SEARCH

SEARCH BY CITATION

Keywords:

  • CNI toxicity;
  • cystic fibrosis;
  • diabetes;
  • lung transplantation;
  • oxalate nephropathy;
  • renal biopsy

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

We have analyzed the evolution of renal status beyond the perioperative period in patients with cystic fibrosis (CF) undergoing lung transplantation and presented histological analysis of 15 patients biopsied for an episode of accelerated renal function loss (RFL).

Episodes of accelerated RFL after the perioperative period occurred in 32.5% of patients and significantly raised the risk of end-stage renal disease (ESRD) (p < 0.001). The histologic lesions associated with these episodes differed according to the time of onset. Early onset (10 cases) was associated with tubulointerstitial lesions in the form of oxalate nephropathy (50%) and/or a pigmented tubulopathy (80%). This latter was correlated with treatment with antiviral agents (p = 0.002) and aminoside and glycopeptide antibiotics (p = 0.03) administered in the month preceding biopsy. Lesions in late episodes of accelerated RFL (5 cases) were principally vascular: arteriosclerosis and arteriolosclerosis (p = 0.007, p = 0.00002), correlated with diabetic glomerulosclerosis or focal segmental glomerulosclerosis in the absence of prominent diabetic changes. Specific calcineurin-inhibitor nephrotoxicity was present in 93.3% of biopsies associated with thrombotic microangiopathy in 46.7% of cases.

The identification of specific etiologies of progressive kidney disease in patients with CF after lung transplantation should permit more effective post-transplant care of these patients.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

End-stage lung disease causes about 80% of deaths among patients with cystic fibrosis (CF) (1). Lung transplantation has become the most frequent and aggressive therapy available for end-stage lung disease in patients with CF (2). Although the effects of lung transplantation on survival and quality of life in patients with CF are still debated (3), retrospective studies have found a survival benefit of lung transplantation for these patients (80.6% at 1 year and 65.2% at 5 years) (4). According to the United Network of Organ Sharing, since 2000 between 140 and 200 patients with CF have been transplanted each year in the United States (5). With increased survivals, the problem of evolution to chronic kidney disease (CKD) has become a major problem. At 5 years, the cumulative incidence of CKD among recipients of lung allografts is 15.8% and continues to increase over time (6). End-stage renal disease (ESRD) occurs in 10% of nonrenal solid organ recipients by 10 years (7–9). Nonetheless, till now data on the renal pathology after lung transplantation in patients with CF have been limited.

Most authors have attributed the development of chronic kidney disease posttransplantation to calcineurin-inhibitor (CNI) toxicity (6,10), a phenomenon clearly demonstrated in cardiac (11), hepatic (12) and renal transplantations (13,14). Other studies have stressed the importance of the perioperative period in the renal evolution of lung-transplant patients, identifying prolonged mechanical ventilation and the utilization of amphotericin B as predictive factors for acute renal failure in the perioperative period (15). But CF patients have further complicating factors, such as diabetes and heavy antibiotic therapy because of the high incidence of bacterial infection. The renal manifestations of diabetes in mucoviscidosis patients after lung transplantation remain as yet undetermined despite a recent study showing that approximately 30% of CF patients receiving lung transplants are already diabetic and another 20% develop diabetes posttransplant (16). Also poorly understood is the role of the antibiotics used in these patients. The only renal pathologic case reports available on these patients have shown an acute tubular toxicity imputed to different antibiotics (17–20). We recently reported the first three observations of irreversible acute renal failure associated with renal oxalosis in nonrenal solid organ transplant patients (21). Two patients had CF, one with lung transplant and the other with combined lung-liver transplant.

This study analyzes the evolution of kidney disease beyond the perioperative period in 15 CF patients receiving lung transplants and presents the histologic analysis of the patients biopsied for an episode of accelerated kidney disease progression.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Patients

This study includes those patients of over 16 years of age with CF who received either a lung, lung-liver or lung-heart transplant who survived the perioperative period (3 months) with meaningful renal function (glomerular filtration rate (GFR)>15 mL/min/1.73m2). Ninety mucoviscidosis patients were transplanted in the thoracic surgery unit of the Hôpital Européen Georges Pompidou in Paris between January, 1998 and December, 2005. Finally included in this study were 77 patients, 67 of whom received a double lung transplantation, nine a combined liver-lung transplantation, and one a combined heart-lung transplantation. All transplant procedures required the use of cardiopulmonary by-pass. Thirteen recipients who died (11 patients) or remained on dialysis (2 patients) at the end of the perioperative period were excluded: 6 died of primary dysfunction of the lung transplant, 1 of a myocardial infarct, 1 of ischemia of the digestive tract and 3 of sepsis with multiorgan failure; 2 patients remained on dialysis, one of them owing to renal cortical necrosis during extracorporeal circulation. This group was followed through December, 2006, the mean follow-up being 40.5 ± 26.6 months (3–102 months).

