Indexed on: 25 Jan '17Published on: 25 Jan '17Published in: Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery
Objective To determine which complications, as defined by the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database, correlate with 30-day mortality in surgery for malignancies of the head and neck. Study Design Retrospective review of prospectively collected national database. Setting NSQIP. Subjects and Methods NSQIP data from 2005 to 2014 were queried for ICD-9 codes head and neck malignancies. Multivariate logistic regression was used to examine the correlation of individual complications with 30-day mortality. Results In total, 15,410 cases met criteria with 3499 complications in 2235 cases. After controlling for patient and surgical variables, postoperative pneumonia ( P = .02; odds ratio [OR], 2.39; 95% confidence interval [CI], 1.15-4.72), progressive renal insufficiency ( P < .001; OR, 21.28; 95% CI, 4.22-87.94), bleeding requiring transfusion ( P = .02; OR, 2.10; 95% CI, 1.12-3.84), sepsis ( P = .02; OR, 2.86; 95% CI, 1.15-6.46), septic shock ( P = .045; OR, 2.87; 95% CI, 0.98-7.81), stroke ( P < .001; OR, 19.81; 95% CI, 6.23-56.03), and cardiac arrest ( P < .001; OR, 135.59; 95% CI, 65.00-286.48) were independently associated with increased odds of 30-day mortality. Conclusion The NSQIP database has been extensively validated and used to examine surgical complications, yet there is little analysis on which complications are associated with death. This study identified complications associated with increased risk of 30-day mortality following head and neck cancer surgery. These associations may be used as a measure of complication severity and should be considered when using the NSQIP database to evaluate outcomes in head and neck surgery.