Obstructive sleep apnea is common in patients with high-impact chronic pain – an exploratory study from an interdisciplinary pain center

Abstract

Abstract Objectives Sleep disturbances are increasingly recognized as a major part of chronic pain pathology. Obstructive sleep apnea (OSA) is a common occurrence in patients with chronic pain attending specialized pain clinics, yet its prevalence remains unclear. Using screening tools such as the Berlin and STOP-BANG questionnaires may aid in early identification of OSA and improve clinical care. This study i) examined the frequency of OSA based on objective sleep monitoring in patients with high-impact chronic pain, ii) explored potential differences in self-reported pain and sleep characteristics between patients with and without OSA, and iii) tested the agreement between OSA classification based on objective assessment and two OSA screening questionnaires. Methods A consecutive cohort of 90 patients (71 women and 19 men; mean age: 47.1 ± 11.0 years) referred for interdisciplinary pain treatment, underwent one night of sleep monitoring using portable respiratory polygraphy (RP), and suspected OSA was confirmed with polysomnography (PSG). Self-reported data on clinical pain (severity, pain drawings and health-related quality of life), sleep characteristics (sleep quality insomnia, sleepiness), and risk of OSA (Berlin and STOP-BANG questionnaires) were collected the day before RP assessment. Results Forty-six (51.1%) patients were classified with OSA according to RP and verified with PSG. Twenty-eight patients (31.1%) had moderate or severe OSA (apnea-hypopnea index [AHI] >15). Patients with OSA reported lower sleep quality compared with patients without OSA. Scores on pain severity, disability, quality of life, insomnia and sleepiness were comparable between patients with and without OSA. Sensitivity and specificity were 78.6 and 45.2% respectively for the Berlin questionnaire, and 71.4 and 58.1% respectively for the STOP-BANG questionnaire. The agreement for both questionnaires with objective assessment was poor-to-fair. Both questionnaires had acceptable negative predictive values but low positive predictive values reducing the clinical utility to identify patients with low OSA-risk in this sample. Conclusions The current study demonstrates a high prevalence of OSA in patients with high-impact chronic pain referred to specialized pain treatment, however the clinical pain profiles were similar in patients with and without OSA. The Berlin and STOP-BANG questionnaires have poor specificity and low-to-fair agreement with RP/PSG questioning their clinical utility in identifying OSA in this sample.