Kidney Transplantation Outcomes across GN Subtypes in the United States

Abstract

Differences in kidney transplantation outcomes across GN subtypes have rarely been studied. From the US Renal Data System, we identified all adult (≥18 years) first kidney transplant recipients (1996-2011) with ESRD attributed to one of six GN subtypes or two comparator kidney diseases. We computed hazard ratios (HRs) for death, all-cause allograft failure, and allograft failure excluding death as a cause (competing risks framework) using Cox proportional hazards regression. Among the 32,131 patients with GN studied, patients with IgA nephropathy (IgAN) had the lowest mortality rates and patients with IgAN or vasculitis had the lowest allograft failure rates. After adjusting for patient- and transplant-related factors, compared with IgAN (referent), FSGS, membranous nephropathy, membranoproliferative GN, lupus nephritis, and vasculitis associated with HRs (95% confidence intervals) for death of 1.57 (1.43 to 1.72), 1.52 (1.34 to 1.72), 1.76 (1.55 to 2.01), 1.82 (1.63 to 2.02), and 1.56 (1.34 to 1.81), respectively, and with HRs for allograft failure excluding death as a cause of 1.20 (1.12 to 1.28), 1.27 (1.14 to 1.41), 1.50 (1.36 to 1.66), 1.11 (1.02 to 1.20), and 0.94 (0.81 to 1.09), respectively. Considering external comparator groups, and comparing with IgAN, autosomal dominant polycystic kidney disease (ADPKD) and diabetic nephropathy associated with higher HRs for mortality [1.22 (1.12 to 1.34) and 2.57 (2.35 to 2.82), respectively], but ADPKD associated with a lower HR for allograft failure excluding death as a cause [0.85 (0.79 to 0.91)]. Reasons for differential outcomes by GN subtype and cause of ESRD should be examined in future research.

Keywords: chronic allograft failure; clinical epidemiology; glomerulonephritis; kidney transplantation; mortality.

Figure 1.

Assembly of the final study cohort of patients with ESRD due to GN, diabetic nephropathy, or autosomal dominant polycystic kidney disease (ADPKD).

Figure 2.

Variation in the rates and causes of allograft failure by cause of ESRD. Cumulative incidence plots of the competing events of death and allograft failure excluding death as a cause. “Graft failure” represents allograft failure due to causes other than death. “Death” represents allograft failure due to death. The cumulative incidence of “Graft failure” and “Death” combined represents the total incidence of allograft failure due to any cause. DN, diabetic nephropathy; TX, transplantation.

Figure 3.

Increased hazards for death and for allograft failure (including or excluding death as a cause) for comparator causes of ESRD compared to IgA nephropathy (IgAN). Model 4 (fully adjusted) HRs with 95% CIs for the primary outcomes of death, all-cause allograft failure, and allograft failure excluding death as a cause (competing risks model), by cause of ESRD, n=107,778. Model 4 stratified by year of transplantation and adjusted for age, age*age, sex, race, ethnicity, geographic region, insurance type, college education, dialysis modality, dialysis vintage, comorbidities (unable to ambulate, coronary heart disease, cancer, congestive heart failure, COPD, CVA/TIA, diabetes, hypertension, current/recent smoker, PVD), BMI group, HCV status, ABO blood group, CIT, donor age, donor sex, donor race, HLA mismatch group, donor type (living/decreased/expanded criteria), PPRA, initial post-transplant immunosuppression (Alemtuzumab, Basiliximab, Daclizumab, or Thymoglobulin induction; Tacrolimus, Ciclosporin, Sirolimus, MMF, Azathioprine and/or steroid maintenance), previous blood transfusion, and DGF. BMI, body mass index; CIT, cold ischemia time; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DN, diabetic nephropathy; HLA, histocompatibility leukocyte antigen; MMF, mycophenolate mofetil; PPRA, peak panel reactive antibody; PVD, peripheral vascular disease; TIA, transient ischemic attack.

Figure 4.

Risk ratios with 95% CIs for the secondary outcome of DGF, showing attenuation of effect estimates with sequential addition of covariates to multivariate Poisson models (n=107,272, after excluding n=560 with missing outcome). Model 1, adjusted for year of transplantation. Model 2, added sociodemographic variables: age, age*age, sex, race, ethnicity, geographic region, insurance type, college. Model 3, added dialysis modality, dialysis vintage, comorbidities (unable to ambulate, coronary heart disease, cancer, congestive heart failure, COPD, CVA/TIA, diabetes, hypertension, current/recent smoker, PVD), BMI group, HCV status. Model 4, added transplant-related variables: ABO blood group, CIT, donor age, donor sex, donor race, HLA mismatch group, donor type (living/decreased/expanded criteria), PPRA, initial post-transplant immunosuppression (Alemtuzumab, Basiliximab, Daclizumab, or Thymoglobulin induction; Tacrolimus, Ciclosporin, Sirolimus, MMF, Azathioprine and/or steroid maintenance), previous blood transfusion. BMI, body mass index; CIT, cold ischemia time; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DN, diabetic nephropathy; HLA, histocompatibility leukocyte antigen; MMF, mycophenolate mofetil; PPRA, peak panel reactive antibody; PVD, peripheral vascular disease; TIA, transient ischemic attack.