Abstract
BACKGROUND AND OBJECTIVES: The four- and six-variable Modified Diet in Renal Disease equations (MDRDEs) are empiric expressions that estimate GFR in patients with kidney disease. No method currently exists to directly compare MDRDE estimates of clearance by the failing biologic kidney to the clearances achieved by dialytic modalities. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: Because the MDRDEs estimate GFR, which in turn determines much of the small molecule clearance by the biologic kidney, the aim of this study was to determine if the MDRDEs would predict the measured creatinine clearances (including residual renal clearance, if any) from the serum creatinine values alone in 454 peritoneal dialysis patients. RESULTS: For the group, the four-variable MDRDE predicted a creatinine clearance of 6.5 +/- 3.7 ml/min/1.73 m(2) when measured clearance was from home collections of dialysate and urine was 6.8 +/- 3.1 ml/min/1.73 m(2). The presence or volume of residual urine had no effect on accuracy. The six-variable MDRDE was similarly accurate. Creatinine appearance rates were similar to those reported in other dialysis populations. CONCLUSIONS: Using serum creatinine values in "standard" fashion, MDRDEs can approximate creatinine clearances achieved by peritoneal dialysis with an accuracy similar to that of the MDRDEs in predialysis populations, perhaps because the MRDREs account exponentially for the increasing fractional gut metabolism of daily creatinine production as the serum creatinine increases. "MDRD-like" equations may provide a much needed method of directly comparing dialytic clearances to those achieved by diseased native kidneys and by kidney transplants.