High prevalence of obstructive sleep apnea among patients with head and neck cancer.

OBJECTIVES: To determine the prevalence of obstructive sleep apnea (OSA) in patients with cancer of the oral cavity and oropharynx scheduled for primary surgical resection. To correlate the presence of OSA and the occurrence of postoperative morbidities in this patient population.

METHODS: This was a prospective study involving 17 patients with malignancies of the oral cavity and oropharynx scheduled for primary surgical resection. Consecutive patients were approached to undergo overnight polysomnography to determine the apnea-hypopnea index (AHI). OSA was defined by an AHI value > or = 20 events per hour. Postoperative morbidities were evaluated in a blinded fashion for the patients completing surgery.

RESULTS: OSA was present in 13 of 17 patients, yielding a striking prevalence of 76% in this patient group. The mean AHI for patients with OSA was 44.7 +/- 3.5 (standard error) events per hour, with a mean nadir oxygen saturation of 88.2 +/- 1.8%, consistent with moderate to severe sleep-disordered breathing. The OSA and non-OSA patients were similar with respect to age and body mass index. The mean size of the primary tumour was 3.3 cm in patients with an AHI < 20 and 3.5 cm in those with an AHI > or = 20 (p = not significant). Overall, postoperative complications, defined as prolonged intensive care unit stay (> 24 hours), need for mechanical ventilation, and cardiopulmonary morbidities, were observed in 67% of OSA and 25% of non-OSA patients.

CONCLUSIONS: These findings point to a strong association between OSA and malignancies of the oral cavity and oropharynx. This relationship was independent of the size of the primary malignancy in this patient population with tumours ranging from 1 to 7 cm (p = not significant). When comparing the two groups (AHI < 20 and AHI > or = 20), there was a tendency for the group with OSA to have an increase in postoperative morbidities. Further research is warranted to further evaluate the postoperative morbidities and mortalities associated with OSA in this patient population and to determine the potential roles for preoperative treatment with continuous positive airway pressure and tracheotomy.