
  • 4The use of calcineurin inhibitor (CNI)–based immunosuppression is an important factor in the development of chronic kidney disease, and the use of mycophenolic acid– or sirolimus-based immunosuppression with reduced-dose CNI may be of benefit. (Liver Transpl 2005;11:S47–S51.)

Factors Affecting Posttransplant Renal Function

Postoperative acute renal failure occurs in 17% to 95% of patients undergoing orthotopic liver transplantation (OLT). The difference in the incidence reported may be due in part to the wide disparity in the criteria used to define “acute renal failure.” Nonetheless the commonly suggested etiologies of postoperative acute renal failure include acute tubular necrosis secondary to ischemic or toxic insult to the kidneys, preexisting hepatorenal syndrome (HRS), and drug-induced interstitial nephritis.1–3 The former may include prolonged hypotension, sepsis or septic shock, sustained prerenal renal failure, and use of nephrotoxic drugs. Acute renal failure or declining renal function associated with the use of cyclosporine or tacrolimus in the posttransplantation period has been well described and is discussed in more detail below. Preoperative renal dysfunction, delayed liver graft function or primary graft nonfunction, and higher serum bilirubin level have also been variably shown to predispose OLT recipients to postoperative acute renal failure.3–5

Chronic renal insufficiency, or chronic kidney disease, has been reported to occur in 4% to more than 80% of OLT recipients.1, 6–8 The wide range in the incidence reported may be partly due to the difference in the criteria used to define chronic renal failure as well as the difference in the duration of follow-up. The commonly suggested causes or risk factors for the development of progressive chronic kidney disease or end-stage renal disease (ESRD) in long-term survivors of OLT include calcineurin inhibitor nephrotoxicity, pre-OLT HRS, preexisting renal insufficiency, and diabetes mellitus.3, 5, 7, 8 Postoperative acute renal failure, dialysis requirement in the pretransplantation and/or posttransplantation period, hepatitis C infection, and age have also been variably shown to be associated with an increased risk for the development of chronic kidney disease.8–11

In a study conducted by Fisher et al., severe chronic renal failure developed in 4% of patients surviving 1 year or more. Progression to ESRD occurred in nearly half of these patients. In almost all patients who underwent a renal biopsy, the histologic findings were suggestive of cyclosporine toxicity. Specific pathologic changes included vascular obliteration, tubular atrophy, interstitial scarring, and glomerular sclerosis.6

In a retrospective study consisting of 834 recipients of liver alone transplantation performed between June 1985 through the end of 1999, chronic severe renal dysfunction during the study period occurred in 10.3% of the patients, more than 50% of whom had ESRD (severe renal dysfunction was defined as serum creatinine >2.5 mg/dL or ESRD requiring dialysis or transplantation). At 10-year follow-up, the total incidence of severe renal dysfunction rose to 14.4%, with more than 50% of these patients having ESRD (7.9%). The presumptive renal diagnoses of those who developed ESRD were calcineurin inhibitor toxicity (73.3%), nonrecovered hepatorenal syndrome (6.66%), focal segmental glomerulosclerosis (6.66%), progression of underlying renal disease (11.1%), and acute tubular necrosis/amphotericin toxicity (2.22%). For those who survived 13 years beyond the OLT, severe renal dysfunction developed in 18.1%.9 It is likely that the incidence of ESRD of both native kidneys (in OLT recipients) and of renal allografts (in cadaveric kidney liver transplants [CKLT] recipients) increases with time after transplantation.

Abbreviations:

CNI, calcineurin inhibitor; OLT, orthotopic liver transplant; HRS, hepatorenal syndrome; ESRD, end-stage renal disease; CKLT, cadaveric kidney liver transplant; MMF, mycophenolate mofetil; ARF, acute renal failure; RRT, renal replacement therapy.

