We evaluated the efficacy of the thickening of human milk by precooked starch in reducing gastroesophageal reflux in preterm infants. Five preterm infants with frequent regurgitations (median gestational age, 28 weeks; range, 27 to 32 weeks; median birth weight, 990 g; range, 570 to 1900 g) were fed alternately during 24 hours with four meals of fortified maternal milk (milk A) and four meals of fortified maternal milk thickened by 1.5 g of precooked starch per 100 mL of milk (milk B). The acidic and buffered refluxes were detected by simultaneous pH monitoring and multiple intraluminal impedance. Eight feeding periods for each baby were recorded. The number of the acidic (34 after milk A vs 36 after milk B) and buffered (112 after milk A vs 134 after milk B) episodes of gastroesophageal reflux did not differ. Thickening human milk by precooked starch is ineffective in reducing gastroesophageal reflux in premature infants.