Maillard reaction and lactose isomerization may be induced during the processing involved in the manufacture of infant formulas. The effects of dehydratation and sterilization in an infant formula on iron and zinc bioavailability were studied. A powder (PIF), previously reconstituted, and an in-bottle-sterilized liquid infant formula (LIF), from the same manufacturer, were evaluated using an in vitro method and in suckling rats. After in vitro digestion the dialyzed and non-dialyzed soluble, and insoluble fractions of iron and zinc were separated. Two-week-old rat pups were fed PIF or LIF in a drinking bottle for 7 days. Infant formula intake (I), body weight and the fecal and urinary excretions were monitored and the following parameters calculated: apparent absorption (A), retention (R), and the coefficients %A/I, %R/A and %R/I. Soluble iron (dialyzed) and zinc (non-dialyzed) were higher (p < 0.001) in LIF than PIF after in vitro digestion. Insoluble iron was similar in both infant formulas but insoluble zinc was lower (p < 0.05) in LIF than PIF. Food intake (p = 0.045) and body weight on day 4 (p < 0.05) and on day 7 (p < 0.001) were lower in LIF compared to PIF. A, R (p < 0.05 for both minerals), %A/I, and %R/I (p < 0.001 and p < 0.05 for iron and zinc, respectively) were significantly lower in rats fed LIF. Similarly, the %R/A of iron was lower (p < 0.001) in this group. Hematocrit and hemoglobin did not show significant differences. Iron and zinc levels in liver, spleen and erythrocytes were similar in both groups, but skin iron concentration was higher in LIF. Therefore, in contrast with the in vitro results, consumption of the in-bottle-sterilized formula determines lower iron and zinc bioavailability compared to the reconstituted powder infant formula.