The medical records were reviewed by two nephrologists to identify episodes of accelerated RFL defined as a loss of GFR >30% over 6 months. We excluded patients with overdosage of CNI (tacrolimus serum levels to trough (T0) >15 ng/mL and cyclosporine serum levels 2 h after the morning dose >1500 ng/mL), pharmacologically proven aminoglycoside overdosage or administrations of nonsteroidal anti-inflammatory drugs or iodinated radiocontrast products at the time of the episode of accelerated RFL.

Among the 25 patients with fast progressive RFL, who constituted our study group, 15 had a renal biopsy. In the remaining 10 patients, renal biopsy was not performed either because of multiorgan failure related to sepsis (5 patients), or because renal failure responded to electrolyte reequilibration (3 patients), or because clinical data established a diagnosis of acute pyelonephritis in three patients, two of whom had associated aminoside toxicity.

Immunosuppression protocols

All patients received immunosuppression according to classical protocols, with induction of polyclonal anti-T-lymphocyte globulin (Thymoglobuline, Genzyme). Maintenance immunosuppression consisted of tacrolimus (Prograf®, Astellas) or cyclosporine (Neoral®, Novartis), mycophenolate-mofetil (MMF) (CellCept®, Roche) and steroids.

Clinical monitoring

Baseline data at the time of transplantation and at 3 months posttransplant were obtained for all patients, including age, sex, past medical history and serum creatinine (SCr). Follow-up data including SCr and cyclosporine/tacrolimus levels were collected at clinical visits every 3 months. At the time of renal biopsy, the following data were recorded: past medical history, comorbidities (diabetes, hypertension), medications (current and in the preceding month), blood pressure, SCr, proteinuria (expressed in g/day), urine sediment.

Hypertension was defined as a systolic blood pressure >140 mmHg, a diastolic blood pressure >90 mmHg, or by the need for antihypertensive medication, and clinical diabetes by the need for hypoglycemic agents or insulin. Nephrotic range proteinuria was defined as proteinuria >3 g/day. GFR was estimated using the abbreviated MDRD equation (22) and chronic kidney disease was classified according to KDIGO guidelines (23).

Renal biopsy processing

Renal biopsies were reviewed by two renal pathologists, blinded to the clinical and laboratory data. All contained at least nine glomeruli. They were stained by standard methods for light microscopy analyses. Immunofluorescence (IF) studies were performed using polyclonal antibodies to IgA, IgG, IgM, fibrin, albumin, C3, C4, C1q (Behringwerke, Marburg, Germany) and anti-kappa and anti-lambda light chains monoclonal antibodies (Dako, Glostrup, Denmark). The presence of crystals of calcium oxalate (CaOx) was sought under polarization.

Histologic lesions

A variety of basic lesions were systematically evaluated on each renal biopsy. Glomeruli were categorized as normal, focal segmental glomerulosclerosis (FSGS), diabetic glomerulosclerosis, ischemic and sclerotic. Results were expressed in percentages of total glomeruli. Diabetic glomerulosclerosis was sought by light microscopy and on IF by specific linear staining for IgG along glomerular basement membranes (GBMs). Arteries were evaluated according to the extent of arteriosclerosis in the form of intimal fibroplasia and thickening, often with medial atrophy. Arterioles were evaluated for the presence of medial necrosis and/or hyaline globules as a sign of CNI toxicity and of subendothelial hyaline deposits typical of simple hyaline arteriolosclerosis, as seen in diabetes and hypertension. Thrombotic microangiopathy (TMA) was also evaluated in glomeruli and in arterioles. The degree of interstitial fibrosis was estimated as a percentage of the parenchyma. Tubular atrophy, tubular epithelial or luminal crystals, tubular vacuolization were evaluated semiquantitatively.

