Trends, factors, and disparities associated with length of stay after lower extremity bypass for tissue loss.

ABSTRACT

OBJECTIVE: The objective of this study was to determine predictors of increased length of stay (LOS) in patients who underwent lower extremity bypass for tissue loss. METHODS: Using 2011 to 2016 National Surgical Quality Improvement Program vascular targeted databases, we compared demographics, comorbidities, procedural characteristics, and 30-day outcomes of patients who had expected LOS vs extended LOS (>75th percentile, 9 days) after nonemergent lower extremity bypass for tissue loss. We also compared factors associated with short LOS (<25th percentile, 4 days) and extended LOS (>75th percentile, 9 days) vs the interquartile range of LOS (4-9 days). Yearly trends and independent predictors were determined by linear and logistic regression. This study was exempt from Institutional Review Board approval. RESULTS: In 4964 analyzed patients, there were no significant yearly trends or changes in LOS in the recent 5 years (P > .05). Overall 30-day mortality, major amputation, and reintervention rates were 1.6%, 4.5%, and 4.8%, respectively, also with no significant yearly trends (all P > .05). On univariate analysis, nonwhite race, dependent functional status, transfers, dialysis, congestive heart failure, hypertension, beta blockers, distal bypass targets, and extended operative time were associated with extended LOS (P < .05). Extended LOS was also associated with higher rates of 30-day major adverse limb and cardiac events, additional procedures related to wound care, deep venous thrombosis, complications (pulmonary, renal, septic, bleeding, and wound), and discharge to facility but lower 30-day readmission rates. After adjustment for covariates, the independent factors for extended LOS included dialysis, beta blockers, prolonged operative time, reintervention, major amputation, additional procedures related to wound care, deep venous thrombosis, complications (pulmonary, renal, septic, bleeding, and wound), and discharge to facility (P < .05). On the other hand, multivariable analysis showed that patients with expected LOS were significantly more likely to have been of white race or readmitted postoperatively (P < .05). CONCLUSIONS: From 2011 to 2016, there were no significant changes in LOS. Efforts to decrease LOS without increasing readmission rates while focusing on some of the identified factors, including preventable postoperative complications and pre-existing socioeconomic factors, may improve the overall vascular care of these challenging patients.