Patient and transplant outcome in infants starting renal replacement therapy before 2 years of age

Patient and transplant outcome in infants starting renal replacement therapy before 2 years of age Abstract Background Despite major technical improvements in the care of children requiring renal replacement therapy (RRT) before 2 years of age, the management of those patients remains challenging and transplantation is generally delayed until the child weighs 10 kg or is 2 years old. In this national cohort study, we studied patient and graft survival in children starting RRT before 2 years of age to help clinicians and parents when deciding on RRT initiation and transplantation management. Methods All children starting RRT before 24 months of age between 1992 and 2012 in France were included through the national Renal Epidemiology and Information Network (REIN) registry. The primary endpoints were patient survival on dialysis and 10-year graft survival. Results A total of 224 patients were included {62% boys, median age 10.5 months [interquartile range (IQR) 5.8–15.6]}. The 10-year survival rate was 84% (IQR 77–89). Suffering from extrarenal comorbidities was the only factor significantly associated with both an increased risk of death on dialysis [hazard ratio 5.9 (95% confidence interval 1.8–19.3)] and a decreased probability of being transplanted. During follow-up, 174 renal transplantations were performed in 171 patients [median age at first transplantation 30.2 (IQR 21.8–40.7) months]. The 10-year graft survival was 74% (IQR 67–81). Factors associated with graft loss in multivariate analysis were the time spent on dialysis before transplantation, donor/recipient height ratio with an increased risk for both small and tall donors and presenting two human leucocyte antigen–antigen D-related mismatches. Conclusions This study confirms the good outcome of children starting RRT before 2 years of age. The main question remains when and how to transplant those children. Our study provides data on the optimal morphological and immunological matching in order to help clinicians in their decisions. epidemiology, graft survival, infants, kidney transplantation, pediatrics INTRODUCTION End-stage renal disease (ESRD) is a rare condition in infants and young children. The incidence of ESRD in children <4 years of age is 5.2 per million age-related inhabitants in France [1] and 6.5 per million age-related inhabitants in Europe [2]. Despite major technical improvements in the care of children requiring renal replacement therapy (RRT) before 2 years of age, the management of these patients remains challenging [3, 4]. Renal transplantation is generally delayed until the child weighs 10 kg or is 2 years old, thus exposing the youngest patients to a prolonged period of dialysis and major issues in terms of nutrition, growth and psychomotor development. Moreover, extrarenal comorbidities are frequent in this group because more than half of the primary renal diseases (PRDs) in this age group are genetic diseases with potential extrarenal involvement [1] and because prenatal renal failure potentially complicated by oligoamnios or anamnios exposes some of these patients to pulmonary hypoplasia, premature birth and its complications. However, major improvement in the survival of these patients has been reported over the last decades. Mitsfenes et al. [5] reported a decrease in the mortality rate from 11.2% in the early 1990s to 8.3% between 2005 and 2010 in the USA, while McDonald and Craig [6], based on data from the Australia and New Zealand Dialysis and Transplant Registry, reported a decrease in the risk ratio of death from 116 to 32 when compared with the age-related mortality of the general population between the 1960s and the 1990s. Despite the progress made, initiating RRT, especially in the youngest children, remains controversial. Geary performed two surveys, one in 1998 and one in 2010: they found that clinical management perspectives have not substantially changed since RRT was offered by 41% of the nephrologists to all infants <1 month and by 53% to all infants between 1 and 12 months in 1998 and by 30 and 50%, respectively, in 2010 [7, 8]. Today, many paediatric nephrologists still consider RRT in infants as an optional rather than a mandatory treatment [9, 10]. Considering renal transplantation, graft survival has been consistently reported to be worse in the youngest children when compared with older ones [1]. Although many studies have investigated the impact of morphological [11–13] and immunological matching [14, 15] on graft survival, major discrepancies in the minimal recipient weight for renal transplantation, the choice of deceased versus living donor and minimal human leucocyte antigen (HLA) matching have been reported between centres and countries [16]. Moreover, no study has investigated the impact of these factors together. Thus we aim to study patient and graft survival in children starting RRT before 2 years of age to help clinicians and parents make decisions on RRT initiation and transplantation programming. MATERIALS AND METHODS Study population and data We included all patients who started RRT before 24 months of age in France between 1 January 1992 and 31 December 2012. In order to be exhaustive, three complementary ways of detecting patients were used: the national transplant database (CRISTAL), which records data on all the patients registered on the waiting list since 1992; the Renal Epidemiology and Information Network Registry, which records data on all patients on RRT and is exhaustive in children since 2005; and all individual hospital databases. Patient characteristics recorded were age at RRT initiation and at renal transplant; sex; PRD, which were classified in four groups [congenital abnormalities of the kidney and the urinary tract (CAKUT), genetic diseases, vascular diseases and others]; first RRT modality [peritoneal dialysis (PD), haemodialysis (HD) or pre-emptive transplantation); comorbidities (including neurological abnormality, liver disease, heart disease and respiratory insufficiency, from which we created the binary variable ‘at least one comorbidity’); and date of death if applicable. Data on the transplantation included age at renal transplant, weight and height of the donor and the recipient, type of donor (living or deceased), HLA matching and cold ischaemia time. Statistical analysis For descriptive analysis, continuous variables are given as median and interquartile range (IQR) and dichotomous variables as number and percentage. Patients’ survival and access to renal transplantation We present the cumulative incidence of death and transplantation overall and stratified by risk factors. In order to assess the factors associated with the risk of death on dialysis, we used univariate and multivariate Cox proportional hazards regression, with the time between the start of RRT and death as the primary outcome and renal transplantation as a censoring event. Patients who received pre-emptive transplantation (n = 7) were excluded from this analysis. All variables with a P-value of 0.20 in the univariate analysis were included in the multivariate analysis. We tested for the presence of a cohort effect by testing the association between the year at RRT initiation and death with the SAS macro RCS-REG (SAS Institute, Cary, NC, USA), which tests both the linearity and the significance of the association using spline modelling. Renal transplant survival The primary outcome was a composite outcome defined as graft loss or death. We used univariate Cox regression to assess the association between each variable and the outcome. The proportional hazards assumption of the models was assessed by graphical methods. All continuous