Free AccessCommentaryObstructive sleep apnea and pulmonary hypertension: the pendulum swings again Jenny Z. Yang, MD, Babak Mokhlesi, MD, Omar A. Mesarwi, MD Jenny Z. Yang, MD Address correspondence to: Jenny Z. Yang, MD, 9300 Campus Point Dr., MC 7381, La Jolla, CA 92037; Email: E-mail Address: [email protected] Division of Pulmonary, Critical Care, and Sleep Medicine and Physiology, University of California, San Diego, La Jolla, California; Search for more papers by this author , Babak Mokhlesi, MD Division of Pulmonary, Critical Care, and Sleep Medicine, Department of Internal Medicine, Rush University Medical Center, Chicago, Illinois Search for more papers by this author , Omar A. Mesarwi, MD Division of Pulmonary, Critical Care, and Sleep Medicine and Physiology, University of California, San Diego, La Jolla, California; Search for more papers by this author Published Online:February 1, 2023https://doi.org/10.5664/jcsm.10454SectionsPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutINTRODUCTIONObstructive sleep apnea (OSA) is highly prevalent1 and has been shown to be an independent risk factor for multiple adverse cardiovascular outcomes, including hypertension, heart failure, coronary heart disease, and arrhythmias.2 However, a clear link demonstrating causality between OSA and pulmonary hypertension (PH) is less well entrenched in the literature. Over 25 years ago, a major study in which 220 patients with OSA were evaluated for PH using right heart catheterization (RHC) concluded that OSA plays only a very minor role in the development of PH.3 The authors suggested that concurrent obstructive lung disease and daytime hypoxemia were more critical for the development of PH than OSA severity. A few years later though, Sajkov et al4 carefully demonstrated that after excluding individuals with lung disease or hypoventilation, there was a persistent and perhaps clinically important association between PH and OSA, though in general OSA was only associated with mild to moderate PH, and elevations in pulmonary arterial pressures improve with the initiation of continuous positive airway pressure therapy.5 Consequently, in 1998 at the Second World Symposium of PH6, sleep-disordered breathing was classified as an associated condition of group 3 (hypoxemia-related) PH. At our institutions and many others worldwide, practice shifted toward diagnosis and treatment of sleep-disordered breathing in those with PH.Since then, multiple studies have re-evaluated the association between OSA and PH, but there remains equipoise on mechanism and controversy on causality. Samhouri et al7 attempted to address some of these issues by performing a retrospective review of PH patients who had both polysomnogram and RHC data. Sleep time spent with peripheral oxyhemoglobin saturation (SpO2) less than 90% (T90) was found to be higher among patients with all subtypes of PH and was associated with higher mean pulmonary artery pressure (mPAP), pulmonary vascular resistance (PVR), and right atrial pressure as well. However, this study was limited by a lack of comprehensive sleep data, and it remained unclear whether a higher T90 in this population was simply a marker of PH severity, instead of—or in addition to—a manifestation of OSA severity.In this issue of the Journal of Clinical Sleep Medicine, Huang et al8 provide some data in order to clarify a few of these points. In this study, patients who were suspected of having PH based on echocardiographic findings were hospitalized and then underwent both an RHC and polysomnogram. Patients were stratified by presence of OSA and PH, which was defined as mPAP ≥ 25 mm Hg. PH was further categorized as pre- or postcapillary, based on a pulmonary artery wedge pressure threshold of 15 mm Hg. Those with predominantly central sleep apnea were excluded, as were those with pulmonary disease requiring supplemental oxygen. Of the 563 patients remaining, 358 did not have OSA and 205 did have OSA. Although the authors primarily focused their attention toward those with OSA, they report data about those without OSA as well. In the OSA cohort, prevalence of PH was 65% (85% of those having precapillary PH), whereas prevalence of PH in the non-OSA cohort was somewhat surprisingly higher (84%). Similar to the study by Samhouri et al,7 T90 was closely correlated with pulmonary hemodynamics. mPAP, PVR, and right ventricular stroke work indices were all worse as a function of increasingly higher tertile of T90, and this relationship held after adjustment for common covariates (age, sex, body mass index, presence of left heart disease, etc). Correspondingly, patients with PH also had more severe hypoxemia on their sleep studies, whether quantified by mean SpO2, T90, nadir SpO2, or apnea time, although respiratory event index, a more traditional measure