Introduction

Noninvasive ventilation (NIV) using bi-level or continuous positive airway pressure (CPAP) at home may improve outcomes in patients with obstructive sleep apneas [1] or chronic obstructive pulmonary disease (COPD) [2]. However, numerous patients admitted to ICU for acute-on-chronic respiratory failure have never undergone sleep studies or pulmonary function tests. We decided to assess the prevalence of sleep apnea and to identify the underlying chronic lung disease among patients admitted to ICU for acute hypercapnic respiratory failure.

Methods

We conducted a prospective single-center cohort study in a French teaching hospital. We included patients admitted to ICU for acute hypercapnic respiratory failure (pH < 7.35 and PaCO2 > 45 mmHg) and treated with invasive mechanical ventilation or NIV. The study was approved by the institutional review board. All patients gave written informed consent. Complete polysomnography and pulmonary function tests were performed 3 months after ICU discharge (after at least 3 days of NIV withdrawal at home). Sleep apnea syndrome was defined according to the 2012 American Academy of Sleep Medicine [3].

Results

Over a 15-month period, 35 patients were included in the study (Table 1). Among them, only 16 (46%) came back for a complete exploration 3 months later while five patients died within the 3 first months after ICU discharge, and 14 patients refused to participate despite initial agreement (40% of enrolled patients). Patients who underwent polysomnography were younger than those who refused to participate [66 years (61–71) vs. 74 (70–75), p = 0.008].

Table 1 Characteristics of the 35 patients included in the study
Full size table

Among the 16 patients studied, all had sleep apnea syndrome with a median apnea–hypopnea index of 33 events per hour of sleep. The prevalence of severe sleep apnea (≥ 30 events/h) was 56% in the whole population (nine out of 16 patients), 50% in obstructive and 71% in restrictive patients. Eleven of the 16 patients studied (69%) were discharged with home NIV.

Discussion

In our study all ICU patients who accepted polysomnography after an acute hypercapnic respiratory failure episode had sleep apnea and more than half of them had severe sleep apnea syndrome. However, this prevalence has not been evaluated on the whole population and our preliminary report needs to be confirmed by more large studies.

To our knowledge, only one other study performed in a single center in Switzerland assessed the prevalence of sleep apnea after ICU discharge in hypercapnic patients [4]. In keeping with our results, the prevalence of severe sleep apnea was 51% in COPD patients and 81% in non-COPD patients [4]. In this study, only 47% of the patients who were planned for sleep exploration actually underwent polysomnography. Similarly, we found that 40% of the included patients (14 out of 35 patients) declined their participation despite initial agreement, highlighting the difficulty for follow-up of these patients.

In conclusion, we found a high prevalence of sleep apnea in patients with acute hypercapnic respiratory failure. As many patients were reluctant to come back for sleep explorations after hospital discharge, sleep could be explored by polysomnography while the patient is still hospitalized or by in-home devices using limited-channel sleep studies [5].