variables were tested for linearity with the SAS macro RCS-REG. We used Kaplan–Meier modelling to present the survival curve. Multivariate Cox regression was used to assess factors associated with graft loss. RESULTS We included 224 patients who started RRT before 24 months of age. Among them, 141 (62.9%) were male, the median age at RRT initiation was 10.4 (IQR 5.8–15.6) months and the most prevalent PRDs were CAKUT and genetic diseases. Of these, 48 patients presented at least one extrarenal comorbidity, including 21 patients with associated neurological abnormalities, 20 patients with associated liver diseases, 14 patients with heart diseases and 5 patients with concomitant respiratory insufficiency. A flow chart of the cohort is presented as Supplementary data, Figure S1 and patients’ characteristics at RRT initiation are presented in Table 1. Table 1. Patients’ characteristics at RRT initiation Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Table 1. Patients’ characteristics at RRT initiation Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  The median follow-up time was 78 (IQR 36–147) months. During follow-up, 174 renal transplantations were performed in 171 patients; 29 patients died (18 before renal transplantation and 11 after renal transplantation) and the mortality rate on dialysis was 40 deaths per 1000 person-years compared with 9 deaths for 1000 person-years after transplantation. The two main causes of death were infections and cardiovascular events (Table 2). Only 10 transplantations were performed from a living donor. Overall the 5- and 10-year survivals were 87% (IQR 82–91) and 84% (IQR 77–89), respectively. Table 2. Number and causes of death in the entire cohort (n = 29) Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Table 2. Number and causes of death in the entire cohort (n = 29) Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Table 3. Cause-specific hazards of death on dialysis (n = 217)   Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5        Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5      Table 3. Cause-specific hazards of death on dialysis (n = 217)   Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5        Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5      Determinants of patient survival on dialysis and access to transplantation Overall, 217 patients started RRT with dialysis and were included in this analysis. Patients’ characteristics found to be associated with an increased risk of death by univariate analysis were having a genetic disease as the PRD [HR 5.6 (95% CI 1.3–25.0)] as compared to patients with CAKUT} and the presence of one or more extra-renal comorbidities [HR 7.7 (95% CI 2.4–24.5)]. There was a trend towards a decreased risk of death with age, although it did not reach statistical significance. Starting RRT on PD or HD did not significantly impact patient survival. Suffering from extrarenal comorbidities was the only factor significantly associated in multivariate analysis with both the hazard of death [HR 5.9 (95% CI 1.8–19.3)] (Table 3) and the hazard of being transplanted (data not shown). Supplementary data, Figures S2 and S3 present the cumulative incidence of death and renal transplantation stratified on the presence of comorbidity and on the different age groups, respectively. Determinants of graft survival The 10-year graft survival was 74% (IQR 67–81). Figure 1 presents the graft survival curve. FIGURE 1: View largeDownload slide Kaplan–Meier curve of graft survival over the 10-year follow-up. FIGURE 1: View largeDownload slide Kaplan–Meier curve of graft survival over the 10-year follow-up. Over the 171 first renal transplantations performed in our patients, the median dialysis time prior to transplantation was 22.1 (IQR 10.1–33.4) months, the median age at transplantation was 30.2 (IQR 21.8–40.7) months and the median recipient weight and height were 12.0 (IQR 10.4–13.6) kg and 85.0 (IQR 80.5–92.5) cm, respectively. Overall, our population of recipients was very homogeneous, with only 5% of the recipients <1 year old. They were very close in terms of weight and height. This homogeneity precluded us from studying the impact of those variables on renal survival. Variables found to be associated with graft survival by univariate analysis were the donor/recipient height ratio (P = 0.05), the number of human leucocyte antigen–antigen D-related (HLA-DR) mismatches (P = 0.02) and the time spent on dialysis prior to renal transplantation (P = 0.001) (Table 4). Table 4. HRs of graft failure or death in transplanted patients (univariate analysis)     Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73        Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73    Ψ, spline variables. Table 4. HRs of graft failure or death in transplanted patients (univariate analysis)     Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73        Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73    Ψ, spline variables. In order to make our results potentially useful in clinical practice, we dichotomized the donor/recipient height ratio in five categories (ratio ≤ 1, 1 < ratio < 1.4, 1.4 ≤ ratio ≤ 1.8, 1.8 < ratio < 2.2 and ratio ≥ 2.2). Figure 2 presents the evolution of the HR of graft loss with the donor/recipient height ratio. FIGURE 2: View largeDownload slide Evolution of the HR of graft loss or death with donor/recipient height ratio. FIGURE 2: View largeDownload slide Evolution of the HR of graft loss or death with donor/recipient height ratio. There was a tendency towards an improved graft with an HR of graft lost of 0.73 (IQR 0.36–1.46) between 2002 and 2012 when compared with the 1992–2002 period. The three factors found significantly associated with graft loss in multivariate analysis were the time spent on dialysis before transplantation, donor/recipient height ratio and presenting two HLA-DR mismatches (Table 5). Table 5. HRs of graft failure or death in transplanted patients (multivariate analysis) Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Table 5. HRs of graft failure or death in transplanted patients (multivariate analysis) Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  DISCUSSION In this national cohort of patients starting RRT before the age of 2 years, we confirmed the overall good patient survival with a 5-year survival of 87%, close to the survival reported in the USA in patients 0–4 years of age (84%) [17]. As reported in many previous studies [5, 6, 18, 19], we found a trend towards an increased risk of death in the youngest children, although age at RRT initiation did not reach statistical significance. Carey et al. [20] did not find a significant difference in survival between patients starting RRT before 1 month and those starting between 1 and 24 months of age; however, because North American Pediatric Renal Trials and Collaborative Studies only includes patients who started dialysis in centres willing to participate in the registry, selection bias of the youngest patients with the most favourable prognosis might explain the absence of effect of age at initiation of RRT in this study. In our study, the only factor significantly associated with patient survival was the presence of extrarenal comorbidities, with a cumulative incidence of death ranging from 4% in patients without extrarenal comorbidities to 30% in patients with at least one comorbidity. This is consistent with the data published by Wedekin et al. [21] in a cohort of patients starting RRT at <1 year of age and with the causes of death reported in The Netherlands in ESRD patients 0–14 years of age treated between 1972 and 1992 [22]. Similarly, a study focusing on patients with CAKUT demonstrated that the major risk factor of death was the association