Predictive Value of Bedside Effective Renal Plasma Flow for Renal Recovery in Severe Acute Renal Failure

Abstract

Single injection, single blood sample, effective renal plasma flow (ERPF) estimated by^l3lI-orthoiodohippurate can be performed accurately and conveniently without urine collection at the bedside. The purpose of this study was to determine if ERPF early in the course of severe acute renal failure (ARF) predicts recovery of renal function in hemodynamically stable patients. Over 18 months, ERPF was determined in 33 such patients with ARF in whom an etiologic diagnosis could be established. Eight patients died within 2 months of onset and while on dialysis, did not have an autopsy, and were not considered further. Six patients (Group A) either remained on dialysis after at least 6 months follow-up or had irreversible renal disease at autopsy. In Group B (19 patients, 13 of whom were dialyzed), serum creatinine returned to less than 2.0 mg/dL (n = 16) or was decreasing without dialysis. Peak serum creatinine (Group A 11.2 ± 1.4; Group B 10.1 ± 1.3 mg/dL) did not differ between groups. Oliguria was present in 75% of Group A patients and in 25% of Group B patients. Initial ERPFs differed (p < 0.001) between Group A (90 ± 11) and Group B (204 ± 20 mL/min). Initial ERPF was greater than 125 mL/min in 15 Group B patients but in no Group A patients; the false-positive rate was 21% and the false-negative rate was 0%. We conclude that at a time when the etiology of ARF is often not established, an initial ERPF of 125 mL/min or greater predicts recovery of renal function and less than 125 mL/min suggests that renal function will not recover. Serial studies improve the diagnostic accuracy of this test.