Oxalate nephropathy was defined as tubulopathy involving primarily the proximal tubules, with numerous cells filled with abundant characteristic crystals of CaOx. Occasional crystals laid free in the tubular lumens, but the majority was intracellular, often bulging into and occluding the lumen, but with a thin layer of cytoplasm overlying the crystals, usually with intact brush border.

Pigmented tubulopathy was characterized by the intracellular accumulation of pigment granules with the characteristics of lipofucsin, since the granules were positive on silver stain, autofluoresced on fluorescence microscopy, and were negative for iron on Perl's stain, ruling out hemosiderin. It occurred in intact tubules, proximal, distal and thick limbs of Henle, initially as very fine granules, so numerous in some cases as to give a brownish cast to the tubules on Masson stain.

Statistical analysis

Statistical methods employed included the Student t test and Pearson product moment correlations for continuous variables and Spearman rank order correlations for categorical variables. A p-value of ≤0.05 was considered significant, although given the small sample size (15 patients) values of <0.10 are retained for reader reference. Software used was Statistica, release 6 (Statsoft, Tulsa, OK).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Baseline characteristics of CF patients undergoing lung transplantation

Seventy-seven CF patients (44 males, 33 females) survived the perioperative period following lung transplantation with meaningful renal function (GFR >15 mL/min/1.73m2). Their mean age was 23 ± 6.6 years (16–45 years). The baseline SCr pretransplantation was 59 ± 30.7 μmol/L (22–230 μmol/L) with a GFR estimated at 169.8 ± 71.6 mL/min/1.73m2. At 3 months posttransplantation, SCr was 95.8 ± 28.2 μmol/L (40–151 μmol/L) and estimated GFR 91.6 ± 46 mL/min/1.73m2. The mean loss of GFR over the first 3 months was 83.5 ± 64 mL/min/1.73m2.

The prevalence of episodes of accelerated RFL after the 3-month perioperative period was 32.5%. Fifteen of the 25 cases with a diminution of GFR > 30% over a 6-month period had a renal biopsy.

Clinical characteristics at the time of renal biopsy in CF patients undergoing lung transplantation

The mean age of the 15 patients (11 males, 4 females) with renal biopsies was 23 ± 6 years (17–37 years). Ten had received lung transplants, 4 combined lung-liver transplants, and 1 heart-lung transplantation. Their baseline SCr pretransplantation was 69 ± 50.7 μmol/L (30–230μmol/L) with an estimated GFR of 168.6 ± 81.9 mL/min/1.73m2, comparable to the general population of CF patients undergoing lung transplantation, as was their estimated GFR at 3 months posttransplantation at 90.5 ± 22.3 mL/min/1.73m2.

The renal biopsies were performed at 47 ± 51 months (4.5–174.6 months) posttransplantation. Clinical diabetes was present in 12 (80%), and 13 (86.7%) were hypertensive at biopsy. Mean GFR at the time of biopsy was 32.2 mL/min/1.73m2. Proteinuria was below 1g/day in 10 cases (66.7%), and only 2 cases (13.3%) had nephrotic range proteinuria. Clinical data are presented in Table 1.

Table 1.  Clinical and laboratory data at the time of renal biopsy for patients with CF undergoing lung transplantation
Characteristics
 N15
 Males73.3%
 Proteinuria (g/d or g/g creatinine)1.7 (0.2–5.7)
 Proteinuria ≤ 1 g/d or g/g creatinine66.7%
 Proteinuria ≥ 3 g/d or g/g creatinine13.3%
 Hematuria13.3%
 Serum creatinine (μmol/L)294.9 (80–650)
 GFR (mL/min/1.73m2)32.2 (10–116)
 GFR < 15 mL/min/1.73m226.7%
 Hypertension86.7%
 Clinical diabetes80%
 Calcineurin-inhibitor therapy100%
 Tacrolimus86.7%
 Cyclosporine A13.3%

Thirteen of the 15 patients (86.7%) were under antibiotic therapy at biopsy or in the month preceding it: bactrim (33.3%), quinolones (40%), macrolides (33.3%), glycopeptides (33.3%), penicillins or cefalosporins (33.3%) and aminosides (26.7%), and 8/15 (53.3%) were under antiviral treatment: valacyclovir (46.7%), gancyclovir followed by valgancyclovir (13.3%). Antifungal agents were employed in 6/15 (40%) voriconazole (33.3%) and amphotericin B (6.7%).

