© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Clin Transplant 2014: 28: 990–994 DOI: 10.1111/ctr.12407
Clinical Transplantation
An analysis of pancreas transplantation outcomes based on age groupings – an update of the UNOS database Siskind E, Maloney C, Akerman M, Alex A, Ashburn S, Barlow M, Siskind T, Bhaskaran M, Ali N, Basu A, Molmenti E, Ortiz J. An analysis of pancreas transplantation outcomes based on age groupings – an update of the UNOS database. Abstract: Introduction: Previously, increasing age has been a part of the exclusion criteria used when determining eligibility for a pancreas transplant. However, the analysis of pancreas transplantation outcomes based on age groupings has largely been based on single-center reports. Methods: A UNOS database review of all adult pancreas and kidneypancreas transplants between 1996 and 2012 was performed. Patients were divided into groups based on age categories: 18–29 (n = 1823), 30–39 (n = 7624), 40–49 (n = 7967), 50–59 (n = 3160), and ≥60 (n = 280). We compared survival outcomes and demographic variables between each age grouping. Results: Of the 20 854 pancreas transplants, 3440 of the recipients were 50 yr of age or above. Graft survival was consistently the greatest in adults 40–49 yr of age. Graft survival was least in adults age 18–29 at one-, three-, and five-yr intervals. At 10- and 15-yr intervals, graft survival was the poorest in adults >60 yr old. Patient survival and age were found to be inversely proportional; as the patient population’s age increased, survival decreased. Conclusion: Pancreas transplants performed in patients of increasing age demonstrate decreased patient and graft survival when compared to pancreas transplants in patients 50 has been considered a relative contraindication for solid organ transplantation at many centers. While it has been established that the outcomes of renal and liver transplants in older patients are comparable to younger recipients (2, 3), no
990
consensus exists on the transplantation of pancreata in elderly patients. Within the last decade, a number of articles have investigated the outcomes of pancreatic transplant in patients over the age of 50 (4–6). The results have been varied with earlier articles citing that outcomes were inferior in elderly recipients due to their increased risk of cardiopulmonary morbidity and infection and that survival was not increased compared to patients on waiting lists (4, 5). More recent articles have reported no difference in patient and graft survival (6, 7). Equivalent rates of surgical complications, nonsurgical infections, and acute rejection have also been reported in the age >50 groups (8). To date, there has been no consensus regarding the viability of transplanting pancreata into recipients of age >50 or 60. The aim
Pancreas transplant age based outcomes: UNOS of our study was to utilize the UNOS database report to analyze national results of pancreas transplantation in all age groups and study the differences in graft and patient survival. Our study includes a significantly larger patient cohort selected from multiple transplant centers and employs a more extensive survival analysis than those used by previous groups. Our hope is to broaden the scope of evidence that is currently available regarding the value of pancreas transplant in patients of all ages.
age category group was used as the stratification variable, and the groups were compared using the log-rank test. One-, three-, five-, 10-, and 15-yr unadjusted patient survival rates and graft survival rates were computed for each group. Time on waitlist reported in days was analyzed using the same survival methods; however, no data were considered censored. A result was considered statistically significant at the p < 0.05 level. All analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA).
Statistical methods
Data from the UNOS database on all adult (≥18 yr old) subjects who received pancreas and kidney-pancreas transplants between 1996 and 2012 were analyzed (n = 20 854). Subjects were broken down into groups based on age categories: 1. 2. 3. 4. 5.
18–29 yr old (n = 1823). 30–39 yr old (n = 7624). 40–49 yr old (n = 7967). 50–59 yr old (n = 3160). ≥60 yr old (n = 280).
