Original Paper Nephron Clin Pract 2007;105:c68–c76 DOI: 10.1159/000097601
Received: May 9, 2006 Accepted: August 31, 2006 Published online: November 29, 2006
Long-Term Outcome of Paediatric Renal Transplantation: Follow-Up of 300 Children from 1973 to 2000 Lesley Rees a Rukshana Shroff a Carol Hutchinson a Oswald N. Fernando b Richard S. Trompeter a a b
Department of Nephro-Urology, Great Ormond Street Hospital for Children NHS Trust, London, and Department of Transplantation, Royal Free Hospital, London, UK
Key Words Renal transplantation Renal replacement therapy Mortality, renal transplantation Transplant survival Living-related kidney donation
Abstract Background/Aim: To report our experience of paediatric renal transplantation at Great Ormond Street and Royal Free Hospitals since the inception of the programme. Methods: Retrospective review of the patient and transplant survival and influencing factors in the 300 children transplanted between 1973 and 2000. Results: 300 children had received a total of 354 transplants; 56 were living-related donations. The median age at transplantation was 10.3 (range 1.4–17.9) years. Forty-four percent had congenital structural abnormalities of the urinary tract. Forty-six children required a second and 8 a third transplant before transfer to an adult unit. The overall patient survival at 5, 10, and 20 years was 97, 94, and 72%, respectively. In the overall cohort, the donor type (deceased donor or living-related donor) did not affect mortality, nor did age at transplantation, but those transplanted before 5 years of age had a significantly shorter post-transplant survival time (p ! 0.0001). Transplant survival (first transplant) for deceased and livingrelated donors was 66 and 87% at 5 years (p ! 0.01), 51 and 54% at 10 years, and 36% at 20 years (deceased-donor trans-
© 2007 S. Karger AG, Basel 1660–2110/07/1052–0068$23.50/0 Fax +41 61 306 12 34 E-Mail
[email protected] www.karger.com
Accessible online at: www.karger.com/nec
plants only). Although the overall transplant survival was inferior in children transplanted before 2 years of age (p ! 0.03), in the most recent cohort (1990–2000), age did not affect the outcome. On multiple regression analysis, the only predictor of transplant survival was the era of transplantation (p ! 0.001). The median final height was within the normal range for males and females; 7 patients received growth hormone after transplantation. Conclusions: The outlook for successful transplantation is improving, and in the last decade was unaffected by age at transplantation. The survival of living-related donor transplants is superior to deceased-donor transplants for the first 5 years. From the above data, we can predict that a 10-year-old child receiving a renal transplant in 2000 and on ciclosporin-based immunosuppression can expect a transplant half-life of 13.1 years from a living-related donor and one of 10.8 years from a deceased-donor transplant. Copyright © 2007 S. Karger AG, Basel
Introduction
Approximately 9 children per million child population start renal replacement therapy (RRT) in the UK each year. The numbers have increased steadily, principally since the 1980s, to reach this level in 2000 and since then have been stable, so that at any time there are be-
Dr. Lesley Rees, MD Department of Nephro-Urology Great Ormond Street Hospital for Children NHS Trust, Great Ormond Street London WC1N 3JH (UK) Tel. +44 207 813 8346, Fax +44 207 829 8841, E-Mail
[email protected]
tween 700 and 800 children on RRT, about two thirds of whom have renal transplants [1]. Papers describing long-term patient survival [2–6], transplant outcome [2, 3, 6–10], and psychosocial adjustment [11, 12] are now emerging, enabling doctors to offer better prognostic advice and guidance to families. Although national registries are able to provide statistical predictions that are not possible from a single centre, more complete and detailed data are often available from individual units. We describe the patient and transplant survival of all children under 18 years of age who received a renal transplant at Great Ormond Street and Royal Free Hospitals between 1973 and the end of 2000, with a minimum follow-up of 2.5 years.
Statistics The results are given as median with range and 95% confidence intervals (CI). The time from transplantation to graft loss or death was determined using the Kaplan-Meier survival analysis and presented as a percentage and 95% CI. The outcomes between LRD and deceased-donor transplants, age at transplantation, transplant number, and era of transplantation were compared. Comparisons between groups were made using the Mann-Whitney U test and the Breslow comparison for survival analysis. Cox regression analysis with time-dependent covariates was used to determine independent predictors of transplant survival. Statistical significance was defined as a two-sided p ! 0.05. All statistical analyses were performed using SPSS version 12.0.1 (SPSS, Chicago, Ill., USA).
