Journal of Pediatric Surgery xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Outcome differences between young children and adolescents undergoing kidney transplantation Iuliana D. Bobanga a, Beth A. Vogt b, Kenneth J. Woodside a,⁎, Devan R. Cote a, Katherine M. Dell b, Robert J. Cunningham III b, Kelly A. Noon a, Edward M. Barksdale a, Vanessa R. Humphreville a, Edmund Q. Sanchez a, James A. Schulak a a b
Department of Surgery, Case Western Reserve University & University Hospitals Rainbow Babies & Children's Hospital, Cleveland, OH Departments of Pediatrics, Case Western Reserve University & University Hospitals Rainbow Babies & Children's Hospital, Cleveland, OH
a r t i c l e
i n f o
Article history: Received 25 February 2015 Accepted 10 March 2015 Available online xxxx Key words: Pediatric Kidney transplantation Graft survival Rejection Adolescence
a b s t r a c t Background/Purpose: Although graft loss remains the biggest challenge for all pediatric kidney transplant (KT) recipients, unique challenges exist within different age groups. We aim to evaluate the different characteristics and graft survival outcomes of young children and adolescents undergoing KT. Methods: Children who underwent isolated KT between 2000 and 2013 at our institution were included in this retrospective analysis. Patient characteristics and outcomes were compared using student’s t-test, chi-square test, Kaplan-Meier curve and Cox proportional hazards model. Results: Of 73 children who underwent KT, 31 were b 12 (young children), and 42 were ≥12 years old (adolescents). Overall patient survival was 100%. The younger group had superior 5-year (100% vs. 75.5%) and 10-year (94.4% vs. 43.8%) graft survival (p = 0.008). Factors predictive of poor graft survival on multivariate analysis were older age at transplantation (HR 1.2, CI 1–1.4, p = 0.047), female gender (HR 9.0, CI 1.9–43, p = 0.006), and acute rejection episodes (HR 13, CI 2–90, p = 0.008). The most common causes of graft loss were acute and chronic rejection episodes and immunosuppression nonadherence. Conclusion: Adolescents undergoing KT have inferior graft survival compared to younger children. In adjusted modeling, children with older age, female gender, and acute rejection episodes have inferior graft survival. © 2015 Elsevier Inc. All rights reserved.
Kidney transplantation (KT) is the treatment of choice for children with end-stage renal disease (ESRD), as it improves growth and increases life expectancy compared to dialysis [1,2]. Significant improvements in pediatric KT outcomes have been made over the past four decades across all recipient subgroups, with approximately a 5% decrease in both risk of patient death and graft loss with each more recent year of transplantation [3]. These improvements are probably due to better immunosuppression, alloantibody detection and infectious disease prevention [4]. Despite these advances, there is still a large discrepancy between patient survival and graft survival, which results in many pediatric recipients eventually requiring retransplantation. Multiple factors may affect patient and graft survival, including recipient and donor age, recipient race and gender, transplant and Abbreviations: KT, kidney transplant; ESRD, end stage renal disease; HLA, human leukocyte antigen; NAPRTCS, North American Pediatric Renal Trials and Collaborative Studies; LD, living donor; DD, deceased donor; HR, hazard ratio; CI, confidence interval; CAKUT, congenital anomalies of the kidney and urinary tract; PD, peritoneal dialysis; HD, hemodialysis; LRRT, living related renal transplant; DDRT, deceased donor renal transplant; FSGS, focal segmental glomerulosclerosis; MPGN, membranoproliferative glomeruloephritis; SRTR, Scientific Registry of Transplant Recipients. ⁎ Corresponding author at: Section of Transplant Surgery, Department of Surgery, 2922 Taubman Center, 1500 E. Medical Center Dr., Ann Arbor, MI 48109-5300. Tel.: +1734 936 8363. E-mail address: [email protected] (K.J. Woodside).
transfusion history, HLA mismatches, episodes of rejection and posttransplant complications, among others [5,6]. While the age most commonly associated with increased risk is infancy, several studies have identified lower graft survival among patients undergoing KT during adolescence and early adulthood [7–9]. A report from the North American Pediatric Renal Trials and Collaborative Studies (NAPRTCS) showed that graft survival rates improved steadily from 1987 to 1999 but remained relatively stable between 2000 to 2010, with the 5-year living donor (LD) graft survival of 84.3% and the 5-year deceased donor (DD) graft survival of 78% [5]. The factors that contributed to this leveling off have not been well defined. The aim of this study was to evaluate patient and graft survival in children and adolescents who underwent kidney transplantation at our institution from 2000 to 2010 with the goal of identifying causes of graft loss and risk factors for poor graft survival. 1. Methods 1.1. Study population We performed a retrospective study of all children (≤18 years old) who underwent kidney transplantation between January 2000 and December 2013 at University Hospitals Rainbow Babies and Children’s
http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
2
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Hospital in Cleveland, Ohio. Patients were excluded if they had a prior or concomitant non-renal solid organ transplant. If patients had undergone a previous kidney transplant prior to 2000, data were collected with the focus on the transplant that occurred after 2000. Patients were divided into two groups based on age at kidney transplantation: young children b12 years, and adolescents ≥12 years. Age 12 was chosen to compare adolescents in middle school and high school that undergo various physiological and social changes from elementary school and younger children [10,11]. Patient demographics, operative and perioperative data, short and long-term outcomes, as well as donor demographics, relationship, and donor-recipient HLA matching data were recorded. The Institutional Review Board of University Hospitals approved the study (IRB #11-13-15). 1.2. Study outcomes The primary outcome of interest was allograft survival, defined as the time between date of transplantation and either date of graft failure (marked by retransplantation or return to dialysis) or last date of follow-up with a functioning graft. Other outcomes of interest were length of stay, delayed graft function, causes of graft loss, disease recurrence and allograft rejection. Delayed graft function was defined as requirement for dialysis within the first week after transplantation. 