Objectives: To demonstrate the effects of amputation on the estimates of urea volume of distribution (V) and KT/V urea in continuous ambulatory peritoneal dialysis (CAPD) patients and to present a method for correcting the errors created by the uncorrected anthropometric formulas estimating V.
Design: (1) A mathematical analysis of the error and the correction proposed was performed. (2) Urea kinetic modeling with uncorrected and corrected estimates utilizing both the Watson and the Hume anthropometric formulas was performed in amputees on CAPD.
Setting: Subjects were recruited from four dialysis units in one city: one Veterans Affairs unit, one university-affiliated unit, and two community units.
Patients: Fourteen amputees on CAPD:12 with unilateral leg amputation and 2 with bilateral leg amputation, at the same length of the leg, were studied.
Interventions: Urea kinetic studies were performed in 24-hour drained dialysate and urine specimens.
Main outcome measures: Uncorrected and corrected estimates of V and KT/V urea were compared to each other and to the predictions of the mathematical model. Body weights corresponding to uncorrected and corrected V estimates were compared to the actual body weights.
Results: (1) The mathematical model predicts that uncorrected estimates by the anthropometric formulas will falsely characterize unilateral amputees as leaner than they are and bilateral amputees as more obese than they are. (2) In unilateral amputees studied with the Watson formulas, uncorrected V was 0.546 +/- 0.023 L/kg and corrected V was 0.520 +/- 0.023 L/kg (p < 0.001). Corresponding weekly KT/V urea values were 1.97 +/- 0.14 and 2.07 +/- 0.14, respectively (p < 0.001). Similar results were obtained with the Hume formulas. In bilateral leg amputees studied with the Watson formulas, uncorrected V was 0.479 +/- 0.022 L/kg and corrected V was 0.514 +/- 0.023 L/kg. Corresponding KT/V estimates were 2.11 +/- 0.45 and 1.96 +/- 0.14, respectively. The differences were even greater with the Hume formulas. Estimates of body weight calculated from corrected V values were equal to actual weight measurements, whereas those calculated from uncorrected V values were lower than actual body weight measurements in unilateral amputees, and much higher than actual body weight measurements in bilateral amputees.
Conclusion: Uncorrected anthropometric estimates falsely characterize unilateral amputees as leaner than they actually are and bilateral amputees, amputated at the same leg length, as more obese than they actually are. Uncorrected KT/V estimates are, therefore, falsely low in unilateral amputees, and falsely high in bilateral amputees. The proposed correction of the anthropometric formulas provides estimates agreeing closely with dietary estimates of body composition. Further studies are needed to define the accuracy of the corrected formulas.