Preparing for the Worst- Predicting Who Will Fail in a Multimodal Enhanced Recovery Pathway

Deborah S Keller1, Irlna Tantchou, MD2, Sergio Ibarra1, Juan R Flores-Gonzalez, MD1, Matt Schultzel, DO2, Eric M Haas, MD, FACS, FASCRS3. 1Colorectal Surgical Associates, 2Colorectal Surgical Associates; University of Texas Medical Center at Houston, 3Colorectal Surgical Associates; Houston Methodist Hospital; University of Texas Medical Center at Houston

Background: Enhanced Recovery Pathways (ERP) have proven to improve outcomes in colorectal surgery. The specific elements in ERPs continue to develop, with multimodal pain management further enhancing recovery. Despite these advances, some patients continue to have unexpected prolonged lengths of stay. Prolonged length of stay (LOS) is associated with higher costs of care and postoperative morbidity. Our goal was to identify the patient and procedural variables associated with failure of an established ERP.

Methods: A prospective departmental database was reviewed to evaluate consecutive patients that underwent minimally invasive colorectal surgery with a multimodal ERP. Patients were stratified into ERP success and ERP failure based on LOS ≥ 5 days (mean + 1SD). Preoperative demographic, perioperative procedural, and postoperative outcome variables were analyzed. Univariate analysis was performed to identify differences in demographic and procedural variables between the ERP success and ERP failure groups. A logistic regression model was used to identify variables predictive of ERP failure.

Results: Between 8/1/13 and 7/31/15, 274 patients met inclusion criteria and were analyzed- 229 ERP successes and 45 failures. The mean age, gender, and BMI were similar across cohorts. The failures had significantly higher rates of preoperative anxiety (p=0.0352), anxiolytic medication (p=0.0103), chronic pain (p=0.0040), prior abdominal surgery (p=0.0313), and chemoradiation (p=0.0301). The main indication for surgery was colon cancer (26.7 failure, 32.8% success) and procedure performed a segmental resection (31.1% failure, 43.2% success) in both groups. The failures had significantly higher conversion rates (p=0.0002), blood loss (p=0.0417), transfusions (p=0.0032), and longer operative times (p=0.0099). Rates of stoma creation were similar. Postoperatively, failures had higher complication rates (p=0.0012), longer time to return of flatus (p=0.0101) and bowel movements (p=0.0005), longer lengths of stay (p=0.0001), and higher reoperation rates (p=0.0182). Total costs for failures were significantly higher than ERP successes (p=0.0182). In the regression model, the variables independently associated with ERP failure were anxiety (OR 2.28, 95% CI 1.04, 4.98, p=0.0389), chronic pain (OR 10.03, 95% CI 2.04, 49.29, p=0.0045), and intraoperative conversion (OR 8.02, 95% CI 1.92, 33.52, p=0.0043).

Conclusions: There are identifiable preoperative and intraoperative factors associated with longer LOS despite following a standardized ERP. By prospectively preparing for these high-risk patients, postoperative clinical and financial results could be improved.


« Return to SAGES 2016 abstract archive