At the end of follow-up

The 77 CF lung transplant patients were followed a mean of 40.5 ± 26.6 months (3–102 months) posttransplantation. Fifteen patients (19.5%) died, 4 among them hemodialyzed in a context of acute multiorgan failure. Eight patients (10.4%) went on to ESRD with dialysis, of whom 3 received renal transplants. Those patients not on dialysis at the end of follow-up had a mean SCr of 115.7 ± 49.3 μmol/L (37–280μmol/L) corresponding to an estimated GFR of 77.5 ± 40.2 mL/min/1.73m2.

The occurrence of an episode of accelerated RFL after the third month posttransplantation significantly increased the risk of evolution to ESRD (p < 0.001). All eight patients on dialysis at the end of follow-up had renal biopsies at the time of the episodes of accelerated RFL. Their clinical and laboratory characteristics at biopsy are presented in Table 2. For the seven biopsied patients not on dialysis, the GFR at the end of follow-up was estimated at 44 ± 12.5 mL/min/1.73m2. The only element definitely associated with poor outcome of renal function was the SCr at the time of biopsy.

Table 2.  Clinical and laboratory data at the time of kidney biopsy for the eight patients reaching ESRD
NAgeTx TypeTime to Bx (mo)SCr pre-Tx (μmol/L)SCr Bx (μmol/L)Proteinuria (g/day)HTADiabetesCalcineurin inhibitorAntibiotics*Other drugs*
  1. LTx = lung transplantation; L-LTx = combined liver and lung transplantation; H-LTx = combined heart and lung transplantation; HTA = hypertension; *= antibiotics and drugs used at time or the month before the biopsy; G = glycopeptide; Q = quinolone; A = aminoside; M = macrolide; P = penicillin; B = bactrim; AV = antivirals; AF = antifungals; CsA = cyclosporine A.

119L-LTx  3,2230510ND++tacrolimusQAF + AV
233LTX  3,9 704500.5++tacrolimusG, Q, A, MAF + AV
317L-LTx  6,1 45 800.3+tacrolimusP, G
418L-LTx  7,7 435000.5++tacrolimusG, P, QAV
522LTx 23,8 622060.2+tacrolimusG, A, MAV
622LTx 72,3 452402++CsABAF
718LTx 84,4 555305.7+tacrolimus
818H-LTx174,6 582121+CsA

Renal histopathologic lesions

The percentage of patients having each of the major lesions is: diabetic glomerulosclerosis (20%), FSGS (26.7%), arteriosclerosis (53.3%), CNI toxicity (93.3%), TMA (46.7%) and tubulopathy (66.7%) including oxalate nephropathy (33.3%) and/or pigmented tubulopathy (53.3%). All biopsies considered, interstitial fibrosis was estimated at 38 ± 27% and the percentage of sclerotic glomeruli 26 ± 31%. The histologic characteristics of the eight patients progressing to ESRD are seen in Table 3. The only histologic variable associated with progression to ESRD was oxalate nephropathy. Summation of the major causes of reversible and irreversible renal dysfunction is seen in Table 4.

Table 3.  Renal morphologic features of patients reaching ESRD
 %Scl. GlDiab. Glscl.1FSGS2Pigmented tubulopathy1Oxalate nephropathy1Arterio sclerosis2Arterial deposits2Arteriolo sclerosis2TMA2Interstit. infl.2Interstit. fibrosis2Principal diagnosis
  1. 1= Categorized as present (1) or absent (0).

  2. 2= Graded semiquantitatively on scale 0–4+.

1 0000100000.51Oxalate nephropathy
2140111211111Pigmented/oxalate nephropathy
3 80011000.500.51Pigmented/oxalate nephropathy
4 000100.5000.50.52Pigmented tubulopathy
5 000101.50.51.5013Pigmented tubulopathy
6781000213112Vasc/glom disease
78403002231.52.53Vasc/glom disease
850020032220.52Vasc/glom disease
Table 4.  Comparison of patients reaching ESRD with patients with relative preservation of function
 ESRD 8 patientsPreserved function 7 patientsp
  1. 1By Chi-square testing.

