CT Transition Zones: Relevance in Management of Small Bowel Obstruction
Modesto Colon*, Dana a. Telem, Debbie Wong, Scott Q. Nguyen, Celia M. Divino
Surgery, The Mount Sinai Hospital, New York, NY
Background: Computed tomography (CT) scans are becoming standard of care in workup of small bowel obstruction (SBO). Radiologists emphasize radiographic transition zones (RTZ), areas of abrupt change from dilated to collapsed bowel on CT, as pathonogmonic for SBO presence and location. The relevance of RTZ in patient management, operative or conservative, remains unknown. The purpose of this study is to determine the surgical predictive value and intraoperative accuracy of RTZ. Methods: Retrospective study of 200 consecutive patients with secondary SBO from 2002-7 who underwent CT. Patients were equally matched for age, gender, comorbidity and clinical presentation. Significance was determined by unpaired t-test and multivariate analysis.Results: Of 200 patients, 150 (75%) had RTZ of which 58 (39%) required surgery and 92 (61%) were managed nonoperatively (p<0.001). For the 50 patients without RTZ, 17 (34%) required surgery and 33 (66%) were managed nonoperatively (p<0.001). RTZ did not increase likelihood of operative intervention (odds ratio (OR) =1.19, 95% CI [0.61-2.32]). Presence of small bowel RTZ conferred an overall 37.9% operative positive predictive value (PPV) with 77% sensitivity. Though patients with proximal versus distal RTZ had increased likelihood of requiring surgical intervention (PPV 47.5% vs. 34.3%), this did not reach significance (p>0.05). (Table 1) Correlation of RTZ and intraoperative location of obstruction occurred in 31 (63%) patients with overall 63.3% PPV and 62% sensitivity (p<0.01). (Table 2)Conclusion: Presence of small bowel RTZ in patients with SBO does not increase likelihood of surgical intervention or predict patients who will need surgery (OR 1.19, PPV 37.9%). Significant correlation exists between radiographic and intraoperative location of SBO making RTZ useful for preoperative planning (PPV 63.3%, p<0.01).
Table 1: Surgical versus Nonoperative Intervention Based on Location of TZ
TransitionZone Location | Surgical(# of patients)n=58 | Nonoperative(# of patients)n=92 | PPV | Sensitivity | Specificity |
All small bowel(n=150) | 58 | 92 | 38.7% | 77% | 26% |
Proximal Small Bowel(n=40) | 19 | 21 | 47.5% | 86% | 34% |
Distal Small Bowel (n=110) | 39 | 71 | 35.4% | 74% | 34% |
Table 2: Correlation Between Radiographic and Intraoperative Location
TransitionZone Location | Correlation(# of patients)n=31 | No Correlation(# of patients)n=18 | PPV | Sensitivity | Specificity |
All small bowel (n=49) | 31 | 18 | 63.3% | 62% | 18% |
Proximal Small Bowel (n=19) | 12 | 7 | 63.2% | 80% | 13% |
Distal Small Bowel(n=30) | 19 | 11 | 63.3% | 66% | 21% |
Back to Program | 2009 Program and Abstracts | 2009 Posters