protein intake was increased from 50 to 60 g/day to approximately 100 g/day. Increasing protein above 100 g/day caused no further change in milk volume. In both studies, however, the samples were very small (N = 6 and 8, respectively), there were no control groups, and their designs may have biased the results. In the Indian study, protein intake was increased in stages from 61 to 99 to 114 g/day in three successive 10-day periods, and milk samples were obtained by complete expression of both breasts at the 10 a.m. feeding during the last 3 days of each period. Because milk intake may increase with the age of the infants and repeated expression of extra milk may stimulate greater milk production (Dewey and Lönnerdal, 1986), milk volumes may have increased during the study for these reasons rather than in response to increased maternal protein intake. In the Nigerian study, total milk volume was calculated by measuring infant intake and by pumping residual milk after each feeding for each of the 7-day measurement periods. The extra stimulation provided by removing residual milk may have increased milk production.
In two small-scale studies of well-nourished women whose usual protein intake ranged from approximately 80 to 100 g/day, no statistically significant changes in milk volume were observed when protein intake was reduced to 8% of total calories or increased to 20% of total calories for 4 days (Forsum and Lönnerdal, 1980) or when daily protein intake was either 1.0 or 1.5 g/kg of body weight for 7 to 10 days (Motil et al., 1986). However, the very small sample sizes (N = 3 and 15, respectively) and short duration of these studies preclude any definitive conclusions regarding the impact of varying protein intake.
It is widely assumed that milk production requires a high fluid intake on the part of the mother, yet the evidence suggests that lactating women can tolerate a considerable amount of water restriction and that supplemental fluids have little effect on milk volume. Lactating women who consumed no food or fluids from 5:00 a.m. to 7:30 p.m. during Ramadan lost 7.6% of their total body water and experienced increases in serum indices of dehydration, although values remained within the normal range (Prentice et al., 1984). Milk volume was unaffected, but changes in milk composition (lower lactose concentrations; increased osmolality due to higher electrolyte concentrations) indicated alterations in mammary cell permeability. Water turnover was very high, in part because the women apparently superhydrated themselves overnight prior to the fasting period.
Two early studies from Germany (Olsen, 1941) and France (Lelong et al., 1949) also showed no influence on milk output when fluid intake ranged from 600 to 2,775 ml/day during 3- to 4-day periods or when total fluid intake from all sources was restricted to 1,765 ml/day for 10 days. Dusdieker et al. (1985) examined the effect produced by increasing fluid intake by at least 25% for 3