Background: Estimated glomerular filtration rate (eGFR) is widely used in follow-up and assessment of patients before start of chronic renal replacement therapy (RRT). Reported data on impact of eGFR decline pattern during pre-dialysis phase on consequent survival on RRT are, however, non-existent.
Methods: Using the database of the Finnish Registry for Kidney Diseases, we conducted a cohort study of all incident adult patients (n = 457) entering chronic RRT in Finland in 1998, with follow-up until 31 December 2008. We included those (n = 319) with three serum creatinine measurements (at ∼12 and 3 months and 1 to 2 weeks prior to RRT start) and calculated their slopes of eGFR using the modification of diet in renal disease formula. According to eGFR slopes (in mL/min/1.73m(2)/year), patients were divided into tertiles: most rapid (>8.5, n = 107), intermediate (3.4-8.5, n = 107) and slowest decline (<3.4, n = 105).
Results: Median survival time was 5.6 (95% confidence interval 4.2-7.0) years. Compared to the patient group with the slowest eGFR decline, age- and gender-adjusted relative risk of death was 1.1 (0.8-1.5) in the intermediate group and 1.7 (1.2-2.4, P = 0.002) in the most rapid decline group. When further adjusting for kidney disease diagnosis, comorbidities, use of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, body mass index, blood haemoglobin and serum albumin, the association was no longer significant.
Conclusions: Rapid decline in eGFR before entering chronic RRT associates with increased mortality on RRT. The elevated mortality appears to be caused by known risk factors for death on RRT.