of OSA severity, was not worse among those with PH.One major strength of this study, compared to previous examinations of the OSA-PH relationship, is that we have some basic information about those patients without OSA too. Having these data is important, as we finally can get some insight about the “chicken or the egg” nature of the interaction between hypoxemia in OSA and PH. Is the hypoxemia in OSA, previously noted to be correlated with PH severity, occurring due to the presence of OSA pushing PA pressures higher, or is this an epiphenomenon of more severe PH? In patients without OSA from this study, there was a statistical but somewhat trivial difference in T90 among those with or without PH (0.35% vs 0%, respectively). By contrast, among those with PH, T90 was markedly higher in those with OSA (9.4%) than in those without OSA (0.3%). This finding seemingly implies that nocturnal hypoxemia resulting from PH alone is minimal, and that OSA is linked perhaps not just by association with worsened pulmonary hemodynamics. But we also note an important caveat: Looking at hypoxemia metrics other than T90 among non-OSA patients with and without PH, there is a significant difference in mean saturation (94% vs 97%, respectively). Though this outcome does not fully explain T90 findings, we believe it adds somewhat to uncertainty about the cause-effect relationship between hypoxemia in OSA and PH; PH itself may be causing mild hypoxemia, contributing somewhat to a false association between hypoxemia in OSA and PH severity, and somewhat clouding the interactions seen in this study.There are a few other limitations worth mentioning. First, all patients enrolled had a suspicion of PH based on echocardiography and were subsequently screened for OSA. Therefore, the findings are likely not generalizable to a standard OSA population, and PH prevalence data were likely higher than one would expect among a general OSA cohort. Indeed, among those even without OSA, the prevalence of PH was very high, suggesting a selection bias. Second, polysomnograms were performed as inpatient studies, which is unusual in many institutions, so this may affect the quality of sleep and recorded data. Third, while 85% of the patients with OSA and PH were deemed to have precapillary PH, when these patients were grouped by tertile, the most severely hypoxemic (tertile 3, T90 ≥ 12.25%) had a mean pulmonary arterial wedge pressure of 20.1 mm Hg, which is more consistent with postcapillary PH. Fourth, and relatedly, the subtypes of precapillary PH are unclear. Precapillary PH encompasses groups 1, 3, 4, and 5 PH, and these conditions have very different underlying pathophysiology. Finally, the 2022 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines on PH were released in September 2022 with some significant changes.9 The mPAP threshold was lowered to > 20 mm Hg, consistent with the Second World Symposium of PH guidelines, but the PVR threshold was also lowered from 3 Wood units to 2. In addition to changes in the hemodynamic definition of PH, there were also changes in the classification of PH; most notably, “sleep disordered breathing” was removed as a potential etiology of group 3 PH and was replaced with “hypoventilation syndromes,” with the note that “[s]ole nocturnal obstructive sleep apnoea is generally not a cause of PH.” This assertion may become more controversial in time with data such as presented by Huang et al. However, we note that this study does not report data on hypercapnia or the possibility of concurrent obesity hypoventilation syndrome.The concept of hypoxemia severity being causally related to PH has merit based on preclinical studies. Chronic hypoxia exposure has been used for decades to produce pulmonary hypertension in rodent models and is particularly successful with a “second hit” of monocrotaline or a vascular endothelial growth factor inhibitor. This exposure faithfully recapitulates hemodynamic and histologic findings of human PH.10 While some studies have demonstrated that severe intermittent hypoxia modeling OSA increases Fulton index (RV mass divided by LV plus septal mass) or right ventricular systolic pressure (RVSP), the severity of PH induced by this model is generally less than that induced by chronic sustained hypoxia.11,12 Indeed, in 1 clinical study of 43 patients with pulmonary arterial hypertension, most patients (70%) had nocturnal hypoxemia (T90 > 10%), but only 2 patients had an apnea index > 5.13 At our institutions (The University of California San Diego and Rush University Medical Center), our practice is to screen for symptoms of sleep-disordered breathing in most PH patients, and to refer symptomatic patients for sleep studies or, at a minimum, nocturnal oximetry. For new PH referrals with mildly elevated RVSP on echocardiography and