Abstract
Introduction
After renal transplantation, decreased renal function is associated with increased risk of cardiovascular disease, graft loss, and mortality. We investigated whether declining renal function was associated with hospitalization after transplantation.
Methods
Adult, first-time, kidney transplant recipients between 2004 and 2006 from the United Network for Organ Sharing database and hospitalizations 1 year after the 6-month posttransplant follow-up visit were examined. Generalized linear models explored the relationship between estimated glomerular filtration rate (eGFR) measured at 6 months and the number of hospitalizations in the following year.
Results
Of 15,778 kidney transplant recipients, 19.1% were admitted in the year after the 6-month follow-up visit. Among those hospitalized, the mean number of hospitalizations was 1.71, which increased with decreasing eGFR. In multivariable models, a decrease in eGFR was significantly associated with increased hospitalizations: for every 10 ml/min per 1.73 m2 decrease in eGFR, there was an 11% increase in hospitalization rate (P < 0.001). Lower eGFR after the first 6 months after transplantation was associated with an increase in late hospitalizations among adult kidney transplant recipients.
Discussion
Identifying patients with declining eGFR and other risk factors may help prevent morbidity and mortality associated with hospitalization after transplantation.
Keywords: estimated glomerular filtration rate, hospital readmission, kidney transplant
The treatment of end-stage renal disease requires renal replacement therapy in the form of dialysis or renal transplantation; transplantation is the preferred treatment, and is associated with lower mortality and improved quality of life.1 Prior studies have shown that estimated glomerular filtration rate (eGFR) is the best predictor of long-term graft function among transplant recipients2, 3 and that declining eGFR after kidney transplantation is associated with higher health care costs;4, 5 however, the association between declining eGFR and posttransplantation hospitalization has not been examined. Hospital readmission can be used as a proxy of poor health outcomes after renal transplantation. According to an analysis of Medicare claims data from 2003 to 2004, 1 of 5 Medicare beneficiaries were rehospitalized within 30 days of discharge at a cost of $17.4 billion.6
Prior studies have found that 30-day readmission or “early readmission” after kidney transplantation is associated with increased morbidity, costs, and transition-of-care errors7 in addition to increased graft loss and mortality among patients with Medicare primary insurance.8, 9 Despite a sizeable amount of research on early hospital readmission among kidney transplant recipients, to our knowledge, no study has examined eGFR as a predictor of hospital readmission beyond 6 months after transplantation. The purpose of this study was to determine whether the eGFR 6 months after renal transplantation was associated with “late” all-cause hospitalization approximately 1 year after the 6-month posttransplant data collection point. In addition, because prior studies of hospitalization among kidney transplant recipients have been conducted among patients receiving Medicare as primary insurance only, we sought to characterize the key risk factors for late hospital readmission among a population that is more generalizable to the entire US kidney transplant recipient population.
Methods
Study Population and Data Sources
The United Network for Organ Sharing (UNOS) is a nonprofit organization that maintains the Organ Procurement and Transplantation Network database. Organ Procurement and Transplantation Network contains information regarding every organ donation and transplant event in the United States since 1 October 1987. For this study, we used the Standard Transplant Analysis and Research files based on Organ Procurement and Transplantation Network data for kidney, pancreas, and kidney-pancreas waiting list and transplant and/or follow-up patients between 1 October 1987 and 31 October 2011.
We restricted our analyses to data from 30 June 2004 to 1 January 2007 for adult (18 years or older), first-time, kidney-only transplant recipients where the transplant was received before 1 January 2006, based on the availability of data regarding hospitalizations during the follow-up period, which was a required field on UNOS transplant follow-up forms until 1 January 2007. Our follow-up time of 2 years after transplantation (18 months after the 6-month “baseline” for this study) was also chosen based on the availability of hospitalization data. Whereas 95% of patients had hospitalization follow-up data at 2 years after transplantation, fewer than half of participants had hospitalization data after this time point. Figure 1 shows exclusion and selection criteria. Analyses were further restricted by excluding patients with missing information on both the exposure—eGFR (n = 302)—and outcome—number of hospitalizations during the year after the 6-month post-renal-transplantation follow-up (n = 318). Next, patients who died (n = 369) or were lost to follow-up (n = 82) during the study period were similarly excluded. Unlikely, values for eGFR (>125 ml/min per 1.73 m2) were set to missing (n = 83). The patients who were excluded for missing data on exposure or outcome were compared with those who were included in this analysis and found to be similar.
Figure 1.
Flow chart of exclusion criteria for study population. eGFR, estimated glomerular filtration rate.
Primary Exposure
The primary exposure was eGFR collected on the UNOS transplant recipient follow-up forms at 6 months, and calculated using the 4-variable Modification of Diet in Renal Disease equation.10 The eGFR was calculated from serum creatinine measured and reported by transplant centers to UNOS at regularly scheduled data collection time points. The 6-month posttransplant follow-up creatinine value served as the baseline time point for calculation of eGFR. Laboratory values reported between 135 and 225 days (6 months ± 45 days) after transplantation were considered to be the 6-month follow-up value. If a patient had multiple follow-up labs reported during this time period, the first value was used as the baseline.
Primary Outcome
The primary outcome for this study was the number of hospitalizations that occurred in the year after the 6-month time point after renal transplantation, that is, hospitalizations that occurred up to 18 months after transplantation. Rehospitalization after 6 months, as opposed to 30 days, is less likely to be the result of surgical complications from the transplant procedure, as patients stabilize over time. Only hospitalizations after this baseline time point were considered for these analyses to ensure that eGFR was measured before rehospitalization.
Statistical Analyses
We performed descriptive analyses of the relationship between the 6-month eGFR and the number of late hospitalizations, controlling for several potential confounders including recipient characteristics (demographics, clinical characteristics, comorbidities), donor characteristics, transplant characteristics, and complications. Bivariate analyses by standard eGFR cut points and by the number of hospitalizations were conducted using t-tests to compare continuous variables and χ2 tests for categorical variables. Covariates considered as potential confounders included recipient characteristics: age, race (white, black, Asian, Hispanic, other), sex, body mass index (continuous),11 education level (high school or less, attended college, associate and/or bachelor degree, post-college graduate degree, missing), primary form of payment (private insurance, public insurance, missing), primary diagnosis (diabetes related, hypertension related, glomerular nephritis, secondary glomerular nephritis, hereditary and/or congenital disease, other), previous dialysis (yes, no), diabetes (yes, no, missing), cancer or malignancy (yes, no, missing), total serum albumin, serum creatinine; donor characteristics: donor age, donor height, donor type (living, deceased, deceased expanded criteria), donor diabetes (yes, no, missing), and donor hypertension (yes, no, missing); transplant characteristics: length of stay (at time of transplant in days), human leukocyte antigen (HLA) match level (zero to six, missing), cold ischemia time (0–10, 11–20, 21–30, >30 hours, missing), panel-reactive antibody (0, 1–20, 21–80, 80+); and complications: previous hospitalization (hospitalization during the first 6 months after transplantation) (yes, no) acute rejection (yes, no), delayed graft function (yes, no). “Missing” was considered as a separate response category for covariates as noted. No patients in our sample had experienced an allograft loss; therefore, loss of allograft was not included in these analyses.
