RESUMO
To evaluate the validity of ultrasound for diagnosis and exclusion of pyloric stenosis in the infants with nonbilious vomiting. In a cross-sectional study, 444 consecutive infants with clinical suspicion of pyloric stenosis were evaluated by ultrasound [US] and categorized as pyloric stenosis or not according measuring parameters as muscle thickness, muscle width and canal length of pylorus. Positive findings were confirmed at surgery; Negative findings were confirmed by means of follow up. Sensitivity, specificity and accuracy were calculated. Sensitivity, specificity and accuracy of ultrasound were 100% if pyloric muscle thickness of >3mm was chosen as diagnostic. When muscle thickness more than 4mm was used, sensitivity, specificity and accuracy were 96%, 100% and 99.32% respectively. Ultrasound is highly sensitive and specific if pyloric muscle thickness 3 mm is used as cut off point. By virtue of direct visualization of the pyloric muscle, ultrasound is method of choice for both the diagnosis and exclusion of pyloric stenosis in infants