1.3. Statistical analysis Descriptive statistics were used to produce frequencies and percentages for dichotomous variables and mean with standard deviations for continuous variables. Categorical variables were compared using the Chi-square test and continuous variables were compared using independent-sample t test. To evaluate factors associated with allograft loss, we developed a Cox Proportional Hazards model and performed a univariate and multivariate analysis. Crude and adjusted hazards ratios (HRs) with 95% confidence interval (95 % CI) and p values are reported. All variables with a p value b0.100 on univariate analysis were included in the multivariate model. Patients who were lost to follow-up were censored at the time of last follow-up. Time to allograft loss between the two age groups was compared using the Kaplan-Meier method and the log-rank test. A p value ≤0.05 was considered statistically significant throughout the analysis. Statistical analysis was performed using SPSS version 21.0 (SPSS, Inc., Chicago, IL). 2. Results 2.1. Baseline characteristics of recipients and donors Seventy-five pediatric patients ≤ 18 years old underwent kidney transplantation between January 2000 and December 2013. Two patients were excluded, as they had a previous or concomitant liver transplant. Of the 73 patients included (38% female, median age of 13 years at transplantation), 31 were b12 years old at KT and 42 were ≥12 years old (Table 1). Five patients had two or more KTs (one in the younger group, four in the older group). Of those, data was captured regarding the first transplant for two patients and regarding the second transplant for three patients, depending on which transplant occurred after the year 2000. The most common cause of ESRD was congenital anomalies of the kidney and urinary tract (CAKUT), which was more common in the young children group compared to the adolescent group (67.7% vs. 40.5%, p = 0.033). Pretransplant peritoneal dialysis (PD) was more frequently performed in the young children group (72% vs. 40%, p = 0.017). Many children had received both PD and hemodialysis (HD) at different time points and 27.4% of the entire cohort underwent preemptive transplantation. The majority of children (64%) received a live related renal transplant (LRRT), usually from one of their parents, with a trend for
Table 1 Baseline characteristics of recipients and donors. Baseline characteristics
Entire group (n = 73)
Age at KT b 12 (n = 31)
Age at KT ≥ 12 (n = 42)
p-value (b12 vs. ≥ 12)
Age at ESRD Age at KT Gender (% female) Weight (kg) Recipient race (%) Caucasian African American Other Diagnosis (%) FSGS Other glomerular CAKUT Other/missing Pretransplant dialysis (%) Hemodialysis Peritoneal dialysis Pre-emptive Extraperitoneal (%) Ureteral stent use (%) Donor (%) Live related Live unrelated Deceased Mean donor age HLA mismatches 1–3 HLA mismatches 4–6
9.7 ± 5.8 11.7 ± 5.3 38.4 36.6 ± 20
4.5 ± 3.8 6.4 ± 3.3 30 19.5 ± 6.8
13.5 ± 3.4 15.5 ± 2 45 51.2 ± 16
b0.001 b0.001 b0.001 b0.001
75 16.6 8.3
74 16 9.6
75 17 7.3
1 1 1
15.1 15.1 52.1 17.8
6.5 9.7 67.7 16.1
21.4 19 40.5 19
0.103 0.335 0.033 1
42.5 53.4 27.4 89 69
35.5 71 25.8 77.4 58.6
47.6 40.5 28.6 97.6 76.2
0.345 0.017 1 0.009 0.128
64 10.7 24 37.5 ± 7 64.2 35.8
74.2 12.9 12.9 34.8 ± 6.8 74.1 25.9
57.1 9.5 33.3 40.4 ± 6 53.8 46.2
0.148 0.716 0.057 0.003 0.158 0.158
ESRD: end stage renal disease. KT: kidney transplant. FSGS: focal segmental glomerulosclerosis. CAKUT: congenital anomalies of kidney and urinary tract. HLA: human leukocyte antigen.
increased LRRT in the young children group and an increased deceased donor renal transplant (DDRT) in the adolescent group. The mean donor age was also lower in the younger group (35 vs. 40 years, p = 0.003). There were no differences in HLA mismatches between the two groups. 2.2. Outcomes and graft survival The mean follow-up time for the entire cohort was 6 years. The average length of stay was 8 days and did not differ significantly between the two groups. Delayed graft function occurred in 8.2% of patients and occurred more commonly in the adolescent group (Table 2). There was no difference in unplanned return to the operating room within 30 days of transplantation. Five patients (6.8%) had disease recurrence, three with focal segmental glomerulosclerosis (FSGS) within 90 days of transplantation, one with membranoproliferative glomerulonephritis (MPGN) seen on transplant nephrectomy three years post-transplantation and one with mesangial proliferative glomerulonephritis along with primary nonfunction immediately posttransplantation. Recurrent disease was evenly distributed between the two groups (Table 1). The incidence of acute biopsy-proven rejection was 31.5% for the entire cohort and did not differ significantly between younger children and adolescents. For one third of the patients (33%), their first acute rejection episode occurred within 6 months of transplantation while 37.5% had their first acute rejection episode within 1 year of transplantation. On univariate logistic regression analysis, black race, FSGS diagnosis, male gender, post-transplant CMV infection, and increased donor age predicted acute rejection episodes, while a diagnosis of CAKUT was associated with a lower risk of acute rejection episodes. In a multivariate analysis, black race (OR 14.9, CI 2.7–83, p = 0.002) and FSGS diagnosis (OR 7.7, CI 1.5–39, p = 0.014) remained significant risk factors. On further review of the medical records, we found that 52% of patients’ medical providers mentioned suspicion or patient
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
3
Table 2 Post-operative outcomes and graft survival. Outcome
Entire group (n = 73)
Age at KT b 12 (n = 31)
Age at KT ≥ 12 (n = 42)
p-value (b12 vs. ≥ 12)
Total f/u time (years) Length of stay (days) Delayed graft function requiring dialysis Unplanned return to OR within 30 days Recurrent disease Acute rejection Pre/post KT hernia Graft survival (%) 1 year 3 years 5 years 10 years
6.0 ± 3.7 8.4 ± 5.8 8.2
6.2 ± 4.1 9.7 ± 8 3.2
5.8 ± 3.3 7.4 ± .3 11.9
0.700 0.100 0.233
8.2
9.7
7.1
1
6.8 31.5 19.2
6.5 29 32
7.1 33 9.5
1 0.801 0.019 0.008
98.6 91.5 85.6 70.9
100 100 100 94.4
97.6 85.7 75.5 43.8
f/u: follow-up. OR: operating room. KT: kidney transplant.