Cyclosporine and Tacrolimus Nephrotoxicity

Although biochemically distinct, cyclosporine and tacrolimus are 2 potent immunosuppressive agents with similar mechanism of action as well as clinical and pathologic patterns of nephrotoxicity. The various clinical and histologic manifestation of cyclosporine and tacrolimus toxicity may include the frequently occurring functional decrease in renal blood flow and GFR and the infrequently occurring thrombotic microangiopathy.12 Cyclosporine and, to a lesser extent, tacrolimus have been shown to cause an acute, dose-related reversible afferent arteriolar vasoconstriction and “preglomerular-type” renal dysfunction. In liver transplant recipients, the fall in GFR occurs immediately after the introduction of cyclosporine, and this effect is exaggerated when the CNI is administered intravenously. Cyclosporine toxicity usually resolves within 24 to 48 hours of a dose reduction, whereas tacrolimus toxicity may take longer to resolve. Nephrotoxicity may also develop at apparently low levels of both drugs, and some degree of toxicity may be intrinsic to their use.12 In contrast to the acute dose-related reversible decrease in glomerular filtration rate, prolonged use of CNIs can cause chronic interstitial fibrosis and irreversible chronic kidney disease. It has been suggested that CNI-induced interstitial fibrosis involves angiotensin-dependent upregulation of profibrotic molecules such as transforming growth factor beta, endothelin-1, and osteopontin, whereas matrix degradation is inhibited, the latter through inhibition of matrix metalloproteinase activity.13, 14 Intense and prolonged vasoconstriction of the renal microcirculation has also been suggested to be a contributing factor.13

Clinical studies comparing the chronic nephrotoxic effects of cyclosporine vs. tacrolimus in organ transplant recipients have yielded variable and conflicting results. Early reports by Fisher et al. revealed a similar incidence of severe CRF in OLT patients receiving tacrolimus or cyclosporine during the same study period. Creatinine levels at 4 years were comparable in both groups.6 In agreement with Fisher et al., Platz et al. found a similar incidence of late renal insufficiency for cyclosporine- and tacrolimus-treated patients.7 In contrast to the findings reported by Fisher et al. and Platz et al., a number of studies suggest that renal function is better preserved with tacrolimus compared to cyclosporine. In a retrospective study conducted to determine long-term renal function in OLT recipients receiving either cyclosporine or tacrolimus-based immunosuppression at discharge, Pham et al.15 have shown that at 5-year follow-up, nondiabetic OLT recipients treated with tacrolimus had better kidney function than those on cyclosporine for the period reviewed (P < 0.01). Similar analysis for diabetics revealed a comparable trend, but statistical significance was not achieved (data obtained from the United Network for Organ Sharing database between April 1, 1994, and December 31, 1997). In a recent large, population-based cohort study involving more than 32,000 recipients of OLT reported to the Scientific Registry of Transplant Recipients, the risk of chronic renal failure (defined as GFR <29 cc/min, or the development of ESRD) associated with the use of calcineurin inhibitor was also found to be higher among patients treated with cyclosporine than among those who were treated with tacrolimus (relative risk, 1.25; P < 0.001). Interestingly, this difference was not seen among patients with other types of solid organ transplants.10 Recently, some but not all studies suggest that in long-term OLT recipients Neoral cyclosporine monitoring using the 2-hour postdose (C2) preserves renal function without increasing the risk of rejection.16–18 Whether short-term or sustained long-term improvement in renal function can be achieved in OLT recipients receiving Neoral using C2 for dosing determination remains to be determined.

Modification of nephrotoxic immunosuppressive regimens to avoid postoperative acute renal failure and/or chronic renal failure has met with variable results. Although there is no well-defined protocol to prevent or minimize cyclosporine or tacrolimus nephrotoxicity, a number of centers advocate the use of a calcineurin-sparing protocol adjusted for the degree of renal dysfunction. Gonwa et al. had previously suggested withholding cyclosporine for recipients with HRS or for those with moderate to severe renal dysfunction (GFR <30 cc/min) and in its place using azathioprine along with steroids. Induction therapy with an antilymphocyte preparation was used only in cases of prolonged renal dysfunction.19 With the advent of the monoclonal antibodies anti-interleukin 2 receptor antibodies (basiliximab and daclizumab), mycophenolate mofetil (MMF), and sirolimus, independent investigators have developed various immunosuppressive protocols that avoid the nephrotoxic side effects associated with CNI therapy while providing adequate immunosuppression. In a small series consisting of 11 adult transplant recipients (7 heart, 2 liver, 2 heart-renal transplants) with established acute renal dysfunction (defined as an increase in serum creatinine to >25% from baseline), withholding cyclosporine in conjunction with the use of basiliximab or daclizumab resulted in an improvement in renal function without an increased risk of acute rejection.20

