Original article
Diagnostic value of computed tomography for detecting anastomotic or staple line leakage after bariatric surgery

https://doi.org/10.1016/j.soard.2018.05.007Get rights and content

Highlights

  • More knowledge of the reliability of abdominal CT is needed to create uniform diagnostic approach when anastomotic or staple line leakage (ASLL) after bariatric surgery is suspected.

  • CT is a useful diagnostic modality to rule out ASLL after bariatric surgery.

  • CT should have a place in the diagnostic process when ASLL is suspected.

  • CT scan can prevent unnecessary performed diagnostic laparoscopy.

Abstract

Background

Postbariatric anastomotic or staple line leakage (ASLL) is a dreaded complication with an incidence up to 1.6% and a leak-associated mortality of 5.0% to 16.7%. Feared low sensitivity of abdominal computed tomography (CT) for detecting ASLL is causing surgeons to omit CT and directly perform a diagnostic laparoscopy in patients with suspected ASLL.

Objectives

To evaluate the diagnostic value of CT in case of suspected ASLL after bariatric procedures and to identify reliable CT characteristics predicting the presence of ASLL.

Setting

A large teaching hospital and bariatric center of excellence.

Methods

All CT scans performed for suspected ASLL after bariatric surgery in the period November 2007 until August 2016 were independently reevaluated by abdominal radiologists. The diagnostic value of CT by means of sensitivity, specificity, and positive and negative predictive value was analyzed comparing results of reevaluation to a standard of reference. Multivariable regression was performed to identify reliable CT characteristics for the presence of ASLL.

Results

A total of 66 CT scans were performed because of suspected leakage. Reevaluation of CT scans revealed a sensitivity of 89% to 100%, a specificity of 69% to 78%, a positive predictive value of 39% to 50%, and a negative predictive value of 97% to 100% of CT for detecting ASLL after bariatric surgery. Multivariable logistic regression of ASLL characteristics on CT revealed ‘air near the anastomosis/staple line’ as the only independent predictor for the presence of ASLL.

Conclusion

With a sensitivity of 89% to 100% and negative predictive value of 97% to 100%, a negative CT can rule out ASLL in patients with a lower suspicion of ASLL.

Section snippets

Study design and study population

All consecutive patients who underwent a bariatric procedure containing an anastomosis or staple line in the period November 2007 until August 2016 were retrospectively reviewed. All patients were operated according to the International Federation for the Surgery of Obesity and Metabolic Disorders criteria for bariatric surgery [9]. The following surgical and revisional procedures were performed by experienced bariatric surgeons or by a resident under their direct supervision: laparoscopic

Patient characteristics

In our study population, 2410 patients underwent a bariatric procedure concerning an anastomosis or staple line (Fig. 1, Table 1). Of the 2410, 121 patients had a clinical suspicion of ASLL. Leakage was diagnosed in 28 (1.2%) patients at a median of 3 (interquartile range, 2–8) days after surgery. Three patients died due to multiorgan failure after ASLL, resulting in a leak-associated mortality of 10.7%.

After laparoscopic Roux-en-Y gastric bypass, ASLL was located at the gastrojejunostomy in 15

Discussion

In the total population, we found a sensitivity of abdominal CT ranging from 77% to 85%, a specificity of 65% to 75%, a PPV of 35% to 45%, and a NPV of 92% to 95%, increasing to 85% to 92%, 66% to 75%, 41% to 45%, and 95% to 97%, respectively, after radiologists received clinical information in the second phase of reevaluation. This increase underlines the importance of providing accurate clinical information to the involved radiologist.

Using original CT reports as the sole reference, Bingham

Conclusion

With an overall sensitivity of 85% to 92% and NPV of 92% to 95%, a negative CT scan can be trusted; it gives high confidence that its negative result is true, especially in patients without the co-existence of tachycardia and tachypnea (sensitivity 89%–100% and NPV of 97%–100%), where a negative CT scan can rule out ASLL. Therefore, CT should be used as a diagnostic modality to triage patients suspected of ASLL and it can subsequently prevent unnecessary performed diagnostic laparoscopy.

Disclosures

The authors have no commercial associations that might be a conflict of interest in relation to this article.

References (10)

There are more references available in the full text version of this article.

Cited by (3)

  • Use of dual energy CT to identify gastrointestinal anastomotic leak by assessment of percutaneous drain contents

    2023, Radiology Case Reports
    Citation Excerpt :

    The 30-day mortality rate is over 3-fold greater for post-operative patients with leaks compared to those without [2]. For patients that remain hemodynamically stable, GI anastomotic leaks are often confirmed radiographically prior to determining treatment, either by fluoroscopy or CT with oral contrast [3,4]. Occasionally, conventional imaging and patient presentation may remain inconclusive.

  • AGA Clinical Practice Update on Evaluation and Management of Early Complications After Bariatric/Metabolic Surgery: Expert Review

    2021, Clinical Gastroenterology and Hepatology
    Citation Excerpt :

    Daily communication should occur among all providers during dynamic phases, either at the bedside or via a portal (secure email, electronic health record, standing phone call, etc.). For example, the interventional radiologist can provide advice on optimal imaging (upper gastrointestinal series vs oral/intravenous contrast computed tomography scan), prospects of placing a percutaneous drain into a perigastric collection, etc.7 The surgeon can advise on the anticipated morbidity of a laparoscopic Roux-en-Y fistulojejunostomy as a treatment for leak post-SG, information critical to ascertaining whether pursuing “out-of-the-box” endoscopic approaches should be considered. Clinicians performing endoscopic management, in addition to reviewing the operative report, should have detailed knowledge and experience with postsurgical anatomy.

View full text