Abstract
Several classifications systems have been developed to predict outcomes of kidney transplantation based on donor variables.
This study aims to identify kidney transplant recipient variables that would predict graft outcome irrespective of donor characteristics.
All U.S. kidney transplant recipients between October 25,1999 and January 1, 2007 were reviewed. Cox proportional hazards regression was used to model time until graft failure. Death-censored and nondeath-censored graft survival models were generated for recipients of live and deceased donor organs. Recipient age, gender, body mass index (BMI), presence of cardiac risk factors, peripheral vascular disease, pulmonary disease, diabetes, cerebrovascular disease, history of malignancy, hepatitis B core antibody, hepatitis C infection, dialysis status, panel-reactive antibodies (PRA), geographic region, educational level, and prior kidney transplant were evaluated in all kidney transplant recipients.
Among the 88,284 adult transplant recipients the following groups had increased risk of graft failure: younger and older recipients, increasing PRA (hazard ratio [HR],1.03–1.06], increasing BMI (HR, 1.04–1.62), previous kidney transplant (HR, 1.17–1.26), dialysis at the time of transplantation (HR, 1.39–1.51), hepatitis C infection (HR, 1.41–1.63), and educational level (HR, 1.05–1.42).
Predictive criteria based on recipient characteristics could guide organ allocation, risk stratification, and patient expectations in planning kidney transplantation.
Keywords: kidney transplant recipients, outcomes predictors, graft failure, comorbidities, patient education, geographic disparities
Kidney transplantation represents the best alternative for survival and improved quality of life for eligible end stage renal disease patients. Although several classifications that estimate outcomes have been developed, they all involve donor features. The Kidney Donor Risk Index (KDRI) and Kidney Donor Profile Index (KDPI),1 2 based on deceased donor age, height, weight, ethnicity, history of hypertension, history of diabetes, cause of death, serum creatinine, hepatitis C virus status, and donation after circulatory death, assess the relative risk of graft failure irrespective of recipient characteristics.
The objective of this study was to identify kidney transplant recipient variables that would predict graft outcome irrespective of donor characteristics. These recipient predictive criteria could constitute an instrument of great potential value and a relevant addition to the current allocation system. They would provide information on expected outcomes not only at the time of evaluation and during wait listing when no donor information is routinely available, but also at the time of organ allocation when they would be complemented by the already existing donor classifications.
Methods
Subjects
Data on 88,284 kidney transplants performed in the United States from October 25, 1999 to January 1, 2007 obtained from the United Network for Organ Sharing (UNOS) were considered in the analysis.
Selection Criteria
There were 119,979 transplants between October 25, 1999 and January 1, 2007. Several variables of interest (drug-treated hypertension, cerebrovascular disease, and angina) had collection end dates of January 1, 2007. Since then, they have become optional data fields and their reporting has been sparse. The start date was chosen because relevant donor-related variables (deceased donor − cardiac arrest post–brain death), although not of primary interest, had collection dates beginning October 25, 1999. Recipients listed for pancreas (n = 3,629) and kidney pancreas (n = 6,719) as well as those with no organ listed (n = 16,173) were excluded. Kidney recipients younger than 18 or with missing age were also excluded (n = 3,857). There were 1,317 adult kidney recipients with multiple transplants in our timeframe of interest. For purposes of our analysis, only the initial transplant between October 25, 1999 and January 1, 2007 was included. Body mass index (BMI) < 15 or > 55, live donor preoperative creatinine > 1.5, and deceased donor terminal creatinine > 6 were deemed unlikely and treated as unknown.
Primary Outcome Variable
The primary outcome considered was (death-censored and nondeath-censored) graft survival, as defined in previous studies.3 In death-censored graft survival, graft survival was censored at the time of death (based on the assumption that death was unrelated to the transplant) or at the time of the last known patient status (if neither failure nor death occurred). In nondeath-censored graft survival, death with a functioning graft was treated as graft failure (under the assumption that death was related to the transplant).
Statistical Analysis
Cox regression was used to model time until graft failure. Recipient risk factors significantly associated with graft failure using univariable screening at the level of p < 0.10 were included in the final multivariable model. Transplant and donor variables known to be highly predictive of graft failure, while not of direct interest, were included in the multivariable model as covariates, regardless of statistical significance. Donor variables included human leukocyte antigen (HLA) mismatch and elements of the KDPI. Although backward elimination was also considered in building the final model, its results were similar to those of univariable screening and are not reported here. Separate models were used for recipients of live and deceased donors. Each model was analyzed using both definitions of graft survival, resulting in four multivariable models. The proportional hazards assumption was evaluated graphically by plotting the log-negative-log of the estimated survival function by the log of time.
Results
Demographics of the 88,284 subjects included in our final sample are detailed in Table 1.