with extrarenal abnormalities [23]. In our study, no difference in survival was found between patients starting RRT with HD or PD. This result is consistent with a recent publication from the European Society of Paediatric Nephrology/European Renal Association–European Dialysis and Transplant Association (ESPN/ERA-EDTA) registry that did not find any difference in survival between HD and PD in patients starting RRT before the age of 1 year [24]. Nevertheless, the overall survival of infants starting RRT remains lower than the survival of older children [17], even though their survival post-transplantation is excellent and not different from the survival of older children [21]. However, many studies have emphasized the issues raised by renal transplantation in this population. Should we favour living kidney donation, usually from adults with major morphological differences? Is HLA matching still important in infants? However, mixing all these results together to guide clinical decisions remains challenging, so that there is a great heterogeneity of practices. In our study, morphological matching, especially in height, was highly associated with graft survival and a donor/recipient height ratio between 1.4 and 1.8 was found to be associated with the best graft survival. One of the main causes of graft loss in patients <2 years of age at transplantation is vascular thrombosis [25, 26], and many studies have reported worse graft outcomes when using age-matched donors. A recent study from the ESPN/ERA-EDTA registry confirmed the increased risk of graft loss in children receiving a transplant from a deceased donor <5 years of age [27]. Moreover, Dick et al. [12] found decreased renal survival among adolescent recipients who received a kidney from a smaller donor (donor/recipient body surface area ratio < 0.9), confirming the risk associated with smaller donors [12]. Pape et al. [28], studying children <10 years of age, questioned the use of kidneys from adult donors, because they found that 3–5 years after transplantation the corrected glomerular filtration rate (GFR) was significantly higher in children who had received a paediatric graft, with grafts also doubling in size, whereas no increase in size was noted in adult grafts. Our study confirms that in the youngest patients the use of height-matched kidneys increases the risk of graft loss and should therefore be avoided. Moreover, the scarcity of small paediatric kidney donors would lead to unacceptable waiting times on dialysis [11]. Our study also shows that a long time on dialysis prior to transplantation is associated with an increased risk of graft loss. Thus the benefit of waiting for an optimal transplant in terms of HLA and size matching might be outweighed by the negative impact of a long waiting time. This finding calls for regular and individualized re-evaluation of the requirements regarding transplant characteristics as time on dialysis increases. The other factor found to be significantly associated with graft survival in our patients was HLA matching for Class 2 antigen DR. Recently, authors have questioned the importance of HLA matching [29] and emphasized the importance of other factors such as prioritizing kidney from young adults for paediatric recipients [14] and minimizing cold ischaemia time. Moreover, not waiting for an HLA-matched kidney could favour rapid access to transplantation and allow pre-emptive transplantation in some children. However, poorer HLA matching, especially in Class 2, harbours the risk of de novo development of panel reactive antibodies (PRA) that could hamper both the ability to receive subsequent transplantation and graft survival [30]. This is a major concern in young children receiving renal transplantation since they will require several transplantations in their lifetime. Moreover, this issue remains unclear, as Gritsch et al. [14] did not find any difference in PRA level between well-matched and unmatched patients. Recently, Tinckam et al. [31] also reported on the risk of repeated HLA mismatch and showed that the presence of HLA Class 2 mismatches and PRA before the second transplantation decreases graft survival. This effect was even stronger in patients who underwent a nephrectomy of the first transplant. Finally, the major issue faced by clinicians when considering the optimal strategy of transplantation is to combine all the data together to make evidence-based decisions. Among adults, several authors have created clinical scores aimed at predicting renal transplantation outcome based on pretransplant [32] and post-transplant data [33–36]. Among children, a study aimed at defining groups of patients by the risk of graft loss has been recently published based on data from the ESPN/ERA-EDTA registry [37]. However, patients starting RRT before 2 years of age require specific studies focusing on this population and will require international collaboration. Based on the data of this national cohort, we were able to confirm and specify risk factors of patient and transplant survival in the youngest children starting RRT. Our study also has some limitations. First, the retrospective design of the study and the absence of termination of pregnancy and conservative treatments induces a selection bias since the study of patient survival does not apply to all patients with ESRD but to those who started RRT. Moreover, practices of RRT initiation differ between countries and centres. Due to the scarcity of patients starting RRT before 2 years of age, we included patients over a 20-year period. Although we did not find any statistically significant cohort effect, modifications of practices could modify the association reported. Similarly, given the scarcity of these patients, we grouped the PRDs into five categories; this did not allow us to study the impact of each specific disease on the outcomes. Finally, the scarcity of transplantation from living donors precluded us from comparing graft survival by the type of donor. CONCLUSION In this national retrospective cohort study, we confirm the overall good outcome of children starting RRT before 2 years of age, especially in the absence of extrarenal comorbidities. Thus the main questions in infants with ESRD remain when and how to transplant them. Our study provides data on the optimal morphological and immunological matching in order to help clinicians in their decisions. By extending this analysis to other cohorts, we aim to develop a tool that is able to predict graft loss and help clinicians when choosing the optimal kidney for their patients. SUPPLEMENTARY DATA Supplementary data are available at ndt online. AUTHORS’ CONTRIBUTIONS J.H. and C.C. participated in the conception of the article, analysis of the data and writing of the manuscript. M.-A.M. participated in the conception of the article, interpretation of the data and writing of the manuscript. J.B., M.C., G.R., R.N., M.T., J.T., T.U., A.G., E.M., J.H.a., I.V., O.D., D.M., E.B. and F.N. participated in data collection, interpretation of the data and writing of the manuscript. All the authors have revised the article and approve the final version. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 Harambat J, Hogan J, Macher M-A et al.   [ESRD in children and adolescents]. Nephrol Ther  2013; 9(Suppl 1): S167– S179 Google Scholar CrossRef Search ADS PubMed  2 Pippias M, Kramer A, Noordzij M et al.   The European Renal Association – European Dialysis and Transplant Association Registry Annual Report 2014: a summary. Clin Kidney J  2017; 10: 154– 169 Google Scholar PubMed  3 Ronco C, Garzotto F, Brendolan A et al.   Continuous renal replacement therapy in neonates and small infants: development and first-in-human use of a miniaturised machine (CARPEDIEM). Lancet  2014; 383: 1807– 1813 Google Scholar CrossRef Search ADS PubMed  4 Coulthard MG, Crosier J, Griffiths C et al.   