Results Patients and Methods The first paediatric kidney transplant in our unit was carried out in 1973 at the Royal Free Hospital in a 9-year-old girl. The transplant was lost 72 h later. There were no more transplants until 1975. Thereafter, over the next 10 years, there were 32 transplants (median 3, range 1–7/year). In 1987 a programme for the transplantation of children !5 years of age was developed at Great Ormond Street Hospital, and after 1996 all transplants were transferred to Great Ormond Street Hospital from the Royal Free Hospital. From 1985 the numbers per year have steadily increased, so that by the end of 2000 a total of 354 transplants had been performed in 300 children !18 years of age resident in the UK. In order to reflect the changes in our transplantation practice, we have divided the study period into three transplantation eras: from 1973 to 1983, from 1984 to 1990, and from 1990 to 2000. The period before 1983 reflects the pre-ciclosporin era, when patients received dual immunosuppression with prednisolone and azathioprine. From 1984, ciclosporin-based immunosuppression was used. In 1986 a programme for living-related kidney donor (LRD) transplantation was established at Great Ormond Street Hospital. In the 1990s, the practice of using small donor kidneys was stopped given their greater risk of vascular thrombosis. Tacrolimus was not used routinely until after the end of the study period. Antithymocyte globulin was given as induction therapy for second transplants. The notes of the patients were reviewed in 2002, so that there was a range of follow-up of 2.5–29 years. Patients who had transferred to adult units were traced using their NHS number and UK Transplant and Renal Registry databases. The following data were collected: date of birth, diagnosis, sex, age at transplant, number of transplants and type (deceased donor or LRD), transplant and patient survival, cause of death (verified by death certificate when possible), and height in those 117 years of age. With full ethical approval, questionnaires were sent to the patients and their adult renal units asking about further transplants, occupation, whether in a relationship, and offspring.
Long-Term Outcome of Paediatric Renal Transplantation
From January 1973 to December 2000, 300 children received renal transplants at our unit. There were 193 boys (64%). The median age at transplantation was 10.3 (range 1.4–17.9) years. Fifty-nine children (19.7%) were !5 years of age at transplant, and 14 of these were !2 years of age. The primary diagnoses were as shown in table 1. A proportion of 44.2% had congenital structural abnormalities (excluding patients with syndromes); 234 (78%) were Caucasian, 58 (18.7%) Asian, 7 (2.3%) Negroid, 2 mixed race, and 1 Chinese. The proportion of patients of Asian origin increased progressively from 7.5% in the 1973–1983 cohort to 22.0% in the last era of transplantation. The 300 children received a total of 354 transplants before the age of 18 years. The median age at first (n = 300), second (n = 46), and third (n = 8) transplants was 10.7 (range 1.4–17.9) years, 11.1 (range 2.5–17.9) years, and 12.2 (range 7.0–17.7) years, respectively; 298 were deceased-donor transplants, and 56 were LRDs. LRD transplants were 54 (18%) first transplants and 2 (4%) second transplants. There were no third transplants. All the LRDs were parents. Follow-Up Data The overall median age of surviving patients at final follow-up was 19.9 (range 5.4–42.5) years. Patient Survival after Transplant The median 5-, 10-, 15-, and 20-year patient survival rates were 97% (95% CI 99–96), 94.5% (95% CI 93–99), 83% (95% CI 80–91), and 71.6% (95% CI 69–76), respectively (fig. 1).
Nephron Clin Pract 2007;105:c68–c76
c69
Table 1. Diagnoses of the patients
Diagnosis
n
%
Dysplasia and/or reflux Solitary dysplastic kidney Posterior urethral valve Non-neuropathic/neuropathic bladder Cloacal anomaly Focal segmental glomerulosclerosis Congenital nephrotic syndrome Alport’s syndrome/hereditary nephritis Cortical necrosis/renal venous thrombosis Crescentic nephritis (Goodpasture’s/Wegener’s/unknown) MPGN/glomerulonephritis of unknown aetiology Henoch-Schönlein purpura/IgA nephropathy Haemolytic-uraemic syndrome Lupus nephritis/polyarteritis Cystinosis Denys-Drash syndrome, Wilms’ tumour Stone disease/hyperoxaluriaa Nephronophthisis Autosomal recessive polycystic kidney disease Other cystic kidney diseases Tubulo-interstitial disease Methylmalonic acidaemia/drugs Unknown Underlying syndromes Eagle-Barrett (prune belly) Branchio-oto-renal VATER/CHARGE Joubert’s/Jeune’s Down’s Undiagnosed
66 10 52 4 1 25 13 6/2 4/1 2/2/2 11/3 4/2 2 2/1 6 6 1/5 6 8 2 5 1/1 20 24
22.0 3.3 17.3 1.3 0.3 8.3 4.3 3.0 1.7 2 4.7 2 0.7 1.0 2 2 2 2 2.7 0.7 1.7 0.7 7 8
Total
9 3 3/1 1/1 1 5 300
100
MPGN = Membranoproliferative glomerulonephritis; VATER = vertebral defects, anal atresia, tracheo-esophageal fistula with esophageal atresia, and radial and renal anomalies; CHARGE = coloboma of the iris, heart defects, atresia choanae, retarded development, genital anomalies, and ear anomalies/deafness. a 4 of the 5 children with hyperoxaluria received a combined liver and kidney transplant.