Table 1. Characteristics of kidney transplant recipients of live and deceased donors and all recipients.
| Factor | All recipients (n = 88,284) | Recipients of live donors (n = 27,117) | Recipients of deceased donors (n = 61,167) |
|---|---|---|---|
| Age, y | 48.90 ± 13.26 | 46.03 ± 13.39 | 50.17 ± 13.00 |
| Gender | |||
| Male no. (%) | 53,174 (60.23) | 16,064 (59.24) | 37,110 (60.67) |
| BMI | |||
| Underweight, no. (%) | 2,432 (2.75) | 775 (2.86) | 1,657 (2.71) |
| Normal weight, no. (%) | 31,722 (35.93) | 9,954 (36.71) | 21,768 (35.59) |
| Overweight, no. (%) | 28,464 (32.24) | 8,677 (32.00) | 19,787 (32.35) |
| Obese class I, no. (%) | 15,349 (17.39) | 4,648 (17.14) | 10,701 (17.49) |
| Obese class II, no. (%) | 4,998 (5.66) | 1,491 (5.50) | 3,507 (5.73) |
| Obese class III, no. (%) | 1,497 (1.70) | 476 (1.76) | 1,021 (1.67) |
| Unknown | 3,822 (4.33) | 1,096 (4.04) | 2,726 (4.46) |
| Previous kidney transplant | |||
| Yes | 9,467 (10.72) | 2,528 (9.32) | 6,939 (11.34) |
| No | 78,817 (89.28) | 24,589 (90.68) | 54,228 (88.66) |
| Previous malignancy | |||
| Yes | 3,043 (3.45) | 955 (3.52) | 2,088 (3.41) |
| No | 74,826 (84.76) | 22,995 (84.80) | 51,831 (84.74) |
| Unknown | 10,415 (11.80) | 3,167 (11.68) | 7,428 (11.85) |
| Angina | |||
| Yes | 8,684 (9.84) | 2,378 (8.77) | 6,306 (10.31) |
| No | 74,075 (83.91) | 22,922 (84.53) | 51,153 (83.63) |
| Unknown | 5,525 (6.26) | 1,817 (6.70) | 3,708 (6.06) |
| Drug-treated hypertension | |||
| Yes | 68,902 (78.05) | 21,341 (78.70) | 47,561 (77.76) |
| No | 14,887 (16.86) | 4,383 (16.16) | 10,504 (17.17) |
| Unknown | 4,495 (5.09) | 1,393 (5.14) | 3,102 (5.07) |
| Peripheral vascular disease | |||
| Yes | 2,981 (3.38) | 837 (3.09) | 2,144 (3.51) |
| No | 79,144 (89.65) | 24,368 (89.86) | 54,776 (89.55) |
| Unknown | 6,159 (6.98) | 1,912 (7.05) | 4,247 (6.94) |
| Drug-treated COPD | |||
| Yes | 761 (0.86) | 217 (0.80) | 544 (0.89) |
| No | 82,384 (93.32) | 25,341 (93.45) | 57,043 (93.26) |
| Unknown | 5,139 (5.82) | 1,559 (5.75) | 3,580 (5.85) |
| Symptomatic cerebrovascular disease | |||
| Yes | 2,046 (2.32) | 560 (2.07) | 1,486 (2.43) |
| No | 80,551 (91.24) | 24,899 (91.82) | 55,652 (90.98) |
| Unknown | 5,687 (6.44) | 1,658 (6.11) | 4,029 (6.59) |
| Hepatitis B surface antigen | |||
| Positive | 1,475 (1.67) | 395 (1.46) | 1,080 (1.77) |
| Negative | 75,098 (85.06) | 23,451 (86.48) | 51,647 (84.44) |
| Not done | 1,949 (2.21) | 562 (2.07) | 1,387 (2.27) |
| Unknown/missing | 9,762 (11.06) | 2,709 (9.99) | 7,053 (11.53) |
| Hepatitis C status | |||
| Positive | 4,875 (5.52) | 799 (2.95) | 4,076 (6.66) |
| Negative | 72,795 (82.46) | 23,554 (86.86) | 49,241 (80.50) |
| Not done | 6,253 (7.08) | 1,592 (5.87) | 4,661 (7.62) |
| Unknown/missing | 4,361 (4.94) | 1,172 (4.32) | 3,189 (5.21) |
| Dialysis at transplant | |||
| Yes | 74,094 (83.93) | 19,780 (72.94) | 54,314 (88.80) |
| No | 12,899 (14.61) | 6,839 (25.22) | 6,060 (9.91) |
| Unknown | 1,291 (1.46) | 498 (1.84) | 793 (1.30) |
| Diabetes at registration | |||
| Yes | 25,270 (28.62) | 7,144 (26.35) | 18,126 (29.63) |
| No | 61,130 (69.24) | 19,434 (71.67) | 41,696 (68.17) |
| Unknown | 1,884 (2.13) | 539 (1.99) | 1,345 (2.20) |
| Current PRA (median [Q1, Q3]) | 0.0 (0.0, 0.0) | 0.00 (0.00, 3.00) | |
| Education | |||
| No high school diploma or GED | 4,765 (5.40) | 994 (3.67) | 3,771 (6.17) |
| High school diploma or GED | 33,091 (37.48) | 9,114 (33.61) | 23,977 (39.20) |
| Attended college/technical school | 17,389 (19.70) | 5,951 (21.95) | 11,438 (18.70) |
| Associate/bachelor's degree | 10,811 (12.25) | 4,176 (15.40) | 6,635 (10.85) |
| Graduate degree | 4,291 (4.86) | 1,831 (6.75) | 2,460 (4.02) |
| Unknown/missing | 17,937 (20.32) | 5,051 (18.63) | 12,886 (21.07) |
| Region | |||
| 1 | 3,722 (4.22) | 1,202 (4.43) | 2,520 (4.12) |
| 2 | 13,544 (15.34) | 4,472 (16.49) | 9,072 (14.83) |
| 3 | 11,145 (12.62) | 2,512 (9.26) | 8,633 (14.11) |
| 4 | 7,443 (8.43) | 1,865 (6.88) | 5,578 (9.12) |
| 5 | 14,055 (15.92) | 4,976 (18.35) | 9,079 (14.84) |
| 6 | 2,960 (3.35) | 628 (2.32) | 2,332 (3.81) |