Haemodialysing babies weighing <8 kg with the Newcastle infant dialysis and ultrafiltration system (Nidus): comparison with peritoneal and conventional haemodialysis. Pediatr Nephrol  2014; 29: 1873– 1881 Google Scholar CrossRef Search ADS PubMed  5 Mitsnefes MM, Laskin BL, Dahhou M et al.   Mortality risk among children initially treated with dialysis for end-stage kidney disease, 1990–2010. JAMA  2013; 309: 1921– 1929 Google Scholar CrossRef Search ADS PubMed  6 McDonald SP, Craig JC. Long-term survival of children with end-stage renal disease. N Engl J Med  2004; 350: 2654– 2662 Google Scholar CrossRef Search ADS PubMed  7 Geary DF. Attitudes of pediatric nephrologists to management of end-stage renal disease in infants. J Pediatr  1998; 133: 154– 156 Google Scholar CrossRef Search ADS PubMed  8 Teh JC, Frieling ML, Sienna JL et al.   Attitudes of caregivers to management of end-stage renal disease in infants. Perit Dial Int  2011; 31: 459– 465 Google Scholar CrossRef Search ADS PubMed  9 Lantos JD, Warady BA. The evolving ethics of infant dialysis. Pediatr Nephrol  2013; 28: 1943– 1947 Google Scholar CrossRef Search ADS PubMed  10 Wightman A, Kett J. Has neonatal dialysis become morally obligatory? Lessons from Baby Doe. Acta Paediatr  2015; 104: 748– 750 Google Scholar CrossRef Search ADS PubMed  11 Goldsmith PJ, Asthana S, Fitzpatrick M et al.   Transplantation of adult-sized kidneys in low-weight pediatric recipients achieves short-term outcomes comparable to size-matched grafts. Pediatr Transplant  2010; 14: 919– 924 Google Scholar CrossRef Search ADS PubMed  12 Dick AAS, Mercer LD, Smith JM et al.   Donor and recipient size mismatch in adolescents undergoing living-donor renal transplantation affect long-term graft survival. Transplantation  2013; 96: 555– 559 Google Scholar CrossRef Search ADS PubMed  13 Donati-Bourne J, Roberts HW, Coleman RA. Donor-recipient size mismatch in paediatric renal transplantation. J Transplant  2014; 2014: 317574 Google Scholar CrossRef Search ADS PubMed  14 Gritsch HA, Veale JL, Leichtman AB et al.   Should pediatric patients wait for HLA-DR-matched renal transplants? Am J Transplant  2008; 8: 2056– 2061 Google Scholar CrossRef Search ADS PubMed  15 Foster BJ, Dahhou M, Zhang X et al.   Impact of HLA mismatch at first kidney transplant on lifetime with graft function in young recipients. Am J Transplant  2014; 14: 876– 885 Google Scholar CrossRef Search ADS PubMed  16 van Huis M, Schoenmaker NJ, Groothoff JW et al.   Policy variation in donor and recipient status in 11 pediatric renal transplantation centers. Pediatr Nephrol  2013; 28: 951– 957 Google Scholar CrossRef Search ADS PubMed  17 United States Renal Data System. USRDS Annual Data Report 2016. https://www.usrds.org/adr.aspx (21 February 2018, date last accessed) 18 Samuel SM, Tonelli MA, Foster BJ et al.   Survival in pediatric dialysis and transplant patients. Clin J Am Soc Nephrol  2011; 6: 1094– 1099 Google Scholar CrossRef Search ADS PubMed  19 Alexander RT, Foster BJ, Tonelli MA et al.   Survival and transplantation outcomes of children less than 2 years of age with end-stage renal disease. Pediatr Nephrol  2012; 27: 1975– 1983 Google Scholar CrossRef Search ADS PubMed  20 Carey WA, Talley LI, Sehring SA et al.   Outcomes of dialysis initiated during the neonatal period for treatment of end-stage renal disease: a North American Pediatric Renal Trials and Collaborative Studies special analysis. Pediatrics  2007; 119: e468– e473 Google Scholar CrossRef Search ADS PubMed  21 Wedekin M, Ehrich JHH, Offner G et al.   Renal replacement therapy in infants with chronic renal failure in the first year of life. Clin J Am Soc Nephrol  2010; 5: 18– 23 Google Scholar CrossRef Search ADS PubMed  22 Groothoff JW, Gruppen MP, Offringa M et al.   Mortality and causes of death of end-stage renal disease in children: a Dutch cohort study. Kidney Int  2002; 61: 621– 629 Google Scholar CrossRef Search ADS PubMed  23 Melo BF, Aguiar MB, Bouzada MCF et al.   Early risk factors for neonatal mortality in CAKUT: analysis of 524 affected newborns. Pediatr Nephrol  2012; 27: 965– 972 Google Scholar CrossRef Search ADS PubMed  24 Vidal E, van Stralen KJ, Chesnaye NC et al.   Infants requiring maintenance dialysis: outcomes of hemodialysis and peritoneal dialysis. Am J Kidney Dis  2017; 69: 617– 625 Google Scholar CrossRef Search ADS PubMed  25 Hwang AH, Cho YW, Cicciarelli J et al.   Risk factors for short- and long-term survival of primary cadaveric renal allografts in pediatric recipients: a UNOS analysis. Transplantation  2005; 80: 466– 470 Google Scholar CrossRef Search ADS PubMed  26 Keller AK, Jorgensen TM, Jespersen B. Identification of risk factors for vascular thrombosis may reduce early renal graft loss: a review of recent literature. J Transplant  2012; 2012: 1 Google Scholar CrossRef Search ADS   27 Chesnaye NC, van Stralen KJ, Bonthuis M et al.   The association of donor and recipient age with graft survival in paediatric renal transplant recipients in a European Society for Paediatric Nephrology/European Renal Association-European Dialysis and Transplantation Association Registry study. Nephrol Dial Transplant  2017; 32: 1949– 1956 Google Scholar CrossRef Search ADS PubMed  28 Pape L, Hoppe J, Becker T et al.   Superior long-term graft function and better growth of grafts in children receiving kidneys from paediatric compared with adult donors. Nephrol Dial Transplant  2006; 21: 2596– 2600 Google Scholar CrossRef Search ADS PubMed  29 Hoyer PF. ‘Tie breaker’ for HLA matching in pediatric renal transplant recipients? Am J Transplant  2008; 8: 1970– 1971 Google Scholar CrossRef Search ADS PubMed  30 Wiebe C, Gibson IW, Blydt-Hansen TD et al.   Evolution and clinical pathologic correlations of de novo donor-specific HLA antibody post kidney transplant. Am J Transplant  2012; 12: 1157– 1167 Google Scholar CrossRef Search ADS PubMed  31 Tinckam KJ, Rose C, Hariharan S et al.   Re-examining risk of repeated HLA mismatch in kidney transplantation. J Am Soc Nephrol  2016; 27: 2833– 2841 Google Scholar PubMed  32 Molnar MZ, Nguyen DV, Chen Y et al.   Predictive score for posttransplantation outcomes. Transplantation  2017; 101: 1353– 1364 Google Scholar CrossRef Search ADS PubMed  33 Foucher Y, Daguin P, Akl A et al.   A clinical scoring system highly predictive of long-term kidney graft survival. Kidney Int  2010; 78: 1288– 1294 Google Scholar CrossRef Search ADS PubMed  34 Kasiske BL, Israni AK, Snyder JJ et al.   A simple tool to predict outcomes after kidney transplant. Am J Kidney Dis  2010; 56: 947– 960 Google Scholar CrossRef Search ADS PubMed  35 Gourishankar S, Grebe SO, Mueller TF. Prediction of kidney graft failure using clinical scoring tools. Clin Transplant  2013; 27: 517– 522 Google Scholar CrossRef Search ADS PubMed  36 Shabir S, Halimi J-M, Cherukuri A et al.   Predicting 5-year risk of kidney transplant failure: a prediction instrument using data available at 1 year posttransplantation. Am J Kidney Dis  2014; 63: 643– 651 Google Scholar CrossRef Search ADS PubMed  37 Lofaro D, Jager KJ, Abu-Hanna A et al.   Identification of subgroups by risk of graft failure after paediatric renal transplantation: application of survival tree models on the ESPN/ERA-EDTA Registry. Nephrol Dial Transplant  2016; 31: 317– 324 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nephrology Dialysis Transplantation Oxford University Press

Loading next page...
 
/lp/ou_press/patient-and-transplant-outcome-in-infants-starting-renal-replacement-tpjDhdYx6m
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
ISSN
0931-0509
eISSN
1460-2385
D.O.I.