Descriptive statistics (mean standard deviation, median, and 25th and 75th percentiles for continuous variables; frequencies and percentages [%] for categorical variables) were calculated by recipient age category group for donor factors and recipient characteristics. The chi-square test was used to compare these groups for each of the categorical variables (i.e., gender, ethnicity, type of transplant). The Kruskal–Wallis test, the nonparametric counterpart to analysis of variance (ANOVA), was used to compare the groups for continuous variables (i.e., age, body mass index [BMI], terminal creatinine). Standard methods of survival analysis (9) were applied to the data. An analysis of patient survival and death-censored graft survival were each conducted separately. In cases where the “event” did not occur, the number of days until last follow-up was used and considered censored. Death-censored graft survival was calculated from the date of transplantation to the date of irreversible pancreas graft failure signified by loss of endocrine function requiring return to exogenous insulin therapy, excision of pancreatic allograft, or retransplantation. In the event of death with a functioning graft, the follow-up period was censored at the date of death (10). Kaplan–Meier/product-limit estimates and their corresponding 95% confidence intervals were computed, using Greenwood’s formula to calculate the standard error (11). Additionally, Kaplan– Meier product limit curves were computed, where
Results
There was a significant difference among the age category groups for donor age, recipient gender, recipient BMI, recipient ethnicity, treatment type, and time on waitlist (Fig. 1 and Table 1). The groups were otherwise comparable for the following donor characteristics: gender, BMI, and deceased donor cause of death (as displayed in Table 1). Kaplan–Meier analysis demonstrated a difference in patient survival (p < 0.0001) and death-censored graft survival (p < 0.0001) among the age category groups (Figs. 2 and 3). Below are Tables 2 and 3 that include one-, three-, five-, 10-, and 15-yr unadjusted patient survival rates and graft survival rates for each age category group. Tables 4 and 5 detail patient and graft survival in calculated in median days. There was also a significant difference in length of stay among the groups (p < 0.0001, Table 6, Fig. 4). Discussion
Solid organ transplantation in patients of increasing age has been a topic of controversy in the transplant literature. While it has become universally
Fig. 1. Time on waitlist (d).
991
Siskind et al. Table 1. Characteristics of donors and recipients 18–29 Donor characteristics Age 25.0 10.1 Gender (male, %) 1239 (68.0) BMI 24.0 4.0 Deceased donor cause of death (%) Anoxia 214 (11.8) Cerebrovascular/stroke 338 (18.6) Head trauma 1212 (66.7) CNS tumor 9 (0.5) Other 45 (2.5) Recipient characteristics Gender (male, %) 769 (42.2) BMI 23.6 3.9 Ethnicity (Hispanic, %) 205 (11.3) Treatment type (%) PAK 206 (11.3) PTA 322 (17.7) PWK 15 (0.8) SKP 1280 (70.2) Time on waitlist (d) 166 (154, 181)
30–39
40–49
50–59
≥60
p Value
25.9 10.5 5099 (66.9) 24.0 4.0
26.3 10.7 5376 (67.5) 24.0 4.0
26.3 10.7 2102 (66.5) 24.1 4.1
26.5 10.9 203 (72.5) 25.7 4.2
Clin Transplant 2014: 28: 990–994 DOI: 10.1111/ctr.12407
Clinical Transplantation
An analysis of pancreas transplantation outcomes based on age groupings – an update of the UNOS database Siskind E, Maloney C, Akerman M, Alex A, Ashburn S, Barlow M, Siskind T, Bhaskaran M, Ali N, Basu A, Molmenti E, Ortiz J. An analysis of pancreas transplantation outcomes based on age groupings – an update of the UNOS database. Abstract: Introduction: Previously, increasing age has been a part of the exclusion criteria used when determining eligibility for a pancreas transplant. However, the analysis of pancreas transplantation outcomes based on age groupings has largely been based on single-center reports. Methods: A UNOS database review of all adult pancreas and kidneypancreas transplants between 1996 and 2012 was performed. Patients were divided into groups based on age categories: 18–29 (n = 1823), 30–39 (n = 7624), 40–49 (n = 7967), 50–59 (n = 3160), and ≥60 (n = 280). We compared survival outcomes and demographic variables between each age grouping. Results: Of the 20 854 pancreas transplants, 3440 of the recipients were 50 yr of age or above. Graft survival was consistently the greatest in adults 40–49 yr of age. Graft survival was least in adults age 18–29 at one-, three-, and five-yr intervals. At 10- and 15-yr intervals, graft survival was the poorest in adults >60 yr old. Patient survival and age were found to be inversely proportional; as the patient population’s age increased, survival decreased. Conclusion: Pancreas transplants performed in patients of increasing age demonstrate decreased patient and graft survival when compared to pancreas transplants in patients 50 has been considered a relative contraindication for solid organ transplantation at many centers. While it has been established that the outcomes of renal and liver transplants in older patients are comparable to younger recipients (2, 3), no
990
consensus exists on the transplantation of pancreata in elderly patients. Within the last decade, a number of articles have investigated the outcomes of pancreatic transplant in patients over the age of 50 (4–6). The results have been varied with earlier articles citing that outcomes were inferior in elderly recipients due to their increased risk of cardiopulmonary morbidity and infection and that survival was not increased compared to patients on waiting lists (4, 5). More recent articles have reported no difference in patient and graft survival (6, 7). Equivalent rates of surgical complications, nonsurgical infections, and acute rejection have also been reported in the age >50 groups (8). To date, there has been no consensus regarding the viability of transplanting pancreata into recipients of age >50 or 60. The aim
Pancreas transplant age based outcomes: UNOS of our study was to utilize the UNOS database report to analyze national results of pancreas transplantation in all age groups and study the differences in graft and patient survival. Our study includes a significantly larger patient cohort selected from multiple transplant centers and employs a more extensive survival analysis than those used by previous groups. Our hope is to broaden the scope of evidence that is currently available regarding the value of pancreas transplant in patients of all ages.