The number of late hospitalizations was modeled using a zero-inflated Poisson regression (generalized linear model using SAS Proc Genmod [SAS Version 9.3, Cary, NC]). For the final multivariable model, we included variables that were significantly associated with both exposure and outcome during descriptive analyses and known potential confounders from a review of the literature.12 The full model included the main exposure variable—eGFR—and the following variables—race, recipient age, sex, body mass index category, insurance type, education, HLA match level, cold ischemia time, length of stay (at the time of transplant), serum albumin, diabetes status, hypertension status, primary diagnosis, delayed graft function, pretransplant dialysis, before hospitalization, donor age, donor type, donor hypertension, and donor diabetes. Collinearity for this model was evaluated by assessing whether condition indices were high (>30) with variance decomposition proportions for variables excluding the intercept greater than 0.50. No variables in the full model violated these assumptions, and thus the full model included all variables stated above.
Sensitivity Analyses
We conducted several sensitivity analyses to examine the robustness of findings. First, we repeated analyses using the chronic kidney disease-Epi formula to calculate the eGFR. We also repeated analyses including patients who (i) died during the study follow-up period, (ii) had previously received a transplant, and (ii) had received a multi-organ transplant.
Statistical Significance
All tests were 2-sided, with an α = 0.05, and analyses were conducted using SAS version 9.3 (SAS Institute, Cary, NC). The Emory Institutional Review Board approved this study.
Results
A total of 15,778 adult, first-time, kidney-only transplant recipients met the inclusion criteria for this analysis. The mean age of the study population was 50 years (±13.5), 24.2% (n = 3812) were black, 60.5% were men (n = 9552), and 57.8% had public insurance (n = 9113).
Figure 2 shows the distribution of patients by eGFR category. Patients in the lowest eGFR category were more likely to be black (40.9% compared with 24.2% in the study sample as a whole), older, overweight, expanded criteria donor recipients, and publicly insured (Table 1). They were also more likely to have been readmitted to the hospital (52.7% compared with 19.1% in the study sample). On average, those with lower eGFR also had a higher total serum albumin, cold ischemia time, serum creatinine, and length of stay at the time of transplant compared with the total study sample. The number of hospital readmissions decreased among those in higher eGFR categories.
Figure 2.
Distribution of study participants by estimated glomerular filtration rate (eGFR) cut point at 6 months after transplantation.
Table 1.
Baseline demographic characteristics by eGFR cut points for adult, first-time, kidney-only recipients (United Network for Organ Sharing 2004–2007)
| Study population |
By eGFR cut points |
P value | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| <15 |
15–29 |
30–44 |
45–59 |
60–89 |
90–125 |
||||||||||
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| 15,778 | 93 | 0.59 | 763 | 4.84 | 3308 | 20.97 | 5437 | 34.46 | 5464 | 34.63 | 713 | 4.52 | |||
| Recipient characteristics | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (yr) | 49.6 | 13.5 | 50.0 | 13.5 | 52.9 | 12.8 | 51.9 | 12.8 | 49.9 | 13.2 | 47.7 | 13.8 | 46.6 | 14.3 | <0.001 |
| BMI (kg/m2) | 27.4 | 5.6 | 27.9 | 5.6 | 28.8 | 6.2 | 28.2 | 5.6 | 27.7 | 5.7 | 26.6 | 5.4 | 25.6 | 5.4 | <0.001 |
| Total serum albumin (g/dl) | 3.9 | 0.6 | 3.9 | 0.5 | 3.9 | 0.6 | 3.9 | 0.6 | 3.9 | 0.6 | 3.9 | 0.6 | 3.9 | 0.7 | <0.001 |
| Serum creatinine (mg/dl) | 1.5 | 0.7 | 6.3 | 2.4 | 2.8 | 0.7 | 1.9 | 0.3 | 1.4 | 0.2 | 1.1 | 0.2 | 0.8 | 0.1 | <0.001 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| Race/ethnicity | <0.001 | ||||||||||||||
| White | 8726 | 55.3 | 38 | 40.9 | 438 | 57.4 | 2042 | 61.7 | 3192 | 58.7 | 2752 | 50.4 | 264 | 37.0 | |
| Black | 3812 | 24.2 | 38 | 40.9 | 186 | 24.4 | 688 | 20.8 | 1189 | 21.9 | 1458 | 26.7 | 253 | 35.5 | |
| Hispanic | 2119 | 13.4 | 12 | 12.9 | 84 | 11.0 | 349 | 10.6 | 684 | 12.6 | 857 | 15.7 | 133 | 18.7 | |
| Asian | 796 | 5.0 | 3 | 3.2 | 41 | 5.4 | 157 | 4.8 | 258 | 4.8 | 285 | 5.2 | 52 | 7.3 | |
| Other | 325 | 2.1 | 2 | 2.2 | 14 | 1.8 | 72 | 2.2 | 114 | 2.1 | 112 | 2.1 | 11 | 1.5 | |
| Sex | <0.001 | ||||||||||||||
| Male | 9552 | 60.5 | 52 | 55.9 | 403 | 52.8 | 1996 | 60.3 | 3422 | 62.9 | 3276 | 60.0 | 403 | 56.5 | |
| Educationa | 2578a | <0.001 | |||||||||||||
| High school (9–12) or less | 7057 | 53.5 | 42 | 55.3 | 351 | 55.1 | 1430 | 51.8 | 2361 | 51.9 | 2512 | 55.0 | 361 | 59.4 | |
| Attended college/technical school | 3169 | 24.0 | 19 | 25.0 | 148 | 23.2 | 650 | 23.5 | 1095 | 24.1 | 1118 | 24.5 | 139 | 22.9 | |
| Associate/bachelor degree | 2089 | 15.8 | 8 | 10.5 | 89 | 14.0 | 467 | 16.9 | 785 | 17.3 | 662 | 14.5 | 78 | 12.8 | |
| Post-college graduate degree | 885 | 6.7 | 7 | 9.2 | 49 | 7.7 | 215 | 7.8 | 307 | 6.8 | 277 | 6.1 | 30 | 4.9 | |
| Insurance | <0.001 | ||||||||||||||
