ABSTRACT
OBJECTIVE: The purpose of this study was to prospectively evaluate the accuracy of MDCT for preoperative determination of the site of surgically proven gastrointestinal tract perforations and to determine the most predictive findings in this diagnosis. SUBJECTS AND METHODS: We prospectively studied 85 consecutive patients with extraluminal air on MDCT who had surgically proven gastrointestinal tract perforations. All patients underwent surgery within 12 hours after MDCT was performed. Two experienced radiologists, blinded to the surgical diagnosis, reached a consensus prediction of the site of the perforation using the following eight MDCT findings: concentration of extraluminal air bubbles adjacent to the bowel wall, free air in supramesocolic or inframesocolic compartments, extraluminal air in both abdomen and pelvis, focal defect in the bowel wall, segmental bowel-wall thickening, perivisceral fat stranding, abscess, and extraluminal fluid. MDCT imaging results were compared with surgical and pathologic findings. Logistic regression analyses were performed to assess the significance of the different radiologic criteria. RESULTS: Analysis of MDCT images was predictive of the site of gastrointestinal tract perforation in 73 (86%) of 85 patients. Logistic regression showed that concentration of extraluminal air bubbles (p < 0.001), segmental bowel wall thickening (p < 0.001), and focal defect of the bowel wall (p = 0.007) were strong predictors of the site of bowel perforation. CONCLUSION: MDCT is highly accurate for predicting the site of gastrointestinal tract perforations. Three of eight CT findings significantly correlate with surgical diagnosis.
