Predictive factors for conversion in laparoscopic colorectal surgery and impact on early postoperative period. An 848 patient case control study.

Jose Gellona, MD, Gonzalo Inostroza, MD, Rodrigo Miguieles, MD, Maria Elena Molina, MD, Felipe Bellolio, MD, Gonzalo Urrejola, MD, Daniella Espinola, MD, Tomas Larach, MD, Jose M Zuniga, Alvaro Zuniga, MD

Colorectal Unit. Department of Digestive Surgery. Pontificia Universidad Catolica de Chile

Introduction: Several factors have been associated with a higher rate of conversion in laparoscopic colorectal surgery (LCS). Conversion to open surgery is not a complication, but there is some evidence that shows a higher morbidity rate in this subgroup of patients. The aim of this study is to identify predictive factors for conversion and their impact on early postoperative recovery in terms of morbidity, hospital stay and reoperation rate.

Methods: Case control study. Patients who underwent LCS from June 2000 until December 2011 were included. Patients (cases) converted to open surgery were identified, analyzed and compared with those (controls) not converted. Univariate and multivariate logistic regression was used to identify possible predictive factors. ROC curve and goodness of fit test was performed to evaluate the model. Analyzed factors were age, sex, body mass index (BMI), malignancy, previous abdominal surgery and American Society of Anesthesiologist (ASA) score. Postoperative lengh of stay, time to first bowel flatus, time to resume a regular diet and pain measured in analog scale were recorded.

Results: 848 patients were included. 466 (55%) female, median age of 58 (8 – 94) years. Main indications for surgery were: colorectal carcinoma (41%), diverticular disease (22%) and inflammatory bowel disease (7%). Most frequently performed surgeries were: sigmoidectomy (39%), anterior resection of the rectum (13%), right colectomy (12%), total colectomy (9%) and Hartmann’s reversal (5.4%). Conversion and overall morbidity rate were 7.2% and 11.6% respectively. On univariate analysis, risk factors for conversion were: male sex (OR 2.92 95% CI 1.49 – 5.73), BMI (OR 1.13, CI 95% 1.01 – 1.25), and malignant disease (OR 1.75 95% CI 1.04 – 2.96). On multivariate analysis, age over 75 years and BMI were independent factors for conversion (OR 7.92, CI 95% 1.89 – 33.01 and 1.16, 95% CI 1.03 – 1.32 respectively), male sex and malignancy did not reached a significant OR. The model has an area under the curve (ROC curve) of 0.7609 and a goodness of fit of 5.17 (p=0.712). Overall morbidity was 11.3% in laparoscopic patients versus 16.4% in converted patients (p = 0.218). Pelvic sepsis was 2.2% and 1.6% (p=NS) in laparoscopic and converted patients respectively. Reoperation rate was 3.3% in converted cases versus 2.5% in controls (p=NS). Median postoperative hospital stay was 5 (2-36) and 9 (4-59) days for non-converted and converted groups respectively (p < 0.001). Postoperative pain, was significantly greater in converted patients from second to fifth postoperative day. Median time to first bowel flatus and time to resume a regular diet also were prolonged in the converted group (2 versus 3 days p<0.001 and 4 versus 5 days p<0.001 respectively). C

onclusion: In these series, preoperative BMI and age over 75 years are independent risk factors for conversion. Conversion did not affect overall morbidity, pelvic sepsis or reoperation rate. Conversion prolongs hospital stay, time to first bowel movement and regular diet.

Session: Poster Presentation

Program Number: P068

« Return to SAGES 2013 abstract archive