&NA;To determine the efficacy of early jejunal hyperalimentation as nutritional support in the head‐injured patient, 32 head‐injured patients with Glasgow Coma Scale scores less than 10 were studied for the first 7 days after injury. The experimental (E) group had nasojejunal feeding tubes placed fluoroscopically. Within 36 hours of injury, they received nutritional support equal to their measured resting energy expenditure. The control (C) group was fed gastrically when bowel sounds returned. There were no significant differences (P> 0.05) in age, Glasgow Coma Scale score, type of neurological injury, or associated injuries between the two groups. The mean resting energy expenditure, serum albumin, glucose, lymphocyte count, body weight, and total nitrogen loss were nearly identical for both groups. With the jejunal feedings, daily caloric (E = 2102 kcal versus C = 1100 kcal) and nitrogen intake (E = 11.1 g versus C = 5.6 g) and daily nitrogen balance (E = ‐4.3 g versus C = ‐11,8 g) improved. The incidence of bacterial infections (E = 3 versus C = 14) and days of intensive care unit hospitalization (E = 6 versus C = 10) were significantly reduced (P< .05). Headinjured patients will tolerate early jejunal hyperalimentation despite the presence of a clinically silent abdomen, and the cost and complications of total parenteral nutrition are avoided. The increased caloric and nitrogen intake and improved nitrogen retention markedly reduced infections and days of stay in the intensive care unit. (Neurosurgery25:729‐735, 1989)