admission of immunosuppression nonadherence, with a significantly higher rate in patients with acute cellular rejection than in patients without acute rejection episodes (74% versus 40%, respectively; p = 0.017). Interestingly, there was an increased incidence of hernias (inguinal, umbilical or incisional) noted in younger children undergoing transplantation. The majority were inguinal hernias (78.5%) and 3 of them (21%) were umbilical hernias. Only one patient of the entire cohort (1.3%) had an incisional hernia post-transplantation. The majority (80%) of those younger children with hernias underwent PD, while only one (25%) of the adolescents with hernia underwent PD, which may have contributed to hernia development. The overall patient survival was 100%. The graft survival at one, three, five and ten years was 98.6%, 91.5%, 85.6% and 70.9%, respectively, and was significantly different between the two age groups (log-rank test, p = 0.008), with younger children having an improved graft survival compared to adolescents (Table 2, Fig. 1). On subgroup analysis of only LRRT, (n = 23 in both groups), there was still a superior graft survival noted for younger children (long-rank test, p = 0.007). Thirteen patients had graft failure, two in the younger age group and 11 adolescents. One graft failure was due to primary non-function, while the remainder occurred between 28 months to 11 years posttransplantation. Most patients had multiple causes attributed for graft failure. The majority had graft failure due to chronic rejection (54%) or acute rejection (46%). Most patients with graft loss had documented suspicion of nonadherence or had discontinued their immunosuppressive medications (11 patients, 85%). There was a significantly higher rate of graft failure in patients with provider suspicion or patient admission of immunosuppression nonadherence than in presumably compliant patients (33% vs. 5%, p = 0.004). Two patients had disease recurrence, one with primary non-function and recurrent mesangial proliferative glomerulonephritis immediately post-transplant and one with chronic rejection and recurrent MPGN on transplant nephrectomy pathology 29 months post-transplantation. Lastly, one patient had graft loss 5.5 years after transplantation due to multiple episodes of pyelonephritis.
2.3. Risk factors for poor graft survival Multiple risk factors reported in the literature to affect graft survival were analyzed in our cohort. On univariate Cox Proportional Hazards Ratio analysis, older age at transplantation female gender, FSGS diagnosis and acute biopsy-proven rejection were associated with increased risk of graft loss, while a diagnosis of CAKUT was predictive of a lower risk of graft loss. On further multivariate Cox Proportional Hazards Ratio analysis, older age at transplantation (HR 1.2, CI 1–1.4, p =
Fig. 1. Kaplan-Meier graft survival curve comparing young children (b12 years) and adolescents (≥12 years) at kidney transplantation. Log rank test, p = 0.008.
0.047), female gender (HR 9.0, CI 1.9–43, p = 0.006) and acute rejection (HR 13, CI 2-90.6, p = 0.008) remained significant predictors of graft loss (Table 3). 3. Discussion We examined the risk factors associated with kidney allograft survival in two pediatric age groups and found that younger children have superior graft survival compared to adolescents. The most frequent causes of graft failure were acute or chronic rejection and nonadherence or discontinuation of immunosuppressive medications. In addition, on multivariate analysis, older age at transplantation, acute rejection episodes and female gender were predictive risk factors for graft failure for all pediatric patients undergoing kidney transplantation. Living donor kidney transplants typically have better graft survival compared to those from deceased donors [2,5]. One significant difference between our two age groups is the higher percent of living donor grafts in the younger group (87% vs. 67%). However, on subgroup analysis including only patients who received a LRRT, younger children still had a superior graft survival compared to adolescents that was statistically significant. Our finding of improved graft survival in the younger age group b12 years old are consistent with other studies showing poorer outcomes in adolescents and young adults undergoing KT [5,7–9,12,13]. Smith et al reviewed the renal transplant outcomes in adolescents using the NAPRTCS database and found that adolescents 13-17 years of age had the poorest long-term graft survival of LD grafts compared to all other pediatric age groups. Furthermore, adolescents had later acute rejection episodes and incomplete rejection reversal [7]. In addition, Van Arendonk et al used the Scientific Registry of Transplant Recipients (SRTR) data from 1987 to 2010 to show that there is a much higher risk of graft loss for patients transplanted at any age while passing through the high-risk age window of 17–24 years old [12]. Possible explanations for this phenomena include higher rates of non-adherence to immunosuppression during late adolescence and early adulthood [14], maturation of the immune system, changes in health insurance, and transitioning from pediatric to adult care. Another hypothesis proposes that there is a period of increased growth during adolescence causing hyperfiltration injury [12].
