The small height of an anastomotic colonic doughnut is an independent risk factor of anastomotic leakage following colorectal resection: results of a prospective study on 154 consecutive cases.

PURPOSE: The aim of this prospective study was to assess the influence of morphological characteristics of anastomotic doughnuts on the risk of anastomotic leakage (AL) after double-stapled colorectal anastomosis. METHODS: This single-center prospective study enrolled all patients undergoing double-stapled colorectal anastomosis between December 2012 and December 2015. Maximal diameter and minimal and maximal heights and widths of both colonic and rectal doughnuts were measured by surgeons in the operating room. Their influence on the risk of AL was analyzed on uni- and multivariate models. RESULTS: One hundred fifty-four patients were included; 92 (59.7%) were operated on for malignancy. Colorectal anastomoses > and <10 cm above the anal verge were performed in 96 (62.3%) and 58 (37.7%) patients, respectively. AL occurred in 17 (11.0%). The minimal height of the colonic doughnut (CD) was the only measurement significantly associated with an increased risk of AL (p = 0.026). A cutoff value of 4.5 mm for the CD determined on the ROC curve (AUC 0.685, p = 0.013) yielded the best sensitivity (61.4%) and specificity (82.4%) to predict AL. On multivariate analysis, a height of the CD <4.5 mm (OR 5.743, 95% IC 1.476-22.346, p = 0.012), malignant disease (OR 8.821, 95% IC 1.051-74.006, p = 0.045), and American Society of Anesthesiologists score >2 (OR 3.408, 95% IC 1.017-11.418, p = 0.047) were the only independent risk factors of AL. CONCLUSION: The CD's minimal height influences the risk of AL. Its routine measurement during operation, along with other risk factors, could help to decide which patients could benefit from a diverting stoma or the creation of a new anastomosis.

Impact of chemotherapy on the accuracy of computed tomography scan for the evaluation of colorectal liver metastases.

BACKGROUND: The aim of this study was to determine the effect of chemotherapy on the accuracy of CT scan for the preoperative evaluation of colorectal liver metastases (LM). METHOD: Between 1999 and 2005, 92 patients with less than six LM at diagnosis underwent surgery within four weeks of CT scan. Of these 92 patients, 60 had received chemotherapy before surgery (chemotherapy group), whereas 32 were operated without preoperative treatment (control group). Histological and surgical findings were compared with helical CT results. RESULTS: On a per lesion basis, the sensitivity of CT scan for the detection of LM was 71% in the chemotherapy group and 76% in the control group. On a per patient basis, the sensitivity of CT scan was 54% in the chemotherapy group and 69% in the control group. Discrepancies between CT scan and surgical or histological findings, including both false-negative and false-positive lesions, were found in 32 patients from the chemotherapy group and 10 patients from the control group (p=0.0471). In multivariate analysis, chemotherapy (p=0.032), number of LM > 3 (p=0.034), and steatosis > 30% (p=0.012) were risk factors for inadequate staging of LM by CT scan. CONCLUSIONS: Chemotherapy reduces the accuracy of CT scan for preoperative evaluation of colorectal LM.