Purpose: The adverse effects of smoking and nicotine on pedicled and microvascular free flaps are well documented in the literature. We set out to determine if a metabolite of nicotine, cotinine, is an objective predictor of patients at increased risk for flap complications. Cotinine was chosen for its reliability, relatively long half-life (18 hours) and ease of measurement.

Methods: Eighty-three patients with stage III and IV SCCHN (squamous cell carcinoma of the head and neck) with 95 myocutaneous, pedicled or free flaps were chosen for evaluation. Patients who underwent flap reconstruction were selected from an existing database of over 500 patients with SCCHN who participated in a prospective epidemiologic study. A retrospective review of the patients' chart and preoperative standardized epidemiologic questionnaire was conducted and the results statistically analyzed. Complication rates, self reported smoking history and serum cotinine levels were compared. Cotinine levels were determined using a competitive micro-plate immunoassay. Complications were defined as any adverse postoperative wound outcome, stricture, fistula, vascular compromise requiring reoperation and partial or complete flap loss. These were further categorized according to the degree of severity.

Results: 48.5 percent (16/33) of patients with elevated (> 10 ng/ml) cotinine levels developed postoperative complications compared with 22.0% (11/50) of those patients with low (< 10 ng/ml) serum cotinine levels (P = 0.012). The relative risk of all wound complications was more than three fold higher in patients with elevated preoperative cotinine levels (odds ratio = 3.34, 95% CI [1.15-9.72]). Furthermore, the risk of wound complications increased with cotinine level (odds ratio = 2.36 and 3.86 for intermediate and high levels, respectively). The risk for complications remained associated with cotinine level after segregation by comorbidities. Furthermore, when the analysis was restricted to incident (newly diagnosed, previously untreated) patients (n=74), findings were consistent (OR = 2.71, p = 0.047).

Discussion: Serum cotinine may be an accurate predictor of increased risk of wound complications in head and neck reconstruction. Cotinine may serve as an objective, easily measured variable available to target those patients who may benefit from a delay in surgery and an aggressive smoking cessation program. This may have implications in both reconstructive and cosmetic surgery. Further study with larger sample size and adjustment for stage and medical comorbidities is underway.