CATABOLISM IN CRITICAL ILLNESS - ESTIMATION FROM UREA NITROGEN APPEARANCE AND CREATININE PRODUCTION DURING CONTINUOUS RENAL REPLACEMENT THERAPY

Thirty-eight intensive care unit (ICU) patients (26 men and 12 women w ith a mean age of 57.0 +/- 16.6 years) with acute renal failure (ARF) treated by venovenous continuous renal replacement therapy (CRRT) were evaluated while in relatively steady metabolic control. Twenty-seven were undergoing continuous venovenous hemodialysis, nine were undergoi ng continuous venovenous hemodiafiltration, and two were undergoing co ntinuous venovenous hemofiltration. Periods of analysis varied between 24 and 408 hours (mean duration, 82.7 +/- 70.6 hours; median, 72 hour s). Their mean Acute Physiology and Chronic Health Evaluation II (APAC HE II) score within 24 hours of admission to the ICU was 21.3 +/- 6.3 and survival rate was 31.6%. Urea nitrogen and creatinine concentratio ns were determined every 6 to 12 hours in both serum (Cun and Cc, resp ectively) and effluent (spent dialysate and/or ultrafiltrate). The mea n effluent rate was 1,472 +/- 580 mL/h and blood flow rate, 166 +/- 32 mL/min. Urine was collected daily for urea nitrogen and creatinine me asurement. Urea nitrogen appearance rate (UnA) and creatinine producti on rate (Pc), calculated from urea nitrogen (UnMR) and creatinine mass removal (CMR) from both the effluent and the urine, using Garred mass balance equations and the Forbes-Bruining formula, allowed normalized protein catabolic rate (nPCR) and estimates of lean body mass (LBM) t o be derived. Creatinine metabolic degradation rate (Dc), estimated by the Mitch formula, was included in the calculation. The lowest body w eight recorded during the study period was considered as dry weight (B W). The creatinine index (CI) was also obtained. For each parameter, t he results are presented as mean, median, and range values: UnMRe (fro m effluent), 13.6 +/- 7.2, 12.5, 1.6 to 32.6 mg/min; UnMRu (from urine ), 0.13 +/- 0.40, 0, 0 to 2.30 mg/min; UnA, 13.6 +/- 7.0, 12.5, 3.8 to 32.1 mg/min; nPCR, 1.75 +/- 0.82, 1.60, 0.61 to 4.23 g/kg/d; CMRe (fr om effluent), 942.0 +/- 362.3, 918.0, 211.2 to 1,641.6 mg/d; CMRu (fro m urine), 44.4 +/- 138.8, 0, 0 to 698.5 mg/d; Dc, 94.6 +/- 49.9, 81.9, 31.0 to 294.1 mg/d; Pc total, 1,067.1 +/- 409.7, 1,053.7, 261.5 to 1, 988.2 mg/d; LBM, 38.3 +/- 11.9, 37.9, 15.0 to 65.0 kg; LBM/BW ratio, 4 9.5% +/- 14.0%, 50.3%, 22.5% to 86.0%; and CI, 13.7 +/- 4.7, 14.2, 4.1 to 25.8 mg/kg/d. When Pc was estimated from the Cockcroft-Gault equat ions (as Pc'), the mean value for Pc and Pc' was similar (1,067.1 +/- 409.7 v 1,284.9 +/- 484.1 mg/d), but there were relatively large diffe rences for the majority of cases. A positive correlation was observed between UnA and Pc (R = 0.42). Serum albumin and LBM/BW correlated poo rly (R = 0.20). Outcome was weakly related to UnA and to nPCR (R = 0.2 9 and R = 0.31, respectively). Urea nitrogen appearance appears widely variable in critically ill ARF patients. This simple approach can pro vide useful information for an easy estimate of net protein catabolism in critically ill patients with ARF undergoing CRRT. (C) 1998 by the National Kidney Foundation, Inc.