|
|
Back to 2018 Posters
PREOPERATIVE ORAL ANTIBIOTICS ARE ASSOCIATED WITH SIGNIFICANTLY LOWER ANASTOMOTIC LEAK RATE AFTER LOW ANTERIOR RESECTION IN RECTUM CANCER PATIENTS
Hiromichi Miyagaki*1,2, Carl Winkler2, Sandhu K. Jaspreet2, Elie Sutton2, H M C Shantha Kumara2, Vesna Cekic2, Xiaohong Yan2, Richard L. Whelan2
1Saiseikai Senri Hospital, Suita, Osaka, Japan; 2Surgery, Mount Sinai West Hospital, New York, NY
Introduction: Additional data regarding surgical methods and MIS conversions and complications is being gathered in the NSQIP data base in reference to colorectal resections including specific information about anastomotic leaks. This review was carried out to assess the risk factors for anastomotic leakage in LAR patients.
Methods: A NSQIP review of LAR for Rectal cancer cases with anastomoses from 2012 to 2015 was carried out. Inclusion criteria were: Indication for resections was 154.1 (ICD9) or C20 (ICD10) and main CPT codes were 44145, 44146, 44207 or 44208.Exclusion criteria were: totally or partially dependent health status, ventilator dependence, ascites or renal failure, dialysis wound infection preoperatively, history of CHF, sepsis, emergency cases and ASA status 5. Cases with missing data were excluded. Demographic parameters, comorbidities, pretreatment, stage of cancer, postoperative complications and anastomotic leakage in Procedure Targeted and Standard Patient User File were assessed. The log regression analysis was used for multivariate analysis regarding risk factors for leakage, treated with intervention means or surgical means with all demographic and operative data.
Results: A total of 2,691 LAR patients met the study criteria (males 61.5%, female 38.5%; mean age 59). The surgical approaches were: MIS. 30.7%; MIS with assist, 32.7%; MIS with conversion to open, 7.4%; open, 29.2%. Splenic flexure mobilization was carried out in 1298 (48.2%) and 1729 pts (64.3%) were proximally diverted (no difference between methods). A mechanical bowel preparation was given to 80.4% while preoperative oral antibiotics were given to 44.1%. Significantly more open patients received oral antibiotics preop vs the MIS group. The anastomotic leak rate was 6.3 %; there was no significant difference between the different surgical methods as regards leaks. 41% of patients with leaks had reoperations while 36% were drained percutaneously (IR, Interventional Radiology); 23% underwent no interventions and were presumably treated medically. As regards the leaks patient who underwent either IR or surgical interventions, a multivariate analysis was carried out that assessed all parameters. The independent risk factors for leaks were: no diversion (p=0.01), no preoperative antibiotics (p=0.0010), and contaminated and dirty/infected wounds (p=0.0025). More open and converted patients received oral antibiotics (p<0.05) There was no difference in the treatment of leaks based on the surgical method used (p>0.05).
Conclusions: MIS methods were used in 70.8% of LAR patients with a 7.4% conversion rate. Leaks developed in 6.3%; 77% of leaks underwent IR drainage or reoperation. The choice of surgical method did not impact the leak rate. Failure to give preop oral antibiotics was found to be an independent risk factor for leaks.
Back to 2018 Posters
|
