Purpose or Case Report: Hypertrophic pyloric stenosis (HPS) is a frequent cause of vomiting in young infants, with an incidence of 1.5 – 4 per 1000 live births, depending on demographic group. The gold standard diagnostic test is abdominal ultrasound targeting the pylorus, and the measurement thresholds for determining an abnormal pylorus are well established. The objective of this study was to assess whether visual, subjective assessment of the pylorus was as accurate in diagnosing HPS as traditional formal measurements.



Methods & Materials: IRB approval was obtained for this retrospective study. All abdominal ultrasounds performed at our institution for evaluation for HPS from August 2014 to August 2015 were obtained through review of the radiology information system. The exams were formatted such that no indelible measurements were present on the images, and were then reviewed by two experienced pediatric radiologists (ER, RS). The ultrasounds were categorized as positive, negative, or equivocal for HPS based solely on the visual appearance of the pylorus. One author (ER) then recorded the formal pylorus measurements obtained for each study. Results were compared to the reference standard of the surgical report in positive cases, and were considered to be truly negative if the patient did not undergo surgery or present for repeat ultrasound within one week.
Results: Subjective assessment of the pylorus without obtaining formal measurements was highly accurate in the diagnosis of HPS, with an accuracy of greater than 95%. There was also a high degree of inter-rater agreement.
Conclusions: Qualitative evaluation of the pylorus is as accurate in diagnosis of HPS as traditional formal measurements. While pediatric surgeons will likely continue to require formal measurements before taking a patient to surgery, the knowledge that the study “looks positive” can add a layer of diagnostic confidence for the radiologist, and potentially increase efficiency in busy practices. Additionally, pyloric measurements can sometimes be inaccurate, if images are obtained in oblique rather than orthogonal planes, or if the incorrect anatomic structure is measured, such as the antrum or duodenum rather than the pylorus.