Modification of Sleep Disordered Breathing after Increase in LVAD Pump Speed in HF Patients

BACKGROUND: Patients affected by heart failure (HF) exhibit a high prevalence of sleep disordered breathing (SDB), responsible of worse prognosis. Central sleep apnoea (CSA) is typical in HF and associated with Cheyne-Stokes breathing. Three factors related to HF play a pivotal role in CSA pathogenesis: hyperventilation, carbon dioxide cerebrovascular abnormal reactivity, and low cardiac output (CO) which leads to an increased circulatory time. Previous experiences proved that in patients wearing left ventricular assistance device (LVAD), a change in pump speed causes a variation in CO, compared to basal speed. In the present analysis, we tested the hypothesis that an increase in CO through pump speed change in patients wearing LVAD, leads to decrease in circulatory time and a to lower occurrence of SDB, as CSA. METHODS: Thirty-one LVAD (Jarvik 2000) patients were enrolled in the study and successfully performed at LVAD speed 2 and at 4, set up randomly, CO assessment at rest by inert gas rebreathing (IGR) technique, circulatory time evaluation, calculated under oxygen saturation control during induced apnoea-rebreathing, and polysomnographic registration to assess presence severity of SDB through the calculation of the apnoea/hypopnoea index (AHI). RESULTS: Changing LVAD pump speed from 2 to 4, CO at rest increased from 3.1±0.8 to 3.7±0.7 L/min (p=0.01), ant this was related to a decrease of circulatory time from 39.6± 8.6 to 33±6.7 (p<0.01). Polysomnographic SDB evaluation showed a AHI decrease from 19.2±15.7 to 12.8±12.1 (p<0.01), CSA index decrease from 6±9.2 to 2.2±4.7 (p<0.01) and hypopnea index from 8.3±5.9 to 4.3±4.5 (p<0.01); on the contrary an increase of obstructive apnoea (OSA) index was observed from 2.7±4.6 to 4.3±6.1 (p=0.01). CONCLUSIONS: In patients affected by HF wearing LVAD, an increase in CO with a higher LVAD pump speed is associated with a shorter circulatory time, which leads to an AHI decrease, showing therefore a consequent SDB reduction. This overall decrease pattern could be therefore considered as the main cause of a longer sleeping time, usually worsened by the presence of SDB. Finally, the increase in OSA remains an uncertain datum to be further analysed in larger HF populations.