Umeå University's logo

Background: Patients with sleep apnoea are at increased risk of postoperative cardiac and respiratory complications, but its prevalence in patients undergoing abdominal cancer surgery is poorly defined. The aim of this study was to estimate the prevalence of sleep apnoea in patients undergoing colorectal cancer surgery compared with community-based controls, and to assess the utility of symptoms, signs, and the STOP-Bang questionnaire in identifying sleep apnoea.

Methods: This prospective observational study included consecutive patients scheduled for colorectal cancer surgery at Umeå University Hospital, Sweden, between 7 January 2015 and 24 May 2020. Overnight polysomnography, including electroencephalogram-based sleep scoring, was undertaken before operation. Prevalence estimates were compared with those of two community-based cohorts from Uppsala, Sweden (female participants from the Sleep and Health in Women cohort; male participants from the Men in Uppsala: A Study of Sleep Apnoea and Cardiometabolic Health cohort). The primary outcome was the prevalence of sleep apnoea.

Results: Of 268 eligible patients, 5 were receiving continuous positive airway pressure therapy and 201 underwent successful polysomnography. Moderate-to-severe sleep apnoea was identified in 98 of 206 patients (48 (95% confidence interval 41 to 54)%) undergoing colorectal cancer surgery, with a similar prevalence in men and women. Among 597 community-based controls, moderate-to-severe sleep apnoea was present in 27 (24 to 31)%. After adjustment for age, sex, and body mass index, the odds ratio for moderate-to-severe sleep apnoea in patients with colorectal cancer was 1.57 (95% confidence interval 1.08 to 2.29; P = 0.019). The STOP-Bang questionnaire demonstrated acceptable sensitivity (85 (95% confidence interval 76 to 91)%) but low specificity (42 (33 to 52)%).

Conclusion: Approximately half of patients undergoing colorectal cancer surgery have moderate-to-severe sleep apnoea, exceeding the prevalence in community-based controls. Clinical symptoms, signs, and STOP-Bang screening are insufficient to reliably identify affected patients, indicating that overnight sleep apnoea assessment is required in this population.