| 7 | 9,398 (10.65) | 3,757 (13.85) | 5,641 (9.22) |
| 8 | 4,944 (5.60) | 1,463 (5.40) | 3,481 (5.69) |
| 9 | 5,579 (6.32) | 1,819 (6.71) | 3,760 (6.15) |
| 10 | 7,567 (8.57) | 2,371 (8.74) | 5,196 (8.49) |
| 11 | 7,927 (8.98) | 2,052 (7.57) | 5,875 (9.60) |
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; GED, general educational development; PRA, panel-reactive antibodies.
Death-Censored Graft Survival
Recipients of Live Donors
Table 2 shows the death-censored multivariable Cox regression analysis for graft survival in recipients of live donors (3,667 graft failures). Younger and older recipients (quadratic term) had an increased risk of graft failure. Recipients with overweight or obese BMI, a previous kidney transplant, hepatitis C, increasing panel-reactive antibodies (PRA), dialysis at the time of transplant, and diabetes also had an increased risk of graft failure as compared to recipients without these characteristics. Recipients with a high school (HS) diploma or general educational development (GED) had an increased risk of graft failure as compared to attendees of college, recipients of associate or bachelor's degrees and graduate degrees, with hazard ratio (HR) ranging from 1.14 to 1.42. Attendees of college had an increased risk of graft failure as compared to recipients of associate or bachelor's degrees (HR,1.13) and graduate degrees. Recipients in UNOS transplant region 2 had an increased risk of graft failure as compared to regions 1, 3, 4, 5, and 10, with HR ranging from 1.20 to 1.27. Malignancy, angina, drug-treated hypertension, drug-treated chronic obstructive pulmonary disease (COPD), peripheral vascular disease, and cerebral vascular disease were not statistically significant at p < 0.10 by univariable screen and were excluded from the final multivariable model.
Table 2. Death-censored graft survival in recipients of live donors and recipients of deceased donors.
| Factora | Recipients of live donorsb
c
HR (95% CI) n = 27,117 |
Recipients of deceased donorsd
e
HR (95% CI) n = 61,167 |
|---|---|---|
| Age | 0.93 (0.92, 0.95) | 0.94 (0.93, 0.94) |
| Age ×age | 1.01 (1.00, 1.01) | 1.00 (1.00, 1.01) |
| Gender | ||
| Male | 0.97 (0.91, 1.04) | 1.07 (1.03, 1.11) |
| Female | Reference | Reference |
| BMI | ||
| Underweight | 1.05 (0.87, 1.26) | 0.95 (0.85, 1.05) |
| Normal weight | Reference | Reference |
| Overweight | 1.09 (1.00, 1.18) | 1.11 (1.07, 1.16) |
| Obese class I | 1.26 (1.14, 1.39) | 1.27 (1.20, 1.33) |
| Obese class II | 1.47 (1.28, 1.68) | 1.37 (1.27, 1.47) |
| Obese class III | 1.62 (1.32, 1.99) | 1.61 (1.43, 1.82) |
| Previous kidney transplant | ||
| Yes | 1.25 (1.13, 1.39) | 1.17 (1.11, 1.24) |
| No | Reference | Reference |
| Peripheral vascular disease | ||
| Yes | 1.13 (1.02, 1.25) | |
| No | Reference | |
| Hepatitis C status | ||
| Positive | 1.56 (1.33, 1.83) | 1.44 (1.34, 1.54) |
| Negative | Reference | Reference |
| Not done | 0.90 (0.75, 1.08) | 1.04 (0.96, 1.13) |
| Dialysis at transplant | ||
| Yes | 1.43 (1.31, 1.56) | 1.51 (1.41, 1.62) |
| No | Reference | Reference |
| Diabetes | ||
| Yes | 1.09 (1.00, 1.18) | 1.02 (0.98, 1.06) |
| No | Reference | |
| Current PRA (in 10-percentage point increments) | 1.06 (1.04, 1.08) | 1.04 (1.03, 1.05) |
| Education | ||
| No high school diploma or GED | 1.33 (1.04, 1.70) | 1.00 (0.89, 1.13) |
| High school diploma or GED | 1.42 (1.20, 1.66) | 1.12 (1.01, 1.23) |
| Attended college/technical school | 1.24 (1.05, 1.47) | 1.06 (0.96, 1.18) |
| Associate/bachelor's degree | 1.09 (0.92, 1.31) | 1.02 (0.92, 1.14) |
| Graduate degree | Reference | Reference |
| Region | ||
| 1 | 1.03 (0.86, 1.23) | 1.07 (0.97, 1.19) |
| 2 | 1.27 (1.11, 1.45) | 1.37 (1.28, 1.47) |
| 3 | 1.06 (0.91, 1.23) | 1.12 (1.04, 1.21) |
| 4 | 1.03 (0.87, 1.21) | 1.18 (1.09, 1.28) |
| 5 | Reference | Reference |
| 6 | 1.00 (0.78, 1.29) | 0.94 (0.84, 1.05) |
| 7 | 1.16 (1.02, 1.33) | 1.11 (1.02, 1.20) |
| 8 | 1.17 (0.99, 1.38) | 1.05 (0.95, 1.15) |
| 9 | 1.08 (0.90, 1.29) | 1.27 (1.17, 1.39) |
| 10 | 1.06 (0.91, 1.23) | 1.13 (1.04, 1.22) |
