Objectives: This study was undertaken to assess the value of Cockroft-Gault Formula (FCG) in the patients of an Internal Medicine ward.
Methods: FCG was compared with creatinine clearance (DC) adjusted for corporal area. Serum creatinine was measured after 24 h urine collection and 8 h fast. By the subtraction of FCG from DC, the variable difference (DIF) was calculated. FCG was calculated using total body weight (PT) and lean body weight (PS). All patients were included unless they had renal failure.
Results: Renal function was calculated in 92 patients, 33 women and 59 men. Mean age was 60.8 +/- 17.9 (13-87). Mean FCG (64.7 +/- 27.4) and mean DC (68.4 +/- 32.6) were not significantly different and a correlation between them was found (r = 0.68; p < 0.001). Using PS to estimate FCG the correlation coefficient was increased to 0.75 (p < 0.001). DC and FCG using PT or PS correlated inversely with age (r = 0.59; r = 0.53; r = 0.66, respectively; p < 00.1). In a cluster analysis model we could find that for extremes of renal function FCG overestimated (DIF = -10.4 +/- 13.3) or underestimated (DIF = 29.8 +/- 29.7) DC. The main influence for the difference between the two methods was body weight (r = -0.69; p < 0.001). The distribution by diagnosis led to two classes of patients for whom DIF was significantly different (p < 0.02), and body weight was the main factor.
Conclusion: Cockroft-Gault formula, although correlating with creatinine clearance, over or underestimates creatinine clearance for extremes of renal function. Age and body weight were the main factors for the differences found between the two methods.