Worse in patients with ESRD
 SCr at biopsy474 ± 226215 ± 48 0.01 
 Oxalate nephropathy3/80/70.071
 Pigmented tubulopathy4/84/70.781
 TMA—arterial or arteriolar6/83/70.201
 Anticalcineurin arteriolopathy1.93 ± 1.121.00 ± 1.000.11 
Worse in patients with preserved function
 Ischemic glomeruli0.62 ± 0.521.21 ± 0.570.06 
 Clinical diabetes5/87/70.071

The histologic lesions associated with episodes of RFL differed according to the time of onset. Early onset was associated with tubulointerstitial lesions (10 cases) in the form of oxalate nephropathy and/or pigmented tubulopathy. The histologic lesions associated with late development of episodes of accelerated RFL (5 cases) were principally vascular and glomerular (Table 5).

Table 5.  Correlations between renal morphologic lesions and the time of biopsy in 15 mucoviscidosis patients receiving lung transplants
Morphologic parameterrp
  1. 1Pearson product moment correlations.

  2. 2Spearman rank order correlations.

  3. 3Among 12 clinically diabetic patients.

% Sclerotic glomeruli1 0.67360.006
Diabetic glomerulosclerosis3,1 0.80770.001
FSGS1 0.63650.01 
Interstitial fibrosis1 0.47510.07 
Arteriosclerosis1 0.64020.01 
Arterial deposits1 0.74110.001
Arteriolosclerosis1 0.70690.003
Anticalcineurin arteriolopathy1 0.2131NS
Pigmented tubulopathy2−0.4021NS
Oxalate nephropathy2−0.69440.004
Pigmented and/or oxalate nephropathy2−0.81830.002

Diabetic nephropathy:  Linear staining of GBMs typical of diabetes was found in 6 patients (40%) of whom 3 (20%) had histologically evident glomerular lesions (diffuse glomerulosclerosis with mesangial hypertrophy in 2 with frank nodular glomerulosclerosis in the third). These histologic lesions were correlated with hyaline arteriolosclerosis (p = 0.003), arterial hyaline deposits (p = 0.004) and FSGS (p = 0.04).

Focal segmental glomerulosclerosis:  FSGS was present in four patients. Three of these had linear IF diagnostic of diabetes.

Arteriosclerosis:  Arteriosclerosis was present in 12/15 patients and was severe (≥2+) in 5 despite their young age. Of the 12 patients, 10 had arterial hyaline deposits, and all those with arterial hyaline deposits had arteriolar deposits as well. Arteriosclerosis was associated only with the presence of hypertension (p = 0.04).

Calcineurin-inhibitor nephrotoxicity:  This was diagnosed specifically on the basis of arteriolopathy. It was present in 93.3% of biopsies, primarily in afferent arterioles but also in efferent arterioles, consisting of hyaline deposits in the media representing necrotic myocytes.

Thrombotic microangiopathy (TMA):  TMA was present in seven patients with organizing thrombotic lesions of the preglomerular arterioles leading to fibrotic and occlusive lesions. All of cases of TMA also had CNI nephrotoxicity lesions (Figure 1). The seven patients with TMA were hypertensive receiving a mean of 2.5 antihypertensive drugs per day. None of them showed biologic signs of intravascular hemolysis, nor of thrombocytopenia. The activity of the cleavage protease of von Willebrand factor was moderately diminished at 53% (40–70%) in the absence of antinuclear factor.

image

Figure 1. TMA and calcineurin-inhibitor nephrotoxicity. Two arterioles (arrows), one showing necrotic myocytes and early thrombus formation, the smaller one showing organizing thrombus. Masson trichrome stain, × 300.

Download figure to PowerPoint

Oxalate nephropathy:  Oxalate crystals were present in tubular epithelium in 9 patients and in 5 of them qualified as an oxalate nephropathy with tubular luminal crystals as well (Figure 2). The three cases with the greatest tubular epithelial load of crystals showed substantial tubular luminal crystals and peculiar vacuolization of Bowman's capsular epithelium, and one had pigmented tubulopathy as well, but they had only modest degrees of chronic lesions such as interstitial fibrosis. Clinically, they were characterized by early posttransplant RFL necessitating biopsy (90 ± 24 days), with high SCr (670 ± 121 μmol/L vs. 274 ± 141 μmol/L, p = 0.0007). All three patients progressed to ESRD.

image

Figure 2. Oxalate nephropathy. Intracellular oxalate crystals in tubular epithelium (arrow), bulging into lumen but with thin overlying layer of intact cytoplasm. Masson trichrome stain, × 400.