Deaths There were 37 deaths (12.3% of the patients) during the follow-up period. Twelve patients died within 6 months after transplant, and there were 14 deaths with a functioning graft. The median age at death was 18.1 (95% CI 3.1–23.9) years, and their median follow-up period after transplantation was 4.0 years. The causes of death are shown in table 2. Twelve patients died within 6 months after transplantation due to: generalized herpes zoster (n = 1), gastrointestinal bleed (n = 2), multi-organ failure after liver and kidney transplantation (n = 1), Pneumocystis and cytoc70
Nephron Clin Pract 2007;105:c68–c76
megalovirus infections (n = 1), small-bowel perforation (n = 1), ruptured graft (n = 1), sepsis (n = 3), intracranial haemorrhage (n = 1), and recurrence of Wilms’ tumour (n = 1). Fourteen patients died with functioning grafts due to: complications of cystinosis (n = 1), cardiovascular disease (CVD) (n = 1), CVD and pneumonia (n = 1), pulmonary haemorrhage (n = 1), intracerebral haemorrhage (n = 1), pneumonia in association with spastic quadriplegia (n = 2; sustained after transplantation 1 and congenital 1), lymphoma (n = 1), brain stem infarct (n = 1), smallbowel infarct (n = 1) and sepsis and multi-organ failure (n =1) (both patients had liver and kidney transplants), Rees /Shroff /Hutchinson /Fernando / Trompeter
Table 2. Causes of death after transplantation Survival function Censored
1.0
Cumulative patient survival
0.8
0.6
Cause of death
n
%
Infections (sepsis, pneumonia, Pneumocystis carinii, cytomegalovirus pneumonitis, herpes zoster)
10
27
Cardiovascular (intracranial bleed or ischaemia, pulmonary oedema or haemorrhage, cardiomyopathy or cardiac conductive disorder)
12
32
3
8
Others (graft rupture, gastro-intestinal complications, drug overdose, non-compliance, treatment withdrawal) 12
32
Malignancy (recurrence of Wilms’ tumor, lymphoma, posttransplant lymphoproliferative disorder)
0.4
0.2 Patients at risk 300 276
173
62
26
6
0 0
5
10 15 20 25 Time after transplantation (years)
30
Patient survival (%) 5 years
10 years
15 years
20 years
97
94.5
83
71.6
Fig. 1. Overall patient survival after transplantation (n = 300).
treatment withdrawal (n = 1), post-transplant lymphoproliferative disease (n = 1), and cause unknown (n = 1). Eleven patients died after their transplant failed due to: accidental drug overdose (n = 1), subdural haemorrhage (n = 1), non-concordance (n = 1), sepsis (n = 2), hyperkalaemia (n= 1), convulsions (n = 1), possible arrhythmia (n = 1), calciphylaxis (n = 1), bowel perforation (n = 1), and unknown cause (n = 1). Effect of LRD and Cadaveric Donation on Mortality There were 5 deaths (9.9%) in patients who had received a kidney from a LRD and 32 deaths (9%) in recipients of deceased-donor kidneys. The patient survival rates for LRD and deceased-donor recipients were 100 versus 98% at 5 years and 90% (95% CI 88–94) versus 93% (95% CI 89–96) at 10 years, respectively (p = 0.4). The 15and 20-year graft survival rates for deceased-donor kidneys were 82% (95% CI 81–89) and 71.6% (95% CI 69–76), respectively. Since our programme of LRD transplantaLong-Term Outcome of Paediatric Renal Transplantation
tion began in 1986, we do not have sufficient numbers for longer-term analysis. There was no difference in the age at death between the two groups: 21.7 years (95% CI 21.5– 21.9) for LRDs and 17.7 years (95%CI 15.5–19.9) for deceased-donor recipients (p = 0.3). Effect of Age on Mortality Children transplanted !5 years of age did not have a higher mortality rate than those 15 years of age at transplantation (13.7 vs. 11.2%), but those that died had a shorter survival time after transplantation (p ! 0.0001); the patient survival rates for those transplanted !5 years of age as compared with those transplanted 15 years were 95% (95% CI 84–99) versus 99% (95% CI 97–100) at 5 years and 84% (95% CI 72–89) versus 96% (95% CI 94– 99) at 10 years, respectively (fig. 2). Transplant Survival The overall median transplant survival was 10.8 (95% CI 6.9–14.6) years. Excluding graft losses during the first 7 days after transplantation (n = 15), the median transplant survival in the remainder of the patients was 12.2 years. The overall transplant survival (including all LRD and deceased-donor grafts and first, second, or third transplants) was 77.7% (95% CI 72–81), 52.5% (95% CI 44–61), 46% (95% CI 39–48), and 37.9% (95% CI 30–39) 5, 10, 15, and 20 years after transplantation, respectively. Effect of Era of Transplantation on Transplant Survival Figure 3 shows the effect of era of transplantation on transplant survival. At 10 years after transplantation, only 6 of the 30 transplants (20%) performed before 1983 Nephron Clin Pract 2007;105:c68–c76
c71
1.0 1.0
1973–1983 (n = 30) 1991–2000 (n = 219) 1984–1990 (n = 105)
>5 years (n = 241) 5 years
6 56
25
0
7
10 15 20 25 Time after transplantation (years)
Age at transplantation