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
4
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Table 3 Univariate and multivariate Cox proportional hazards ratios for predictors of graft loss. Variable
Univariate Cox proportional hazards ratio cHR
Age at KT (per year) Female Black race FSGS CAKUT Living related donor Acute rejection Deceased donor Extraperitoneal Donor age Previous KT
1.1 3.8 3.1 4.2 0.1 0.8 6.5 2.2 0.9 1 1.7
95% CI
p-value
1.0–1.3 1.2–12.2 0.9–10.5 1.4–12.9 0.03–0.58 0.24–2.53 1.8–23.7 0.60–8.46 0.17–4.38 0.9–1.15 0.2–13.6
0.026 0.027 0.066 0.012 0.008 0.672 0.005 0.231 0.859 0.834 0.597
Multivariate Cox Proportional Hazards Ratio aHR Age at KT (per year) Female Black race FSGS CAKUT Acute rejection
1.2 9.0 1.4 1.1 0.4 13
95% CI 1.0–1.4 1.9–43 0.3–6.7 0.2–4.9 0.1–2.5 2.0–90.6
p-Value 0.047 0.006 0.672 0.904 0.364 0.008
KT: kidney transplant. FSGS: focal segmental glomerulosclerosis. CAKUT: congenital anomalies of kidney and urinary tract. HLA: human leukocyte antigen. cHR: crude hazards ratio. CI: confidence interval. aHR: adjusted hazards ratio.
Our study also found a high rate of non-adherence to immunosuppression among adolescents who experienced graft loss. Interestingly, we saw no difference in the frequency of acute biopsy-proven rejection episodes between the two cohorts (Table 2), but a much higher rate of graft loss in the adolescent group attributed to acute or chronic rejection. This finding is consistent with the observation by Smith et al that adolescents have a harder time recovering from rejection episodes, with fewer complete rejection reversals and a greater number of partial reversals that other age groups [7]. While there have been dramatic improvements in pediatric KT outcomes in all recipient subgroups over the past four decades, graft survival improvements have been the lowest in adolescents, females, those treated with pretransplant dialysis and those with a diagnosis of FSGS [3]. The NAPRTCS report summarizing outcome benchmarks from 1987 to 2010 also showed female gender as a predictor of lower graft survival. In our study, we found older age at transplantation to be predictive of graft failure on multivariate analysis. Also consistent with prior studies, female gender and acute rejection were significant predictors of graft failure on multivariate analysis. There is no clear explanation in the literature as to why females undergoing renal transplant have lower graft survival. This study has limitations. Retrospective studies are limited in their ability to prove causative factors, and only show associations. The small study size limited our ability to explore subgroup analysis com-
paring patients with living donor versus deceased donor transplants, which are known to have different outcomes. Furthermore, causes of graft failure recorded in the medical records as chronic rejection or patient nonadherence are partially subjective and at the discretion of the clinician. Additionally, there is likely bias in reporting of nonadherence, with greater attention to patients who experienced graft loss or rejection, limiting the utility of this variable for a retrospective study such as ours. As this important factor has significant bias and is best investigated prospectively, we did not include it in the regression analysis. In conclusion, we show that children 12–18 years old have inferior kidney allograft survival compared to younger children. Older age at transplantation, female gender and acute rejection episodes are risk factors predictive of graft failure. The most common causes of graft loss were acute and chronic rejection and immunosuppression nonadherence. Our results have implications for risk stratification, social resources, and immune surveillance of the pediatric population undergoing kidney transplantation, with differing risk based on age and gender. Further studies should evaluate the likely multifactorial causes for the age and gender disparities in pediatric renal transplantation outcomes. References [1] Gillen DL, Stehman-Breen CO, Smith JM, et al. Survival advantage of pediatric recipients of a first kidney transplant among children awaiting kidney transplantation. Am J Transplant 2008;8:2600–6. [2] Van Arendonk KJ, Orandi BJ, James NT, et al. Living unrelated renal transplantation: a good match for the pediatric candidate? J Pediatr Surg 2013;48:1277–82. [3] Van Arendonk KJ, Boyarsky BJ, Orandi BJ, et al. National trends over 25 years in pediatric kidney transplant outcomes. Pediatrics 2014;133:594–601. [4] Gulati A, Sarwal MM. Pediatric renal transplantation: an overview and update. Curr Opin Pediatr 2010;22:189–96. [5] Smith JM, Martz K, Blydt-Hansen TD. Pediatric kidney transplant practice patterns and outcome benchmarks, 1987-2010: a report of the North American Pediatric Renal Trials and Collaborative Studies. Pediatr Transplant 2013;17:149–57. [6] Yamada A, Tashiro A, Hiraiwa T, et al. Long-term outcome of pediatric renal transplantation: A single center study in Japan. Pediatr Transplant 2014;18:453–62. [7] Smith JM, Ho PL, McDonald RA, et al. Renal transplant outcomes in adolescents: a report of the North American Pediatric Renal Transplant Cooperative Study. Pediatr Transplant 2002;6:493–9. [8] Kiberd JA, Acott P, Kiberd BA. Kidney transplant survival in pediatric and young adults. BMC Nephrol 2011;12:54–9. [9] Ishitani M, Isaacs R, Norwood V, et al. Predictors of graft survival in pediatric livingrelated kidney transplant recipients. Transplantation 2000;70:288–92. [10] Li Y, Raychowdhury S, Tedders SH, et al. Association between increased BMI and severe school absenteeism among US children and adolescents: findings from a national survey, 2005–2008. Int J Obes (Lond) 2012;36:517–23. [11] Rhodes ET, Prosser LA, Lieu TA, et al. Preferences for type 2 diabetes health states among adolescents with or at risk of type 2 diabetes mellitus. Pediatr Diabetes 2011;12:724–32. [12] Van Arendonk KJ, James NT, Boyarsky BJ, et al. Age at graft loss after pediatric kidney transplantation: exploring the high-risk age window. Clin J Am Soc Nephrol 2013;8: 1019–26. [13] Andreoni KA, Forbes R, Andreoni RM, et al. Age-related kidney transplant outcomes: health disparities amplified in adolescence. JAMA Intern Med 2013;173:1524–32. [14] Dobbels F, Ruppar T, De Geest S, et al. Adherence to the immunosuppressive regimen in pediatric kidney transplant recipients: a systematic review. Pediatr Transplant 2010;14:603–13.
