Objective: To compare three low-lactose milk formulas differing in osmolality and degree of protein hydrolysis in the treatment of diarrhoea and malnutrition in subjects with high rates of lactose intolerance, osmotic diarrhoea and a tropical/environmental enteropathy.

Methods: A randomized double-blind trial of 180 Aboriginal children under 3 years of age admitted with acute diarrhoea and/or malnutrition was carried out. The intervention milk formulas were: (i) De-Lact, a low-osmolality lactose-free formula; (ii) O-Lac, a lactose-free formula; and (iii) Alfaré, a partially hydrolysed formula. Outcome measures were diarrhoeal severity, weight gain, formula palatability and changes in intestinal permeability (L/R ratios).

Results: The duration of diarrhoea in days (mean; 95% confidence interval) was significantly longer on Alfaré (8.5; 7.0-10.0) compared to De-Lact (6.1; 5.0-7.2) and O-Lac (6.9; 5.6-8.1; P = 0.04). There were no differences in mean intake between formulas, but palatability of Alfaré was significantly worse (P < 0.01) than the other formulas. Over the trial 5 days, improvement in L/R ratios was significantly greater (P = 0.05) for De-Lact (18.6; 10.6-26.6) than for Alfaré (8.5; 2.1-14.9). Weight gain was not significantly different between the three formulas, except in a malnourished subgroup who had better weight gain on De-Lact (P = 0.05).

Conclusions: In these Aboriginal children with diarrhoea and growth failure, a low osmolality milk was associated with better outcomes and a partially hydrolysed formula with less improvement in mucosal recovery, suggesting that cow's milk protein intolerance is not contributing to greater diarrhoeal severity or enteropathy in Aboriginal children.