| Private insurance | 6,629 | 42.0 | 26 | 28.0 | 271 | 35.5 | 1446 | 43.7 | 2383 | 43.8 | 2276 | 41.7 | 227 | 31.8 | |
| Public insurance | 9113 | 57.8 | 67 | 72.0 | 492 | 64.5 | 1859 | 56.2 | 3045 | 56.0 | 3169 | 58.0 | 481 | 67.5 | |
| Other | 36 | 0.2 | – | – | – | – | 3 | 0.1 | 9 | 0.2 | 19 | 0.4 | 5 | 0.7 | |
| Primary diagnosis | 1719a | <0.001 | |||||||||||||
| Diabetes related | 3825 | 24.3 | 23 | 21.5 | 209 | 26.9 | 683 | 24.5 | 1102 | 24.3 | 1100 | 23.6 | 168 | 25.8 | |
| Hypertension related | 3052 | 22.6 | 31 | 29.0 | 167 | 21.5 | 609 | 21.8 | 998 | 22.0 | 1080 | 23.2 | 167 | 25.6 | |
| Glomerular nephritis | 2521 | 18.7 | 15 | 14.0 | 128 | 16.5 | 509 | 18.2 | 903 | 19.9 | 855 | 18.4 | 111 | 17.0 | |
| Secondary glomerular nephritis | 596 | 4.4 | 2 | 1.9 | 38 | 4.9 | 106 | 3.8 | 179 | 4.0 | 234 | 5.0 | 37 | 5.7 | |
| Hereditary and/or congenital diseases | 1863 | 13.8 | 12 | 11.2 | 93 | 12.0 | 441 | 15.8 | 654 | 14.4 | 598 | 12.9 | 65 | 10.0 | |
| Other | 2202 | 16.3 | 24 | 22.4 | 141 | 18.2 | 445 | 15.9 | 701 | 15.5 | 787 | 16.9 | 104 | 16.0 | |
| Previous dialysisa | 349a | <0.001 | |||||||||||||
| Yes | 12,118 | 79.1 | 92 | 79.3 | 702 | 85.4 | 2512 | 78.9 | 4025 | 77.1 | 4174 | 79.4 | 613 | 85.5 | |
| Previous malignancy | 0.29 | ||||||||||||||
| Yes | 562 | 3.8 | 2 | 2.5 | 32 | 4.8 | 139 | 4.8 | 206 | 4.4 | 183 | 3.8 | 24 | 3.8 | |
| Diabetesa | 303a | 0.31 | |||||||||||||
| Yes | 4828 | 31.2 | 25 | 27.5 | 258 | 34.7 | 1014 | 31.5 | 1644 | 30.8 | 1660 | 30.9 | 227 | 32.2 | |
| Cerebrovascular diseasea | 878a | 0.27 | |||||||||||||
| Yes | 392 | 2.6 | 5 | 6.2 | 17 | 2.4 | 74 | 2.4 | 129 | 2.5 | 147 | 2.8 | 20 | 3.0 |
| Donor characteristics | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Donor age (yr) | 39.2 | 14.55 | 43.6 | 15.7 | 48.4 | 14.2 | 45.5 | 14.0 | 40.3 | 13.7 | 34.3 | 13.4 | 29.9 | 12.9 | 0.50 |
| Donor height (cm) | 169.4 | 14.4 | 166.1 | 16.7 | 168.1 | 13.1 | 168.2 | 13.4 | 169.4 | 13.1 | 170.3 | 15.6 | 170.8 | 18.2 | <0.001 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| Donor type | <0.001 | ||||||||||||||
| Living | 6839 | 43.4 | 24 | 25.8 | 187 | 24.5 | 1348 | 40.8 | 2595 | 47.7 | 2424 | 44.4 | 261 | 36.6 | |
| Deceased standard criteria | 7374 | 46.7 | 46 | 49.5 | 351 | 46.0 | 1369 | 41.4 | 2354 | 43.3 | 2817 | 51.6 | 437 | 61.3 | |
| Deceased expanded criteria | 1565 | 9.9 | 23 | 24.7 | 225 | 29.5 | 591 | 17.9 | 488 | 9.0 | 223 | 4.1 | 15 | 2.1 | |
| Donor diabetesa | 310a | <0.001 | |||||||||||||
| Yes | 452 | 2.88 | 7 | 5.9 | 64 | 7.6 | 154 | 4.7 | 125 | 2.3 | 88 | 1.6 | 14 | 1.9 | |
| Donor hypertensiona | 355a | <0.001 | |||||||||||||
| Yes | 2274 | 14.51 | 32 | 27.1 | 298 | 35.5 | 733 | 22.5 | 693 | 13.0 | 465 | 8.7 | 53 | 7.3 |
| Transplant characteristics | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cold ischemic time (h) | 12.9 | 10.76 | 15.5 | 9.0 | 16.8 | 11.0 | 13.8 | 11.3 | 12.2 | 10.5 | 12.3 | 10.6 | 13.2 | 9.9 | <0.001 |
| Length of stay at transplant (d) | 7.1 | 11.89 | 10.2 | 9.9 | 9.0 | 7.6 | 7.7 | 18.1 | 6.8 | 7.0 | 6.6 | 8.5 | 7.7 | 24.3 | <0.001 |
| Peak PRA | 8.9 | 22.0 | 15.6 | 29.6 | 10.5 | 24.0 | 9.2 | 22.4 | 7.9 | 20.4 | 9.1 | 22.6 | 10.1 | 23.2 | <0.001 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| PRA | <0.001 | ||||||||||||||
| 0 | 10,712 | 68.36 | 73 | 61.9 | 538 | 64.1 | 2186 | 67.1 | 3725 | 69.6 | 3714 | 69.1 | 476 | 65.1 | |
| 1–20 | 3081 | 19.66 | 19 | 16.1 | 183 | 21.8 | 670 | 20.6 | 1048 | 19.6 | 1009 | 18.8 | 152 | 20.8 | |
| 21–80 | 1235 | 7.88 | 15 | 12.7 | 75 | 8.9 | 265 | 8.1 | 391 | 7.3 | 422 | 7.9 | 67 | 9.2 | |
| 80+ | 643 | 4.1 | 11 | 9.3 | 43 | 5.1 | 138 | 4.2 | 185 | 3.5 | 230 | 4.3 | 36 | 4.9 | |
| HLA match levela | 53a | <0.001 | |||||||||||||
| Zero | 2759 | 17.6 | 21 | 22.8 | 176 | 23.1 | 589 | 17.9 | 947 | 17.5 | 906 | 16.7 | 120 | 16.9 | |
| One | 4267 | 27.1 | 28 | 30.4 | 233 | 30.5 | 985 | 29.9 | 1457 | 26.9 | 1368 | 25.1 | 196 | 27.5 | |
| Two | 2737 | 17.4 | 14 | 15.2 | 136 | 17.8 | 566 | 17.2 | 943 | 17.4 | 946 | 17.4 | 132 | 18.5 | |
| Three | 3211 | 20.4 | 15 | 16.3 | 117 | 15.3 | 628 | 19.1 | 1163 | 21.5 | 1153 | 21.2 | 135 | 19.0 | |
| Four | 1150 | 7.3 | 5 | 5.4 | 31 | 4.1 | 227 | 6.9 | 378 | 7.0 | 454 | 8.3 | 55 | 7.7 | |
| Five | 844 | 5.4 | 4 | 4.4 | 44 | 5.8 | 167 | 5.1 | 276 | 5.1 | 314 | 5.8 | 39 | 5.5 | |
| Six | 757 | 4.8 | 5 | 5.4 | 26 | 3.4 | 134 | 4.1 | 257 | 4.7 | 300 | 5.5 | 35 | 4.9 |
| Complications | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Posttransplant hospital readmissions (number) | 0.3 | 0.9 | 1.2 | 1.7 | 0.7 | 1.3 | 0.4 | 1.0 | 0.3 | 0.8 | 0.3 | 0.7 | 0.3 | 0.8 | <0.001 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | <0.001 | |
| Previous hospitalization | |||||||||||||||
| Yes | 3815 | 24.3 | 75 | 63.6 | 379 | 45.1 | 992 | 30.4 | 1156 | 21.6 | 1055 | 19.6 | 158 | 21.6 | |
| Delayed graft function | <0.001 | ||||||||||||||
| Yes | 2534 | 14.3 | 49 | 32.2 | 314 | 32.0 | 681 | 18.7 | 750 | 12.6 | 642 | 10.6 | 98 | 11.3 | |
| Acute rejection | 370a | <0.001 | |||||||||||||
| Yes | 816 | 5.3 | 36 | 28.6 | 160 | 19.2 | 274 | 8.6 | 204 | 3.9 | 128 | 2.4 | 14 | 1.9 |
BMI, body mass index; eGFR, estimated glomerular filtration rate; HLA, human leukocyte antigen; PRA, panel-reactive antibody.
