Surgical site infection after primary surgery for epithelial ovarian cancer: predictors and impact on survival.

OBJECTIVE: Surgical site infection (SSI) following epithelial ovarian cancer (EOC) primary surgery (PS) occurs in 10-15% of women. Perioperative factors associated with SSI and impact of SSI on survival were determined. METHODS: EOC cases that underwent PS from 1/2/2003 to 12/30/2011 were retrospectively reviewed. SSIs were defined according to ACS NSQIP. Logistic regression models were fit to identify factors associated with SSI. Cox proportional hazards models were utilized to evaluate the association of patient and perioperative characteristics with overall survival (OS) and disease-free survival (DFS). RESULTS: Among 888 cases, 96 (10.8%) developed SSI: 32 superficial, 2 deep, and 62 organ/space. Factors independently associated with superficial SSI were increasing BMI (odds ratio 1.41 [95% confidence interval, 1.12, 1.76] per 5kg/m(2)), increasing operative time (1.24 [1.02, 1.50] per hour), and advanced stage (III/IV) (10.22 [1.37, 76.20]). Factors independently associated with organ/space SSI were history of gastroesophageal reflux disease (2.13 [1.23, 3.71]), surgical complexity (intermediate 3.11 [1.02, 9.49]; high 8.07 [2.60, 25.09]; referent: low), and residual disease (RD) (measureable ≤1cm 1.77 [0.96, 3.27]; suboptimal >1cm (3.36 [1.48, 7.61]; referent: microscopic). Occurrence of superficial (hazard ratio 1.69 [1.12, 2.57]) or organ/space (1.46 [1.07, 2.00]) SSI was independently associated with worse OS. SSI occurrence was not independently associated with DFS. CONCLUSIONS: SSI after PS is associated with decreased OS. Most risk factors for SSI are not modifiable. Alternative measures to lower rates of SSIs are needed as this may improve OS. Preoperative identification of SSI risk factors may assist in risk-assessment and operative planning.

Risk-scoring model for prediction of non-home discharge in epithelial ovarian cancer patients.

BACKGROUND: Identification of preoperative factors predictive of non-home discharge after surgery for epithelial ovarian cancer (EOC) may aid counseling and optimize discharge planning. We aimed to determine the association between preoperative risk factors and non-home discharge. STUDY DESIGN: Patients who underwent primary surgery for EOC at Mayo Clinic between January 2, 2003 and December 29, 2008 were included. Demographic, preoperative, and intraoperative factors were retrospectively abstracted. Logistic regression models were fit to identify preoperative factors associated with non-home discharge. Multivariable models were developed using stepwise and backward variable selection. A risk-scoring system was developed for use in preoperative counseling. RESULTS: Within our cohort of 587 EOC patients, 12.8% were not discharged home (61 went to a skilled nursing facility, 1 to a rehabilitation facility, 1 to hospice, and there were 12 in-hospital deaths). Median length of stay was 7 days (interquartile range [IQR] 5, 10 days) for patients dismissed home compared with 11 days (IQR 7, 17 days) for those with non-home dismissals (p < 0.001). In multivariable analyses, patients with advanced age (odds ratio [OR] 3.75 95% CI [2.57, 5.48], p < 0.001), worse Eastern Cooperative Oncology Group (ECOG) performance status (OR 0.92 [95% CI 0.43, 1.97] for ECOG performance status 1 vs 0 and OR 5.40 (95% CI 2.42, 12.03) for score of 2+ vs 0; p < 0.001), greater American Society of Anesthesiologists (ASA) score (OR 2.03 [95% CI 1.02, 4.04] for score ≥3 vs < 3, p = 0.04), and higher CA-125 (OR 1.28 [95% CI 1.12, 1.46], p < 0.001) were less likely to be discharged home. The unbiased estimate of the c-index was excellent at 0.88, and the model had excellent calibration. CONCLUSIONS: Identification of preoperative factors associated with non-home discharge can assist patient counseling and postoperative disposition planning.