Recurrence of haemolytic-uraemic syndrome in renal transplants: a ...

Nephrol Dial Transplant ( 1997) 12: 1425–1430

Nephrology Dialysis Transplantation

Original Article

Recurrence of haemolytic–uraemic syndrome in renal transplants: a single-centre report R. B. Miller1, B. A. Burke2, W. J. Schmidt3, K. J. Gillingham3, A. J. Matas3, M. Mauer1 and C. E. Kashtan1 University of Minnesota, Departments of 1Pediatrics, 2Laboratory Medicine and Pathology and 3Surgery

Abstract Background. The incidence of recurrence of haemolytic–uraemic syndrome ( HUS) in renal allografts appears to vary by centre, with the highest rates reported from the University of Minnesota. It is possible that the high rate of HUS recurrence at this institution reflects a transplant population skewed towards patients with a form of HUS that is more likely to recur in the allograft. Methods. This study examined whether the initial episode of HUS in the native kidneys was preceded by a diarrhoeal prodrome (‘classical HUS’) or not (‘atypical HUS’), and evaluated transplant outcomes in 24 patients who received 36 transplants at the University of Minnesota between 31 May 1972 and 31 December 1994. Results. Eighteen of the 24 patients had atypical HUS, three had classical HUS, and in three patients the presence or absence of a diarrhoeal prodrome could not be determined. Recurrent HUS, defined as microangiopathic haemolytic anaemia, thrombocytopenia, renal insuciency, and allograft biopsy findings compatible with HUS, occurred 16 times in 14 grafts in 11 patients. Nine of these patients had atypical HUS, one had classical HUS, and in one the nature of the prodrome could not be determined. Eleven of the 14 initial recurrences took place within 2 months of transplant. Recurrence was not more frequent in patients who received cyclosporin or antilymphocyte preparations. Actuarial analysis using matched controls showed poorer graft survival in patients with a primary diagnosis of HUS (P=0.007), due to the high frequency of graft loss in HUS patients with recurrence. Conclusion. Based upon these data and a review of the literature, it can be concluded that the risk of recurrence of HUS in the allograft is confined almost entirely to patients with atypical forms of HUS.

Correspondence and o
print requests to: Cli
ord E. Kashtan MD, University of Minnesota, Department of Pediatrics, Box 491, 420 Delaware St. SE, Minneapolis, Minnesota, USA 55455.

Key words: acute rejection; haemolytic–uraemic syndrome; kidney allograft survival; kidney transplantation; recurrent disease

Introduction Haemolytic–uraemic syndrome (HUS ) is a common cause of acute renal failure in infants and children, usually occurring as a sequela of gastrointestinal infection with toxin-producing E. coli 0157: H7 [1]. Chronic renal failure develops in 10–15% of children when all forms of HUS are considered, but is more common in atypical forms of the syndrome [2]. HUS in adults is a more heterogeneous disorder and also has a worse prognosis, with up to 50% of patients developing chronic renal failure [3]. In children, a distinction has been made between classical HUS, which is regularly preceded by a diarrhoeal prodrome, and atypical HUS, in which a diarrhoeal prodrome is absent [2 ]. Patients with atypical HUS frequently exhibit a relapsing course and are more likely to develop chronic renal failure than patients with classical HUS [2 ]. Some cases of atypical HUS are familial; both autosomal dominant and autosomal recessive inheritance have been reported [4,5 ]. Numerous case reports of recurrence of HUS in renal allografts have appeared [6–15 ]. In previous studies of the outcome of renal transplantation for HUS at the University of Minnesota, recurrence was observed in 41% of patients [16,17]. Other centres have reported much lower rates of HUS recurrence (0–10%) [18–22]. The reasons for the unusually high rate of HUS recurrence at this institution are unclear. It is possible that the transplant population includes a disproportionate number of patients with atypical HUS, in whom recurrence of disease after transplantation may occur more frequently. In the present study the outcome of renal transplantation for HUS at the University of Minnesota was reviewed, adding patients who have undergone transplantation since the last report. The presence or absence of a diarrhoeal

© 1997 European Renal Association–European Dialysis and Transplant Association

1426

R. B. Miller et al.

prodrome preceding the onset of HUS in the native kidneys was assessed for each patient, in order to determine whether the risk of recurrence in the allograft is influenced by the character of the primary episode of HUS.