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Outcome differences between young children and adolescents undergoing kidney transplantation Iuliana D. Bobanga a, Beth A. Vogt b, Kenneth J. Woodside a,⁎, Devan R. Cote a, Katherine M. Dell b, Robert J. Cunningham III b, Kelly A. Noon a, Edward M. Barksdale a, Vanessa R. Humphreville a, Edmund Q. Sanchez a, James A. Schulak a a b
Department of Surgery, Case Western Reserve University & University Hospitals Rainbow Babies & Children's Hospital, Cleveland, OH Departments of Pediatrics, Case Western Reserve University & University Hospitals Rainbow Babies & Children's Hospital, Cleveland, OH
a r t i c l e
i n f o
Article history: Received 25 February 2015 Accepted 10 March 2015 Available online xxxx Key words: Pediatric Kidney transplantation Graft survival Rejection Adolescence
a b s t r a c t Background/Purpose: Although graft loss remains the biggest challenge for all pediatric kidney transplant (KT) recipients, unique challenges exist within different age groups. We aim to evaluate the different characteristics and graft survival outcomes of young children and adolescents undergoing KT. Methods: Children who underwent isolated KT between 2000 and 2013 at our institution were included in this retrospective analysis. Patient characteristics and outcomes were compared using student’s t-test, chi-square test, Kaplan-Meier curve and Cox proportional hazards model. Results: Of 73 children who underwent KT, 31 were b 12 (young children), and 42 were ≥12 years old (adolescents). Overall patient survival was 100%. The younger group had superior 5-year (100% vs. 75.5%) and 10-year (94.4% vs. 43.8%) graft survival (p = 0.008). Factors predictive of poor graft survival on multivariate analysis were older age at transplantation (HR 1.2, CI 1–1.4, p = 0.047), female gender (HR 9.0, CI 1.9–43, p = 0.006), and acute rejection episodes (HR 13, CI 2–90, p = 0.008). The most common causes of graft loss were acute and chronic rejection episodes and immunosuppression nonadherence. Conclusion: Adolescents undergoing KT have inferior graft survival compared to younger children. In adjusted modeling, children with older age, female gender, and acute rejection episodes have inferior graft survival. © 2015 Elsevier Inc. All rights reserved.
Kidney transplantation (KT) is the treatment of choice for children with end-stage renal disease (ESRD), as it improves growth and increases life expectancy compared to dialysis [1,2]. Significant improvements in pediatric KT outcomes have been made over the past four decades across all recipient subgroups, with approximately a 5% decrease in both risk of patient death and graft loss with each more recent year of transplantation [3]. These improvements are probably due to better immunosuppression, alloantibody detection and infectious disease prevention [4]. Despite these advances, there is still a large discrepancy between patient survival and graft survival, which results in many pediatric recipients eventually requiring retransplantation. Multiple factors may affect patient and graft survival, including recipient and donor age, recipient race and gender, transplant and Abbreviations: KT, kidney transplant; ESRD, end stage renal disease; HLA, human leukocyte antigen; NAPRTCS, North American Pediatric Renal Trials and Collaborative Studies; LD, living donor; DD, deceased donor; HR, hazard ratio; CI, confidence interval; CAKUT, congenital anomalies of the kidney and urinary tract; PD, peritoneal dialysis; HD, hemodialysis; LRRT, living related renal transplant; DDRT, deceased donor renal transplant; FSGS, focal segmental glomerulosclerosis; MPGN, membranoproliferative glomeruloephritis; SRTR, Scientific Registry of Transplant Recipients. ⁎ Corresponding author at: Section of Transplant Surgery, Department of Surgery, 2922 Taubman Center, 1500 E. Medical Center Dr., Ann Arbor, MI 48109-5300. Tel.: +1734 936 8363. E-mail address: [email protected] (K.J. Woodside).
transfusion history, HLA mismatches, episodes of rejection and posttransplant complications, among others [5,6]. While the age most commonly associated with increased risk is infancy, several studies have identified lower graft survival among patients undergoing KT during adolescence and early adulthood [7–9]. A report from the North American Pediatric Renal Trials and Collaborative Studies (NAPRTCS) showed that graft survival rates improved steadily from 1987 to 1999 but remained relatively stable between 2000 to 2010, with the 5-year living donor (LD) graft survival of 84.3% and the 5-year deceased donor (DD) graft survival of 78% [5]. The factors that contributed to this leveling off have not been well defined. The aim of this study was to evaluate patient and graft survival in children and adolescents who underwent kidney transplantation at our institution from 2000 to 2010 with the goal of identifying causes of graft loss and risk factors for poor graft survival. 1. Methods 1.1. Study population We performed a retrospective study of all children (≤18 years old) who underwent kidney transplantation between January 2000 and December 2013 at University Hospitals Rainbow Babies and Children’s
http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
2
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Hospital in Cleveland, Ohio. Patients were excluded if they had a prior or concomitant non-renal solid organ transplant. If patients had undergone a previous kidney transplant prior to 2000, data were collected with the focus on the transplant that occurred after 2000. Patients were divided into two groups based on age at kidney transplantation: young children b12 years, and adolescents ≥12 years. Age 12 was chosen to compare adolescents in middle school and high school that undergo various physiological and social changes from elementary school and younger children [10,11]. Patient demographics, operative and perioperative data, short and long-term outcomes, as well as donor demographics, relationship, and donor-recipient HLA matching data were recorded. The Institutional Review Board of University Hospitals approved the study (IRB #11-13-15). 1.2. Study outcomes The primary outcome of interest was allograft survival, defined as the time between date of transplantation and either date of graft failure (marked by retransplantation or return to dialysis) or last date of follow-up with a functioning graft. Other outcomes of interest were length of stay, delayed graft function, causes of graft loss, disease recurrence and allograft rejection. Delayed graft function was defined as requirement for dialysis within the first week after transplantation. 