Screening Strategies for Sleep-Disordered Breathing in Patients With Spinal Cord Injury in a Tertiary Care Rehabilitation Center

BACKGROUND: Sleep-disordered breathing (SDB) is frequent in patients with spinal-cord injury (SCI). However, SDB is frequently underdiagnosed due to limited access to diagnostic testing and knowledge about the condition. Moreover, SDB heterogeneity (sleep apnea, obstructive sleep apnea or central sleep apnea and nocturnal alveolar hypoventilation) implies complex evaluation of both nocturnal respiratory effort and hypercapnia. The aim of this study was to compare different screening strategies for an SDB diagnosis in patients with SCI. METHODS: This was a retrospective analysis of data from subjects with SCI followed up in a tertiary-care rehabilitation center with a specialized sleep unit. Subjective (questionnaires) and objective data (polysomnography [PSG]), extracted from the PSG, morning blood gases, and nocturnal transcutaneous CO<INF>2</INF> (P<INF>tcCO<INF>2</INF></INF>) were collected and analyzed. A retrospective comparison of different strategies for SDB screening was carried out. Each strategy was compared (alone and in combination) with the standard of care for sleep apnea (PSG) and nocturnal alveolar hypoventilation (P<INF>tcCO<INF>2</INF></INF>) diagnosis. The performance of the usual cutoff and visual analysis was studied. RESULTS: Among 190 subjects with SCI who underwent a full night's PSG, data were available for 104 questionnaires and 162 with oximetry. Nocturnal alveolar hypoventilation was screened by P<INF>tcCO<INF>2</INF></INF> and blood gases in 52 subjects with SCI. Questionnaires (the modified Screening for Obstructive Sleep Apnea in Tetraplegia and the Epworth Sleepiness Scale) had poor performance for identifying sleep apnea and did not identify nocturnal alveolar hypoventilation. (oxygen desaturation index score >= 13) and visual analysis of were good at identifying sleep apnea but insufficient to identify nocturnal alveolar hypoventilation. Diurnal blood gases were poor predictors of nocturnal alveolar hypoventilation. CONCLUSIONS: Questionnaires were of limited use in subjects with SCI, but the oxygen desaturation index derived from oximetry performed well for sleep apnea screening. Both diurnal blood gases and oximetry visual analysis were insufficient for nocturnal alveolar hypoventilation screening. P<INF>tcCO<INF>2</INF></INF> monitoring should be mandatory and ideally combined with PSG given the heterogeneity of SDB phenotypes and associated sleep comorbidities of patients with SCI.