Subjects and methods Between 31 May 1972 and 31 December 1994, 24 patients with a primary diagnosis of HUS underwent renal transplantation at the University of Minnesota ( Table 1 ). The medical records of these patients were reviewed and the initial presentation of HUS was defined for each patient. Those with a history of a diarrhoeal prodrome within 14 days of presentation were classified as having classical HUS, while those in whom a diarrhoeal prodrome was absent at presentation were classified as having atypical HUS. Diagnosis of recurrent HUS required documentation of (a) microangiopathic haemolytic anaemia, ( b) thrombocytopenia, (c) renal insuciency, and (d ) allograft biopsy or nephrectomy findings compatible with HUS (endothelial

swelling, endothelial sloughing, and RBC fragmentation with glomerular and/or arteriolar fibrin thrombi (Figure 1 )). All biopsies were reviewed by one of the authors (BB), who was not aware of the clinical information. Each renal biopsy was also evaluated for evidence of rejection, utilizing the Ban
criteria [23]. At transplant, immunosuppression consisted of prednisone and azathioprine ( AZA) in all 36 allografts. Minnesota Anti-Lymphocyte Globulin ( MALG) was used in 25 transplants for induction, while OKT3 (OrthocloneA, Ortho Pharmaceuticals) was used for induction in six cases (one patient was changed to OKT3 from MALG for induction therapy because of an adverse reaction to MALG). In three transplants MALG was not used for induction but was administered later for treatment of acute rejection. Cyclosporin A (SandimmunA, Sandoz Pharmaceuticals) (CsA) was used in eight transplants as part of the initial immunosuppression, and in five other transplants CsA was added after episodes of acute rejection. In one transplant tacrolimus (PrografA, Fujisawa) was substituted for CsA as rescue therapy for rejection. Renal allograft survival was examined using Kaplan–Meier analysis. The generalized Wilcoxon (Gehan’s) test was used

Table 1. Patients with a primary diagnosis of HUS transplanted at the University of Minnesota, 1972–1994 Patient

Age at onset ( years)

Diarrhoeal prodrome

Age at transplant

Immunosuppression Initial

1

17.8

no

2

1.4

no

3

0.5

unknown

4

0.8

no

5 6 7 8 9 10

10.4 3.5 5.4 0.5 0.9 0.7

no unknown no no no no

11 12 13 14 15

0.8 8.8 1.0 7.6 16

no no yes no no

16 17 18 19 20 21 22

9.4 0.7 3.2 51 20 37 25

no no yes no unknown yes no

23 24

38 21

no no

18 18.7 1.9 3.6 7.2 8.6 22.8 1.7 2.4 4 13 18.2 12 6.1 (#3)b 5.7 3.7 1.8 2.6 4.3 5.5 1.4 31 12 10.1 16.9 22.9 12.4 1.7 12.9 (#3)b 52 24 38 28 (#3 )b 32.7 38.3 24

M,P,A M,P,A M,P,A M,P,A P,A M,P,A M,C,P,A M,P,A M,P,A P,A M,P,A C,P,A M,P,A M,C,P,A M,P,A M,P,A M,P,A M,P,A M,P,A O,P,A M,P,A M,P,A M,P,A O,P,A M,C,P,A C,P,A O,P,A O,C,P,A M,O,C,P,A P,A,M M,P,A O,P,A M,P,A C,P,A M,P,A P,A

Recurrence Rejectiona

M

C T C

C

M C M,C

yes no no yes no no no no no no no yes no yes no no no yes yes yes yes no yes no no no no no no yes no no yes yes yes yes

M, Minnesota antilymphocyte globulin; C, cyclosporin; T, tacrolimus; P, prednisone; A, azathioprine; O, OKT3. aThe indicated agent was used in the treatment of rejection ( MALG) or added to the maintenance regimen after a rejection episode (cyclosporin, tacrolimus). bEarlier transplants performed elsewhere.

Recurrence of HUS in renal transplants: a single-centre report

Fig 1. (a) Renal allograft biopsy from patient no. 23 demonstrates thrombi within the glomerular capillary tuft (arrow) (H&E,×200 ). ( b) Allograft biopsy from patient no. 24 shows a thrombus within an arteriole extending into the glomerulus (trichrome,×200).

to compare graft survival in HUS patients with and without recurrence, and to compare graft survival in HUS patients to graft survival in patients with a primary diagnosis other than HUS who were matched for age, donor source, and transplant number. Two non-HUS patients were available for comparison with each HUS patient. The Mann–Whitney test was utilized to compare the interval between the onset of HUS and the first renal transplant in HUS patients with and without recurrence. The incidence of glomerular capillary, arteriolar, and arterial thrombi in grafts with acute rejection was evaluated to determine whether the presence of renal vascular thrombi in HUS patients could be attributable to acute rejection. Biopsies with acute rejection from HUS patients with and without recurrence and from non-HUS patients matched for age, donor source, HLA match, and era were compared using Fisher’s exact test. Three non-HUS patients were available for comparison with each HUS patient.