In another small series consisting of 19 adults, long-term (>1 year) OLT recipients with renal dysfunction (defined as a decreased creatinine clearance >25% compared with the first month posttransplant), Cantarovich et al. have shown that the introduction of MMF followed by tapering of cyclosporine A to a very low dose (25 mg twice a day) resulted in a significant improvement in renal function. At 1-year follow-up, serum creatinine decreased from 141 ± 24 to 105 ± 22 mmol/L, P = 0.002, and glomerular filtration rate increased from 40 ± 13 to 64 ± 18 mL/min, P = 0.002. However, acute rejection occurred in 29% of the subjects studied, suggesting that this strategy may be associated with a risk of acute rejection.21 In contrast to the results reported by Cantarovich et al. and Neau-Cransac et al. failed to demonstrate any significant improvement in renal function in OLT recipients with biopsy-proven chronic CNI nephrotoxicity despite cyclosporine or tacrolimus withdrawal and institution of either MMF or azathioprine. On the other hand, there was no increase in the incidence of graft rejection.22

Sirolimus is a new and potent immunosuppressant with a mechanism of action and a side effect profile distinct from that of calcineurin inhibitors. When used as base-therapy without a calcineurin inhibitor, sirolimus has been shown to be devoid of nephrotoxicity. In a retrospective study consisting of 16 long-term (>3 years) OLT recipients with different degrees of renal insufficiency ranging from mild (CCr >70 mL/min) to severe (CCr 20-40 mL/min), conversion from cyclosporine or tacrolimus to sirolimus-based immunosuppression resulted in variable improvement in renal function and no rejections at 6-month follow-up.23

Due to the lack of large prospective controlled trials and mixed results obtained from small series of patients, manipulation of immunosuppressive therapy to avoid nephrotoxicity should be best tailored to each patient. In patients with HRS, MMF in conjunction with low-dose tacrolimus and standard steroid therapy appears to be safe and effective (unpublished observation). Although the use of interleukin 2 receptor blocker induction therapy in a calcineurin-sparing protocol has been reported to result in improvement in renal function without an increased risk of rejection, anecdotal reports have suggested that interleukin 2 receptor blockers in combination with MMF or rapamycin increases the risk of viral reactivation and/or the development of more severe hepatitis C recurrence after liver transplantation.24, 25 Interestingly, an increased incidence of hepatitis C viral reactivation associated with interleukin 2 receptor blockers has also been observed at our center, reemphasizing that modification of immunosuppressive therapy should be individualized. Although early studies suggest that MMF may have ribavirin-like antiviral effect and may provide synergism when use with interferon-alfa, its use in the posttransplant period has not been consistently shown to be beneficial or deleterious. Studies on the association between an increased incidence and/or severity of hepatitis C virus recurrence and the use of polyclonal antilymphocyte preparations and/or anti-OKT3 monoclonal antibody have also resulted in contradictory results. In the authors' opinion, these agents should be reserved for patients with delayed graft function and for the treatment of acute rejection. Their routine use in a CNI-sparing protocol as prophylactic therapy is not recommended. In patients with chronic renal insufficiency who have unrelenting renal failure despite drastic CNI dose reduction or withdrawal, the options available to prevent further decline in renal function remain contentious. Although angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers have been suggested to retard the progression of interstitial fibrosis, the role of these agents in halting or alleviating the progression of chronic CNI nephrotoxicity remains to be determined.13

Drug Interactions

Well-substantiated potentiation of renal impairment has been described when amphotericin, aminoglycosides, nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme inhibitors, and/or angiotensin receptor antagonists are used in patients receiving calcineurin inhibitor therapy. More recently, exacerbation of nephrotoxicity has been observed in renal transplant recipients receiving sirolimus and cyclosporine combination therapy. Two phase III clinical trials (The Global and U.S. Rapamune Study Group) have shown that concomitant administration of cyclosporine and sirolimus potentiate cyclosporine-induced nephrotoxicity.26, 27 There has been substantial evidence suggesting that cyclosporine exposure is increased by a pharmacokinetic interaction with sirolimus. In rat animal models, sirolimus has also been shown to increase partitioning into renal tissue to a greater extent than it increases whole blood concentrations.28 When combination therapy is used, a reduction in therapeutic cyclosporine level is desirable, particularly when there is an unexplained rise in SCr level. The pharmacologic interaction between sirolimus and tacrolimus has been less rigorously studied. Coadministration of tacrolimus and sirolimus has been shown to result in reduced exposure to tacrolimus at sirolimus doses of ≥ 2 mg/day.29 However, in recipients of renal transplant, cases of acute renal allograft failure following sirolimus-tacrolimus therapy have been reported.30 Caution should be exercised when combination immunosuppressive agents are used.

