Mark Katory1, Ross McLean2, Khalid Osman2, Mukhtar Ahmad3, Tracey Hughes4, Mike Newby4, Christopher Dennison4, Paul O'Loughlin2. 1. Department of Colorectal Surgery, Queen Elizabeth Hospital, Queen Elizabeth Avenue, Gateshead, NE9 6SX, UK. mark.katory@ghnt.nhs.uk. 2. Department of Colorectal Surgery, Queen Elizabeth Hospital, Queen Elizabeth Avenue, Gateshead, NE9 6SX, UK. 3. Department of Colorectal Surgery, Poole Hospital NHS Trust, Poole, Longfleet Road, Dorset, BH15 2JB, UK. 4. Department of Radiology, Queen Elizabeth Hospital, Queen Elizabeth Avenue, Gateshead, NE9 6SX, UK.
Abstract
PURPOSE: Interpretation of water-soluble contrast enema following laparoscopic low anterior resection can be very challenging for both radiologists and colorectal surgeons. Discriminating the radiological appearances secondary to anastomotic configuration from those caused by actual anastomotic dehiscence is a common problem and may be made worse with the advent of laparoscopic surgery. The aim of this study is to identify potential novel appearances of the water-soluble contrast enema (WSCE) images of rectal anastomosis following laparoscopic low anterior resection to radiologists and surgeons. METHODS: We enrolled 45 patients who underwent laparoscopic low anterior resection with proximal de-functioning loop ileostomy within a specialized colorectal unit. The water-soluble contrast enema reports were reviewed. Two blinded colorectal radiologists independently reviewed the images of patients suspected of anastomotic leak. All of these patients also underwent a flexible sigmoidoscopy to confirm or exclude anastomotic leak before reversal of loop ileostomy. Inter-observer concordance was calculated. RESULTS: Seven out of eighteen patients (38.9%) were found to have true anastomotic leaks on flexible sigmoidoscopy (15% overall leak rate). In the remaining eleven patients the image appearances were attributed to the appearance of the anastomotic 'dog-ear effect', created by the anastomotic configuration due to multiple firing of the intra-corporeal laparoscopic stapling device. Radiologist inter-observer concordance was 83%. Sensitivity was 100%, specificity 71%, positive-predictive value (38.9%) and negative-predictive value (100%). CONCLUSIONS: The novel appearances of laparoscopic-stapled rectal anastomoses in WSCE can be mistaken for anastomotic leak. To avoid delay in reversal of ileostomy, a flexible sigmoidoscopy can be used to confirm or exclude a leak.
PURPOSE: Interpretation of water-soluble contrast enema following laparoscopic low anterior resection can be very challenging for both radiologists and colorectal surgeons. Discriminating the radiological appearances secondary to anastomotic configuration from those caused by actual anastomotic dehiscence is a common problem and may be made worse with the advent of laparoscopic surgery. The aim of this study is to identify potential novel appearances of the water-soluble contrast enema (WSCE) images of rectal anastomosis following laparoscopic low anterior resection to radiologists and surgeons. METHODS: We enrolled 45 patients who underwent laparoscopic low anterior resection with proximal de-functioning loop ileostomy within a specialized colorectal unit. The water-soluble contrast enema reports were reviewed. Two blinded colorectal radiologists independently reviewed the images of patients suspected of anastomotic leak. All of these patients also underwent a flexible sigmoidoscopy to confirm or exclude anastomotic leak before reversal of loop ileostomy. Inter-observer concordance was calculated. RESULTS: Seven out of eighteen patients (38.9%) were found to have true anastomotic leaks on flexible sigmoidoscopy (15% overall leak rate). In the remaining eleven patients the image appearances were attributed to the appearance of the anastomotic 'dog-ear effect', created by the anastomotic configuration due to multiple firing of the intra-corporeal laparoscopic stapling device. Radiologist inter-observer concordance was 83%. Sensitivity was 100%, specificity 71%, positive-predictive value (38.9%) and negative-predictive value (100%). CONCLUSIONS: The novel appearances of laparoscopic-stapled rectal anastomoses in WSCE can be mistaken for anastomotic leak. To avoid delay in reversal of ileostomy, a flexible sigmoidoscopy can be used to confirm or exclude a leak.