10.1093/ndt/gfy040
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Despite major technical improvements in the care of children requiring renal replacement therapy (RRT) before 2 years of age, the management of those patients remains challenging and transplantation is generally delayed until the child weighs 10 kg or is 2 years old. In this national cohort study, we studied patient and graft survival in children starting RRT before 2 years of age to help clinicians and parents when deciding on RRT initiation and transplantation management. Methods All children starting RRT before 24 months of age between 1992 and 2012 in France were included through the national Renal Epidemiology and Information Network (REIN) registry. The primary endpoints were patient survival on dialysis and 10-year graft survival. Results A total of 224 patients were included {62% boys, median age 10.5 months [interquartile range (IQR) 5.8–15.6]}. The 10-year survival rate was 84% (IQR 77–89). Suffering from extrarenal comorbidities was the only factor significantly associated with both an increased risk of death on dialysis [hazard ratio 5.9 (95% confidence interval 1.8–19.3)] and a decreased probability of being transplanted. During follow-up, 174 renal transplantations were performed in 171 patients [median age at first transplantation 30.2 (IQR 21.8–40.7) months]. The 10-year graft survival was 74% (IQR 67–81). Factors associated with graft loss in multivariate analysis were the time spent on dialysis before transplantation, donor/recipient height ratio with an increased risk for both small and tall donors and presenting two human leucocyte antigen–antigen D-related mismatches. Conclusions This study confirms the good outcome of children starting RRT before 2 years of age. The main question remains when and how to transplant those children. Our study provides data on the optimal morphological and immunological matching in order to help clinicians in their decisions. epidemiology, graft survival, infants, kidney transplantation, pediatrics INTRODUCTION End-stage renal disease (ESRD) is a rare condition in infants and young children. The incidence of ESRD in children <4 years of age is 5.2 per million age-related inhabitants in France [1] and 6.5 per million age-related inhabitants in Europe [2]. Despite major technical improvements in the care of children requiring renal replacement therapy (RRT) before 2 years of age, the management of these patients remains challenging [3, 4]. Renal transplantation is generally delayed until the child weighs 10 kg or is 2 years old, thus exposing the youngest patients to a prolonged period of dialysis and major issues in terms of nutrition, growth and psychomotor development. Moreover, extrarenal comorbidities are frequent in this group because more than half of the primary renal diseases (PRDs) in this age group are genetic diseases with potential extrarenal involvement [1] and because prenatal renal failure potentially complicated by oligoamnios or anamnios exposes some of these patients to pulmonary hypoplasia, premature birth and its complications. However, major improvement in the survival of these patients has been reported over the last decades. Mitsfenes et al. [5] reported a decrease in the mortality rate from 11.2% in the early 1990s to 8.3% between 2005 and 2010 in the USA, while McDonald and Craig [6], based on data from the Australia and New Zealand Dialysis and Transplant Registry, reported a decrease in the risk ratio of death from 116 to 32 when compared with the age-related mortality of the general population between the 1960s and the 1990s. Despite the progress made, initiating RRT, especially in the youngest children, remains controversial. Geary performed two surveys, one in 1998 and one in 2010: they found that clinical management perspectives have not substantially changed since RRT was offered by 41% of the nephrologists to all infants <1 month and by 53% to all infants between 1 and 12 months in 1998 and by 30 and 50%, respectively, in 2010 [7, 8]. Today, many paediatric nephrologists still consider RRT in infants as an optional rather than a mandatory treatment [9, 10]. Considering renal transplantation, graft survival has been consistently reported to be worse in the youngest children when compared with older ones [1]. Although many studies have investigated the impact of morphological [11–13] and immunological matching [14, 15] on graft survival, major discrepancies in the minimal recipient weight for renal transplantation, the choice of deceased versus living donor and minimal human leucocyte antigen (HLA) matching have been reported between centres and countries [16]. Moreover, no study has investigated the impact of these factors together. Thus we aim to study patient and graft survival in children starting RRT before 2 years of age to help clinicians and parents make decisions on RRT initiation and transplantation programming. MATERIALS AND METHODS Study population and data We included all patients who started RRT before 24 months of age in France between 1 January 1992 and 31 December 2012. In order to be exhaustive, three complementary ways of detecting patients were used: the national transplant database (CRISTAL), which records data on all the patients registered on the waiting list since 1992; the Renal Epidemiology and Information Network Registry, which records data on all patients on RRT and is exhaustive in children since 2005; and all individual hospital databases. Patient characteristics recorded were age at RRT initiation and at renal transplant; sex; PRD, which were classified in four groups [congenital abnormalities of the kidney and the urinary tract (CAKUT), genetic diseases, vascular diseases and others]; first RRT modality [peritoneal dialysis (PD), haemodialysis (HD) or pre-emptive transplantation); comorbidities (including neurological abnormality, liver disease, heart disease and respiratory insufficiency, from which we created the binary variable ‘at least one comorbidity’); and date of death if applicable. Data on the transplantation included age at renal transplant, weight and height of the donor and the recipient, type of donor (living or deceased), HLA matching and cold ischaemia time. Statistical analysis For descriptive analysis, continuous variables are given as median and interquartile range (IQR) and dichotomous variables as number and percentage. Patients’ survival and access to renal transplantation We present the cumulative incidence of death and transplantation overall and stratified by risk factors. In order to assess the factors associated with the risk of death on dialysis, we used univariate and multivariate Cox proportional hazards regression, with the time between the start of RRT and death as the primary outcome and renal transplantation as a censoring event. Patients who received pre-emptive transplantation (n = 7) were excluded from this analysis. All variables with a P-value of 0.20 in the univariate analysis were included in the multivariate analysis. We tested for the presence of a cohort effect by testing the association between the year at RRT initiation and death with the SAS macro RCS-REG (SAS Institute, Cary, NC, USA), which tests both the linearity and the significance of the association using spline modelling. Renal transplant survival The primary outcome was a composite outcome defined as graft loss or death. We used univariate Cox regression to assess the association between each variable and the outcome. The proportional hazards assumption of the models was assessed by graphical methods. All continuous variables were tested for linearity with the SAS macro RCS-REG. We used Kaplan–Meier modelling to present the survival curve. Multivariate Cox regression was used to assess factors associated with graft loss. RESULTS We included 224 patients who started RRT before 24 months of age. Among them, 141 (62.9%) were male, the median age at RRT initiation was 10.4 (IQR 5.8–15.6) months and the most prevalent PRDs were CAKUT and genetic diseases. Of these, 48 patients presented at least one extrarenal comorbidity, including 21 patients with associated neurological abnormalities, 20 patients with associated liver diseases, 14 patients with heart diseases and 5 patients with concomitant respiratory insufficiency. A flow chart of the cohort is presented as Supplementary data, Figure S1 and patients’ characteristics at RRT initiation are presented in Table 1. Table 1. Patients’ characteristics at RRT initiation Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Table 1. Patients’ characteristics at RRT initiation Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  Characteristics (n = 224)  Value  Gender (male), n (%)  141 (62.9)  Age at RRT start (months), mean (IQR)  10.4 (5.8–15.6)  Weight at RRT start, mean (IQR)  8.2 (6.3–10.0)  PRD, n (%)     CAKUT  76 (34.1)   Genetic diseases  85 (35.1)   Vascular diseases  33 (14.8)   Others  29 (13.0)  Comorbidities, n (%)     None  96 (42.9)   At least one  45 (20.1)   Unknown  83 (37.1)  Initial treatment modality, n (%)     Peritoneal dialysis  144 (64.3)   Haemodialysis  61 (27.2)   Dialysis not specified  12 (5.4)   Pre-emptive transplantation  7 (3.1)  The median follow-up time was 78 (IQR 36–147) months. During follow-up, 174 renal transplantations were performed in 171 patients; 29 patients died (18 before renal transplantation and 11 after renal transplantation) and the mortality rate on dialysis was 40 deaths per 1000 person-years compared with 9 deaths for 1000 person-years after transplantation. The two main causes of death were infections and cardiovascular events (Table 2). Only 10 transplantations were performed from a living donor. Overall the 5- and 10-year survivals were 87% (IQR 82–91) and 84% (IQR 77–89), respectively. Table 2. Number and causes of death in the entire cohort (n = 29) Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Table 2. Number and causes of death in the entire cohort (n = 29) Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Cause of death  n  %  Infection  6  21  Cardiovascular  9  31  Treatment withdrawal  0  0  Neoplasia  2  7  Liver failure  2  7  Multiple organ failure  3  10  Unknown  7  24  Table 3. Cause-specific hazards of death on dialysis (n = 217)   Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5        Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5      Table 3. Cause-specific hazards of death on dialysis (n = 217)   Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5        Univariate analysis   Multivariate analysis   Characteristics  HR  95% CI  HR  95% CI  Age at RRT start (months)   0–6  6.1  0.8–48.8  7.7  0.9–63.1   6–12  3.0  0.4–25.0  3.8  0.4–31.9   12–18  1.8  0.2–20.0  2.2  0.2–24.6   18–24  1    1    Gender   Boys  1    1     Girls  1.6  0.6–3.9  0.8  0.3–2.2  PRD   CAKUT  1    1     Genetic diseases  5.6  1.3–25.0  3.9  0.8–18.8   Vascular diseases  1.0  0.1–11.7  0.9  0.1–9.9   Other  3.9  0.7–22.7  3.9  0.6–26.5  Comorbidities   None  1    1     At least one  7.7  2.4–24.5  6.2  1.9–20.1   Unknown  1.8  0.4–8.3  1.5  0.3–6.8  Initial treatment modality   Haemodialysis  2.2  0.9–5.6       Peritoneal dialysis  1         Dialysis not specified  1.3  0.2–10.5      Determinants of patient survival on dialysis and access to transplantation Overall, 217 patients started RRT with dialysis and were included in this analysis. Patients’ characteristics found to be associated with an increased risk of death by univariate analysis were having a genetic disease as the PRD [HR 5.6 (95% CI 1.3–25.0)] as compared to patients with CAKUT} and the presence of one or more extra-renal comorbidities [HR 7.7 (95% CI 2.4–24.5)]. There was a trend towards a decreased risk of death with age, although it did not reach statistical significance. Starting RRT on PD or HD did not significantly impact patient survival. Suffering from extrarenal comorbidities was the only factor significantly associated in multivariate analysis with both the hazard of death [HR 5.9 (95% CI 1.8–19.3)] (Table 3) and the hazard of being transplanted (data not shown). Supplementary data, Figures S2 and S3 present the cumulative incidence of death and renal transplantation stratified on the presence of comorbidity and on the different age groups, respectively. Determinants of graft survival The 10-year graft survival was 74% (IQR 67–81). Figure 1 presents the graft survival curve. FIGURE 1: View largeDownload slide Kaplan–Meier curve of graft survival over the 10-year follow-up. FIGURE 1: View largeDownload slide Kaplan–Meier curve of graft survival over the 10-year follow-up. Over the 171 first renal transplantations performed in our patients, the median dialysis time prior to transplantation was 22.1 (IQR 10.1–33.4) months, the median age at transplantation was 30.2 (IQR 21.8–40.7) months and the median recipient weight and height were 12.0 (IQR 10.4–13.6) kg and 85.0 (IQR 80.5–92.5) cm, respectively. Overall, our population of recipients was very homogeneous, with only 5% of the recipients <1 year old. They were very close in terms of weight and height. This homogeneity precluded us from studying the impact of those variables on renal survival. Variables found to be associated with graft survival by univariate analysis were the donor/recipient height ratio (P = 0.05), the number of human leucocyte antigen–antigen D-related (HLA-DR) mismatches (P = 0.02) and the time spent on dialysis prior to renal transplantation (P = 0.001) (Table 4). Table 4. HRs of graft failure or death in transplanted patients (univariate analysis)     Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73        Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73    Ψ, spline variables. Table 4. HRs of graft failure or death in transplanted patients (univariate analysis)     Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73        Univariate analysis   Characteristics  Value  HR  P-value  Recipients  Female gender, n  67  1.20  0.62  PRD, n      0.93   CAKUT  59  1.00     Genetic diseases  61  0.64     Vascular diseases  27  0.90     Other  24  0.95    Comorbidities, n      0.28   None  101  1.00     At least one  26  2.00     Unknown  44  1.00    Initial treatment type, n      0.91   Haemodialysis  43  1.00     Peritoneal dialysis  111  0.74     Dialysis Not Specified  11  0.83     Pre-emptive transplantation  6  0.83    Time on dialysis prior to transplantation (months), median (IQR)  21.3 (9.7–32.7)  1.20  0.001  Donor characteristics, median (IQR)   Age (years)  11.9 (6–15)  0.99  0.78   Weight (kg)  38.2 (20.0–52.0)  1  0.85   Height (cm)  139.7 (120–162)  1  0.82  Donor–recipient matching, median (IQR)   Donor/recipient body weight ratio  3.82 (1.71–4.17)  Ψ  0.33   Donor/recipient body height ratio  1.62 (1.33–1.87)  Ψ  0.05  DR mismatch number, n      0.02   2  25  1.00     1  104  0.32     0  30  0.48    B mismatch number, n      0.58   2  69  1.00     1  81  1.49     0  9  0.00    A mismatch number, n      0.19   2  50  1.00     1  87  0.56     0  22  0.80    Year of transplantation, n      0.37   1992–2002  63  1.00     2002–12  108  0.73    Ψ, spline variables. In order to make our results potentially useful in clinical practice, we dichotomized the donor/recipient height ratio in five categories (ratio ≤ 1, 1 < ratio < 1.4, 1.4 ≤ ratio ≤ 1.8, 1.8 < ratio < 2.2 and ratio ≥ 2.2). Figure 2 presents the evolution of the HR of graft loss with the donor/recipient height ratio. FIGURE 2: View largeDownload slide Evolution of the HR of graft loss or death with donor/recipient height ratio. FIGURE 2: View largeDownload slide Evolution of the HR of graft loss or death with donor/recipient height ratio. There was a tendency towards an improved graft with an HR of graft lost of 0.73 (IQR 0.36–1.46) between 2002 and 2012 when compared with the 1992–2002 period. The three factors found significantly associated with graft loss in multivariate analysis were the time spent on dialysis before transplantation, donor/recipient height ratio and presenting two HLA-DR mismatches (Table 5). Table 5. HRs of graft failure or death in transplanted patients (multivariate analysis) Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Table 5. HRs of graft failure or death in transplanted patients (multivariate analysis) Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  Variable  HR  95% CI  Donor/recipient body height ratio ≤1  4.54  1.34–15.41  1 < ratio < 1.4  1.89  0.44–8.11  1.4 ≤ ratio ≤ 1.8  1.00    1.8 < ratio < 2.2  1.83  0.41–8.25  ≥2.2  9.53  1.58–57.67  Number of HLA-DR mismatches       0–1  1.00     2  2.29  1.08–4.86  Time on dialysis prior to transplantation  1.02  1.01–1.04  DISCUSSION In this national cohort of patients starting RRT before the age of 2 years, we confirmed the overall good patient survival with a 