1.3. Statistical analysis Descriptive statistics were used to produce frequencies and percentages for dichotomous variables and mean with standard deviations for continuous variables. Categorical variables were compared using the Chi-square test and continuous variables were compared using independent-sample t test. To evaluate factors associated with allograft loss, we developed a Cox Proportional Hazards model and performed a univariate and multivariate analysis. Crude and adjusted hazards ratios (HRs) with 95% confidence interval (95 % CI) and p values are reported. All variables with a p value b0.100 on univariate analysis were included in the multivariate model. Patients who were lost to follow-up were censored at the time of last follow-up. Time to allograft loss between the two age groups was compared using the Kaplan-Meier method and the log-rank test. A p value ≤0.05 was considered statistically significant throughout the analysis. Statistical analysis was performed using SPSS version 21.0 (SPSS, Inc., Chicago, IL). 2. Results 2.1. Baseline characteristics of recipients and donors Seventy-five pediatric patients ≤ 18 years old underwent kidney transplantation between January 2000 and December 2013. Two patients were excluded, as they had a previous or concomitant liver transplant. Of the 73 patients included (38% female, median age of 13 years at transplantation), 31 were b12 years old at KT and 42 were ≥12 years old (Table 1). Five patients had two or more KTs (one in the younger group, four in the older group). Of those, data was captured regarding the first transplant for two patients and regarding the second transplant for three patients, depending on which transplant occurred after the year 2000. The most common cause of ESRD was congenital anomalies of the kidney and urinary tract (CAKUT), which was more common in the young children group compared to the adolescent group (67.7% vs. 40.5%, p = 0.033). Pretransplant peritoneal dialysis (PD) was more frequently performed in the young children group (72% vs. 40%, p = 0.017). Many children had received both PD and hemodialysis (HD) at different time points and 27.4% of the entire cohort underwent preemptive transplantation. The majority of children (64%) received a live related renal transplant (LRRT), usually from one of their parents, with a trend for
Table 1 Baseline characteristics of recipients and donors. Baseline characteristics
Entire group (n = 73)
Age at KT b 12 (n = 31)
Age at KT ≥ 12 (n = 42)
p-value (b12 vs. ≥ 12)
Age at ESRD Age at KT Gender (% female) Weight (kg) Recipient race (%) Caucasian African American Other Diagnosis (%) FSGS Other glomerular CAKUT Other/missing Pretransplant dialysis (%) Hemodialysis Peritoneal dialysis Pre-emptive Extraperitoneal (%) Ureteral stent use (%) Donor (%) Live related Live unrelated Deceased Mean donor age HLA mismatches 1–3 HLA mismatches 4–6
9.7 ± 5.8 11.7 ± 5.3 38.4 36.6 ± 20
4.5 ± 3.8 6.4 ± 3.3 30 19.5 ± 6.8
13.5 ± 3.4 15.5 ± 2 45 51.2 ± 16
b0.001 b0.001 b0.001 b0.001
75 16.6 8.3
74 16 9.6
75 17 7.3
1 1 1
15.1 15.1 52.1 17.8
6.5 9.7 67.7 16.1
21.4 19 40.5 19
0.103 0.335 0.033 1
42.5 53.4 27.4 89 69
35.5 71 25.8 77.4 58.6
47.6 40.5 28.6 97.6 76.2
0.345 0.017 1 0.009 0.128
64 10.7 24 37.5 ± 7 64.2 35.8
74.2 12.9 12.9 34.8 ± 6.8 74.1 25.9
57.1 9.5 33.3 40.4 ± 6 53.8 46.2
0.148 0.716 0.057 0.003 0.158 0.158
ESRD: end stage renal disease. KT: kidney transplant. FSGS: focal segmental glomerulosclerosis. CAKUT: congenital anomalies of kidney and urinary tract. HLA: human leukocyte antigen.
increased LRRT in the young children group and an increased deceased donor renal transplant (DDRT) in the adolescent group. The mean donor age was also lower in the younger group (35 vs. 40 years, p = 0.003). There were no differences in HLA mismatches between the two groups. 2.2. Outcomes and graft survival The mean follow-up time for the entire cohort was 6 years. The average length of stay was 8 days and did not differ significantly between the two groups. Delayed graft function occurred in 8.2% of patients and occurred more commonly in the adolescent group (Table 2). There was no difference in unplanned return to the operating room within 30 days of transplantation. Five patients (6.8%) had disease recurrence, three with focal segmental glomerulosclerosis (FSGS) within 90 days of transplantation, one with membranoproliferative glomerulonephritis (MPGN) seen on transplant nephrectomy three years post-transplantation and one with mesangial proliferative glomerulonephritis along with primary nonfunction immediately posttransplantation. Recurrent disease was evenly distributed between the two groups (Table 1). The incidence of acute biopsy-proven rejection was 31.5% for the entire cohort and did not differ significantly between younger children and adolescents. For one third of the patients (33%), their first acute rejection episode occurred within 6 months of transplantation while 37.5% had their first acute rejection episode within 1 year of transplantation. On univariate logistic regression analysis, black race, FSGS diagnosis, male gender, post-transplant CMV infection, and increased donor age predicted acute rejection episodes, while a diagnosis of CAKUT was associated with a lower risk of acute rejection episodes. In a multivariate analysis, black race (OR 14.9, CI 2.7–83, p = 0.002) and FSGS diagnosis (OR 7.7, CI 1.5–39, p = 0.014) remained significant risk factors. On further review of the medical records, we found that 52% of patients’ medical providers mentioned suspicion or patient
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
3
Table 2 Post-operative outcomes and graft survival. Outcome
Entire group (n = 73)
Age at KT b 12 (n = 31)
Age at KT ≥ 12 (n = 42)
p-value (b12 vs. ≥ 12)
Total f/u time (years) Length of stay (days) Delayed graft function requiring dialysis Unplanned return to OR within 30 days Recurrent disease Acute rejection Pre/post KT hernia Graft survival (%) 1 year 3 years 5 years 10 years
6.0 ± 3.7 8.4 ± 5.8 8.2
6.2 ± 4.1 9.7 ± 8 3.2
5.8 ± 3.3 7.4 ± .3 11.9
0.700 0.100 0.233
8.2
9.7
7.1
1
6.8 31.5 19.2
6.5 29 32
7.1 33 9.5
1 0.801 0.019 0.008
98.6 91.5 85.6 70.9
100 100 100 94.4
97.6 85.7 75.5 43.8
f/u: follow-up. OR: operating room. KT: kidney transplant.
