Diagnostic Performance of Ultrasound for Diagnosing Midgut Malrotation and Volvulus in Children: A Multiinstitutional Retrospective Review.

BACKGROUND. The reported diagnostic performance of ultrasound (US) for midgut malrotation and volvulus varies. OBJECTIVE. The purpose of this study was to evaluate the diagnostic performance of US for midgut malrotation and volvulus separately and to assess individual sonographic signs for each diagnosis. METHODS. This multicenter, retrospective, cross-sectional study included children (who were defined as individuals 0-18 years old) who had US performed as the first imaging test for evaluation of suspected midgut malrotation or volvulus from January 1, 2018, to June 30, 2021. Clinical data were extracted from medical records. Blinded reviewers measured the proximal duodenum and assessed for malrotation and volvulus as well as for specific sonographic signs. Nondiagnostic and equivocal studies were excluded. The reference standards used for malrotation included surgery, upper gastrointestinal (UGI) series, and CT and/or MRI. Volvulus reference standards included the same plus clinical follow-up (minimum, 28 days). Sensitivity, specificity, and accuracy were calculated to evaluate diagnostic performance; the Wilcoxon rank sum test was used to compare median values. RESULTS. Malrotation analysis included 384 examinations; median patient age was 32 days (IQR, 5-182 days). Volvulus analysis included 900 examinations; median patient age was 60 days (IQR, 9-573 days). The sensitivity and specificity of US for malrotation were 93% (95% CI, 81-98%) and 96% (95% CI, 92-98%) by original report and 97% (95% CI, 87-100%) and 99% (95% CI, 97-100%) by blinded research review. The sensitivity and specificity of US for volvulus were 97% (95% CI, 85-100%) and 98% (95% CI, 96-99%) by original report and 97% (95% CI, 86-100%) and 99% (95% CI, 98-100%) by blinded research review. The most accurate sonographic signs were the intraperitoneal position of the third portion of the duodenum for malrotation (accuracy = 98%) and the whirlpool sign for volvulus (accuracy = 99%). The median proximal duodenal diameter was greater in children with volvulus (13 mm [IQR, 7-18 mm] versus 6 mm [IQR, 4-8 mm], p < .001). CONCLUSION. The performance of US for diagnosing both midgut malrotation and volvulus is excellent in the setting of a diagnostic imaging study. CLINICAL IMPACT. US can be used as the first-line imaging modality for diagnosing midgut malrotation and volvulus, with UGI series reserved for nondiagnostic or equivocal examinations.