Microsurgical free flaps have largely supplanted pedicled flaps as the gold standard for head and neck cancer reconstruction. However, incidence of postoperative complications after accounting for patient comorbidities based on choice of reconstruction has not been well-defined in the literature in recent years. Patients undergoing head and neck reconstruction were identified in the 2011-2016 ACS-NSQIP (American College of Surgeons National Surgical Quality Improvement Program) database and stratified into groups by free flap, myocutaneous pedicled flap, and other reconstruction. Demographics were analyzed and covariates balanced using overlap propensity score-based weighting. Logistic regression was used for binary outcomes and Gamma generalized linear model was used for length of stay. A total of 4,712 patients met inclusion criteria out of which 1,297 patients (28%) underwent free flap, 208 patients (4%) pedicled flap, and 3,207 patients (68%) had other, or no reconstruction performed. After adjusting for patient and disease-specific factors, pedicled flap reconstruction was associated with a higher risk of deep vein thrombosis (odds ratio [OR] = 2.64, confidence interval [CI] 1.02-6.85, p = 0.045), sepsis (OR = 2.95, CI 1.52-5.71, p = 0.001), and infection (OR = 2.03, CI 1.39-2.96, p <0.001) compared with free flap reconstruction. Free flaps had the longest mean operative time compared with the other two groups (unadjusted 578 vs. 440 vs. 326, p <0.001). Pedicled flaps had a lower incidence of bleeding requiring transfusion (adjusted OR = 0.65, CI 0.50-0.85, p = 0.002), and lower incidence of prolonged mechanical ventilation (adjusted OR = 0.33, CI 0.12-0.92, p = 0.034) compared with free flaps. There was no difference in rates of reoperation, hospital readmission, or hospital length-of-stay between pedicled and free flaps. Myocutaneous pedicled flaps are associated with higher overall short-term postoperative complications compared with free flaps in head and neck reconstruction, which demonstrate a more favorable morbidity profile without significantly impacting hospital readmission, reoperation, or length-of-stay.
Read full abstract