admission of immunosuppression nonadherence, with a significantly higher rate in patients with acute cellular rejection than in patients without acute rejection episodes (74% versus 40%, respectively; p = 0.017). Interestingly, there was an increased incidence of hernias (inguinal, umbilical or incisional) noted in younger children undergoing transplantation. The majority were inguinal hernias (78.5%) and 3 of them (21%) were umbilical hernias. Only one patient of the entire cohort (1.3%) had an incisional hernia post-transplantation. The majority (80%) of those younger children with hernias underwent PD, while only one (25%) of the adolescents with hernia underwent PD, which may have contributed to hernia development. The overall patient survival was 100%. The graft survival at one, three, five and ten years was 98.6%, 91.5%, 85.6% and 70.9%, respectively, and was significantly different between the two age groups (log-rank test, p = 0.008), with younger children having an improved graft survival compared to adolescents (Table 2, Fig. 1). On subgroup analysis of only LRRT, (n = 23 in both groups), there was still a superior graft survival noted for younger children (long-rank test, p = 0.007). Thirteen patients had graft failure, two in the younger age group and 11 adolescents. One graft failure was due to primary non-function, while the remainder occurred between 28 months to 11 years posttransplantation. Most patients had multiple causes attributed for graft failure. The majority had graft failure due to chronic rejection (54%) or acute rejection (46%). Most patients with graft loss had documented suspicion of nonadherence or had discontinued their immunosuppressive medications (11 patients, 85%). There was a significantly higher rate of graft failure in patients with provider suspicion or patient admission of immunosuppression nonadherence than in presumably compliant patients (33% vs. 5%, p = 0.004). Two patients had disease recurrence, one with primary non-function and recurrent mesangial proliferative glomerulonephritis immediately post-transplant and one with chronic rejection and recurrent MPGN on transplant nephrectomy pathology 29 months post-transplantation. Lastly, one patient had graft loss 5.5 years after transplantation due to multiple episodes of pyelonephritis.
2.3. Risk factors for poor graft survival Multiple risk factors reported in the literature to affect graft survival were analyzed in our cohort. On univariate Cox Proportional Hazards Ratio analysis, older age at transplantation female gender, FSGS diagnosis and acute biopsy-proven rejection were associated with increased risk of graft loss, while a diagnosis of CAKUT was predictive of a lower risk of graft loss. On further multivariate Cox Proportional Hazards Ratio analysis, older age at transplantation (HR 1.2, CI 1–1.4, p =
Fig. 1. Kaplan-Meier graft survival curve comparing young children (b12 years) and adolescents (≥12 years) at kidney transplantation. Log rank test, p = 0.008.
0.047), female gender (HR 9.0, CI 1.9–43, p = 0.006) and acute rejection (HR 13, CI 2-90.6, p = 0.008) remained significant predictors of graft loss (Table 3). 3. Discussion We examined the risk factors associated with kidney allograft survival in two pediatric age groups and found that younger children have superior graft survival compared to adolescents. The most frequent causes of graft failure were acute or chronic rejection and nonadherence or discontinuation of immunosuppressive medications. In addition, on multivariate analysis, older age at transplantation, acute rejection episodes and female gender were predictive risk factors for graft failure for all pediatric patients undergoing kidney transplantation. Living donor kidney transplants typically have better graft survival compared to those from deceased donors [2,5]. One significant difference between our two age groups is the higher percent of living donor grafts in the younger group (87% vs. 67%). However, on subgroup analysis including only patients who received a LRRT, younger children still had a superior graft survival compared to adolescents that was statistically significant. Our finding of improved graft survival in the younger age group b12 years old are consistent with other studies showing poorer outcomes in adolescents and young adults undergoing KT [5,7–9,12,13]. Smith et al reviewed the renal transplant outcomes in adolescents using the NAPRTCS database and found that adolescents 13-17 years of age had the poorest long-term graft survival of LD grafts compared to all other pediatric age groups. Furthermore, adolescents had later acute rejection episodes and incomplete rejection reversal [7]. In addition, Van Arendonk et al used the Scientific Registry of Transplant Recipients (SRTR) data from 1987 to 2010 to show that there is a much higher risk of graft loss for patients transplanted at any age while passing through the high-risk age window of 17–24 years old [12]. Possible explanations for this phenomena include higher rates of non-adherence to immunosuppression during late adolescence and early adulthood [14], maturation of the immune system, changes in health insurance, and transitioning from pediatric to adult care. Another hypothesis proposes that there is a period of increased growth during adolescence causing hyperfiltration injury [12].
