Diagnosing sleep disordered breathing in patients with chronic pulmonary disease: which test for which patient?

Obstructive pulmonary disease

Patients with COPD usually report poor sleep quality [4]. The prevalence of OSA in COPD patients is similar to that in the general population. However, patients with COPD–OSA overlap syndrome have a poorer prognosis [5] and lower quality of life [6] than those with COPD alone. Undiagnosed OSA increases the risk of acute exacerbation [7] and readmission rates [3, 8]. In patients with severe to very severe COPD, nocturnal hypoventilation occurs regardless of daytime hypercapnia [9]. However, screening for nocturnal hypoventilation remains limited to research purposes, as no trial has yet evaluated the benefit of correcting such nocturnal hypoventilation.

As for the general population, patients with asthma are more and more frequently overweight [10] and can therefore suffer comorbid OSA. Comorbid OSA is associated with an increased risk of an asthma exacerbation [11]. Untreated OSA may contribute to worsening daytime control of the disease [12] and should therefore be screened in patients with poorly controlled asthma. Initiating continuous positive airway pressure (CPAP) in patients with asthma and comorbid OSA improves the asthma-related quality of life [13] as well as nocturnal symptoms [14].

Diagnosis and management of sleep disordered breathing in obstructive pulmonary disease is further discussed in two recently published reviews [15, 16].

Restrictive pulmonary diseases

Restrictive pulmonary diseases are often associated with sleep disordered breathing [17, 18]. Indeed, functional residual capacity decreases during sleep [19], which can promote obstructive respiratory events [20] and contribute to hypoventilation [21].

Patients with restrictive pulmonary disease due to obesity are at high risk of OSA and should therefore be screened for OSA, especially before undergoing bariatric surgery. In this population, daytime arterial carbon dioxide tension (P_aCO2) measurement is also recommended to rule out obesity hypoventilation syndrome [22].

Patients with restrictive pulmonary disease due to a neuromuscular disorder are at high risk of developing hypoventilation and require close follow-up, particularly those suffering from a rapidly progressive disease such as amyotrophic lateral sclerosis [23]. Prevalence of OSA varies according to the underlying restrictive disease [24, 25]. In addition, patients with restrictive disease can develop central events related to heart failure [26].

Patients with restrictive disease secondary to an interstitial lung disease have a high prevalence of OSA [27]. However, the use of CPAP may have limited impact on disease progression and survival [28]. In patients with end-stage interstitial lung disease, hypoventilation may occur. However, NIV is rarely used.

Overnight oximetry (type 4 sleep monitoring device)

Overnight oximetry should be considered as a screening tool rather than a diagnostic tool, as it cannot distinguish obstructive and central respiratory events. However, by continuously monitoring peripheral oxygen saturation (S_pO2), it can provide useful information, including the frequency of oxygen desaturation, the mean oxygen saturation, and the time spent with a saturation <90%. Hence, apnoeas or hypopnoeas can be suspected on a nocturnal oximetry by the presence of rapid and repeated desaturation and recovery episodes over time (figure 1b). In patients who do not have a ventilation/perfusion mismatch, overnight oximetry can also identify nocturnal hypoventilation, as highlighted in figure 1a where prolonged desaturations suggest alveolar hypoventilation. In addition to oxygen measurements, overnight oximetry provides information on heart rate. Oximetry monitoring devices are cheap, easy to use and can be performed in the home setting without supervision.

Overnight capnography (type 4 sleep monitoring device)

Overnight capnography allows continuous monitoring of transcutaneous carbon dioxide level (P_tcCO2) in addition to S_pO2. Overnight capnography is easy to use and provides additional information compared with oximetry, including mean P_tcCO2 level overnight and times spent above P_tcCO2 thresholds (50 or 55 mmHg). Therefore, overnight capnography is essential for the diagnosis of nocturnal hypoventilation, as shown in figure 1c for a patient with amyotrophic lateral sclerosis displaying significant alveolar hypoventilation when looking at the P_tcCO2 trace with a close to normal oximetry trace. The limitations of overnight capnography include its cost (the test is more expensive than oximetry), a high failure rate, and a lack of consensus for the cut-offs defining hypoventilation.