Download figure to PowerPoint

Pigmented tubulopathy:  Pigmented tubulopathy was found in eight patients (Figures 3 and 4). They were characterized by a shorter time to manifestations eliciting biopsy (673 ± 536 vs. 2265 ± 1875 days, p = 0.038) and by their correlation with drugs administered in the month preceding biopsy: antiviral agents valacyclovir in 7/8 (p = 0.002), aminosides in 4 (p = 0.03), glycopeptides in 4 (p = 0.03). The number of anti-infectious agents administered in the month preceeding the diagnosis of pigmented tubulopathy was much greater than in the other patients: 4.38 ±1.06 versus 1.43 ± 1.13 agents, p = 0.0002.

image

Figure 3. Pigmented tubulopathy—early. There are isolated small Ag-positive granules in intact tubules (upper left). These coalesce into larger accumulations of small granules in individual cells, particularly in tubules with early atrophy (center, upper right). Marinozzi silver stain, × 400.

Download figure to PowerPoint

image

Figure 4. Pigmented tubulopathy—advanced. While some relatively intact tubules show fine granules (upper center), granules in atrophic tubules have coalesced into fewer but much larger structures. Occasional silver-positive granules are seen in the interstitium. Marinozzi silver stain, × 400.

Download figure to PowerPoint

Categorization of cases

Analysis reveals the patients in this study separated into two groups, 10 patients with tubulopathy, presenting generally in the first 1–2 years following transplantation, and five patients with chronic vascular and glomerular lesions, biopsied 5 to 14 years after lung transplantation (Table 6).

Table 6.  Dividing lesions into tubulopathy and other cases
 Tubulopathy 10 patientsOther cases 5 patientsp
  1. 1Differences in date of TX = 1727 days = 4.73 years.

  2. 2Differences in time to biopsy = 2579 days = 7.06 years.

  3. 3By chi-square testing.

Sclerotic glomeruli (%)10 ± 2058 ± 240.0013 
Diabetic glomerulosclerosis00.6 ± 0.50.003  
FSGS0.1 ± 0.31.8 ± 1.80.01   
Pigmented tubules1.4 ± 0.80.6 ± 0.50.07   
Ag + pigment in tubules2.2 ± 1.01.3 ± 0.70.10   
Pigment + Ag + granules0.8 ± 0.400.001  
Tubular epithelial crystals0.3 ± 0.40.NS
Hyaline casts1.3 ± 0.62.0 ± 0.60.06   
Interstitial fibrosis1.8 ± 0.82.6 ± 0.50.07   
Arteriosclerosis1.0 ± 1.02.6 ± 0.80.007  
Arterial deposits0.4 ± 0.42.1 ± 0.90.0001 
Arteriolosclerosis0.8 ± 0.52.8 ± 0.60.00001
Date transplant137851 ± 634  35854 ± 853  0.001  
Time to biopsy2556 ± 5333135 ± 14010.0002 
Proteinuria at biopsy (g/d)0.66 ± 0.572.54 ± 1.950.016  
Aminosides4 patients0 patients0.103  
Glycopeptides4 patients0 patients0.103  
Antiviral agents8 patients0 patients0.0033

The tubulopathy group consisted of 5 cases with pigmented tubulopathy, 2 cases with oxalate nephropathy and 3 cases with features of both. They were accompanied by rather striking clinical differences. Those with tubulopathy were transplanted an average of 4.7 years more recently than the other cases and came to biopsy much more rapidly (18.5 ± 17.8 months vs. 104.5 ± 46.7 months, p = 0.0001).