Results Twenty-four patients with a primary diagnosis of HUS received 36 transplants (21 primary transplants, 6 second transplants, and 9 third, fourth or fifth grafts). Three patients received their first two transplants at

1427

other institutions. Twenty-five of the transplants were from living related donors, and 11 were from cadaver donors. Eighteen of the 24 patients were less than 18 years old when they received their first transplants. The median age at first transplant was 12.1 years (range,1.8–52 years). Eighteen patients lacked a diarrhoeal prodrome with the onset of HUS (atypical HUS ), while three patients had a diarrhoeal prodrome (classical HUS ). Three patients could not be classified because of insucient data. Fifteen of the 18 patients who lacked a diarrhoeal prodrome and two of the three patients with a diarrhoeal prodrome were less than 18 years old at the onset of HUS. One patient’s sibling donor developed HUS 1 week after donation [24]; the family history was negative or unavailable in the other 23 patients. There were 16 episodes of recurrence of HUS involving 14 allografts in 11 patients. Nine of these 11 patients had atypical HUS, one had classical HUS, and one could not be classified ( Table 2 ). Only one of 14 allografts with recurrence is presently functioning. Eleven were lost to recurrent HUS. Two were lost to death from sepsis, 3 and 134 months respectively, after recurrence of HUS. Recurrence began 1 day to 9.4 years after transplantation; 11 of 14 initial recurrences developed within 2 months post-transplant. Two patients each had two episodes widely separated in time ( 8 and 28 months respectively); both of these grafts were lost after the second recurrence. The median time to loss of the graft after recurrence was 3 days (range 0 day to 1 month). There was no significant di
erence between the recurrence and non-recurrence patients in the interval between the onset of HUS and the first renal transplant (median 1 year (range 2 months to 11 years) in patients with recurrence vs 2 years (range 2 months to 22 years) in patients without recurrence). Allograft biopsies at recurrence showed glomerular thrombi in 12 of 14 grafts; in five grafts thrombi were present in >50% of the glomeruli. Arteriolar thrombi were present in 11 grafts. Arterial thrombi were present in three grafts; these biopsies were performed 3 weeks, 12 weeks, and 29 months after transplantation. Nine grafts had glomerular and arteriolar thrombi; all three grafts with arterial thrombi had glomerular and arteriolar thrombi. Eighteen biopsies from HUS patients with acute rejection and 54 biopsies of non-HUS patients with acute rejection were examined in order to assess whether the presence of renal vascular thrombi in HUS patients could be attributable to acute rejection. Glomerular, arteriolar, or arterial thrombi were found in eight of the 18 HUS grafts with acute rejection and only three of 54 non-HUS grafts with acute rejection (P=0.0003). There was no clear relationship of immunosuppressive drugs to HUS recurrence. CsA was used in 10 of 22 transplants ( 45%) without recurrence and three of 14 transplants ( 21%) with recurrence (two were receiving CsA at the time of recurrence). With the discontinuation of CsA, one patient had clinical resolution of the

1428

R. B. Miller et al.

Table 2. Characteristics of HUS recurrence in 11 patients Onset of

Location of renal thrombi

Graft #

recurrence

Glomerular

Arteriolar

Arterial

Rejection

1 2 4

1 2 5

2 months 8 days 9 months

no yes yes

yes yes yes

yes no no

6

3

1 day

yes

no

no

mild ATI moderate ATI mild ATI mild CV mild ATI

10

15 months 2 months 1 week 9.4 years 1 week 4 days

yes yes yes yes yes yes

no yes no yes no yes

no no no no no no

no mild ATI mild ATI mild ATI no no

dipyridamole plasma infusion

11 13 19

1 2 3 1 1 1

22

3 4 1 1

yes yes yes yes yes yes

yes yes yes yes yes yes

no yes no no no no

no no no no no mild ATI

plasmapheresis plasmapheresis plasmapheresis

23 24

1 week 2.5 years 1 week 2 days 1 week 8 months

Patient

Therapy

Outcome graft lost Cr 1.9 mg/dl @ 5 years

T held C held

plasmapheresis

plasmapheresis plasmapheresis

graft lost Cr 1.0 mg/dl @ death 3 mos later (sepsis) graft lost graft lost graft lost graft lost graft lost Cr 0.7 mg/dl @ death 12 years later (sepsis) good function for 2 years graft lost graft lost graft lost good function for 7 mos graft lost

