In order to evaluate the influence of mode of delivery on perinatal morbidity and mortality in vertex infants weighing less than 1500 g (VLBW), we made a retrospective study of 152 singleton newborns, in vertex presentation, with a birthweight of less than 1500 g, delivered in the Cruces Hospital (Vizcaya, Spain), a major perinatal referral center, between 1 January 1987 and 31 December 1989. Twins and infants with lethal congenital anomalies or gross intrauterine growth deviations were excluded from the study (n = 71). Of the infants studied (n = 81), 37 were delivered by cesarean section (mean weight 1120 +/- 206 g, range: 680-1495 g) and 44 were delivered vaginally (mean weight 1029 +/- 283 g, range: 530-1475 g). The patients were divided into four groups: Group A: 500-749 g (n = 10); Group B: 750-999 g (n = 21); Group C: 1000-1249 g (n = 27); and Group D: 1250-1499 g (n = 23). The percentages of cesarean sections in each group were 10%, 42%, 66% and 39%, respectively. A comparison within each group of immediate perinatal outcome (Apgar score and umbilical vein cord pH), as well as mortality and sequelae up to 1 year of age did not yield any significant differences between cesarean and vaginal birth. We conclude that cesarean delivery does not appear to offer improved outcome over vaginal delivery in live births without congenital anomalies. For this reason, we believe that fetal weight should not be the only obstetrical variable considered when deciding whether or not to perform a cesarean section in these circumstances.