Patterns and predictors of kidney function decline in the last year prior to dialysis

Establishing a comprehensive characterization of kidney function decline before dialysis is necessary to predict dialysis onset and prepare patients for replacement therapy.

To investigate kidney function as measured by pattern and rate of decline in glomerular filtration rate (GFR) over the year preceding dialysis and to identify factors associated with a nonlinear GFR decline.

We enrolled patients beginning dialysis in Lorraine (France) in 2005 and 2006, who were referred to a nephrologist more than 4 months before dialysis and had received more than 3 predialysis serum creatinine tests. From medical records, we retrospectively collected demographic and clinical data, as well as biological data during nephrologist follow-up, limited to 1 year before dialysis. A curve of GFR evolution by time was drawn for each patient and his linearity was evaluated graphically and confirmed by R2 > 0.7. Factors associated with a nonlinear decline in GFR were identified by logistic regression.

A total of 342 patients were included; the mean length of predialysis nephrologist care was 10.0 +/- 9.7 months and the median number of serum creatinine tests per patient was 9 . Among these patients, 185 (54.1%) showed a linear decline in GFR and 157 (45.9%) a nonlinear decline. Patients with cardiovascular disease were 2.6 times more likely to show a nonlinear than linear decline in GFR (p < 0.0001).

For patients with a linear decline in GFR, but not those with a nonlinear decline, date of dialysis onset can be estimated.