Includes missing values.
A total of 19.1% of patients were hospitalized at least once within the year after the 6-month posttransplant follow-up visit, and 0.7% were hospitalized 5 or more times in this time period. Black race, Asian race, female sex, length of stay at the time of transplant, public insurance (vs. private insurance), standard and expanded criteria deceased donor (vs. living donors), diabetic status, and increasing donor age were associated with increased hospitalization (Table 2). Additional covariates considered are provided in Supplemental Table S1. Among those hospitalized, the mean number of hospitalizations was 1.71 (range, 0–12) and increased monotonically with decreasing eGFR. Patients who were hospitalized (n = 3009) during the year after the 6-month posttransplant follow-up had a significantly lower eGFR than those who were not hospitalized (n = 12,769) (48.79 vs. 57.45 ml/min per 1.73 m2, respectively) (Figure 3).
Table 2.
Basic demographic characteristics by the number of hospitalizations in the year after the 6-month visit for adult, first-time, kidney-only recipients (United Network for Organ Sharing 2004–2007)
| Study population (N = 15,778) |
By number of hospitalizations |
P value | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Zero (N = 12,769) |
One (N = 1868) |
Two (N = 623) |
Three (N = 278) |
Four (N = 128) |
Five or more (N = 217) |
||||||||||
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | ||
| Recipient characteristics | |||||||||||||||
| eGFR (ml/min per 1.73 m2) | 56.6 | 18.02 | 57.4 | 17.62 | 54.2 | 18.25 | 52.1 | 20.33 | 49.9 | 20.73 | 48.7 | 18.83 | 46.8 | 22.13 | <0.001 |
| Mean age (yr) | 49.6 | 13.50 | 49.5 | 13.42 | 49.7 | 13.48 | 49.7 | 14.13 | 50.3 | 14.48 | 49.6 | 14.86 | 48.9 | 15.34 | 0.90 |
| Mean BMI (kg/m2) | 27.4 | 5.64 | 27.3 | 5.60 | 27.6 | 5.76 | 27.7 | 5.68 | 27.6 | 6.06 | 27.1 | 5.80 | 27.7 | 5.92 | 0.40 |
| Total serum albumin (g/dl) | 3.9 | 0.60 | 3.9 | 0.59 | 3.9 | 0.66 | 3.8 | 0.59 | 3.8 | 0.52 | 3.9 | 0.52 | 3.9 | 0.50 | <0.001 |
| Serum creatinine (mg/dl) | 1.5 | 0.65 | 1.5 | 0.57 | 1.6 | 0.70 | 1.7 | 1.08 | 1.8 | 1.20 | 1.8 | 0.80 | 2.0 | 1.25 | <0.001 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| Ethnicity | <0.001 | ||||||||||||||
| White | 8726 | 55.3 | 7122 | 55.8 | 1008 | 54.0 | 338 | 54.3 | 143 | 51.4 | 66 | 51.6 | 49 | 43.8 | |
| Black | 3812 | 24.2 | 2953 | 23.1 | 523 | 28.0 | 164 | 26.3 | 81 | 29.1 | 44 | 34.4 | 47 | 42.0 | |
| Hispanic | 2119 | 13.4 | 1749 | 13.7 | 227 | 12.2 | 83 | 13.3 | 35 | 12.6 | 15 | 11.7 | 10 | 8.9 | |
| Asian | 796 | 5.0 | 675 | 5.3 | 72 | 3.9 | 29 | 4.7 | 15 | 5.4 | 3 | 2.3 | 2 | 1.8 | |
| Other | 325 | 2.1 | 270 | 2.1 | 38 | 2.0 | 9 | 1.4 | 4 | 1.4 | 4 | 3.6 | |||
| Sex | 0.04 | ||||||||||||||
| Male | 9552 | 60.5 | 7796 | 61.1 | 1103 | 59.1 | 368 | 59.1 | 158 | 56.8 | 71 | 55.5 | 56 | 50.0 | |
| Educationa | 2578a | 0.69 | |||||||||||||
| High school (9–12) or less | 7057 | 53.5 | 5671 | 53.1 | 867 | 54.9 | 287 | 56.4 | 129 | 53.3 | 57 | 53.8 | 46 | 53.5 | |
| Attended college/technical school | 3169 | 24.0 | 2578 | 24.1 | 373 | 23.6 | 111 | 21.8 | 54 | 22.3 | 34 | 32.1 | 19 | 22.1 | |
| Associate/bachelor degree | 2089 | 15.8 | 1712 | 16.0 | 234 | 14.8 | 78 | 15.3 | 41 | 16.9 | 9 | 8.5 | 15 | 17.4 | |
| Post-college graduate degree | 885 | 6.7 | 717 | 6.7 | 105 | 6.7 | 33 | 6.5 | 18 | 7.4 | 6 | 5.7 | 6 | 7.0 | |
| Insurance | <0.001 | ||||||||||||||
| Private insurance | 6629 | 42.0 | 5481 | 42.9 | 715 | 38.3 | 245 | 39.3 | 110 | 39.6 | 40 | 31.3 | 38 | 33.9 | |
| Public insurance | 9113 | 57.8 | 7256 | 56.8 | 1150 | 61.6 | 378 | 60.7 | 167 | 60.1 | 88 | 68.8 | 74 | 66.1 | |
| Other | 36 | 0.2 | 32 | 0.3 | 3 | 0.2 | – | – | 1 | 0.4 | – | – | – | – | |
| Primary diagnosis | 1719a | <0.001 | |||||||||||||
| Diabetes related | 3825 | 24.3 | 2337 | 23.2 | 530 | 25.2 | 223 | 30.6 | 106 | 31.0 | 53 | 36.6 | 36 | 25.9 | |
| Hypertension related | 3052 | 22.6 | 2284 | 22.7 | 482 | 22.9 | 162 | 22.2 | 77 | 22.5 | 18 | 12.4 | 29 | 20.9 | |
| Glomerular nephritis | 2521 | 18.7 | 1952 | 19.4 | 363 | 17.2 | 107 | 14.7 | 58 | 17.0 | 21 | 14.5 | 20 | 14.4 | |