ATI, acute tubulointerstitial; CV, chronic vascular; Cr, creatinine; C, cyclosporin; T, tacrolimus.

HUS episode, while the other patient showed no improvement and lost the allograft to recurrence. Tacrolimus was substituted for CsA in patient no. 4 and was temporally associated with recurrence of HUS in the patient’s fifth allograft (four previous allografts were lost to rejection). MALG was used in 10 of 14 transplants (71%) with recurrence and 18 of 22 transplants (81%) without recurrence. Among those patients with recurrent HUS who received MALG, five had their episodes of recurrence weeks to months after MALG was discontinued, while another had an initial recurrence while receiving MALG and a second recurrence 32 months later. One of six patients who received OKT3 immediately following transplant developed recurrent HUS while on OKT3. Allograft survival was significantly poorer in all HUS allografts than in patients matched for age, donor source, and transplant number (P=0.007) (not shown). This di
erence is largely explained by the poorer allograft survival in HUS patients with recurrence compared to the control group (P=0.0004) ( Figure 2). There was a trend toward poorer graft survival in HUS patients without recurrence compared to the control group, but this did not reach statistical significance (P=0.094). The di
erence in graft survival between HUS patients with recurrence and those without recurrence was significant (P=0.04).

Fig 2. Renal allograft survival in patients transplanted for HUS, with and without recurrence, compared to non-HUS patients matched for age, donor source, and transplant number ( HUS (no recurrence) vs Control, P=0.094; HUS (recurrence) vs Control: P= 0.0004; HUS (no recurrence) vs HUS (recurrence): P=0.04).

Discussion Previous reports from this institution documented a high rate of recurrence of HUS in renal allografts [16,17]. In contrast, studies from other centres found that recurrence was never or only rarely observed [18–22]. One potential explanation for this discrepancy

in outcome is that the population of HUS transplant patients at the University of Minnesota has been skewed toward those in whom the risk of recurrence is high. Therefore the information regarding the character of the initial HUS episode was sought for each patient.

Recurrence of HUS in renal transplants: a single-centre report

This study utilized a strict definition of recurrent HUS, requiring the presence of MAHA, thrombocytopenia and acute graft dysfunction in association with histopathological evidence of intrarenal thrombosis for the diagnosis of recurrence. Using these criteria recurrence of HUS was observed in 11 of 24 patients (46%), an incidence similar to previous findings [16,17]. The clinical triad of MAHA, thrombocytopenia, and allograft dysfunction may be produced by acute vascular rejection [25 ]. None of the biopsy or nephrectomy specimens reviewed in this study showed acute vascular rejection, although the possibility of a sampling error cannot be entirely excluded. Although eight grafts with recurrent HUS also showed acute tubulointerstitial rejection, the infrequency of thrombi in acutely rejecting grafts in patients with an original diagnosis other than HUS suggests that acute tubulointerstitial rejection per se is rarely sucient to produce intrarenal microvascular thrombosis in patients without HUS. Eighteen of 24 patients (75%) in this study had atypical HUS and nine of these 18 ( 50%) experienced recurrent HUS in their allografts. The predominance of atypical HUS di
erentiates the patients in this study from those reported by centres where recurrent HUS was rarely if ever observed [18–22]. Eleven of 12 previously reported cases of recurrent HUS involved patients with atypical HUS [6–15]. Combining these reports with the present study, 25 of 26 patients with recurrent HUS did not have a diarrhoeal illness preceding the original episode of HUS (in five patients with recurrence information on diarrhoea was not available or not provided ). It appears, therefore, that HUS recurrence is particularly, although not exclusively, observed in patients with atypical HUS, and that the predominance of these patients among those transplanted at the University of Minnesota may explain the relatively high recurrence rate. In this study, as in other reports, recurrence of HUS was usually an early event, with 11 of 14 initial episodes occurring within 2 months after transplantation. However, recurrence may occur years after transplant, as illustrated by patient no. 11 in the present study, suggesting that the predisposition to HUS is a virtually permanent characteristic. Allograft survival was inferior in patients transplanted for HUS as compared to matched patients with other diagnoses. This di
erence arose largely from the poorer allograft survival in patients with recurrence of HUS compared to those HUS patients without recurrence. Although CsA was used more frequently in transplants without recurrence of HUS (10 of 22) than in transplants with HUS recurrence (3 of 14), the fact that none of the transplants with recurrence was lost to rejection makes it unlikely that the di
erence in CsA use accounts for the di
erence in graft survival. However, di
erences in immunosuppression between the recurrence and non-recurrence groups may have contributed to the di
erences in graft survival. Although graft survival was not statistically di
erent in HUS patients without recurrence compared to controls, the trend toward poorer graft survival in HUS