| 11 | 1.15 (0.99, 1.33) | 1.35 (1.26, 1.46) |
Abbreviations: BMI, body mass index; CI, confidence interval; COPD, chronic obstructive pulmonary disease; GED, general educational development; HR, hazard ratio; PRA, panel-reactive antibodies.
Recipients of live/deceased donors: Unknown or missing categories are not presented in the table since valid inferences cannot be drawn from them.
Recipients of live donors: Model is also adjusted for recipient hepatitis B surface antigen. However, the results are not displayed due to nonsignificance in the multivariable model.
Recipients of live donors: Model is also adjusted for donor age, donor BMI, donor gender, donor ethnicity, HLA mismatch level, donor preoperative creatinine, and donor hepatitis C antibody.
Recipients of deceased donors:Model is also adjusted for recipient angina status, cerebrovascular disease, and hepatitis B surface antigen. However, the results are not displayed due to nonsignificance in the multivariable model.
Recipients of deceased donors: Model is also adjusted for donor age, donor BMI, donor gender, donor ethnicity, HLA mismatch level, donor terminal laboratory creatinine, donor hepatitis C antibody, donor history of diabetes, donor history of hypertension, donor cause of death, and donor cardiac arrest postbrain death.
Recipients of Deceased Donors
Table 2 shows the death-censored multivariable Cox regression analysis for graft survival in recipients of deceased donors (13,361 graft failures). Younger and older recipients had an increased risk of graft failure. Males, recipients with overweight or obese BMI, a previous kidney transplant, peripheral vascular disease, hepatitis C, increasing PRA, and dialysis at the time of transplant also had an increased risk of graft failure as compared to recipients without these characteristics. Recipients with a HS diploma or GED had an increased risk of graft failure as compared to recipients without a HS diploma, attendees of college, recipients of associate or bachelor's degrees, and graduate degrees, with HR ranging from 1.05 to 1.12. Region 5 and 6 recipients had a reduced risk of graft failure as compared to all other regions, excluding region 8, with HR ranging from 0.68 to 0.90. Region 2 recipients had an increased risk of graft failure as compared to all other regions, excluding 9 and 11, with HR ranging from 1.16 to 1.46. Region 9 recipients had an increased risk of graft failure as compared to regions 1, 3, 5, 6, 7, 8, and 10, with HR ranging from 1.13 to 1.36. Malignancy, drug-treated hypertension, and drug-treated COPD were not statistically significant at p < 0.10 by univariable screen and were excluded from the final multivariable model.
Nondeath-Censored Graft Survival
Recipients of Live Donors
Table 3 shows the noncensored multivariable Cox regression graft survival analysis for recipients of live donors (6,026 graft failures, inclusive of recipients who died with functioning grafts). Younger and older recipients had an increased risk of graft failure. Recipients with underweight or obese BMI, a previous kidney transplant, malignancy, drug-treated COPD, angina, peripheral vascular disease, cerebral vascular disease, hepatitis C, increasing PRA, dialysis at the time of transplant, and diabetes also had an increased risk of graft failure as compared to recipients without these characteristics. Recipients without a HS diploma or GED had an increased risk of graft failure as compared to recipients of associate or bachelor's degrees (HR,1.22) and graduate degrees. Recipients with a HS diploma or GED had an increased risk of graft failure as compared to attendees of college, recipients of associate or bachelor's degrees, and graduate degrees with HR ranging from 1.11 to 1.36. Attendees of college had an increased risk of graft failure as compared to recipients of associate or bachelor's degrees (HR,1.14) and graduate degrees. Gender was not statistically significant at p < 0.10 by univariable screen and was excluded from the final multivariable model.
