Electrical Stimulation in Obstructive Sleep Apnoea: Treatment Effect on the Apnoea-Hypopnoea-Index in Randomised Controlled Trials

<bold>Introduction:</bold> Hypoglossal nerve stimulation (HNS) is a novel treatment for patients with obstructive sleep apnoea (OSA). Despite its use in public healthcare there are limited published data from randomised controlled trials (RCT). <bold>Methods:</bold> We evaluated RCTs of HNS published up to 01/01/2024 related to HNS, and assessed the between-group difference in the apnoea-hypopnoea-index (AHI). Data are reported as treatment effect (delta) with the 95% confidence interval (CI). <bold>Results:</bold> There are five RCTs on HNS reporting the between group treatment effect (Table 1). The BLAST-OSA trial (Eastwood et al, <italic>Eur Resp J</italic> 2020), a single-arm trial that did not contain a control group, reported a treatment effect on the AHI of -10.8 (95% -14.6; -7.0) hour⁻¹. <table-wrap><object-id>erj;64/suppl_68/PA869/TB1</object-id><object-id>T1</object-id><object-id>TB1</object-id><table><colgroup><col></col><col></col></colgroup><tbody><tr><td><bold>RCT (author, publication)</bold></td><td><bold>Treatment Effect AHI (events x hour⁻¹)</bold></td></tr><tr><td>STAR trial (Strollo et al, <italic>NEJM</italic> 2014)^*</td><td>-17.3 (95%CI -20.7; -14.9)⁼</td></tr><tr><td>TESLA trial (Pengo et al, Thorax 2016)^{^}</td><td>-9.1 (95%CI -16.2; -2.0)⁺</td></tr><tr><td>Wu et al (Thorax, 2022)^{^}</td><td> -7.3 (95%CI -18.5; 3.9)⁺ -13.3 (95%CI -23.5; -3.1)^% </td></tr><tr><td>THN3 trial (Schwartz et al, JAMA Otolaryngol 2023)*</td><td> -14.4 (95%CI -18.0; -10.5)⁼ </td></tr><tr><td>TESLA home trial (Ratneswaran et al, <italic>eClinicalMedicine</italic> 2023)^{^}</td><td> -11.5 (95%CI -20.7; -2.3)⁺ unadjusted -7.0 (95%CI -15.7; 1.8)⁺ adjusted </td></tr></tbody></table></table-wrap> ^*unilateral invasive; ^{^}bilateral transcutaneous; ⁼triggered; ⁺continuous; ^%intermittent <bold>Conclusion:</bold> There are limited data from RCTs on HNS reporting the AHI treatment effect. Data suggest a greater reduction in the AHI using the invasive rather than the transcutaneous method. However, there is substantial overlap in the confidence intervals of the treatment effect, and different approaches in responder selection need to be considered.