1429

patients has been previously recognized by Gagnadoux et al. [26 ] who postulated an increased susceptibility to chronic vascular rejection. The findings of the present study indicate that HUS recurrence has a detrimental e
ect on graft survival, and that a primary determinate of the outcome of transplantation for HUS is the presence or absence of recurrence. An HUS-like clinical picture associated with glomerular thrombosis has been described in patients receiving tacrolimus [12,27], CsA [21], and antilymphocyte preparations [13,15,25 ]. In this series the single patient who received tacrolimus (patient no. 4 ) had a recurrence temporally related to administration of this agent. Only two of 13 patients receiving CsA had recurrence of HUS. None of the other patients with recurrence of HUS was receiving CsA at the time of recurrence. These results suggest that CsA does not increase the risk of HUS recurrence and need not be avoided in patients transplanted for HUS. While MALG had been administered in 10 cases of recurrent HUS, the recurrence of HUS was clearly unrelated to MALG in six of these cases. Whether MALG contributed to the development of recurrent HUS in the other four cases is unknown. MALG was not administered to other reported patients with recurrent HUS [6–15,18–22 ]. One patient in the present study was receiving OKT3 when HUS recurrence took place; two other patients who developed recurrence while receiving OKT3 have been reported [13,15]. The results of the present study suggest that a particular immunosuppressive agent may provoke recurrence in an individual patient, but the overall contribution of these drugs to recurrent HUS after transplantation is small. Atypical HUS most probably encompasses a heterogeneous group of disorders. Some patients with atypical HUS appear to be at high risk of recurrence of HUS after transplantation. Other patients with atypical HUS can be successfully transplanted without recurrence of the disease. A patient’s response to a first transplant does not always predict the response to retransplantation, since HUS may recur in a second transplant when it did not recur in the first transplant, or may recur in the first but not subsequent transplants. Unless and until specific markers for the recurrent form of the disease become available, transplant physicians will have to rely on their clinical judgment in determining strategies for renal transplantion in these patients. Acknowledgements. This study was supported in part by NIH grants no. AM13083, no. DK07087, and no. AM07087.

References 1. Karmali MA, Petric M, Lim C, Fleming PC, Arbus GS, Lior H. The association between idiopathic hemolytic uremic syndrome and infection by vero-producing Escherichia coli. J Infect Dis 1985; 151: 775–782 2. Tonsho
B, Sammet A, Sanden I, Mehls O, Waldherr R, Scharer K. Outcome and prognostic determinants in the hemolytic uremic syndrome of children. Nephron 1994; 68: 63–70