Table 3. Nondeath-censored graft survival in recipients of live donors and recipients of deceased donors.
| Factora | Recipients of live donorsb
c
HR (95% CI) n = 27,117 |
Recipients of deceased donorsd
e
HR (95% CI) n = 61,167 |
|---|---|---|
| Age | 0.92 (0.91, 0.93) | 0.93 (0.93, 0.94) |
| Age ×age | 1.01 (1.00, 1.01) | 1.00 (1.00, 1.01) |
| Gender | ||
| Male | 1.12 (1.09, 1.15) | |
| Female | Reference | |
| BMI | ||
| Underweight | 1.19 (1.02, 1.38) | 1.07 (0.98, 1.16) |
| Normal weight | Reference | Reference |
| Overweight | 1.01 (0.95, 1.08) | 1.04 (1.01, 1.07) |
| Obese class I | 1.10 (1.02, 1.38) | 1.17 (1.12, 1.21) |
| Obese class II | 1.32 (1.18, 1.47) | 1.21 (1.14, 1.28) |
| Obese class III | 1.46 (1.23, 1.73) | 1.54 (1.40, 1.69) |
| Previous kidney transplant | ||
| Yes | 1.26 (1.15, 1.37) | 1.19 (1.14, 1.24) |
| No | Reference | Reference |
| Malignancy | ||
| Yes | 1.28 (1.13, 1.44) | 1.10 (1.03, 1.18) |
| No | Reference | Reference |
| Drug-treated COPD | ||
| Yes | 1.54 (1.24, 1.92) | 1.10 (0.97, 1.25) |
| No | Reference | Reference |
| Angina | ||
| Yes | 1.11 (1.02, 1.21) | 1.18 (1.13, 1.23) |
| No | Reference | Reference |
| Peripheral vascular disease | ||
| Yes | 1.27 (1.12, 1.43) | 1.30 (1.22, 1.38) |
| No | Reference | Reference |
| Cerebral vascular disease | ||
| Yes | 1.21 (1.04, 1.41) | 1.05 (0.97, 1.14) |
| No | Reference | Reference |
| Hepatitis C status | ||
| Positive | 1.63 (1.44, 1.84) | 1.41 (1.33, 1.49) |
| Negative | Reference | Reference |
| Not done | 0.95 (0.82, 1.10) | 1.02 (0.95, 1.08) |
| Dialysis at transplant | ||
| Yes | 1.46 (1.36, 1.56) | 1.39 (1.32, 1.46) |
| No | Reference | Reference |
| Diabetes | ||
| Yes | 1.42 (1.34, 1.50) | 1.32 (1.28, 1.36) |
| No | Reference | Reference |
| Current PRA (in 10-percentage point increments) | 1.05 (1.04, 1.07) | 1.03 (1.03, 1.04) |
| Education | ||
| No high school diploma or GED | 1.32 (1.10, 1.57) | 1.05 (0.97, 1.15) |
| High school diploma or GED | 1.36 (1.21, 1.53) | 1.15 (1.07, 1.23) |
| Attended college/technical school | 1.23 (1.09, 1.39) | 1.08 (1.01, 1.17) |
| Associate/bachelor's degree | 1.08 (0.95, 1.23) | 1.00 (0.93, 1.09) |
| Graduate degree | Reference | Reference |
| Region | ||
| 1 | 1.04 (0.91, 1.20), | 1.07 (0.99, 1.15) |
| 2 | 1.18 (1.07, 1.31) | 1.22 (1.16, 1.29) |
| 3 | 1.09 (0.97, 1.23) | 1.13 (1.07, 1.19) |
| 4 | 0.98 (0.80, 1.19) | 1.16 (1.10, 1.23) |
| 5 | Reference | Reference |
| 6 | 0.98 (0.80, 1.19) | 0.97 (0.90, 1.06) |
| 7 | 1.11 (1.00, 1.23) | 1.13 (1.06, 1.20) |
| 8 | 1.09 (0.95, 1.24) | 1.05 (0.98, 1.12) |
| 9 | 1.07 (0.93, 1.23) | 1.18 (1.11, 1.26) |
| 10 | 1.08 (0.97, 1.21) | 1.16 (1.10, 1.24) |
| 11 | 1.20 (1.07, 1.34) | 1.27 (1.20, 1.35) |
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; CI, confidence interval; GED, general educational development; HR, hazard ratio; PRA, panel-reactive antibodies.
Recipients of live/deceased donors: Unknown or missing categories are not presented in the table since valid inferences cannot be drawn from them.
