SummaryA controlled therapeutic trial was performed in newborns with birthweight of 750‐1 250 g. 40 infants (the “treatment” group) received an i.v. infusion with glucose and NaHCO3 from the 1st to the 4th‐7th day of life, and increasing amounts of human milk from the 2nd day of life. They also received, during the newborn period, 3 antibiotics (sodium penicillin G 500 000 U + methicillin 100 mg + colistin sulf. 8 mg per kg/day), a gamma globulin preparation (0.5 ml/day for 3 days) and fresh human plasma (3 ml every 2nd day in the first 5 days of birth). 40 infants (the “control” group) received only oral feeding with 10% glucose and human milk, starting in the 2nd day of life, and i.m. kanamycin (20 mg/kg/day for 5 days).No difference between treated and control babies was observed with respect to the following findings: neonatal mortality rate and survival curve on the whole series; neonatal mortality rate in babies with birthweight above or below 1.04 kg, with gestational age below 28 weeks, with above or below median respiratory rate or arterial pH on admission, with or without marked abnormality on chest film, and in males or females (but, in subjects with gestational age above 27 weeks, a significantly lower mortality rate was found in controls); postmortem findings; incidence of apnoeic spells, tremors and cloni, and pattern of muscular tonus. Bradycardia associated with 2: 1 atrioventricular block was observed in 6 controls in the 2nd day of birth. No adverse effects presumably due to the therapeutic procedures so contrasted were demonstrated.It was concluded that in the present series of newborns with very low weight: (a) the massive anti‐infectious therapy was not superior to the administration of a single broadspectrum antibiotic; (b) the administration of fresh human plasma was not effective in preventing hemorrhage (but further evaluation is needed); (c) the routine i.v. infusion with glucose and NaHCO3 did not improve neonatal survival as compared with oral feeding.However, previous studies, as well as observations on the present series reported elsewhere (11), have shown that several biochemical abnormalities commonly seen in orally fed newborns with very low weight can be prevented or corrected, at least in part, by the early infusion with glucose and NaHCO3. Since some of these abnormalities are potentially dangerous to the central nervous system, the final evaluation of this therapy must await long‐term follow‐up studies in survivors. It is therefore suggested that, before such studies are available, the early i.v. infusion with glucose and NaHCO3 should be given routinely to newborns with very low weight.