Abstract
Limited attention is given to the efficacy of protocols for the estimation of infant intake of milk components when investigating their impact on infant outcomes. We compared the actual measured intake of human milk components with estimations derived from 15 protocols to determine the most reliable approach for estimating intake of HM leptin, adiponectin, insulin, glucose, and total lipid. Twenty mothers who were 3-5 months postpartum completed a 24 h milk profile study with pre-/post-feed milk samples collection. The true infant intake (control group) based on 24 h milk intake (MI) was compared to estimated infant intakes using concentrations from five sampling protocols that were multiplied by one of true infant MI, considered mean MI (800 mL), or global mean MI (766 mL). The mean measured concentrations of six samples (three sets of pre- and post-feed samples, from morning (06:00-09:00), afternoon (13:00-16:00), and evening (19:00-22:00)) multiplied by the true infant MI, mean considered MI, and global mean MI produced the most accurate estimates of infant intake of these components. Therefore, in the absence of 24 h measurements and sampling, a sampling protocol comprising three sets of pre-/post-feed samples provides the most reliable infant intake estimates of HM leptin, adiponectin, insulin, glucose, and total lipid.