Recipients of live donors: Model is also adjusted for recipient drug-treated hypertension and hepatitis B surface antigen. However, the results are not displayed due to nonsignificance in the multivariable model or significance in the unknown category.
Recipients of live donors: Model is also adjusted for donor age, donor BMI, donor gender, donor ethnicity, HLA mismatch level, donor preoperative creatinine, and donor hepatitis C antibody.
Recipients of deceased donors: Model is also adjusted for recipient drug-treated hypertensionand hepatitis B surface antigen. However, the results are not displayed due to nonsignificance in the multivariable model or significance in the unknown category.
Recipients of deceased donors: Model is also adjusted for donor age, donor BMI, donor gender, donor ethnicity, HLA mismatch level, donor terminal laboratory creatinine, donor hepatitis C antibody, donor history of diabetes, donor history of hypertension, donor cause of death, and donor cardiac arrest postbrain death.
Recipients of Deceased Donors
Table 3 shows the noncensored multivariable Cox regression graft survival analysis for recipients of deceased donors (23,231 graft failures inclusive of patients who died with a functioning graft). Younger and older recipients had an increased risk of graft failure. Males and recipients with overweight or obese BMI, a previous kidney transplant, malignancy, angina, peripheral vascular disease, hepatitis C, increasing PRA, dialysis at the time of transplant, and diabetes also had an increased risk of graft failure as compared to recipients without these characteristics. Recipients with a HS diploma or GED had an increased risk of graft failure as compared to attendees of college, recipients of associate or bachelor's degrees and graduate degrees, with HR ranging from 1.06 to 1.15. Attendees of college had an increased risk of graft failure as compared to recipients of associate or bachelor's degrees (HR,1.08) and graduate degrees. Recipients from regions 5, 6, and 8 had a reduced risk of graft failure compared to all other regions, excluding 1, with HR ranging from 0.76 to 0.93. Recipients from region 2 had an increased risk of graft failure compared to regions 1, 3, 5, 6, 7, and 8, with HR ranging from 1.08 to 1.26. Recipients from region 11 had an increased risk of graft failure compared to all other regions, excluding 2, with HR ranging from 1.08 to 1.31. All variables were statistically significant at p < 0.10 by univariable screen and all were included in the final multivariable model.
Discussion
Our series identified recipient age, increasing PRA, BMI, prior transplant, dialysis at the time of transplantation, hepatitis C infection, and education as variables associated with increased risk for graft failure in adult kidney transplant recipients regardless of survival definition or donor type.
Younger and elderly recipients had an increased risk of graft failure. A higher degree of immune responsiveness, immunosuppressive medication noncompliance, financial, and social factors have been postulated to contribute to a greater incidence of early graft loss in the young.4 5 6 Previous studies reported elderly recipients to have a high incidence of comorbidities, frailty, and death with functioning grafts, as well as a greater impact of rejection on graft loss.7
PRA is an immunological test that quantifies the percentage of the population against which an individual reacts via preformed antibodies. Recipients with PRA ≥ 80% have a higher risk of acute rejection leading to graft failure, and are given additional points in the organ allocation algorithm.8
BMI is a measure of body fat based on height and weight.9 Patients with BMI ≥ 30 (obese classes I, II, and III) had an increased risk of graft failure when compared with those with lower BMI. An analysis of adult renal transplant patients registered in U.S. Renal Data System had demonstrated that BMI < 18 and > 36 were associated with worse patient and graft survival.10 Surgical-site infections, delayed graft function, prolonged hospitalization, acute rejection, and decreased overall graft survival had previously been found to be more prevalent in obese individuals.11 12
Prior smaller studies had reported conflicting results on retransplantation outcomes.13 14 15 In our series retransplantation was associated with decreased graft survival.
The process of dialysis is associated with micro- and macrovascular pathologies, malnutrition, chronic systemic inflammation, and decreased renal clearance.16 Studies have shown that patients who are preemptively transplanted have improved graft survival.17 Longer transplant wait times while on dialysis negatively impact posttransplant graft function.16 End-stage renal disease and chronic dialysis can be associated with up to a 72% increase in mortality risk.16
Our study demonstrated that recipients with a diagnosis of hepatitis C before transplantation had an increased risk of graft failure, an observation consistent with prior reports.18 This finding was consistent in all four models. Chronic active hepatitis and cirrhosis should be thoroughly evaluated prior to kidney transplantation.19
Previous studies showed that education could have an association with improved outcomes.20 21 22 We observed increased graft survival among HS graduates irrespective of definition or donor type.
Prior malignancy, peripheral vascular disease, angina, and diabetes were associated with an increased risk of graft failure in recipients of both live and deceased donors only when using the nondeath-censored definition (that treats all deaths as graft failures). This difference could potentially be attributable to deaths associated with malignancies, cardiovascular disease, and diabetic complications.23 24 UNOS and the Organ Procurement and Transplantation Network (OPTN) facilitate organ procurement, allocation, and transplantation by dividing the nation into 11 regions that loosely correspond to U.S. census regions.25 Variations in waiting list times, availability of organs, and transplant rates have been previously reported26 and could account for some of the regional geographic differences observed.
