A prospective study was performed to investigate the validity of renal failure index (RFI) or fractional excretion of sodium (FENa) in preterm infants. The subjects were 128 newborn infants, 72 with oliguria and 56 without renal dysfunction (control). Oliguric infants were divided into two categories: acute renal failure (ARF) and prerenal failure (PRF), according to creatinine clearance (Ccr). Furthermore, all subjects were divided into four groups according to gestation, that is, 38 infants with gestational age of 25-28 weeks (group 1), 28 with 29-30 weeks (group 2), 38 with 31-36 weeks (group 3) and 24 of > 37 weeks (group 4). As a result, differentiation between ARF and PRF was valid when the RFI or FENa was used in infants of > 29 weeks gestation (groups 2, 3 and 4). Although infants of > 31 weeks gestation (groups 3 and 4) who present with an RFI > 3 or an FENa > 3% may be diagnosed as having ARF, infants in group 2 with an RFI of > 8 or an FENa of > 6% may be diagnosed as having ARF. For the infants in group 1, the application of RFI or FENa for diagnosis of ARF may be limited because of some overlap among the groups.