Analysis of cardiopulmonary coupling to assess adaptive servo-ventilation success in complex sleep apnea management

Background: Adaptive servo-ventilation (ASV) is used to treat complex sleep apnea syndrome (CompSAS), but with variable success. Factors influencing success are poorly understood. ASV devices determine their output based upon characteristics of a given breath and on proprietary algorithms that assume a periodic breathing pattern. Periodic breathing patterns produce elevated narrow band low-frequency cardiopulmonary coupling (eNB-LFC). Therefore, we hypothesized that ASV success would correlate with elevated proportions of periodic breathing as marked by eNB-LFC on cardiopulmonary coupling (CPC) analysis. Methods: This was a retrospective study of 106 consecutive patients presenting to an academic tertiary care sleep center with CompSAS between July 2008 and July 2009 who underwent ASV titration with polysomnographic signals amenable to CPC analysis. Results: The study included 89 males (84 %) and 17 females (16 %), with mean age of 63.3 years. Median diagnostic apnea-hypopnea index (AHI) was 38 (21, 56)/h, and on continuous positive airway pressure (CPAP), the median residual AHI (CompSAS) was 36.5 (23, 58)/h, with central apneas occurring on average 22.5 (13, 39)/h. ASV brought AHI to 11.0 ± 13.0, with success in 81.1 % of patients, as defined by an AHI of <10/h. NB-LFC was elevated (>0) in 45.3 %; however, the percentage of eNB-LFC did not correlate with ASV treatment success (p = 0.518). No clinical factors were found to be associated with ASV success. Conclusion: ASV was successful in 81 % of patients with CompSAS. However, eNB-LFC calculated from CPC, a marker for periodic breathing, did not correlate with ASV success and therefore may not be a useful tool to predict ASV success.