What Is the Proportion of Undiagnosed Obstructive Sleep Apnea in a Surgical Population?

Abstract

Introduction: Obstructive sleep apnea (OSA) affects approximately 2-26% of general population (1), about 90% of whom remain undiagnosed (2). Since OSA is associated with an increased risk of perioperative complications, we conducted a single-centre cross-sectional study to evaluate the proportion of undiagnosed OSA in surgical patients. Methods: Following REB approval, patients visiting preoperative clinics were approached. The 840 who gave consent subsequently underwent laboratory polysomnography (PSG) (n=240) or portable PSG at home with Embletta X-100 (n=600) preoperatively. Apnea-hypopnea index (AHI) was determined from PSG data, and severity of OSA categorized as mild (AHI >5 to <15), moderate (AHI >15 to <30) or severe OSA (AHI >30). These patients' charts were also reviewed to ascertain whether they had been diagnosed with OSA by surgeons and anesthesiologists. Of 840, 723 patients with complete data were included in the analysis. “Definite OSA” was defined as prior documented diagnosis of OSA based on a formal PSG study performed exclusive of this study, and “Probable OSA” as features suggesting presence of OSA based on clinical suspicion. Results: The sample, of which 52% (n=380) were female, had a mean age of 59 years (SD 13) and mean BMI of 31 kg/m2 (SD 7). The types of included surgery were mainly orthopedic (49%) and general (18%). Of the 723 patients included, 65% (n=473) had OSA based on PSG data (AHI>5). Notably, 84% (n=398) of these patients had reported at least one symptom either of snoring, daytime sleepiness or observed apnea while asleep. Of the 473 patients who were found to have OSA on PSG testing, surgeons diagnosed 9.7% (n=46) with OSA, with 8.4% (n=40) having definite OSA and 1.3% (n=6) having probable OSA. Definite OSA was identified by surgeons in 5.5% (12/219), 9.8% (13/132) and 12.3% (15/122) patients having mild, moderate and severe OSA, respectively. Conversely, probable OSA was identified in only 0.9% (2/219), 0.8% (1/132) and 2.5% (3/122) patients with mild, moderate and severe OSA, respectively. By comparison, of the 473 patients with OSA on PSG testing, anesthesiologists diagnosed 40.4% (n=191) with OSA, with 20.3% (n=96) having definitive OSA, and 20.1% (n=95) having probable OSA. Anesthesiologists identified definite OSA in 15.5% (34/219), 18.2% (24/132) and 31.2% (38/122) patients having mild, moderate and severe OSA, respectively. Conversely, they identified probable OSA in 16.4% (36/219), 21.2% (28/132) and 25.4% (31/122) patients having mild, moderate and severe OSA, respectively. Overall, 90.3% (n=427) of patients with OSA on PSG testing were not diagnosed by surgeons, whereas 59.6% (n=282) were not diagnosed by anesthesiologists. Discussion: Proportion of undiagnosed OSA remains high among patients awaiting elective surgery, ranging from 59.6% (diagnosis by anesthesiologists) to 90.3% (diagnosis by surgeons). More than 80% of the included patients had at least one symptom suggestive of OSA. Screening questionnaires may be useful in reducing the percentage of undiagnosed OSA among surgical patients.