Association between obstructive sleep apnea and cancer incidence in a large multicenter Spanish cohort.

Rationale: Obstructive sleep apnea (OSA) has been associated with increased cancer mortality, but whether it is also associated with cancer incidence is unknown.

Objectives: To investigate whether OSA is associated with increased cancer incidence in a large clinical cohort.

Methods: A multicenter, clinical cohort study including consecutive patients investigated for suspected OSA between 2003 and 2007 in seven Spanish teaching hospitals. Apnea-hypopnea index (AHI) and percent nighttime with oxygen saturation less than 90% (TSat(90)) were used as surrogates of OSA severity, both as continuous variables and categorized by tertiles. Cox proportional hazards regression analyses were used to calculate hazard ratio (HR) and 95% confidence interval (CI) for cancer incidence after adjusting for confounding variables.

Measurements And Main Results: A total of 4,910 patients were analyzed (median follow-up, 4.5 yr; interquartile range, 3.4-5.2). Compared with the lower TSat(90) category (<1.2%), the adjusted hazards (95% CI) of cancer incidence for increasing categories were 1.58 (1.07-2.34) for TSat(90) 1.2-12% and 2.33 (1.57-3.46) for TSat(90) greater than 12%. Continuous TSat(90) was also associated with cancer incidence (adjusted HR, 1.07 [1.02-1.13] per 10-unit increase in TSat(90)). In stratified analyses, TSat(90) was associated with cancer incidence in patients younger than 65 years (adjusted HR, 1.13 [95% CI, 1.06-1.21] per 10-unit increase in TSat(90)) and males (adjusted HR, 1.11 [95% CI, 1.04-1.17] per 10-unit increase in TSat(90)). AHI was not associated with cancer incidence in the adjusted analyses, except for patients younger than 65 years (adjusted HR for AHI >43 vs. <18.7, 1.66; 95% CI, 1.04-2.64).

Conclusions: Increased overnight hypoxia as a surrogate of OSA severity was associated with increased cancer incidence. This association seems to be limited to men and patients younger than 65 years of age.