1430 3. Morel-Maroger L, Kaufer A, Solez K, Sraer JD, Richet G. Prognostic importance of vascular lesions in acute renal failure with microangiopathic hemolytic anemia ( hemolytic uremic syndrome). Clinicopathological study in 20 adults. Kidney Int 1979; 15: 548–558 4. Matto TK, Majmood MA, Al-harbi MS, Mikail I. Familial, recurrent hemolytic–uremic syndrome. J Pediatr 1989; 114: 814–816 5. Merril RH, Knupp CL, Jennette JC. Familial thrombotic microangiopathy. Q J Med 1985; 57: 749–759 6. Folman R, Arbus GS, Churchil B, Gaum L, Huber J. Recurrence of the hemolytic uremic syndrome in a 3 1/2-year-old child, 4 months after second renal transplant. Clin Nephrol 1978; 10: 121–127 7. Hauglustaine D, Van Damme B, Vanrenterghem Y, Michielsen P. Recurrent hemolytic uremic syndrome during oral contraception. Clin Nephrol 1981; 15: 148–153 8. Leithner C, Sinzinger H, Pohanka E, Schwarz M, Kretschmer G, Syre G. Recurrence of haemolytic uraemic syndrome triggered by cyclosporin A after renal transplant. Lancet 1982; 1: 1470 9. Strom TB, McCluskey RT. Renal disorder 13 months after renal transplantation for hemolytic uremic syndrome. N Engl J Med 1986; 314: 1032–1040 10. Grino JM, Caralps A, Carreras L et al. Apparent recurrence of hemolytic uremic syndrome in azathioprine-treated allograft recipients. Nephron 1988; 49: 301–304 11. Springate J, Rildes R, Anthone S et al. Recurrent hemolytic syndrome after renal transplantation. Transplant Proc 1988; 20: 559–561 12. McCauley J, Shapiro R, Bronster O et al. Renal transplantation under FK506 in patients with previous loss of renal function due to hemolytic uremic syndrome. Transplant Proc 1991; 23: 3068–3070 13. Goodman DJ, Walker RG, Birchall IE, d’Apice AJF, Powell JR, Kincaid-Smith P. Recurrent haemolytic uraemic syndrome in a transplant recipient on Orthocone (OKT . Pediatr Nephrol 3] 1991; 5: 240–241 14. Monchon M, Kaiser deChadarevian JP, Polinsky MS, Valuarte HJ. Cerebral infarct with recurrence of hemolytic-uremic syndrome in a child following renal transplantation. Pediatr Nephrol 1992; 6: 550–552

R. B. Miller et al. 15. Doutrelepont JM, Abramowicz D, Florquin S et al. Early recurrent of hemolytic uremic syndrome in a renal transplant recipient during prophylactic OKT therapy. Transplantation 3 1992; 53: 1378–1379 16. Hebert D, Kim E, Sibley RK, Mauer SM. Post-transplant outcome of patients with hemolytic-uremic syndrome: update. Pediatr Nephrol 1991; 5: 162–167 17. Hebert D, Sibley RK, Mauer SM. Recurrence of hemolytic uremic syndrome in renal transplant recipients. Kidney Int 1986; 30: S51-S58 18. Van den Berg-Wolf MG, Kootte AMM, Weening JJ, Paul LC. Recurrent hemolytic uremic syndrome in a renal transplant recipient and review of the Leiden experience. Transplantation 1988; 45: 248–251 19. Eijgenraam FJ, Donckerwolcke RA, Monnens LA, Proesmans W, Wol
ED, Van Damme B. Renal transplantation in 20 children with hemolytic-uremic syndrome. Clin Nephrol 1990; 33: 87–93 20. Bassani C, Ferraris J, Gianantonio CA, Ruiz S, Ramirez J. Renal transplantation in patients with classical hemolytic-uremic syndrome. Pediatr Nephrol 1991; 5: 607–611 21. Schwarz A, Krause PH, O
erman G, Keller F. Recurrent and de novo renal disease after kidney transplantation with or without cyclosporine A. Am J Kidney Dis 1991; 17: 524–531 22. Gagnadoux MF, Habib R, Broyer M. Outcome of renal transplantation in 34 cases of childhood hemolytic–uremic syndrome and the role of cyclosporine. Transplant Proc 1994; 26: 269–270 23. Solez K, Axelsen R, Benediktsson H et al. International standardization of criteria for the histologic diagnosis of renal allograft rejection: the Ban
working classification of kidney transplant pathology. Kidney Int 1993; 44: 411–422 24. Bergstein J, Michael AF, Kjellstrand C, Simmons R, Najarian JS. Hemolytic uremic syndrome in adult sisters. Transplantation 1974; 17: 487–490 25. Matas AJ, Sibley R, Mauer M, Sutherland DER, Simmons RL, Najarian JS. The value of needle renal allograft biopsy. Ann Surg 1983; 197: 226–237 26. Gagnadoux MF, Broyer M, Habib R. Renal transplantation in hemolytic-uremic syndrome. Report on 31 cases (abstract). Pediatr Nephrol 1989; 3: C184 27. Schmidt R, Venkat K, Dumler F. Hemolytic-uremic syndrome in a renal transplant recipient on FK506 immunosuppression. Transplant Proc 1991; 23: 3156–3157 Received for publication: 17.11.96 Accepted in revised form: 11.3.97