| Secondary glomerular nephritis | 596 | 4.4 | 442 | 4.4 | 95 | 4.5 | 30 | 4.1 | 12 | 3.5 | 6 | 4.1 | 11 | 7.9 | |
| Hereditary/congenital diseases | 1863 | 13.8 | 1418 | 14.1 | 288 | 13.7 | 86 | 11.8 | 42 | 12.3 | 20 | 13.8 | 9 | 6.5 | |
| Other | 2202 | 16.3 | 1624 | 16.2 | 348 | 16.5 | 122 | 16.7 | 47 | 13.7 | 27 | 18.6 | 34 | 24.5 | |
| Diabetesa | 303a | <0.0001 | |||||||||||||
| Yes | 4828 | 31.2 | 3782 | 30.2 | 608 | 33.2 | 230 | 37.8 | 110 | 40.7 | 50 | 41.0 | 48 | 44.0 | |
| Previous malignancy | 1984a | 0.02 | |||||||||||||
| Yes | 562 | 3.8 | 465 | 4.2 | 62 | 3.7 | 26 | 4.7 | 20 | 8.1 | 8 | 7.1 | 5 | 5.2 | |
| Previous dialysisa | 349 | <0.0001 | |||||||||||||
| Yes | 12,118 | 79.09 | 8875 | 77.99 | 1951 | 80.89 | 695 | 83.73 | 319 | 83.73 | 143 | 87.2 | 135 | 86.54 | |
| Cerebrovascular diseasea | 878a | 0.15 | |||||||||||||
| Yes | 392 | 2.6 | 301 | 2.5 | 52 | 3.0 | 24 | 4.2 | 7 | 2.7 | 5 | 4.2 | 3 | 2.9 | |
| Donor characteristics | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Donor age (yr) | 39.2 | 14.55 | 38.9 | 14.42 | 40.0 | 15.02 | 41.6 | 14.90 | 41.6 | 15.02 | 41.5 | 14.43 | 44.7 | 15.16 | <0.001 |
| Donor height (cm) | 169.4 | 14.4 | 169.5 | 14.5 | 169.4 | 14.2 | 169.4 | 13.3 | 169.6 | 14.9 | 168.8 | 13.3 | 166.8 | 15.3 | 0.32 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| Donor type | <0.001 | ||||||||||||||
| Living | 6839 | 43.4 | 5674 | 44.4 | 733 | 39.2 | 243 | 39.0 | 97 | 34.9 | 50 | 39.1 | 42 | 37.5 | |
| Deceased standard criteria | 7374 | 46.7 | 5915 | 46.3 | 924 | 49.5 | 289 | 46.4 | 142 | 51.1 | 59 | 46.1 | 45 | 40.2 | |
| Deceased expanded criteria | 1565 | 9.9 | 1180 | 9.2 | 211 | 11.3 | 91 | 14.6 | 39 | 14.0 | 19 | 14.8 | 25 | 22.3 | |
| Donor diabetesa | 310 | <0.001 | |||||||||||||
| Yes | 452 | 2.88 | 298 | 2.61 | 80 | 3.33 | 37 | 4.49 | 20 | 5.26 | 4 | 2.44 | 13 | 8.39 | |
| Donor hypertensiona | 355 | <0.001 | |||||||||||||
| Yes | 2274 | 14.51 | 1539 | 13.51 | 407 | 16.94 | 164 | 19.88 | 82 | 21.75 | 39 | 23.78 | 43 | 27.56 |
| Transplant characteristics | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cold ischemic time (h) | 12.9 | 10.76 | 12.7 | 10.78 | 13.3 | 10.65 | 13.2 | 10.40 | 15.3 | 11.10 | 12.2 | 9.64 | 15.2 | 12.23 | <0.001 |
| Length of stay at transplant (d) | 7.1 | 11.89 | 6.9 | 12.46 | 7.5 | 10.07 | 8.2 | 7.54 | 8.2 | 6.76 | 8.8 | 6.24 | 8.7 | 6.84 | <0.001 |
| Peak PRA | 8.9 | 22.0 | 8.7 | 21.7 | 9.1 | 22.4 | 9.8 | 23.1 | 9.9 | 23.9 | 11.3 | 24.8 | 11.5 | 25.3 | 0.16 |
| N | % | N | % | N | % | N | % | N | % | N | % | N | % | ||
| PRA | 0.36 | ||||||||||||||
| 0 | 107,12 | 68.4 | 8048 | 68.7 | 1651 | 67.7 | 554 | 66.0 | 260 | 67.9 | 104 | 62.7 | 95 | 60.1 | |
| 1–20 | 3081 | 19.7 | 2268 | 19.4 | 493 | 20.2 | 169 | 20.1 | 73 | 19.1 | 36 | 21.7 | 42 | 26.6 | |
| 21–80 | 1235 | 7.9 | 907 | 7.8 | 189 | 7.8 | 76 | 9.1 | 30 | 7.8 | 19 | 11.5 | 14 | 8.9 | |
| 80+ | 643 | 4.1 | 464 | 4.0 | 105 | 4.3 | 40 | 4.8 | 20 | 5.2 | 7 | 4.2 | 7 | 4.4 | |
| HLA match levela | 53a | 0.05 | |||||||||||||
| Zero | 2759 | 17.6 | 2204 | 17.3 | 320 | 17.2 | 122 | 19.7 | 62 | 22.3 | 26 | 20.3 | 25 | 22.3 | |
| One | 4267 | 27.1 | 3409 | 26.8 | 543 | 29.2 | 171 | 27.5 | 86 | 30.9 | 30 | 23.4 | 28 | 25.0 | |
| Two | 2737 | 17.4 | 2200 | 17.3 | 336 | 18.1 | 110 | 17.7 | 45 | 16.2 | 27 | 21.1 | 19 | 17.0 | |
| Three | 3211 | 20.4 | 2620 | 20.6 | 362 | 19.5 | 134 | 21.6 | 37 | 13.3 | 31 | 24.2 | 27 | 24.1 | |
| Four | 1150 | 7.3 | 958 | 7.5 | 121 | 6.5 | 39 | 6.3 | 19 | 6.8 | 6 | 4.7 | 7 | 6.3 | |
| Five | 844 | 5.4 | 704 | 5.5 | 90 | 4.8 | 25 | 4.0 | 15 | 5.4 | 5 | 3.9 | 5 | 4.5 | |
| Six | 757 | 4.8 | 631 | 5.0 | 88 | 4.7 | 20 | 3.2 | 14 | 5.0 | 3 | 2.3 | 1 | 0.9 | |
| Complications | N | % | N | % | N | % | N | % | N | % | N | % | N | % | |
| Delayed graft function | <0.001 | ||||||||||||||
| Yes | 2534 | 14.3 | 1658 | 12.7 | 460 | 16.5 | 215 | 21.7 | 106 | 23.5 | 53 | 25.6 | 42 | 22.2 | |
| Acute rejection | 370a | <0.001 | |||||||||||||
| Yes | 816 | 5.3 | 478 | 4.1 | 167 | 7.0 | 77 | 9.4 | 40 | 10.3 | 20 | 11.9 | 34 | 15.7 |
BMI, body mass index; eGFR, estimated glomerular filtration rate; HLA, human leukocyte antigen; PRA, panel-reactive antibody.
