Abstract: Background: Wisconsin Sleep Cohort recently found an association between sleep apnea (SA) and cancer mortality in a community-based study. Objective: To investigate whether SA is associated with increased cancer mortality in a large clinic cohort. Methods: Multicenter study of 5,618 patients referred to 7 Spanish Sleep Clinics for suspected SA. Apnea-hypopnea index (AHI) and percent sleep time spent with 02sat<90% (hypoxemia index [HI]) were used as a surrogate of SA severity, both as quantitative values or divided into tertiles (first tertile as a reference risk value). Cox proportional hazards model was used to calculate full-adjusted (age, gender, alcohol intake, body mass index and smoke habit) OR [95%CI] p value- for cancer mortality. Results: 5,467 patients were analyzed. Mean age 55.2 yrs; 65% males; median follow-up: 4.5 years with 375 deaths (92 from cancer). A greater tertile of HI (>14%), but not of AHI, was associated with increased full-adjusted cancer mortality: OR 1.94, [1.001 to 3.75]; p=0.049. This association was stronger in the 3,232 patients without CPAP treatment or poor compliance (<4 hrs/day): OR 2.56 [1.08 to 6.06], p=0.032. When stratified by gender and age (cut-off 65 years), only men (OR 2.56 [1.08 to 6.06], p=0.032) and younger patients (OR 7.3 [1.6 to 33], p=0.01) for greater HI tertiles and AHI as a quantitative value (only in younger patients, OR 1.02 [1 to 1.02], p=0.047) were associated with an increased risk of cancer mortality.. Conclusions: Sleep apnea severity was associated with increased cancer mortality. This association seems stronger in males and younger patients.

Authorship

Download reference