Auto-trilevel versus bilevel positive airway pressure ventilation for hypercapnic overlap syndrome patients

Although bilevel positive airway pressure (Bilevel PAP) therapy is usually used for overlap syndrome (OS), there is still a portion of OS patients in whom Bilevel PAP therapy could not simultaneously eliminate residual apnea events and hypercapnia. The current study was expected to explore whether auto-trilevel positive airway pressure (auto-trilevel PAP) therapy with auto-adjusting end expiratory positive airway pressure (EEPAP) can serve as a better alternative for these patients.

From January of 2014 to June of 2016, 32 hypercapnic OS patients with stable chronic obstructive pulmonary diseases (COPD) and moderate-to-severe obstructive sleep apnea syndrome (OSAS) were recruited. Three variable modes of positive airway pressure (PAP) from the ventilator (Prisma25ST, Weinmann Inc., Germany) were applicated for 8 h per night. We performed the design of each mode at each night with an interval of two nights with no PAP treatment as a washout period among different modes. In Bilevel-1 mode (Bilevel-1), the expiratory positive airway pressure (EPAP) delivered from Bilevel PAP was always set as the lowest PAP for abolishment of snoring. For each patient, the inspiratory positive airway pressure (IPAP) was constantly set the same as the minimal pressure for keeping end-tidal CO₂ (ETCO₂) ≤45 mmHg for all three modes. However, the EPAP issued by Bilevel PAP in Bilevel-2 mode (Bilevel-2) was kept 3 cmH₂O higher than that in Bilevel-1. In auto-trilevel mode (auto-trilevel) with auto-trilevel PAP, the initial part of EPAP was fixed at the same PAP as that in Bilevel-1 while the EEPAP was automatically regulated to rise at a range of ≤4 cmH₂O based on nasal airflow wave changes. Comparisons were made for parameters before and during or following treatment as well as among different PAP therapy modes. The following parameters were compared such as nocturnal apnea hypopnea index (AHI), minimal SpO₂ (minSpO₂), arousal index, sleep structure and efficiency, morning PaCO₂, and daytime Epworth Sleepiness Scale (ESS).

Compared with the parameters before PAP therapies, during each mode of PAP treatment, significant reduction was detected in nocturnal AHI, arousal index, morning PaCO₂, and daytime ESS while significant elevation was revealed in nocturnal minSpO₂ and sleep efficiency (all P < 0.01). Comparison among three PAP modes indicated that under the same IPAP, the auto-trilevel PAP mode could result in the lowest arousal index, daytime ESS, and the highest sleep efficiency. Compared with Bilevel-1, it was detected that (a) AHI was lower but minSpO₂ was higher in both Bilevel-2 and auto-trilevel (all P < 0.05) and (b) morning PaCO₂ showed no statistical difference from that in auto-trilevel but displayed higher in Bilevel-2 (P < 0.05). Compared with Bilevel-2, in auto-trilevel, both AHI and minSpO₂ showed no obvious changes (all P > 0.05) except with a lower morning PaCO₂ (P < 0.05).

Auto-trilevel PAP therapy was superior over conventional Bilevel PAP therapy for hypercapnic OS patients with their OSAS moderate to severe, since auto-trilevel PAP was more efficacious in synchronous elimination of residual obstructive apnea events and CO₂ retention as well as in obtaining a better sleep quality and milder daytime drowsiness.