The timeframe of October 25, 1999 through January 1, 2007 was chosen because some variables of interest had a collection end date of January 1, 2007. These variables are currently optional data fields and become very sparse in more recent years. Additionally, important donor-related variables, while not of primary interest for this study, had collection dates that began on October 25, 1999. Certain assumptions were made in regard to plausible values for BMI and creatinine. Specifically, BMI < 15 or > 55, live donor preoperative creatinine > 1.5 and deceased donor terminal creatinine > 6 were deemed unlikely values due to common donor selection practices and were interpreted as missing or unknown. Some important predictors of graft failure were excluded due to large amounts of missing data, such as warm and cold ischemia times. With limitations related to quality control, variables which were found to be nonsignificant may not have been appropriately registered or followed up over time in the database and should be considered further in detail. Donor variables for inclusion in the multivariable model were based on HLA mismatch and on the KDPI. Since KDPI addresses only deceased donors, variables such as diabetes, cause of death, and cardiac arrest postbrain death were not relevant for live donors.
Conclusions
We propose a model based on the recipient characteristics, and independent of donor variables. Age, increasing PRA, BMI, prior transplant, dialysis at the time of transplantation, hepatitis C infection, and education were found to be associated with an increased risk for graft failure in adult kidney transplant recipients regardless of survival definition and donor type. These recipient predictive criteria could further optimize outcomes and organ allocation27 by providing information not only at the time of evaluation and during wait listing when no donor information is routinely available, but also at the time of organ allocation when they would be complemented by the already existing donor grading classifications.
Acknowledgments
The authors of this article have no conflicts of interest to disclose. This work was supported in part by Health Resources and Services Administration contract 231–00–0015. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. Lisa Rosen, ScM, Feinstein Institute of Medical Research, conducted the data analysis, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the analysis. The authors would like to acknowledge Bruce Shepperson, SAS analyst at UNOS, for his assistance with all data-related queries. We would also like to thank Mr. John Raggio and Dr. Donna Kesselman, the Janet and John Raggio Nephrology Institute, and the Michael and Jamie Mazzei family for their kind support.
References
- 1.Rao P S, Schaubel D E, Guidinger M K. et al. A comprehensive risk quantification score for deceased donor kidneys: the kidney donor risk index. Transplantation. 2009;88(2):231–236. doi: 10.1097/TP.0b013e3181ac620b. [DOI] [PubMed] [Google Scholar]
- 2.Akkina S K, Asrani S K, Peng Y, Stock P, Kim W R, Israni A K. Development of organ-specific donor risk indices. Liver Transpl. 2012;18(4):395–404. doi: 10.1002/lt.23398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hariharan S, Johnson C P, Bresnahan B A, Taranto S E, McIntosh M J, Stablein D. Improved graft survival after renal transplantation in the United States, 1988 to 1996. N Engl J Med. 2000;342(9):605–612. doi: 10.1056/NEJM200003023420901. [DOI] [PubMed] [Google Scholar]
- 4.Joosten S A, Sijpkens Y W, van Kooten C, Paul L C. Chronic renal allograft rejection: pathophysiologic considerations. Kidney Int. 2005;68(1):1–13. doi: 10.1111/j.1523-1755.2005.00376.x. [DOI] [PubMed] [Google Scholar]
- 5.Andreoni K A, Forbes R, Andreoni R M, Phillips G, Stewart H, Ferris M. Age-related kidney transplant outcomes: health disparities amplified in adolescence. JAMA Intern Med. 2013;173(16):1524–1532. doi: 10.1001/jamainternmed.2013.8495. [DOI] [PubMed] [Google Scholar]
- 6.Kasiske B L, Cangro C B, Hariharan S. et al. The evaluation of renal transplantation candidates: clinical practice guidelines. Am J Transplant. 2001;1 02:3–95. [PubMed] [Google Scholar]
- 7.Furian L, Rigotti P. Transplantation: Optimizing outcomes in elderly kidney transplant recipients. Nat Rev Nephrol. 2013;9(7):382–384. doi: 10.1038/nrneph.2013.105. [DOI] [PubMed] [Google Scholar]