Includes missing values.
Figure 3.
Mean estimated glomerular filtration rate (eGFR) at 6 months after renal transplantation by the total number of hospitalizations in 1 year after this visit among adult, first-time, kidney-only transplant recipients, United Network for Organ Sharing 2004–2006. Mean eGFR values are listed for each hospitalization number. The red line indicates the trend line for mean eGFR values from 0 to 5 or more late hospitalizations (P < 0.0001). The boxes represent the interquartile range of eGFR, and the horizontal line represents the median eGFR values.
In multivariable-adjusted analyses, a 10 ml/min per 1.73 m2 eGFR was significantly associated with the total number of late hospitalizations (Rate Ratio [RR]: 0.89, 95% confidence interval [CI]: 0.89–0.89) within the year after the 6-month visit; thus, for every 10-unit increase in eGFR, there was an 11% decrease in the rate of hospitalization (Table 3). After adjusting for other covariates, transplant recipients who were diabetic, who received dialysis before transplantation, or who experienced delayed graft function were at increased risk for hospitalization compared with those who were not. Patients who had been hospitalized in the first months after transplantation were more than twice as likely to experience later hospitalization (RR: 2.33, 95% CI: 2.20–2.48), and those who were on public insurance were hospitalized 12% more than those on private insurance (RR: 1.12, 95% CI: 1.09–1.37). Those who had only attended college and/or technical school or high school or less were also at increased risk for hospitalization compared with those with a post-college degree (RR: 1.16, CI: 1.01–1.33 and RR: 1.13, CI: 0.99–1.29, respectively) (Table 3). There were no differences by donor type (living vs. deceased standard vs. deceased expanded criteria) on the number of hospitalizations.
Table 3.
Multivariable zero-inflated Poisson modeling results for association of eGFR with number of hospitalizations among adult, first-time, kidney-only transplant recipients (United Network for Organ Sharing 2004–2007)
| Adj. rate ratio | 95% CI |
P value | ||
|---|---|---|---|---|
| Lower | Upper | |||
| eGFR (per 10 ml/min per 1.73 m2) | 0.89 | 0.89 | 0.89 | <0.0001 |
| Ethnicity | ||||
| White (reference) | 1 | – | – | – |
| Black | 1.03 | 0.97 | 1.11 | 0.33 |
| Hispanic | 0.93 | 0.85 | 1.01 | 0.09 |
| Asian | 0.93 | 0.81 | 1.07 | 0.31 |
| Time | 0.94 | 0.76 | 1.15 | 0.53 |
| Sex | ||||
| Male (reference) | 1 | – | – | – |
| Female | 1.09 | 1.03 | 1.15 | 0.004 |
| Recipient age (per 1 yr) | 1.00 | 0.99 | 1.00 | <0.0001 |
| Education level | ||||
| High school or less | 1.13 | 0.99 | 1.29 | 0.06 |
| Attended college or technical school | 1.16 | 1.01 | 1.33 | 0.04 |
| Associate/bachelor degree | 1.10 | 0.95 | 1.27 | 0.20 |
| Post-college graduate degree (reference) | 1.00 | – | – | – |
| Education level missing | 1.22 | 1.06 | 1.41 | <0.001 |
| Insurance | ||||
| Private insurance (reference) | 1.00 | – | – | – |
| Public insurance | 1.12 | 1.05 | 1.19 | 0.004 |
| Other | 1.12 | 0.54 | 2.33 | 0.75 |
| Length of stay (d) | 1.00 | 1.00 | 1.01 | <0.001 |
| HLA match level | ||||
| 0 | 1.22 | 1.05 | 1.42 | 0.01 |
| 1 | 1.08 | 0.93 | 1.25 | 0.31 |
| 2 | 1.15 | 0.99 | 1.34 | 0.08 |
| 3 | 1.18 | 1.01 | 1.37 | 0.03 |
| 4 | 1.08 | 0.90 | 1.28 | 0.41 |
| 5 | 1.04 | 0.86 | 1.25 | 0.67 |
| 6 (reference) | 1 | – | – | – |
| Diabetes | ||||
| No (reference) | 1 | – | – | – |
| Yes | 1.18 | 1.25 | 1.11 | <0.0001 |
| Pretransplant dialysis | ||||
| Yes (reference) | 1 | – | – | – |
| No | 0.87 | 0.80 | 0.94 | <0.001 |
| Missing | 0.45 | 0.34 | 0.61 | <0.0001 |
| Delayed graft function | ||||
| No (reference) | 1 | – | – | – |
| Yes | 1.19 | 1.27 | 1.11 | <0.0001 |
| Acute rejection episode | ||||
| No (reference) | 1 | – | – | – |
| Yes | 1.09 | 1.19 | 0.99 | 0.07 |
| Donor hypertension | ||||
| Yes (reference) | 1 | – | – | – |
| No | 0.81 | 0.64 | 1.03 | 0.08 |
| Missing | 0.92 | 0.85 | 0.98 | 0.02 |
| Previous hospitalizations (in first 6 mo) | ||||
| Yes (reference) | 1 | – | – | – |
| No | 0.43 | 0.45 | 0.40 | <0.0001 |
CI, confidence interval; eGFR, estimated glomerular filtration rate, HLA, human leukocyte antigen.