Overnight respiratory polygraphy (type 3 sleep monitoring device)

Overnight respiratory polygraphy adds respiratory efforts and airflow to oximetry. This allows you to distinguish obstructive from central respiratory events. Overnight respiratory polygraphy is widely available and can be easily performed in the home setting. However, as it cannot evaluate sleep stages or micro-arousals, and the frequency of respiratory events can be underestimated.

Overnight polysomnography (type 1 or 2 sleep monitoring device)

Overnight polysomnography is the gold standard to assess sleep disordered breathing as it provides an assessment of sleep stages (using electroencephalography, electro-oculography and chin electromyography), sleep structure, sleep efficiency and sleep disruption (arousals, wake time after sleep onset). Type 2 polysomnography is performed in the home setting and is unsupervised, whereas type 1 polysomnography is performed in hospital with a sleep technician monitoring the patient. The main limitations of overnight polysomnography are its costs and its availability given the prevalence of respiratory diseases.

Respiratory extended overnight polysomnography (combined with P_tcCO2 and respiratory electromyography)

In patients with chronic pulmonary disease, polysomnography recordings should also include transcutaneous capnography due to the limits of S_pO2 measurement in chronic hypoxaemic diseases. Polysomnographic recordings identify during which sleep stage hypoventilation occurs (usually during rapid eye movement sleep, during which activity of accessory respiratory muscle is lost (figure 1d)). In addition, one may assess the inspiratory muscles activity during sleep, mostly using neck inspiratory muscles (scalene or sterno-cleido-mastoidian muscles) as the diaphragm muscle activity is poorly accessible at the body surface [30, 31].

Diagnostic strategy in patients with obstructive pulmonary disease

In patients with COPD, overnight oximetry poorly identifies comorbid OSA [32]. Therefore, it should not be used. In patients where OSA is not suspected, assessment of the time spent with a saturation <90% using overnight oximetry is not clinically relevant as no benefit of overnight oxygen therapy has been shown [33, 34]. When overnight polysomnography is not available, home polygraphy can be used to diagnose comorbid OSA in COPD patients [35]. A similar approach may be applied for patients with asthma; however, no study has assessed any diagnostic strategy in this population. In patients with end-stage COPD, there is no indication to perform overnight capnography to detect nocturnal hypoventilation, as until now all randomised controlled trials suggesting a benefit of NIV used daytime P_aCO2 (and not overnight capnography) as a criterion for NIV initiation [36, 37].

Diagnostic strategy in patients with restrictive pulmonary disease

In patients with restrictive pulmonary disease due to obesity, overnight polygraphy should be used as a first-line diagnostic tool. However, overnight polysomnography followed by multiple sleep latency tests should be performed in patients with rare (mostly hypothalamic) causes of obesity, including type 1 narcolepsy, craniopharyngioma, neurosarcoidosis and Prader–Willi syndrome (who suffer from comorbid central hypersomnia and narcolepsy) [38]. In the general obese population, there is no indication for capnography monitoring alone.

Overnight capnography is probably only useful in patients with restrictive pulmonary disease associated with a neuromuscular disorder, and especially in those with a rapidly progressive disease for which the regular recommended overnight monitoring cannot be performed by overnight polysomnography [23]. However, given the technical challenges associated with home capnography, a pragmatic approach is to perform regular home oximetry in patients with rapidly progressive neuromuscular disease. When overnight capnography can be performed, in this clinical condition, the cut-off for significant nocturnal hypoventilation is debatable [39] and may have different clinical impacts according to the underlying disease [40]. In patients with restrictive pulmonary disease associated with an interstitial lung disease, given the lack of dedicated trials, OSA should primarily be screened with the use of overnight polygraphy.

Finally, overnight polysomnography (followed by multiple sleep latency tests) needs to be performed in some groups of patients including those suffering from myotonic dystrophy who, in addition to weak muscles, have impaired central breathing control and secondary narcolepsy [41].