|
Back to 2016 Annual Meeting
Do anti-TNF agents affect postoperative outcomes in patients undergoing surgery for Inflammatory Bowel Disease?
Celia M. Divino, Chaya Shwaartz*, Adam C. Fields, Maximiliano Sobrero, Brian Cohen
Ichan school of medicine, New York, NY
Introduction: Since 1998, biologic therapy with anti-TNF-alpha agents has been an integral part in the treatment of inflammatory bowel disease (IBD). However, a subset of IBD patients ultimately require surgery. Up to 30% of IBD patients who undergo surgery have been trialed on anti-TNF-alpha agents. Studies assessing the effect of anti-TNF-alpha agents on postoperative outcomes have been inconsistent. The aim of this study is to assess postoperative morbidity in IBD patients who underwent surgery with anti-TNF-alpha therapy prior to surgery.
Methods: This study is a retrospective review of 282 patients with ulcerative colitis or Crohn's disease undergoing intestinal surgery between 2013 and 2015. Patients were divided into two groups based on treatment with anti-TNF-alpha agents (infliximab, adalimumab, certolizumab) within 8 weeks of surgery. Thirty-day postoperative outcomes including anastomotic leak, intra-abdominal abscess, wound infection, extra-abdominal infection, readmission, and mortality were recorded. Univariate and multivariate statistical analyses were carried out.
Results: 73 patients were treated with anti-TNF-alpha therapy within 8 weeks of surgery while 209 patients did not have exposure (Table 1). Thirty-day anastomotic leak (5.5% vs. 5.3%, p = 1.00), intra-abdominal abscess (5.5% vs. 1.9%, p = 0.21), wound infection (6.8% vs. 4.3%, p = 0.37), extra-abdominal infection (4.1% vs. 1.9%, p = 0.38), readmission (8.2% vs. 5.3%, p = 0.39), and mortality (0%) rates were not significantly different between the two groups (Table 2). Preoperative steroid use (p = 0.04) and hand-sewn anastomosis (p = 0.009) were independent predictors of anastomotic leak in multivariate logistic regression.
Conclusion: The use of anti-TNF-alpha medications in IBD patients within two months of intestinal surgery is not associated with an increased risk of thirty-day postoperative complications.
Table 1. Patient and procedure characteristics
| Variables | TNF
(N = 73) | No TNF
(N = 209) | p value | | Demographics | | Age, y | 36.6 ± 14.5 | 42.6 ± 15.8 | 0.004 | | Female gender | 31 (42.5%) | 107 (51.2%) | 0.22 | | Clinical characteristics | | ASA 1 or 2 | 26 (35.6%) | 52 (24.9%) | 0.08 | | BMI, kg/m2 | 22.3 ± 4.3 | 23.8 ± 7.4 | 0.10 | | Smoker | 6 (8.2%) | 16 (7.7%) | 0.88 | | Steroid use | 26 (35.6%) | 50 (23.9%) | 0.07 | | Other immunosuppression use | 29 (39.7%) | 55 (26.3%) | 0.04 | | Comorbidities | | Crohn | 60 (82.2%) | 110 (52.6%) | <0.01 | | Hypertension | 5 (6.8%) | 28 (13.4%) | 0.20 | | Diabetes | 2 (2.7%) | 3 (1.4%) | 0.47 | | Coronary artery disease | 1 (1.4%) | 3 (1.4%) | 0.97 | | Procedure Characteristics | | Procedure | | Ileoanal pouch | 8 (11.0%) | 46 (22.0%) | 0.001 | | Ileocolic resection | 43 (58.9%) | 72 (34.4%) | | | Partial colectomy | 2 (2.7%) | 12 (5.7%) | | | Ileostomy reversal | 9 (12.3%) | 56 (26.8%) | | | Small bowel resection | 7 (9.6%) | 19 (9.1%) | | | Ileosigmoid/ileorectal anastomosis | 4 (5.5%) | 4 (1.9%) | | | Diverting loop ileostomy | 9 (12.3%) | 49 (23.4%) | 0.05 | | Laparoscopic | 46 (63.0%) | 119 (56.9%) | 0.41 | | Stapled anastomosis | 63 (86.3%) | 180 (86.1%) | 1.00 | | Preoperative albumin level, g/dL | 3.6 +/- 0.8 | 3.7 +/- 0.7 | 0.21 |
Table 2. Thirty-day postoperative outcomes | Variables | TNF
(N=73) | No TNF
(N = 209) |
p value | | Anastomotic leak | 4 (5.5%) | 11 (5.3%)
| 1.00 | | Intra-abdominal abscess | 4 (5.5%) | 4 (1.9%)
| 0.21 | | Wound infection | 5 (6.8%) | 9 (4.3%)
| 0.37 | | Extra-abdominal infection | 3 (4.1%) | 4 (1.9%)
| 0.38 | Readmission
| 6 (8.2%) | 11 (5.3%) | 0.39 | | Mortality | 0 | 0 | - |
Back to 2016 Annual Meeting
|