5-year survival of 87%, close to the survival reported in the USA in patients 0–4 years of age (84%) [17]. As reported in many previous studies [5, 6, 18, 19], we found a trend towards an increased risk of death in the youngest children, although age at RRT initiation did not reach statistical significance. Carey et al. [20] did not find a significant difference in survival between patients starting RRT before 1 month and those starting between 1 and 24 months of age; however, because North American Pediatric Renal Trials and Collaborative Studies only includes patients who started dialysis in centres willing to participate in the registry, selection bias of the youngest patients with the most favourable prognosis might explain the absence of effect of age at initiation of RRT in this study. In our study, the only factor significantly associated with patient survival was the presence of extrarenal comorbidities, with a cumulative incidence of death ranging from 4% in patients without extrarenal comorbidities to 30% in patients with at least one comorbidity. This is consistent with the data published by Wedekin et al. [21] in a cohort of patients starting RRT at <1 year of age and with the causes of death reported in The Netherlands in ESRD patients 0–14 years of age treated between 1972 and 1992 [22]. Similarly, a study focusing on patients with CAKUT demonstrated that the major risk factor of death was the association with extrarenal abnormalities [23]. In our study, no difference in survival was found between patients starting RRT with HD or PD. This result is consistent with a recent publication from the European Society of Paediatric Nephrology/European Renal Association–European Dialysis and Transplant Association (ESPN/ERA-EDTA) registry that did not find any difference in survival between HD and PD in patients starting RRT before the age of 1 year [24]. Nevertheless, the overall survival of infants starting RRT remains lower than the survival of older children [17], even though their survival post-transplantation is excellent and not different from the survival of older children [21]. However, many studies have emphasized the issues raised by renal transplantation in this population. Should we favour living kidney donation, usually from adults with major morphological differences? Is HLA matching still important in infants? However, mixing all these results together to guide clinical decisions remains challenging, so that there is a great heterogeneity of practices. In our study, morphological matching, especially in height, was highly associated with graft survival and a donor/recipient height ratio between 1.4 and 1.8 was found to be associated with the best graft survival. One of the main causes of graft loss in patients <2 years of age at transplantation is vascular thrombosis [25, 26], and many studies have reported worse graft outcomes when using age-matched donors. A recent study from the ESPN/ERA-EDTA registry confirmed the increased risk of graft loss in children receiving a transplant from a deceased donor <5 years of age [27]. Moreover, Dick et al. [12] found decreased renal survival among adolescent recipients who received a kidney from a smaller donor (donor/recipient body surface area ratio < 0.9), confirming the risk associated with smaller donors [12]. Pape et al. [28], studying children <10 years of age, questioned the use of kidneys from adult donors, because they found that 3–5 years after transplantation the corrected glomerular filtration rate (GFR) was significantly higher in children who had received a paediatric graft, with grafts also doubling in size, whereas no increase in size was noted in adult grafts. Our study confirms that in the youngest patients the use of height-matched kidneys increases the risk of graft loss and should therefore be avoided. Moreover, the scarcity of small paediatric kidney donors would lead to unacceptable waiting times on dialysis [11]. Our study also shows that a long time on dialysis prior to transplantation is associated with an increased risk of graft loss. Thus the benefit of waiting for an optimal transplant in terms of HLA and size matching might be outweighed by the negative impact of a long waiting time. This finding calls for regular and individualized re-evaluation of the requirements regarding transplant characteristics as time on dialysis increases. The other factor found to be significantly associated with graft survival in our patients was HLA matching for Class 2 antigen DR. Recently, authors have questioned the importance of HLA matching [29] and emphasized the importance of other factors such as prioritizing kidney from young adults for paediatric recipients [14] and minimizing cold ischaemia time. Moreover, not waiting for an HLA-matched kidney could favour rapid access to transplantation and allow pre-emptive transplantation in some children. However, poorer HLA matching, especially in Class 2, harbours the risk of de novo development of panel reactive antibodies (PRA) that could hamper both the ability to receive subsequent transplantation and graft survival [30]. This is a major concern in young children receiving renal transplantation since they will require several transplantations in their lifetime. Moreover, this issue remains unclear, as Gritsch et al. [14] did not find any difference in PRA level between well-matched and unmatched patients. Recently, Tinckam et al. [31] also reported on the risk of repeated HLA mismatch and showed that the presence of HLA Class 2 mismatches and PRA before the second transplantation decreases graft survival. This effect was even stronger in patients who underwent a nephrectomy of the first transplant. Finally, the major issue faced by clinicians when considering the optimal strategy of transplantation is to combine all the data together to make evidence-based decisions. Among adults, several authors have created clinical scores aimed at predicting renal transplantation outcome based on pretransplant [32] and post-transplant data [33–36]. Among children, a study aimed at defining groups of patients by the risk of graft loss has been recently published based on data from the ESPN/ERA-EDTA registry [37]. However, patients starting RRT before 2 years of age require specific studies focusing on this population and will require international collaboration. Based on the data of this national cohort, we were able to confirm and specify risk factors of patient and transplant survival in the youngest children starting RRT. Our study also has some limitations. First, the retrospective design of the study and the absence of termination of pregnancy and conservative treatments induces a selection bias since the study of patient survival does not apply to all patients with ESRD but to those who started RRT. Moreover, practices of RRT initiation differ between countries and centres. Due to the scarcity of patients starting RRT before 2 years of age, we included patients over a 20-year period. Although we did not find any statistically significant cohort effect, modifications of practices could modify the association reported. Similarly, given the scarcity of these patients, we grouped the PRDs into five categories; this did not allow us to study the impact of each specific disease on the outcomes. Finally, the scarcity of transplantation from living donors precluded us from comparing graft survival by the type of donor. CONCLUSION In this national retrospective cohort study, we confirm the overall good outcome of children starting RRT before 2 years of age, especially in the absence of extrarenal comorbidities. Thus the main questions in infants with ESRD remain when and how to transplant them. Our study provides data on the optimal morphological and immunological matching in order to help clinicians in their decisions. By extending this analysis to other cohorts, we aim to develop a tool that is able to predict graft loss and help clinicians when choosing the optimal kidney for their patients. SUPPLEMENTARY DATA Supplementary data are available at ndt online. AUTHORS’ CONTRIBUTIONS J.H. and C.C. participated in the conception of the article, analysis of the data and writing of the manuscript. M.-A.M. participated in the conception of the article, interpretation of the data and writing of the manuscript. J.B., M.C., G.R., R.N., M.T., J.T., T.U., A.G., E.M., J.H.a., I.V., O.D., D.M., E.B. and F.N. participated in data collection, interpretation of the data and writing of the manuscript. All the authors have revised the article and approve the final version. