Возможности диагностики и частота выявления нарушений дыхания во время сна у больных с имплантированными кардиовертерами-дефибрилляторами

Objective. To assess diagnostic efficacy of screening methods for sleep-disordered breathing — questionnaires (assessing the risk of obstructive sleep apnea and daytime sleepiness), оvernight pulse oximetry and apnea detection algorithm embedded in implantable cardioverter-defibrillators (ICDs) in patients with ICDs. Material and methods. The study included 67 patients aged 34-83 years (median 65). There were 54 (81%) men, median body mass index was 29.4 kg/m². All patients underwent CRM to detect sleep-disordered breathing. Screening performance of various diagnostic methods for sleep apnea detection was assessed in 39 patients using STOP-Bang questionnaire and Epworth Sleepiness Scale (ESS), in 20 patients — using computer-based pulse oximetry (РО), in 7 patients — via interpretation of respiratory disturbance index obtained from ICDs of specific models. Results of these screening methods were compared with CRM data. Results. According to CRM data, sleep-disordered breathing was detected in 64 (95.5%) patients (mild in 23 (34.5%), moderate in 22 (33%), and severe in 19 (28%) cases). In all cases, sleep-disordered breathing was obstructive, but 19 (28%) patients had combination with significant (more than 5 events/hour) episodes of central apnea. Compared to CRM data, sensitivity of STOP-BANG questionnaire in diagnosis of sleep-disordered breathing was 92% (specificity 33%). Sensitivity of ESS was 44%. Sensitivity of computer-based pulse oximetry for detecting sleep-disordered breathing was 100%. In 3 cases, computer-based pulse oximetry substantially underestimated severity of sleep-disordered breathing compared to CRM. Overestimation was noted in other 3 cases. In all 7 patients in whom respiratory disturbance indices were obtained from ICDs, mean overnight values indicated sleep-disordered breathing that was confirmed by CRM. In 3 patients with mild SDB confirmed by CRM, apnea detection algorithm in ICD markedly overestimated respiratory event index. Conclusion. In patients with ICDs, SDB is diagnosed in 95.5% of cases with moderate-to-severe forms identified in 61% patients. Compared to CRM data, STOP-BANG and ESS scales in patients with ICDs are not reliable in identifying SDB and cannot be recommended as the main screening methods. High sensitivity of computer-based pulse oximetry in detecting SDB, when compared with CRM data, supports recommendation of this method for screening in patients with ICDs.