Predictors of in-hospital length of stay after laparoscopic ventral hernia repair: results of multivariate logistic regression analysis

To identify variables that predict in-hospital length of stay (LOS) after laparoscopic ventral hernia repair (LVHR). Univariate analysis of patient and intraoperative variables was conducted on an operating room database of LVHRs performed from April 2001 to April 2009. Analysis was performed using either chi-square or linear trend analysis, as appropriate. A multivariate logistic regression model was created manually, to determine independent variables that predict LOS. p Value < 0.05 was considered significant. A total of 221 patients, with mean age of 56 years (range 25-88 years) underwent LVHR, for a total of 121 incisional and 100 primary ventral hernias. Of patients, 40% had incarcerated hernias and 25% had complex hernias (defined as multiple points of weakness on the anterior abdominal wall). The overall conversion rate to open operation was 6%. Mean LOS was 1.54 days (range 0-22 days). Eighty-six patients (39%) were discharged on the day of the procedure. Variables associated with significantly longer LOS on univariate analysis were incisional hernia (p = 0.000009), mesh size (p = 0.00007), complex hernia (p = 0.00009), incarcerated hernia (p = 0.0004), patient age (p = 0.0006), need for lysis of adhesions (p = 0.001), and female gender (p = 0.01). American Society of Anesthesiologists (ASA) grade > 2, conversion to open procedure, and recurrent hernia were not associated with longer LOS. Four factors were independently associated with significant longer length of stay on multivariate logistic regression analysis (p < 0.05): mesh size (p = 0.00005), incarcerated hernia (p = 0.002), patient age (p = 0.018), and complex hernia (p = 0.035). Mesh size, incarcerated hernia, patient age, and complex hernia predict longer length of stay after laparoscopic ventral hernia repair.