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
4
I.D. Bobanga et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Table 3 Univariate and multivariate Cox proportional hazards ratios for predictors of graft loss. Variable
Univariate Cox proportional hazards ratio cHR
Age at KT (per year) Female Black race FSGS CAKUT Living related donor Acute rejection Deceased donor Extraperitoneal Donor age Previous KT
1.1 3.8 3.1 4.2 0.1 0.8 6.5 2.2 0.9 1 1.7
95% CI
p-value
1.0–1.3 1.2–12.2 0.9–10.5 1.4–12.9 0.03–0.58 0.24–2.53 1.8–23.7 0.60–8.46 0.17–4.38 0.9–1.15 0.2–13.6
0.026 0.027 0.066 0.012 0.008 0.672 0.005 0.231 0.859 0.834 0.597
Multivariate Cox Proportional Hazards Ratio aHR Age at KT (per year) Female Black race FSGS CAKUT Acute rejection
1.2 9.0 1.4 1.1 0.4 13
95% CI 1.0–1.4 1.9–43 0.3–6.7 0.2–4.9 0.1–2.5 2.0–90.6
p-Value 0.047 0.006 0.672 0.904 0.364 0.008
KT: kidney transplant. FSGS: focal segmental glomerulosclerosis. CAKUT: congenital anomalies of kidney and urinary tract. HLA: human leukocyte antigen. cHR: crude hazards ratio. CI: confidence interval. aHR: adjusted hazards ratio.
Our study also found a high rate of non-adherence to immunosuppression among adolescents who experienced graft loss. Interestingly, we saw no difference in the frequency of acute biopsy-proven rejection episodes between the two cohorts (Table 2), but a much higher rate of graft loss in the adolescent group attributed to acute or chronic rejection. This finding is consistent with the observation by Smith et al that adolescents have a harder time recovering from rejection episodes, with fewer complete rejection reversals and a greater number of partial reversals that other age groups [7]. While there have been dramatic improvements in pediatric KT outcomes in all recipient subgroups over the past four decades, graft survival improvements have been the lowest in adolescents, females, those treated with pretransplant dialysis and those with a diagnosis of FSGS [3]. The NAPRTCS report summarizing outcome benchmarks from 1987 to 2010 also showed female gender as a predictor of lower graft survival. In our study, we found older age at transplantation to be predictive of graft failure on multivariate analysis. Also consistent with prior studies, female gender and acute rejection were significant predictors of graft failure on multivariate analysis. There is no clear explanation in the literature as to why females undergoing renal transplant have lower graft survival. This study has limitations. Retrospective studies are limited in their ability to prove causative factors, and only show associations. The small study size limited our ability to explore subgroup analysis com-
paring patients with living donor versus deceased donor transplants, which are known to have different outcomes. Furthermore, causes of graft failure recorded in the medical records as chronic rejection or patient nonadherence are partially subjective and at the discretion of the clinician. Additionally, there is likely bias in reporting of nonadherence, with greater attention to patients who experienced graft loss or rejection, limiting the utility of this variable for a retrospective study such as ours. As this important factor has significant bias and is best investigated prospectively, we did not include it in the regression analysis. In conclusion, we show that children 12–18 years old have inferior kidney allograft survival compared to younger children. Older age at transplantation, female gender and acute rejection episodes are risk factors predictive of graft failure. The most common causes of graft loss were acute and chronic rejection and immunosuppression nonadherence. Our results have implications for risk stratification, social resources, and immune surveillance of the pediatric population undergoing kidney transplantation, with differing risk based on age and gender. Further studies should evaluate the likely multifactorial causes for the age and gender disparities in pediatric renal transplantation outcomes. References [1] Gillen DL, Stehman-Breen CO, Smith JM, et al. Survival advantage of pediatric recipients of a first kidney transplant among children awaiting kidney transplantation. Am J Transplant 2008;8:2600–6. [2] Van Arendonk KJ, Orandi BJ, James NT, et al. Living unrelated renal transplantation: a good match for the pediatric candidate? J Pediatr Surg 2013;48:1277–82. [3] Van Arendonk KJ, Boyarsky BJ, Orandi BJ, et al. National trends over 25 years in pediatric kidney transplant outcomes. Pediatrics 2014;133:594–601. [4] Gulati A, Sarwal MM. Pediatric renal transplantation: an overview and update. Curr Opin Pediatr 2010;22:189–96. [5] Smith JM, Martz K, Blydt-Hansen TD. Pediatric kidney transplant practice patterns and outcome benchmarks, 1987-2010: a report of the North American Pediatric Renal Trials and Collaborative Studies. Pediatr Transplant 2013;17:149–57. [6] Yamada A, Tashiro A, Hiraiwa T, et al. Long-term outcome of pediatric renal transplantation: A single center study in Japan. Pediatr Transplant 2014;18:453–62. [7] Smith JM, Ho PL, McDonald RA, et al. Renal transplant outcomes in adolescents: a report of the North American Pediatric Renal Transplant Cooperative Study. Pediatr Transplant 2002;6:493–9. [8] Kiberd JA, Acott P, Kiberd BA. Kidney transplant survival in pediatric and young adults. BMC Nephrol 2011;12:54–9. [9] Ishitani M, Isaacs R, Norwood V, et al. Predictors of graft survival in pediatric livingrelated kidney transplant recipients. Transplantation 2000;70:288–92. [10] Li Y, Raychowdhury S, Tedders SH, et al. Association between increased BMI and severe school absenteeism among US children and adolescents: findings from a national survey, 2005–2008. Int J Obes (Lond) 2012;36:517–23. [11] Rhodes ET, Prosser LA, Lieu TA, et al. Preferences for type 2 diabetes health states among adolescents with or at risk of type 2 diabetes mellitus. Pediatr Diabetes 2011;12:724–32. [12] Van Arendonk KJ, James NT, Boyarsky BJ, et al. Age at graft loss after pediatric kidney transplantation: exploring the high-risk age window. Clin J Am Soc Nephrol 2013;8: 1019–26. [13] Andreoni KA, Forbes R, Andreoni RM, et al. Age-related kidney transplant outcomes: health disparities amplified in adolescence. JAMA Intern Med 2013;173:1524–32. [14] Dobbels F, Ruppar T, De Geest S, et al. Adherence to the immunosuppressive regimen in pediatric kidney transplant recipients: a systematic review. Pediatr Transplant 2010;14:603–13.
Please cite this article as: Bobanga ID, et al, Outcome differences between young children and adolescents undergoing kidney transplantation, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.03.021