Sensitivity Analyses
When main multivariable modeling results were repeated using the chronic kidney disease-Epi equation instead of Modification of Diet in Renal Disease, results were similar to the main effect estimates (RR: 0.9877, 95% CI: 0.9857–0.9897). Similar results were also observed when we examined the association between eGFR and hospitalization in multivariable models for (i) including patients who died (RR: 0.9872, 95% CI: 0.9844–0.9900), (ii) retransplant patients (RR: 0.9865, 95% CI: 0.9839–0.9891), and (iii) multi-organ transplant recipients (RR: 0.9828, 95% CI: 0.9803–0.9854).
Discussion
In this national, longitudinal study of more than 15,000 kidney transplant recipients across the United States, we found that lower eGFR was associated with a higher hospitalization rate among adult, first-time kidney-only transplant recipients, where a 10-unit decrease in the 6-month eGFR was associated with an 11% increased risk of hospitalization in the year after the 6-month transplant follow-up visit. Additional risk factors for hospital readmission included female sex, lower educational attainment, diabetes, delayed graft function, previous hospitalization, and length of stay at transplant. To our knowledge, this is the first study that identified risk factors for hospital readmission at 6 months after transplantation among a population of both publicly and privately insured kidney transplant recipients. These findings have significant implications for the identification of patients at increased risk for late hospitalization—a proxy for transplant-related poor health outcomes—such as loss of graft function and mortality. As expected, some of these risk factors, such as delayed graft function, length of stay at transplant, and previous hospitalization, may reflect complications early on that lead to later hospitalizations. Our finding that patients with fewer HLA mismatches have lower hospitalization rates may have important clinical implications; further studies are needed to determine whether policies and practices to encourage greater HLA matching may reduce health care resource utilization.
Our results are consistent with other studies that identified a series of risk factors for early hospital readmission. McAdams-Demarco et al.7 found that among 32,961 Medicare primary kidney transplant recipients, 31% were readmitted within 30 days of initial discharge from the transplant center. In both this study and our study, risk factors for early hospital readmission included older age, various patient comorbidities, longer length of stay, and transplant characteristics such as increased HLA mismatches. However, unlike previous studies, we did not find that donor type impacted the risk of late hospitalization. We also identified the 6-month eGFR as a novel marker for late hospitalization. Information about both early and late hospitalization may be useful in risk stratifying patients at the time of either discharge after surgery or at the 6-month posttransplant visit to monitor more closely and potentially prevent poor adverse outcomes.
This study contributes to the growing body of research investigating the association of eGFR and transplant-related health outcomes, such as allograft function13, 14 and mortality.15, 16, 17 A meta-analysis of 105,872 participants from 14 studies found that eGFR < 60 ml/min per 1.73 m2 is an independent predictor of all-cause mortality in the general population.15 Furthermore, a retrospective study of 332 deceased donor kidney transplant recipients found that patients with kidneys from older donors or who experienced delayed graft function and acute rejection episodes were more likely to have deteriorating eGFR between 6 and 24 months after transplantation.16 A similar retrospective study of 428 kidney transplant recipients with a mean follow-up of 10 years found that changes in eGFR over the first year relates to poorer long-term renal outcomes.17
Risk factors for decreased posttransplant renal function include increasing donor age, receipt of a deceased donor kidney, incidence of acute rejection, and delayed graft function.18, 19 As posttransplant renal function is an independent risk factor for long-term renal graft function, increased clinical scrutiny of patients with low eGFR may help reduce hospitalizations, and both decrease the incidence of poor health outcomes after transplantation and reduce cost. Closely monitoring patients with low eGFR, specifically during the first year after transplantation, may prove especially beneficial, as adverse events during that time period have been shown to predict long-term graft survival.20 Because most transplant recipients attend a follow-up visit around 6 months posttransplant at their transplant center, identifying patients with declining eGFR at this time may allow risk stratification of patients and the potential to intervene to prevent future readmissions, graft failure, and mortality.
There were limitations to these analyses. Because hospitalization data were only collected during a limited time frame by UNOS (through 2007), we were unable to examine more recent data or hospitalizations over a longer time period. Furthermore, we could not assess early (30 days or less after surgery) versus late hospital readmissions or the length of each hospitalization because we did not have access to hospitalization date. We also did not have admission or discharge diagnosis, and therefore cannot determine whether patients were admitted to the hospital for reasons unrelated to their kidney transplantation (e.g., elective procedures, pregnancy-related admissions, etc.). There is also the possibility that hospitalizations may be underreported to UNOS. There may be a number of potential factors that also contribute to hospitalization risk that were not captured in our multivariable analyses, such as acute rejection, immunosuppression medication, or other comorbidities that were not measured or available in the UNOS Standard Transplant Analysis and Research file data. Nevertheless, an increasing number of hospitalizations between follow-up visits may still reflect poorer health status of the patient as well as increased unplanned health costs. Exclusion of patients who died or who had incomplete follow-up may have resulted in selection bias; however, these patients had similar characteristics compared with the study sample, and sensitivity analyses including these patients showed a similar effect between eGFR and increasing hospitalization risk. Differential access to medical treatment may result in bias when hospital data are used as proxy measures21; however, the vast majority of transplant patients in our sample were insured. Finally, some covariates included a high number of missing values; thus, for selected covariates, we included missing values as a category.
Despite these limitations, this research had numerous strengths. The use of a nationally representative surveillance database, containing information regarding every organ donation and transplant event in the United States, resulted in a large sample size with a comprehensive set of covariates. This large sample size allowed us to detect small differences in the rate of hospitalization. In addition, risk prediction models for hospitalization status based on eGFR can be used to identify and treat patients at risk. Finally, because these data are based on a nationally representative surveillance system, they can be generalized to the US population. Other research examining hospitalization data using Medicare claims data may not be more robust; however, they are limited to a Medicare-insured population only and thus are not generalizable to the US transplant population. This is the first study that we are aware of that has examined the association between eGFR and posttransplantation hospitalization among patients with both public and private insurance, and thus results are more generalizable to all transplant recipients.
Lower eGFR measured 6 months after transplantation was significantly associated with an increase in the number of hospitalizations over time—even when adjusting for a variety of covariates—in patients who survived past 6 months. It is unclear whether there is a clinically meaningful cut point for eGFR that would help predict negative health outcomes, and whether there is an optimal time frame after transplantation to screen for changes in eGFR. As such, the findings warrant further investigation into specific cut points for eGFR that can be used as a clinical predictor of future hospitalization among kidney transplant recipients. Prediction of such preventable hospitalizations and early intervention to prevent hospitalization based on serial eGFR measurement may have significant implications in cost reduction and the prevention of morbidity and mortality associated with hospitalization after transplantation.
