RESUMO
OBJECTIVE: To determine the preoperative risk factors most predictive of prolonged length of stay (LOS) or admission to a skilled nursing facility (SNF) or inpatient rehabilitation center (IPR) after free flap reconstruction of the head and neck. STUDY DESIGN: Retrospective cohort study. SETTING: Tertiary academic medical center. METHODS: Retrospective review of 1008 patients who underwent tumor resection and free flap reconstruction of the head and neck at a tertiary referral center from 2002 to 2019. RESULTS: Of 1008 patients (65.7% male; mean age of 61.4 years, SD 14.0 years), 161 (15.6%) were discharged to SNF/IPR, and the median LOS was 7 days. In multiple linear regression analysis, Charlson Comorbidity Index (CCI; P < .001), American Society of Anesthesiologists (ASA) classification (P = .021), female gender (P = .023), and inability to tolerate oral diet preoperatively (P = .006) were statistically significantly related to increased LOS, whereas age, body mass index (BMI), modified frailty index (MFI), a history of prior radiation or chemotherapy, and home oxygen use were not. Multiple logistic regression analysis demonstrated that CCI (odds ratio [OR] = 1.119, confidence interval [CI] 1.023-1.223), age (OR = 1.082, CI 1.056-1.108), and BMI <19.0 (OR = 2.141, CI 1.159-3.807) were the only variables statistically significantly related to posthospital placement in an SNF or IPR. CONCLUSION: Common tools for assessing frailty and need for additional care may be inadequate in a head and neck reconstructive population. CCI appears to be the best of the aggregate metrics assessed, with significant relationships to both LOS and placement in SNF/IPR.