age category group was used as the stratification variable, and the groups were compared using the log-rank test. One-, three-, five-, 10-, and 15-yr unadjusted patient survival rates and graft survival rates were computed for each group. Time on waitlist reported in days was analyzed using the same survival methods; however, no data were considered censored. A result was considered statistically significant at the p < 0.05 level. All analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA).
Statistical methods
Data from the UNOS database on all adult (≥18 yr old) subjects who received pancreas and kidney-pancreas transplants between 1996 and 2012 were analyzed (n = 20 854). Subjects were broken down into groups based on age categories: 1. 2. 3. 4. 5.
18–29 yr old (n = 1823). 30–39 yr old (n = 7624). 40–49 yr old (n = 7967). 50–59 yr old (n = 3160). ≥60 yr old (n = 280).
Descriptive statistics (mean standard deviation, median, and 25th and 75th percentiles for continuous variables; frequencies and percentages [%] for categorical variables) were calculated by recipient age category group for donor factors and recipient characteristics. The chi-square test was used to compare these groups for each of the categorical variables (i.e., gender, ethnicity, type of transplant). The Kruskal–Wallis test, the nonparametric counterpart to analysis of variance (ANOVA), was used to compare the groups for continuous variables (i.e., age, body mass index [BMI], terminal creatinine). Standard methods of survival analysis (9) were applied to the data. An analysis of patient survival and death-censored graft survival were each conducted separately. In cases where the “event” did not occur, the number of days until last follow-up was used and considered censored. Death-censored graft survival was calculated from the date of transplantation to the date of irreversible pancreas graft failure signified by loss of endocrine function requiring return to exogenous insulin therapy, excision of pancreatic allograft, or retransplantation. In the event of death with a functioning graft, the follow-up period was censored at the date of death (10). Kaplan–Meier/product-limit estimates and their corresponding 95% confidence intervals were computed, using Greenwood’s formula to calculate the standard error (11). Additionally, Kaplan– Meier product limit curves were computed, where
Results
There was a significant difference among the age category groups for donor age, recipient gender, recipient BMI, recipient ethnicity, treatment type, and time on waitlist (Fig. 1 and Table 1). The groups were otherwise comparable for the following donor characteristics: gender, BMI, and deceased donor cause of death (as displayed in Table 1). Kaplan–Meier analysis demonstrated a difference in patient survival (p < 0.0001) and death-censored graft survival (p < 0.0001) among the age category groups (Figs. 2 and 3). Below are Tables 2 and 3 that include one-, three-, five-, 10-, and 15-yr unadjusted patient survival rates and graft survival rates for each age category group. Tables 4 and 5 detail patient and graft survival in calculated in median days. There was also a significant difference in length of stay among the groups (p < 0.0001, Table 6, Fig. 4). Discussion
Solid organ transplantation in patients of increasing age has been a topic of controversy in the transplant literature. While it has become universally
Fig. 1. Time on waitlist (d).
991
Siskind et al. Table 1. Characteristics of donors and recipients 18–29 Donor characteristics Age 25.0 10.1 Gender (male, %) 1239 (68.0) BMI 24.0 4.0 Deceased donor cause of death (%) Anoxia 214 (11.8) Cerebrovascular/stroke 338 (18.6) Head trauma 1212 (66.7) CNS tumor 9 (0.5) Other 45 (2.5) Recipient characteristics Gender (male, %) 769 (42.2) BMI 23.6 3.9 Ethnicity (Hispanic, %) 205 (11.3) Treatment type (%) PAK 206 (11.3) PTA 322 (17.7) PWK 15 (0.8) SKP 1280 (70.2) Time on waitlist (d) 166 (154, 181)
30–39
40–49
50–59
≥60
p Value
25.9 10.5 5099 (66.9) 24.0 4.0
26.3 10.7 5376 (67.5) 24.0 4.0
26.3 10.7 2102 (66.5) 24.1 4.1
26.5 10.9 203 (72.5) 25.7 4.2