high risk for OSA (eg, older males with obesity and/or cardiovascular comorbidities), we often pursue diagnostic testing for OSA, though RHC is usually not needed. And in our experience, when OSA is treated, the RVSP is typically reduced to normal on repeat echocardiography if there are no other contributing conditions. From the sleep medicine side, in the absence of clinical signs of right heart failure, or obesity hypoventilation syndrome, we do not routinely screen for PH in all OSA patients.Huang et al’s work is commendable and adds significantly to our understanding of the relationship between PH and OSA, though there is more to be said about this complex interaction. Future studies could re-evaluate T90 and invasive pulmonary hemodynamics after continuous positive airway pressure therapy specifically in OSA patients with more severe PH. Understanding the contribution of intermittent and/or sustained hypercapnia (in rodents and in humans) would also help us understand biological mechanisms in a more robust manner. We also need clinical guidelines about who specifically to refer for PH assessment in those with OSA, and who to refer for OSA assessment in those with PH. On both sides, it is becoming clear that referral of all patients is probably unnecessary. The new ESC/ERS PH guidelines will likely lead to more PH diagnoses but the ramifications of changes to the PH definition and classification, specifically in relation to those with OSA, remain to be seen. The pendulum has swung more than once on this topic, and we will likely see it happen again.DISCLOSURE STATEMENTAll authors have seen and approved this manuscript. The authors have no financial support to declare, nor any conflict of interest which could impact the manuscript as submitted.REFERENCES1. Benjafield AV, Ayas NT, Eastwood PR, et al.. Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis. Lancet Respir Med. 2019;7(8):687–698 . CrossrefGoogle Scholar2. Bradley TD, Floras JS. Obstructive sleep apnoea and its cardiovascular consequences. Lancet. 2009;373(9657):82–93 . CrossrefGoogle Scholar3. Chaouat A, Weitzenblum E, Krieger J, Oswald M, Kessler R. Pulmonary hemodynamics in the obstructive sleep apnea syndrome. Results in 220 consecutive patients. Chest. 1996;109(2):380–386 . CrossrefGoogle Scholar4. Sajkov D, Wang T, Saunders NA, Bune AJ, Neill AM, McEvoy RD. Daytime pulmonary hemodynamics in patients with obstructive sleep apnea without lung disease. Am J Respir Crit Care Med. 1999;159(5):1518–1526 . CrossrefGoogle Scholar5. Sajkov D, Wang T, Saunders NA, Bune AJ, McEvoy RD. Continuous positive airway pressure treatment improves pulmonary hemodynamics in patients with obstructive sleep apnea. Am J Respir Crit Care Med. 2002;165(2):152–158 . CrossrefGoogle Scholar6. Rich S, ed. Primary pulmonary hypertension: executive summary from the World Symposium - Primary Pulmonary Hypertension 1998. http://www.wsphassociation.org/wp-content/uploads/2019/04/Primary-Pulmonary-Hypertension-Evian-1998.pdf. Accessed January 13, 2023. Google Scholar7. Samhouri B, Venkatasaburamini M, Paz Y Mar H, Li M, Mehra R, Chaisson NF. Pulmonary artery hemodynamics are associated with duration of nocturnal desaturation but not apnea-hypopnea index. J Clin Sleep Med. 2020;16(8):1231–1239 . LinkGoogle Scholar8. Huang Z, Duan A, Hu M, et al.. Implication of prolonged nocturnal hypoxemia and obstructive sleep apnea for pulmonary hemodynamics in patients being evaluated for pulmonary hypertension: a retrospective study. J Clin Sleep Med. 2023;19(2):213–223 . LinkGoogle Scholar9. Humbert M, Kovacs G, Hoeper MM, et al. ESC/ERS Scientific Document Group. 2022 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Respir J. 2022:2200879 . Google Scholar10. Maarman G, Lecour S, Butrous G, Thienemann F, Sliwa K. A comprehensive review: the evolution of animal models in pulmonary hypertension research; are we there yet? Pulm Circ. 2013;3(4):739–756 . CrossrefGoogle Scholar11. Fagan KA. Selected contribution: pulmonary hypertension in mice following intermittent hypoxia. J Appl Physiol. 2001;90(6):2502–2507 . CrossrefGoogle Scholar12. Zhen X, Moya EA, Gautane M, et al.. Combined intermittent and sustained hypoxia is a novel and deleterious cardio-metabolic phenotype. Sleep. 2022;45:zsab290 . CrossrefGoogle Scholar13. Minai OA, Pandya CM, Golish JA, et al.. Predictors of nocturnal oxygen desaturation in pulmonary arterial hypertension. Chest. 2007;131(1):109–117 . CrossrefGoogle Scholar Next article FiguresReferencesRelatedDetails Volume 19 • Issue 2 • February 1, 2023ISSN (print): 1550-9389ISSN (online): 1550-9397Frequency: Monthly Metrics History Submitted for publicationDecember 15, 2022Accepted for publicationDecember 15, 2022Published onlineFebruary 1, 2023 Information© 2023 American Academy of Sleep MedicinePDF download