The five remaining patients had more severe proteinuria (2.54 ± 1.95 vs. 0.66 ± 0.57 g/d), in accordance with their higher rate of FSGS (p = 0.009). Histologic diabetic glomerulosclerosis was confined to this group, being found in all three diabetic patients. This group also had much worse arteriosclerosis (p = 0.007), arterial hyaline deposits (p = 0.0001) and hyaline arteriolosclerosis (p = 0.00001) as well as greater interstitial inflammation, interstitial fibrosis and pseudothryoidization of tubules.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

As with earlier studies, we observed a considerable loss of renal function after lung transplantation in CF patients. Moreover, this study describes and analyzes for the first time the renal pathologic features in patients presenting progressive loss of renal function, at risk for ESRD. The histologic lesions associated with rapid RFL differ according to the time of its appearance after transplantation. Early loss (18.5 ± 17.8 months) of function is associated with tubulointerstitial lesions, either as oxalate nephropathy, pigmented tubulopathy or a combination of both. The lesions associated with late episodes (104.5 ± 46.7 months) of accelerated renal functional decline are primarily chronic vascular and glomerular lesions. Some lesions, such as CNI nephrotoxicity, are comparable to those seen in other solid organ transplants. Other lesions appear to be more specifically linked to CF, such as oxalate nephropathy, pigmented tubulopathy and the precocious appearance of the glomerular and arteriolar lesions of diabetes to which these patients are subject.

In our cohort, 10.4% of the patients surviving the perioperative period with satisfactory renal function were on dialysis at the end of follow up (40.5 ± 26.6 months) and an additional 29.9% had a GFR < 60 mL/min/1.73m2. These values are at the high end of current results for cardiac and lung transplantation, which reveal that 3–10% of patients reach ESRD (6,7,24). This excess of ESRD after pulmonary transplantation in patients with CF suggests there are specific effects linked to the CF. Two studies (25,26) have demonstrated that both the early renal functional decline and the long-term rate of renal function loss posttransplant in CF patients were more pronounced than in patients with other pulmonary diagnoses.

In our study, a low GFR at the time of diagnosis was associated with progression to ESRD, reemphasizing the importance of nephrologic monitoring posttransplantation in these patients and the importance of creatinine-based estimates of GFR. The Levey estimation (22) is currently considered to be the most sensitive method for detecting loss of GFR and is the preferred method for estimating renal function following pulmonary transplant (27). Numerous retrospective studies have identified risk factors for the development of progressive chronic renal disease following pulmonary or cardiac transplantation (6,7,9,26,28,29). Among the risk factors specific for transplantation is the level of pretransplant function as well as that after the perioperative period. In our study, the pretransplant GFR and that at 3 months in those patients later presenting rapid progression was similar to that of the general population of CF transplant patients. There is, nonetheless, a sharp reduction of GFR during the perioperative period, which we found to be 34.1 ± 28.1% in our group. Two patients arrived at ESRD 3 months posttransplant.

In this study, 66.7% of the patients biopsied for accelerated kidney disease progression, and 62.5% of the patients going on to ESRD, had tubulopathy. We have defined two separate histologic types of tubulopathy, appearing to have different etiologies and pathophysiologic mechanisms.

Oxalate nephropathy was present in 33.3% of those biopsied for progressive kidney disease and 37.5% of those ending in ESRD. Hyperoxaluria is very common in patients with CF, with frequent urinary CaOX calculi and medullary nephrocalcinosis (30). It is principally due to malabsorption of fatty acids as a result of pancreatic insufficiency, leading to increased oxalate bioavailability in the colon. However, recent studies (31,32) have stressed the role of diminution or absence of Oxalobacter formigenes in digestive hyperabsorption of oxalate. The importance of O. formigenes was stressed by the demonstration of a correlation between the existence of hyperoxaluria and the absence of this organism in patients with CF (31). A likely reason is that CF patients receive prolonged antibiotic treatment capable of interfering with colonic flora. Retention of CaOX on the surface of proximal tubular cells must be preceded by some stress to the epithelium (33). There are multiple potential stressors to the posttransplant epithelial cell: hemodynamic factors, hypoxia, antibiotics or other nephrotoxic agents. The recognition that oxalate nephropathy is an etiology of progressive renal disease in CF patients receiving lung transplants permits intervention at each of the three stages leading to oxalate nephropathy.