- 8.Organ Procurement and Transplantation Network (OPTN) Allocation calculators Available at: http://optn.transplant.hrsa.gov/converge/resources/allocationcalculators.asp?index=75. Accessed August 26, 2013
- 9.The World Health Organization (WHO) Global database on body mass index Available at: http://apps.who.int/bmi/index.jsp?introPage=intro_3.html. Accessed August 26, 2013
- 10.Meier-Kriesche H-U, Arndorfer J A, Kaplan B. The impact of body mass index on renal transplant outcomes: a significant independent risk factor for graft failure and patient death. Transplantation. 2002;73(1):70–74. doi: 10.1097/00007890-200201150-00013. [DOI] [PubMed] [Google Scholar]
- 11.Lynch R J, Ranney D N, Shijie C, Lee D S, Samala N, Englesbe M J. Obesity, surgical site infection, and outcome following renal transplantation. Ann Surg. 2009;250(6):1014–1020. doi: 10.1097/SLA.0b013e3181b4ee9a. [DOI] [PubMed] [Google Scholar]
- 12.Gore J L, Pham P T, Danovitch G M. et al. Obesity and outcome following renal transplantation. Am J Transplant. 2006;6(2):357–363. doi: 10.1111/j.1600-6143.2005.01198.x. [DOI] [PubMed] [Google Scholar]
- 13.Ott U, Busch M, Steiner T, Schubert J, Wolf G. Renal retransplantation: a retrospective monocentric study. Transplant Proc. 2008;40(5):1345–1348. doi: 10.1016/j.transproceed.2008.01.068. [DOI] [PubMed] [Google Scholar]
- 14.Wang D, Xu T Z, Chen J H. et al. Factors influencing second renal allograft survival: a single center experience in China. Transpl Immunol. 2009;20(3):150–154. doi: 10.1016/j.trim.2008.09.010. [DOI] [PubMed] [Google Scholar]
- 15.Horovitz D, Caumartin Y, Warren J. et al. Outcome of third renal allograft retransplants versus primary transplants from paired donors. Transplantation. 2009;87(8):1214–1220. doi: 10.1097/TP.0b013e31819f0f5c. [DOI] [PubMed] [Google Scholar]
- 16.Meier-Kriesche H U, Port F K, Ojo A O. et al. Effect of waiting time on renal transplant outcome. Kidney Int. 2000;58(3):1311–1317. doi: 10.1046/j.1523-1755.2000.00287.x. [DOI] [PubMed] [Google Scholar]
- 17.Meier-Kriesche H-U, Kaplan B. Waiting time on dialysis as the strongest modifiable risk factor for renal transplant outcomes: a paired donor kidney analysis. Transplantation. 2002;74(10):1377–1381. doi: 10.1097/00007890-200211270-00005. [DOI] [PubMed] [Google Scholar]
- 18.Morales J M, Aguado J M. Hepatitis C and renal transplantation. Curr Opin Organ Transplant. 2012;17(6):609–615. doi: 10.1097/MOT.0b013e32835a2bac. [DOI] [PubMed] [Google Scholar]
- 19.Carbone M, Mutimer D, Neuberger J. Hepatitis C virus and nonliver solid organ transplantation. Transplantation. 2013;95(6):779–786. doi: 10.1097/TP.0b013e318273fec4. [DOI] [PubMed] [Google Scholar]
- 20.Schaeffner E S, Mehta J, Winkelmayer W C. Educational level as a determinant of access to and outcomes after kidney transplantation in the United States. Am J Kidney Dis. 2008;51(5):811–818. doi: 10.1053/j.ajkd.2008.01.019. [DOI] [PubMed] [Google Scholar]
- 21.Waterman A D, Barrett A C, Stanley S L. Optimal transplant education for recipients to increase pursuit of living donation. Prog Transplant. 2008;18(1):55–62. doi: 10.1177/152692480801800111. [DOI] [PubMed] [Google Scholar]
- 22.Goldfarb-Rumyantzev A S, Koford J K, Baird B C. et al. Role of socioeconomic status in kidney transplant outcome. Clin J Am Soc Nephrol. 2006;1(2):313–322. doi: 10.2215/CJN.00630805. [DOI] [PubMed] [Google Scholar]
- 23.Morath C, Mueller M, Goldschmidt H, Schwenger V, Opelz G, Zeier M. Malignancy in renal transplantation. J Am Soc Nephrol. 2004;15(6):1582–1588. doi: 10.1097/01.asn.0000126194.77004.9b. [DOI] [PubMed] [Google Scholar]
- 24.Kasiske B L, Zeier M G, Chapman J R. et al. KDIGO clinical practice guideline for the care of kidney transplant recipients: a summary. Kidney Int. 2010;77(4):299–311. doi: 10.1038/ki.2009.377. [DOI] [PubMed] [Google Scholar]
- 25.Organ Procurement and Transplantation Network (OPTN) Regions Available at: http://optn.transplant.hrsa.gov/converge/members/regions.asp. Accessed August 26, 2013
- 26.Ellison M D, Edwards L B, Edwards E B, Barker C F. Geographic differences in access to transplantation in the United States. Transplantation. 2003;76(9):1389–1394. doi: 10.1097/01.TP.0000090332.30050.BA. [DOI] [PubMed] [Google Scholar]
- 27.Ross L F, Parker W, Veatch R M, Gentry S E, Thistlethwaite J R Jr. Equal Opportunity Supplemented by Fair Innings: equity and efficiency in allocating deceased donor kidneys. Am J Transplant. 2012;12(8):2115–2124. doi: 10.1111/j.1600-6143.2012.04141.x. [DOI] [PubMed] [Google Scholar]