Analysis of reciprocal creatinine plots in renal failure

The linearity that has frequently been observed when the reciprocal of the serum creatinine is plotted against time of follow-up has been used to predict the future course of patients with renal disease. In the present study, we have analyzed reciprocal creatinine plots with a computer program that identifies breakpoints in a linear plot and calculates the statistical significance of the fit provided by two intersecting lines instead of a single straight line. When applied to cases of chronic renal failure, the program revealed breakpoints that were not temporally related to recognized major clinical events. The computer program may prove useful for evaluating the validity of data derived from clinical trials of therapies to moderate the course of declining kidney function.