Trends in mortality and graft failure for renal transplant patients

Abstract

Background: Several important advances in general medical management both before and after renal transplantation have occurred over the last 5-15 years, however, few studies have formally examined trends in the outcomes of renal transplantation. We, therefore, aimed to determine the degree to which these advances have resulted in improved outcomes such as survival of patient and graft.

Methods: We analyzed the rates of death and graft failure among the 11,482 Canadians with end-stage renal disease who received a kidney transplant in 1981-98. Patients were followed from the date of transplantation to the date of graft failure, the date of death or the end of the observation period, namely, Dec. 31, 1998, depending on which was the earliest. Rate ratios for mortality and graft failure--ratios of the rate for each calendar period to the rate for the arbitrarily chosen reference period, 1981-85--were estimated with a piece-wise exponential model that adjusted for age, sex, ethnicity, primary renal diagnosis, follow-up time and donor-organ source.

Results: The rates and adjusted rate ratios for death and graft failure decreased significantly and steadily over time. Relative to 1981-85, the adjusted mortality rate ratios were 0.70 (95% confidence interval [CI] 0.54-0.89), 0.65 (95% CI 0.52-0.82) and 0.53 (95% CI 0.41-0.67) for 1986-89, 1990-94 and 1995-98 respectively, and the adjusted graft failure rate ratios were 0.68 (95% CI 0.60-0.78), 0.62 (95% CI 0.54-0.70) and 0.51 (95% CI 0.44-0.58) respectively. The decrease was mostly among the cadaveric-organ recipients. Calendar period was as important a predictor of outcome as well-known prognostic factors such as age and primary renal diagnosis.

Interpretation: Decreases in mortality rates are probably related to refinements in patient management. Decreases in graft failure rates are probably the result of a combination of improved immunotherapy and better management of nonimmunologic conditions such as hypertension and hyperlipidemia.

Fig. 1: Fitted 5-year survival probability by calendar period for patients (top panel) and grafts (bottom panel), as predicted for reference Canadian renal transplant patients (women, white, aged 45–64 years, with glomerulonephritis as the primary renal diagnosis and a cadaveric-donor organ).

Fig. 2: Covariate-adjusted rate ratio (RR) — the ratio of each calendar period–specific rate to the rate for the reference period, 1981–85 — and 95% confidence interval (CI) by calendar period for mortality (top panel) and graft failure (bottom panel) in the reference patients.