Pigmented tubulopathy was found in 53.3% of patients biopsied for progressive kidney disease and in 50% of those going on to ESRD. The pigment behaved as lipofucsin, appearing as golden granules, silver-positive and iron-negative on special stain, and showing typical autofluorescence on fluorescence microscopy. Such pigment is classically thought of as ‘aging’ pigment, thought to be the result of abnormal degradation of lysosomes and is found in limited quantities in normal aging kidneys and in chronic renal diseases of a variety of causes. What distinguishes it here is its precocious appearance in young patients in such large quantities that still-intact tubular cells often have a brownish aspect and are filled with small positive granules on silver stain (Figures 3 and 4). It seems likely that the pigment here is not toxic per se, but is simply the result of, and marker for, severe tubular metabolic dysfunction. As seen below, it functions coherently as a clinical entity.

Accumulation of lipofucsin in the kidney as a function of aging has been confirmed in the rat (34). It is also seen experimentally as a function of renal toxicity of a variety of agents (35,36). In man, however, renal lipofucsin accumulations have only been noted in association with phenacetin abuse in analgesic abuse nephropathy in the older literature (37). Our study marks the first time that lipofucsin has been documented as an element of anti-infectious agent administration. No single agent can be incriminated here, since no one agent was taken by all of the patients with pigmented tubulopathy. However, the association between this tubulopathy and the antibiotic and/or antiviral treatment in the month preceding the biopsy is very strong. Although this association does not prove the nephrotoxicity of these products, it underlines the possible impact of these treatments on renal function, particularly in CF patients with lung transplants who classically receive a large number of drugs. The use of these possibly nephrotoxic drugs, either singly or in combination, necessitates close monitoring to assure optimal efficacity while maintaining minimum renal risk.

Diabetes

In our study, 12 patients (80%) were clinically diabetic. Because of this large percentage, there were no significant morphologic or clinical differences between diabetics and nondiabetics. Of the 12 diabetics, 6 had the linear IgG GBM staining typical of diabetes. They differed from those that were negative, only by the presence in three of diabetic glomerulosclerosis, with no other significant differences in clinical or histologic parameters. The three patients with diabetic glomerulosclerosis had worse vascular lesions (arteriosclerosis, arteriolosclerosis). They did not differ significantly, however, in terms of SCr or proteinuria and only one patient with diabetic glomerulosclerosis finished in ESRD.

Up until now, because of their shortened life span, microvascular diabetic complications have only been sporadically reported in patients with CF-related diabetes (38,39). A recent study of such patients, whether they had lung transplants or not, has shown, for the first time, an elevated incidence of diabetic retinopathy (27%), approaching the levels found in type I diabetics with the same duration of diabetes (40). These results underline the importance of as close diabetologic monitoring in these CF diabetic patients as in those with more typical type I diabetes.

Calcineurin-inhibitor nephrotoxicity

This is a well-described clinical and histologic entity in solid organ transplantation (10–14), which may be responsible for progressive degradation of renal function following transplant. Our renal biopsy study among pulmonary recipients with fast progressive RFL has shown that CNI-related injury is a common finding. It was present in 93.3% of biopsies, associated with TMA in 46.7% of cases. One unique feature in our series was the large percentage of patients with arterial deposits 10/15 (66.7%) patients, of whom 7 were diabetic. Arterial (as opposed to arteriolar) deposits are infrequent in typical CNI nephrotoxicity, and it is tempting to attribute a role for diabetes in this increase.

The importance of CNI nephropathy in pulmonary recipients creates the opportunity for the nephrologists to initiate discussion about lowering CNI exposure to the minimum level for effective immunosuppression. Lowering cumulative CNI exposure has beneficial effects on renal function and may lead to improvements in blood pressure, glucose regulation, volume status and dyslipidemia (41). Different strategies of CNI-sparing therapy were employed to prevent progressive renal dysfunction posttransplant. Only a few small series of cardiac or pulmonary transplants have reported the diminution of CNI therapy following addition or replacement by sirolimus with variable results (42–44). Studies in cardiac transplantation are focusing on an approach using MMF with small doses of cyclosporine (45,46). Additionally, the use of renin-angiotensin blockers might be an effective option in slowing the progression of renal disease in these hypertensive patients with extensive vascular lesions.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

This study describes three patterns of histologic lesions with likely different etiologies responsible for accelerated decline in renal function after the perioperative period following pulmonary transplantation in patients with CF. The recognition of these different patterns could improve the posttransplant management of at-risk patients. Early nephrologic consultation would permit establishment of the diagnosis of the renal lesion and point the way to therapeutic options. The objective of this more specific approach is to minimize kidney injury and to slow progression of renal disease following pulmonary transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References