Introduction: Thirty-day readmission has become an increasingly scrutinized event in the field of surgery, especially in light of projected cuts in reimbursement. Our objective was to determine if any factors are predictive of 30-day readmission in patients undergoing ventral hernia repair.
Methods: We retrospectively reviewed consecutive charts of patients who underwent laparoscopic or open ventral hernia repair from September 1st, 2005 through June 30th, 2009 by one surgeon at our institution. Perioperative variables included: age, gender, ASA score, BMI, # comorbidities (smoking, diabetes mellitus, COPD, immunosupression), number of prior abdominal surgeries, type of surgery and active abdominal infection at the time of surgery. The primary outcome measure was all cause 30-day readmission. Univariate analysis consisted of Mann-Whitney and Fischer’s or Chi-squared statistics. Multiple logistic regression was performed on significant variables followed by the likelihood ratio test to assess model significance. A p-value of <0.05 was considered significant for all tests.
Results: During the study period, 344 patients were identified (163 open repair and 181 laparoscopic repair). The incidence of 30-day readmission was 8.4% (29 of 344). Readmission indications included: 48% wound infection (14/29), 28% GI complaints (8/29) and 24% other (7/29). (hematologic n=3, cardiac n=2, pulmonary n=1, hepatobiliary n=1). On univariate analysis, patients readmitted within 30 days were older (61 vs 56; p=0.05), predominantly female (79% vs 57%; p=0.03), had greater prior abdominal surgeries (3 vs 2; p=0.01), had open repair (66% vs 46%; p=0.04) and had active abdominal infection at the time of surgery (31% vs 12%; p=0.009). Remaining variables (ASA score, BMI, comorbidities) were not predictive of 30-day readmission. On multivariate analysis, independent predictors of 30-day readmission included age (OR 1.03; 95% CI: 1.00-1.06), female gender (OR 3.03; 95% CI: 1.18-7.80) and active abdominal infection (OR 2.8; 95% CI: 1.17-6.74). Likelihood ratio test had a p=0.007.
Variable |
OR Univariate | 95% CI | OR Multivariate ** | 95% CI |
Age | 1.02 | 1.0 – 1.1 | 1.03 | 1.00-1.06 |
Female | 2.88 | 1.1 – 7.3 | 3.03 | 1.18-7.80 |
Prior Abdominal Surgeries | 1.14 | 1.0 – 1.3 | ||
Open Case | 2.26 | 1.02 – 5.01 | ||
Active Abdominal Infection | 3.28 | 1.40-7.73 | 2.80 | 1.17-6.74 |
** Likelihood ratio test for model: p-value 0.007
Conclusion: Patients undergoing ventral hernia repair can represent a complex and challenging group. In our practice age, female gender and active abdominal infection are all significant independent risk factors for 30-day readmission after ventral hernia repair. Preoperative recognition of these high risk patients can help focus hospital resource utilization to increase surveillance and possible early intervention to reduce readmissions and limit future revenue loss.
Session: Podium Presentation
Program Number: S028