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 Harambat J, Hogan J, Macher M-A et al.   [ESRD in children and adolescents]. Nephrol Ther  2013; 9(Suppl 1): S167– S179 Google Scholar CrossRef Search ADS PubMed  2 Pippias M, Kramer A, Noordzij M et al.   The European Renal Association – European Dialysis and Transplant Association Registry Annual Report 2014: a summary. Clin Kidney J  2017; 10: 154– 169 Google Scholar PubMed  3 Ronco C, Garzotto F, Brendolan A et al.   Continuous renal replacement therapy in neonates and small infants: development and first-in-human use of a miniaturised machine (CARPEDIEM). Lancet  2014; 383: 1807– 1813 Google Scholar CrossRef Search ADS PubMed  4 Coulthard MG, Crosier J, Griffiths C et al.   Haemodialysing babies weighing <8 kg with the Newcastle infant dialysis and ultrafiltration system (Nidus): comparison with peritoneal and conventional haemodialysis. Pediatr Nephrol  2014; 29: 1873– 1881 Google Scholar CrossRef Search ADS PubMed  5 Mitsnefes MM, Laskin BL, Dahhou M et al.   Mortality risk among children initially treated with dialysis for end-stage kidney disease, 1990–2010. JAMA  2013; 309: 1921– 1929 Google Scholar CrossRef Search ADS PubMed  6 McDonald SP, Craig JC. Long-term survival of children with end-stage renal disease. N Engl J Med  2004; 350: 2654– 2662 Google Scholar CrossRef Search ADS PubMed  7 Geary DF. Attitudes of pediatric nephrologists to management of end-stage renal disease in infants. J Pediatr  1998; 133: 154– 156 Google Scholar CrossRef Search ADS PubMed  8 Teh JC, Frieling ML, Sienna JL et al.   Attitudes of caregivers to management of end-stage renal disease in infants. Perit Dial Int  2011; 31: 459– 465 Google Scholar CrossRef Search ADS PubMed  9 Lantos JD, Warady BA. The evolving ethics of infant dialysis. Pediatr Nephrol  2013; 28: 1943– 1947 Google Scholar CrossRef Search ADS PubMed  10 Wightman A, Kett J. Has neonatal dialysis become morally obligatory? Lessons from Baby Doe. Acta Paediatr  2015; 104: 748– 750 Google Scholar CrossRef Search ADS PubMed  11 Goldsmith PJ, Asthana S, Fitzpatrick M et al.   Transplantation of adult-sized kidneys in low-weight pediatric recipients achieves short-term outcomes comparable to size-matched grafts. Pediatr Transplant  2010; 14: 919– 924 Google Scholar CrossRef Search ADS PubMed  12 Dick AAS, Mercer LD, Smith JM et al.   Donor and recipient size mismatch in adolescents undergoing living-donor renal transplantation affect long-term graft survival. Transplantation  2013; 96: 555– 559 Google Scholar CrossRef Search ADS PubMed  13 Donati-Bourne J, Roberts HW, Coleman RA. Donor-recipient size mismatch in paediatric renal transplantation. J Transplant  2014; 2014: 317574 Google Scholar CrossRef Search ADS PubMed  14 Gritsch HA, Veale JL, Leichtman AB et al.   Should pediatric patients wait for HLA-DR-matched renal transplants? Am J Transplant  2008; 8: 2056– 2061 Google Scholar CrossRef Search ADS PubMed  15 Foster BJ, Dahhou M, Zhang X et al.   Impact of HLA mismatch at first kidney transplant on lifetime with graft function in young recipients. Am J Transplant  2014; 14: 876– 885 Google Scholar CrossRef Search ADS PubMed  16 van Huis M, Schoenmaker NJ, Groothoff JW et al.   Policy variation in donor and recipient status in 11 pediatric renal transplantation centers. Pediatr Nephrol  2013; 28: 951– 957 Google Scholar CrossRef Search ADS PubMed  17 United States Renal Data System. USRDS Annual Data Report 2016. https://www.usrds.org/adr.aspx (21 February 2018, date last accessed) 18 Samuel SM, Tonelli MA, Foster BJ et al.   Survival in pediatric dialysis and transplant patients. Clin J Am Soc Nephrol  2011; 6: 1094– 1099 Google Scholar CrossRef Search ADS PubMed  19 Alexander RT, Foster BJ, Tonelli MA et al.   Survival and transplantation outcomes of children less than 2 years of age with end-stage renal disease. Pediatr Nephrol  2012; 27: 1975– 1983 Google Scholar CrossRef Search ADS PubMed  20 Carey WA, Talley LI, Sehring SA et al.   Outcomes of dialysis initiated during the neonatal period for treatment of end-stage renal disease: a North American Pediatric Renal Trials and Collaborative Studies special analysis. Pediatrics  2007; 119: e468– e473 Google Scholar CrossRef Search ADS PubMed  21 Wedekin M, Ehrich JHH, Offner G et al.   Renal replacement therapy in infants with chronic renal failure in the first year of life. Clin J Am Soc Nephrol  2010; 5: 18– 23 Google Scholar CrossRef Search ADS PubMed  22 Groothoff JW, Gruppen MP, Offringa M et al.   Mortality and causes of death of end-stage renal disease in children: a Dutch cohort study. Kidney Int  2002; 61: 621– 629 Google Scholar CrossRef Search ADS PubMed  23 Melo BF, Aguiar MB, Bouzada MCF et al.   Early risk factors for neonatal mortality in CAKUT: analysis of 524 affected newborns. Pediatr Nephrol  2012; 27: 965– 972 Google Scholar CrossRef Search ADS PubMed  24 Vidal E, van Stralen KJ, Chesnaye NC et al.   Infants requiring maintenance dialysis: outcomes of hemodialysis and peritoneal dialysis. Am J Kidney Dis  2017; 69: 617– 625 Google Scholar CrossRef Search ADS PubMed  25 Hwang AH, Cho YW, Cicciarelli J et al.   Risk factors for short- and long-term survival of primary cadaveric renal allografts in pediatric recipients: a UNOS analysis. Transplantation  2005; 80: 466– 470 Google Scholar CrossRef Search ADS PubMed  26 Keller AK, Jorgensen TM, Jespersen B. Identification of risk factors for vascular thrombosis may reduce early renal graft loss: a review of recent literature. J Transplant  2012; 2012: 1 Google Scholar CrossRef Search ADS   27 Chesnaye NC, van Stralen KJ, Bonthuis M et al.   The association of donor and recipient age with graft survival in paediatric renal transplant recipients in a European Society for Paediatric Nephrology/European Renal Association-European Dialysis and Transplantation Association Registry study. Nephrol Dial Transplant  2017; 32: 1949– 1956 Google Scholar CrossRef Search ADS PubMed  28 Pape L, Hoppe J, Becker T et al.   Superior long-term graft function and better growth of grafts in children receiving kidneys from paediatric compared with adult donors. Nephrol Dial Transplant  2006; 21: 2596– 2600 Google Scholar CrossRef Search ADS PubMed  29 Hoyer PF. ‘Tie breaker’ for HLA matching in pediatric renal transplant recipients? Am J Transplant  2008; 8: 1970– 1971 Google Scholar CrossRef Search ADS PubMed  30 Wiebe C, Gibson IW, Blydt-Hansen TD et al.   Evolution and clinical pathologic correlations of de novo donor-specific HLA antibody post kidney transplant. Am J Transplant  2012; 12: 1157– 1167 Google Scholar CrossRef Search ADS PubMed  31 Tinckam KJ, Rose C, Hariharan S et al.   Re-examining risk of repeated HLA mismatch in kidney transplantation. J Am Soc Nephrol  2016; 27: 2833– 2841 Google Scholar PubMed  32 Molnar MZ, Nguyen DV, Chen Y et al.   Predictive score for posttransplantation outcomes. Transplantation  2017; 101: 1353– 1364 Google Scholar CrossRef Search ADS PubMed  33 Foucher Y, Daguin P, Akl A et al.   A clinical scoring system highly predictive of long-term kidney graft survival. Kidney Int  2010; 78: 1288– 1294 Google Scholar CrossRef Search ADS PubMed  34 Kasiske BL, Israni AK, Snyder JJ et al.   A simple tool to predict outcomes after kidney transplant. Am J Kidney Dis  2010; 56: 947– 960 Google Scholar CrossRef Search ADS PubMed  35 Gourishankar S, Grebe SO, Mueller TF. Prediction of kidney graft failure using clinical scoring tools. Clin Transplant  2013; 27: 517– 522 Google Scholar CrossRef Search ADS PubMed  36 Shabir S, Halimi J-M, Cherukuri A et al.   Predicting 5-year risk of kidney transplant failure: a prediction instrument using data available at 1 year posttransplantation. Am J Kidney Dis  2014; 63: 643– 651 Google Scholar CrossRef Search ADS PubMed  37 Lofaro D, Jager KJ, Abu-Hanna A et al.   Identification of subgroups by risk of graft failure after paediatric renal transplantation: application of survival tree models on the ESPN/ERA-EDTA Registry. Nephrol Dial Transplant  2016; 31: 317– 324 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nephrology Dialysis TransplantationOxford University Press

Published: Mar 29, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off