The Clinical Characteristics of Patients With Pulmonary Hypertension Combined With Obstructive Sleep Apnoea

Lu Yan; Zhihui Zhao; Qing Zhao; Qi Jin; Yi Zhang; Xin Li; Anqi Duan; Qin Luo; Zhihong Liu


BMC Pulm Med. 2021;21(378) 

In This Article


Sleep-disordered breathing is an important risk factor for a variety of cardiovascular and metabolic diseases, and it has a high prevalence in patients with PH.[15]

OSA is difficult to detect with a routine blood gas analysis on admission because patients often have no typical complaints, such as snoring and daytime drowsiness. Fadia et al. found that sleep apnoea was very common in patients with PH. Their multivariate logistic regression analysis suggested that symptoms did not predict the occurrence of sleep-disordered breathing.[16] In our study, the Epworth score was not significantly different in PH patients with and without OSA. Most clinicians do not perform polysomnography or respiratory polygraphy if the patient does not have the typical complaint of daytime sleepiness, but this may result in a significant underestimation of OSA incidence. Very few OSA patients have daytime drowsiness, which may be associated with the high sympathetic activity in OSA.[17]

Ulrich et al. analysed 38 patients with PH, mainly IPAH, CHD-PAH and CTEPH. When AHI ≥ 10/h was defined as the cut-off value, 11% of the patients with PH presented OSA.[18] Florian et al. showed that sleep-disordered breathing was common in PAH and CTEPH patients and reflected disease severity.[19] Our study demonstrated a high incidence of OSA in patients with PH, as OSA was present in 25% (35/140) of them, which is higher than found abroad. At the same time, the use of oxygen therapy, diuretics, and targeted drugs for PH may alleviate OSA symptoms in these patients, so the incidence of OSA may still be underestimated. The differences between the studied patient populations may account for the differences in the results.

When we compared patients with and without OSA, older males were more likely to have OSA, which is consistent with earlier studies. A unique finding of our study was that a decrease in the partial pressure of arterial blood oxygen during the day could indicate an increased risk of OSA in PH patients.

When comparing the relationships between different types of PH and OSA, we found that OSA had the highest incidence in patients with lung disease– or hypoxia-related PH (76%); in CTEPH patients, the incidence of OSA was 50%. The presence of OSA in patients with CTEPH is not clearly explained, but current studies report a high incidence of OSA in patients with acute pulmonary embolism and deep vein thrombosis.[20,21]

According to our study, it is not yet possible to quantify the correlation between OSA and the severity of PH. Our study did not find that PH patients who also had OSA had more severe heart damage signs, including BNP, 6MWD, and CI. Minai et al. performed nocturnal oximetry tests on 43 patients with IPAH and CTD-PAH and found that the sleep-disordered breathing group had a higher BNP, higher mRAP, higher mPAP, higher PVR, and lower CI than the no-OSA group, suggesting that sleep-disordered breathing is associated with the progression of PH and the dysregulation of right ventricular function.[22] In our study, only mPAP showed the same pattern; that is, in PH patients with OSA, mPAP was significantly lower than that in patients without OSA, while we found no significant differences in other haemodynamic indicators. We also found no significant difference in any of the echocardiographic indices between the two groups. It cannot be concluded that OSA is related to the severity of PH, the destruction of cardiac structure or the decline of function. We also compared the CPET indicators between the two groups of patients and found no significant difference. Whether the patients had OSA did not have a significant impact on the cardiopulmonary exercise capacity of these PH patients.

Regarding the link between sleep-disordered breathing and pulmonary hypertension, the mechanism by which OSA influences the left heart is complicated. First, hypoxia causes pulmonary vasoconstriction, and pulmonary vascular resistance increases, leading to an increase in precapillary pulmonary artery pressure. In addition, during an obstructive event, the increases in chest negative pressure, venous return, right ventricular preload and stroke volume lead to an increase in pulmonary artery blood flow, which can also lead to an increase in pulmonary artery pressure. At the same time, OSA can increase the afterload of the patient's left ventricle by increasing the transmural pressure of the left ventricle and the negative pressure in the chest cavity. These factors can damage the patient's left heart function and cause an increase in pulmonary venous pressure.[23]

Sajkov et al.[24] showed that PH associated with obstructive sleep apnoea is usually mild to moderate, can be reversed by positive airway pressure therapy, and is usually associated with significant hypoxic vascular reactivity (i.e., pulmonary artery pressure and lack of oxygen exposure). In patients with fibrotic lung disease, including that which may occur in scleroderma, pulmonary artery pressure may increase due to hypoxic pulmonary vasoconstriction and destruction of the lung parenchyma. Pulmonary artery expansion does not necessarily correspond to increased pulmonary artery pressure, but expansion is related to sleep-disordered breathing.[25,26] In patients with systemic sclerosis, we should pay close attention to the association between sleep-disordered breathing and pulmonary artery dilation.[27] About parenchymal lung disease, especially chronic obstructive pulmonary disease (COPD), an important paper in the New England Journal of Medicine suggests that dilation of the pulmonary arteries may indicate the worsening of COPD.[28] We want to know whether OSA has an important causal relationship in this association, especially considering recent evidence that positive airway pressure can reduce the risk of rehospitalization in patients with frequent COPD worsening.[13] As mentioned above, the definition of PH has recently changed (i.e., the mean pulmonary arterial pressure is 20 mmHg instead of 25 mmHg), suggesting that earlier studies may have underestimated the true burden of PH. Therefore, we highly support further efforts to explore the link between PH and OSA, especially in patients with pulmonary parenchymal disease. Our study did not investigate whether PH could be alleviated by the correction of OSA. Large, prospective cohort studies are needed to further explore these issues.