Disclosure
All the authors declared no competing interests.
Acknowledgments
A portion of this work was presented at the American Transplant Congress in Seattle, WA, on 19 May 2013 (Abstract #B1089). The data reported here have been supplied by the United Network for Organ Sharing as the contractor for the Organ Procurement and Transplantation Network. The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by the OPTN or the US Government. REP and SOP are both supported in part by R24MD008077 and RMD010290A through the National Institute on Minority Health and Health Disparities.
Footnotes
Table S1. Additional baseline demographic characteristics by eGFR cut points and hospitalization for adult, first-time, kidney-only recipients (United Network for Organ Sharing, 2004-2007).
Supplementary material is linked to the online version of the paper at www.kireports.org.
Supplementary Material
Additional baseline demographic characteristics by eGFR cut points and hospitalization for adult, first-time, kidney-only recipients (United Network for Organ Sharing, 2004-2007).
References
- 1.Tonelli M., Wiebe N., Knoll G. Systematic review: kidney transplantation compared with dialysis in clinically relevant outcomes. Am J Transplant. 2011;11:2093–2109. doi: 10.1111/j.1600-6143.2011.03686.x. [DOI] [PubMed] [Google Scholar]
- 2.Salvadori M., Rosati A., Bock A. Estimated one-year glomerular filtration rate is the best predictor of long-term graft function following renal transplant. Transplantation. 2006;81:202–206. doi: 10.1097/01.tp.0000188135.04259.2e. [DOI] [PubMed] [Google Scholar]
- 3.Lenihan C.R., O'Kelly P., Mohan P. MDRD-estimated GFR at one year post-renal transplant is a predictor of long-term graft function. Ren Fail. 2008;30:345–352. doi: 10.1080/08860220801947686. [DOI] [PubMed] [Google Scholar]
- 4.Schnitzler M.A., Gheorghian A., Axelrod D. The cost implications of first anniversary renal function after living, standard criteria deceased and expanded criteria deceased donor kidney transplantation. J Med Econ. 2013;16:75–84. doi: 10.3111/13696998.2012.722571. [DOI] [PubMed] [Google Scholar]
- 5.Gheorghian A., Schnitzler M.A., Axelrod D.A. The implications of acute rejection and reduced allograft function on health care expenditures in contemporary US kidney transplantation. Transplantation. 2012;94:241–249. doi: 10.1097/TP.0b013e318255f839. [DOI] [PubMed] [Google Scholar]
- 6.Jencks S.F., Williams M.V., Coleman E.A. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med. 2009;360:1418–1428. doi: 10.1056/NEJMsa0803563. [DOI] [PubMed] [Google Scholar]
- 7.McAdams-Demarco M.A., Grams M.E., Hall E.C. Early hospital readmission after kidney transplantation: patient and center-level associations. Am J Transplant. 2012;12:3283–3288. doi: 10.1111/j.1600-6143.2012.04285.x. [DOI] [PubMed] [Google Scholar]
- 8.McAdams-DeMarco M.A., Grams M.E., King E. Sequelae of early hospital readmission after kidney transplantation. Am J Transplant. 2014;14:397–403. doi: 10.1111/ajt.12563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lum H.D., Studenski S.A., Degenholtz H.B., Hardy S.E. Early hospital readmission is a predictor of one-year mortality in community-dwelling older Medicare beneficiaries. J Gen Intern Med. 2012;27:1467–1474. doi: 10.1007/s11606-012-2116-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Levey A.S., Stevens L.A., Schmid C.H. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604–612. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. World Health Organization Technical Report Series No. 854. 1995:1–452. [PubMed]
- 12.Kleinbaum D.G., Kupper L.L., Nizam A., Muller K.E. 4th ed. Brooks/Cole; Australia/Belmont, CA: 2007. Applied Regression Analysis and Other Multivariable Methods. xxi, 906pp. [Google Scholar]
- 13.Schnitzler M.A., Lentine K.L., Axelrod D. Use of 12-month renal function and baseline clinical factors to predict long-term graft survival: application to BENEFIT and BENEFIT-EXT trials. Transplantation. 2012;93:172–181. doi: 10.1097/TP.0b013e31823ec02a. [DOI] [PubMed] [Google Scholar]
- 14.Levy A.R., Briggs A.H., Johnston K. Projecting long-term graft and patient survival after transplantation. Value Health. 2014;17:254–260. doi: 10.1016/j.jval.2014.01.001. [DOI] [PubMed] [Google Scholar]
- 15.Chronic Kidney Disease Prognosis Consortium. Matsushita K., van der Velde M., Astor B.C. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet. 2010;375:2073–2081. doi: 10.1016/S0140-6736(10)60674-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Magott-Procelewska M., Boratynska M., Janczak D. Estimated glomerular filtration rate evolution between 6 and 24 months predicts long-term kidney transplant survival among patients with inferior graft function. Transplant Proc. 2009;41:3028–3032. doi: 10.1016/j.transproceed.2009.07.105. [DOI] [PubMed] [Google Scholar]
- 17.Park J.S., Oh I.H., Lee C.H. The rate of decline of glomerular filtration rate is a predictor of long-term graft outcome after kidney transplantation. Transplant Proc. 2013;45:1438–1441. doi: 10.1016/j.transproceed.2012.10.052. [DOI] [PubMed] [Google Scholar]
- 18.Resende L., Guerra J., Santana A. First year renal function as a predictor of kidney allograft outcome. Transplant Proc. 2009;41:846–848. doi: 10.1016/j.transproceed.2009.01.066. [DOI] [PubMed] [Google Scholar]
- 19.Siddiqi N., McBride M.A., Hariharan S. Similar risk profiles for post-transplant renal dysfunction and long-term graft failure: UNOS/OPTN database analysis. Kidney Int. 2004;65:1906–1913. doi: 10.1111/j.1523-1755.2004.00589.x. [DOI] [PubMed] [Google Scholar]
- 20.Hariharan S., McBride M.A., Cherikh W.S. Post-transplant renal function in the first year predicts long-term kidney transplant survival. Kidney Int. 2002;62:311–318. doi: 10.1046/j.1523-1755.2002.00424.x. [DOI] [PubMed] [Google Scholar]
- 21.Tuchsen F., Andersen O., Olsen J. Referral bias among health workers in studies using hospitalization as a proxy measure of the underlying incidence rate. J Clin Epidemiol. 1996;49:791–794. doi: 10.1016/0895-4356(95)00554-4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Additional baseline demographic characteristics by eGFR cut points and hospitalization for adult, first-time, kidney-only recipients (United Network for Organ Sharing, 2004-2007).