Length of Stay after Thoracic Endovascular Aortic Repair Depends on Indication and Acuity.

BACKGROUND: Length of stay (LOS) is a commonly used metric to optimize value in medical care. Although pathways have been developed for some procedures in vascular surgery to reduce LOS, they do not yet exist for thoracic endovascular aortic repair (TEVAR). The purpose of this study is to identify and define the risk factors for prolonged LOS in patients undergoing TEVAR to facilitate pathway development. METHODS: We included TEVAR patients in the National Surgical Quality Improvement Program database from 2005 to 2015. Prolonged LOS was defined as LOS > 75th percentile of the overall cohort (11 days). Because initial analysis revealed the distinct clinical differences between dissection and aneurysm patients, further analysis was stratified by aortic pathology. Student's t-test and Chi-square tests were used to compare demographic and perioperative variables between dissection and aneurysm patients, respectively. Multivariable logistic regression was used to evaluate the predictors for prolonged LOS. RESULTS: A total of 3,021 patients underwent TEVAR, with 858 patients (28.4%) undergoing TEVAR for dissection and 2,163 (71.6%) undergoing TEVAR for aneurysm. An initial analysis with logistic regression identified dissection indication (odds ratio [OR], 2.87; 95% confidence interval [CI], 1.1-7.3) as an independent predictor of prolonged LOS. Further analysis for prolonged LOS was subsequently performed separating dissection and aneurysm patients. Aneurysm patients were older (71.2 ± 11.7 vs. 63.1 ± 13.6 years, P < 0.001), more often Caucasian (76.8% vs. 61.8%, P < 0.001), and had more medical comorbidities (chronic obstructive pulmonary disease, cardiac history, diabetes, peripheral vascular disease, transient ischemic attack [TIA], P < 0.001). In contrast, dissection patients had higher American Society of Anesthesiology (ASA) classification score (58.5% had >3 ASA vs. 45.5%, P < 0.001), longer hospitalizations (10.2 ± 9.3 vs. 8.5 ± 10.4 days, P < 0.001), were more likely to have been transferred from another hospital or emergency room (58.4% vs. 48.3%, P < 0.001), and were more often emergent (32.4% vs. 15.4%, P < 0.001). In dissection patients, ASA classification score (OR, 1.49; 95% CI, 1.1-2.1) and dialysis (OR, 1.98; 95% CI, 1.0-3.9) were independent predictors for prolonged LOS. In aneurysm patients, dependent functional status (OR, 2.03; 95% CI, 1.4-2.8), diabetes (OR, 1.75; 95% CI, 1.1-2.8), cardiac history (OR, 1.37; 95% CI, 1.0-1.9), emergency status (OR, 1.98; 95% CI, 1.4-2.8), and dialysis (OR, 2.08; 95% CI, 1.2-3.7) predicted prolonged LOS. Postoperative complications including stroke/TIA; failure to wean from ventilator, sepsis, and pneumonia; and need for reoperation similarly increased LOS in both dissection and aneurysm patients. CONCLUSIONS: Dissection and aneurysm patients undergoing TEVAR are comprised of different patient populations, with dissection patients more often enduring prolonged hospitalizations. In contrast, TEVAR performed for nonemergent aneurysm repair had the shortest LOS. These data support the development of separate pathways defined by indication and acuity for patients undergoing TEVAR.