Breast-fed infants have significantly more stools per day than formula-fed ones and only a few have hard stools (1.1% of exclusively breast-fed infants vs. 9.2% of formula-fed infants [20, 21]). The frequency of painful defecation in constipated children varies from 89%  to 40-50% [23, 24] depending on the series reported. Fewer bowel movements due to prolonged transit time may result in distension of the gut, irritability and crying, and this may be a reason for inclusion in the associated group of infantile colic [2–4]. Tunc et al. report that, in the first 2 months of life, the median number of stools is lower in colicky infants than in those who do not have colic (p = 0.0001) .
Several hypotheses have been proposed to explain the softer stools associated with breast feeding. First, increased levels of gastric inhibitory polypeptide, motilin, neurotensin, vasoactive intestinal peptide secretions in formula-fed infants compared to breast-fed infants may explain their slower intestinal transit . Second, more frequent breast feedings may stimulate the gastro-colic reflex resulting in more frequent defecation. Third, human milk contains large amounts of prebiotic oligosaccharides. Fourth, fat composition may play a role in softer stools. In breast milk and in most infant formulae, fats represent nearly 50% of energy content. The main saturated fatty acid in breast milk is palmitic acid (C16:0), with more than 70% esterified to the sn-2 position, whereas in regular infant formulae 88-94% of the palmitic acid is esterified to the sn-1,3 position. Lipolysis at the sn-1 and sn-3 positions requires pancreatic lipase, which is deficient in the first 6 months of life. The result is relative fat malabsorption, which may react with luminal calcium to form calcium soaps producing hard stools [25, 26].
Previous NIRA studies have shown lower stool fat in breast-fed infants: 2.6% in breast-fed infants versus 7-11% in formula-fed infants younger than 6 months and less than 5% in formula-fed infants older than 6 months [9, 10]. Our study confirmed the relative steatorrhoea in formula-fed infants [18, 19]. Feeding a formula containing a high concentration of sn-2 palmitic acid caused softer stools but no change in stool frequency in one study  and decreased fatty acid soaps in another study . We saw no change in stool fat content in the current study, suggesting that the improvement in constipation was not related to changes in fat metabolism.
A clinical trial by Chao et al. of 93 infants (average age 3.8 months) showed that the same formula we used in the current study (Novalac AE/IT) improved infant constipation. The infants receiving this formula had increased stool weight and decreased abdominal distension and irritability compared to infants receiving placebo . Our study adds to these results and includes objective NIRA data.
The two additives to the adapted formula likely impact stool water content through different mechanisms. Lactose has prebiotic effects in that it is not completely absorbed in young infants and stimulates the growth of commensal bacteria. Non-hydrolyzed lactose reaches the colon, where it is metabolized by anaerobic microorganisms, producing an osmotic laxative effect as it attracts water into the intestinal lumen [29–32]. Magnesium, due to its osmotic properties, increases the laxative effect and stimulates intestinal motility by inducing cholecystokinin secretion .
In conclusion, this non-randomized, non-placebo-controlled preliminary study demonstrates that the use of an infant formula with high lactose and magnesium concentrations may increase stool hydration, which softens the stools with a corresponding clinical improvement. Thus, constipated infants who present hard stools may benefit from a change from standard formula to this specifically adapted formula. Larger randomized clinical trials on the efficacy of this formula are needed.