The Impact of Acute Renal Failure or Renal Insufficiency on Patient and Allograft Outcome Following Orthotopic Liver Transplantation

Studies on the impact of acute renal failure (ARF) or renal insufficiency on patient and allograft outcomes have yielded variable and conflicting results. This section provides an overview of the literature on the clinical implications of ARF/renal insufficiency on patient and allograft survival in OLT. Based on the literature, the authors' view of the possible impact of renal insufficiency on survival in patients undergoing OLT is discussed.

Early studies by Cuerva-Mons et al. showed that a preoperative serum creatinine level of either less than or greater than 1.72 mg/dL accurately predicted survival or death in 79% of cases.31 Similarly, a strong correlation between preoperative renal dysfunction and postoperative patient survival was later demonstrated by Rimola et al.32 In their series of 102 patients studied, 26 (25%) had renal impairment at the time of OLT. The causes of renal failure were HRS in 21 patients, acute tubular necrosis in 3 patients, and unclassified in 2 patients. Following OLT, 68 patients (67%) experienced renal dysfunction. Twenty-five patients died during the observation period (range, 4-167 days). Renal failure was a major contributory cause of death in 13 (52%). Multivariate risk factor analysis identified serious postoperative infection, graft failure, and preoperative renal function to be independent predictors of mortality.

In contrast to the results reported by Cuerva-Mons et al. and Rimola et al., Gonwa et al. found no difference in graft and patient survival at up to 5 years in non-HRS OLT recipients with different levels of pretransplant renal dysfunction.19 In a large retrospective study, the same group of investigators demonstrated that patients who developed ARF requiring renal replacement therapy (RRT) postoperatively, regardless of the treatment modality, had a significantly lower 1-year survival rate compared with those who were started on RRT preoperatively (41% vs. 73.6%, respectively, P = 0.03).33 Further analysis revealed that mortality was highest among patients who developed acute renal failure requiring postoperative continuous venovenous hemodialysis. The 90-day mortality for those who required hemodialysis both pre- and post-OLT vs. those who required continuous venovenous hemodialysis both pre- and post- vs. those who required only continuous venovenous hemodialysis postoperatively were 25% vs. 27.7% vs. 50%, respectively (P = not significant between groups). Sepsis, primary graft nonfunction, and hepatic artery thrombosis were commonly observed in patients who developed postoperative ARF requiring RRT.

Fraley et al. have previously shown that both pre- and post-OLT ARF were associated with an increased mortality. When ARF was stratified by pre- vs. post-OLT and by subgroups who required hemodialysis vs. CRRT vs. no dialysis, highest mortality rates were seen among patients with postoperative ARF requiring CRRT (primarily in the form of continuous venovenous hemodialysis), a finding similar to that of Gonwa et al. (mortality in ARF pre-OLT: no dialysis vs. HD vs. CRRT: 0% vs. 10% vs. 44%, respectively, compared with ARF post-OLT: 15% vs. 22% vs. 67%, respectively). The authors further demonstrated that the number of comorbid conditions, most notably sepsis, encephalopathy, respiratory failure, and disseminated intravascular coagulation correlated best with a worse outcome.34

An association between postoperative acute renal failure requiring RRT and increased morbidity and mortality was also demonstrated by Gainza et al.35 In their series consisting of 259 consecutive liver transplantation performed in 251 patients, 4 of whom underwent combined liver-kidney transplantation, the mortality rate of patients requiring RRT was 52.1%, compared with 6.77% of that of the total population studied (P < 0.00001). A higher Child-Turcotte-Pugh score and previous renal insufficiency were identified as strong risk factors for the development of postoperative acute renal failure. Other risk factors included the use of calcineurin inhibitors, sepsis, liver dysfunction, and nephrotoxic antimicrobials, among others.

In conclusion, although the literature on the impact of renal insufficiency and patient and allograft survival are inconsistent, the commonly identified factors predicting a worse outcome appear to be renal failure associated with sepsis and/or renal failure requiring RRT particularly among those who required CRRT(commonly performed because of hemodynamic instability associated with sepsis as a major comorbid condition) in the postoperative period.

Ancillary