Susan Jeffrey

August 28, 2016

ROME, ITALY ( updated ) — A randomized trial shows no reduction in cardiovascular events with continuous positive airway pressure (CPAP) therapy in patients with moderate to severe obstructive sleep apnea and established cardiovascular disease[1].

There was no difference in the primary composite end point, including death from cardiovascular causes, MI, stroke, or hospitalization for unstable angina, heart failure, or transient ischemic attack (TIA), the researchers, with lead author Dr R Doug McEvoy (Flinders University and Southern Adelaide Local Health Network, Australia), report.

There were still benefits from CPAP, though, on secondary end points such as snoring and daytime sleepiness and improvement in health-related quality of life and mood, the authors note.

Dr R Doug McEvoy

"At the moment, we could not recommend, on the basis of our study findings, that CPAP be used for the prevention of cardiovascular disease in patients, particularly those who are asymptomatic from sleep apnea," McEvoy told heartwire from Medscape. "For those who are symptomatic, we would strongly urge cardiologists and cardiovascular physicians to be vigilant to identify patients with obstructive sleep apnea and treat them adequately for their sleep apnea in terms of well-being for their patients."

The results from the Sleep Apnea Cardiovascular Endpoints (SAVE) study were presented here at the European Society of Cardiology (ESC) Congress 2016 to coincide with their publication online August 28, 2016 in the New England Journal of Medicine.

Studies show an association between obstructive sleep apnea and the risk for cardiovascular events, particularly stroke, the authors write. Randomized trials have shown treatment with CPAP reduces systolic blood pressure and improves endothelial function and insulin sensitivity, and observational studies have suggested treated patients have lower rates of cardiovascular complications and death from cardiovascular causes, they note.

"However, currently randomized evidence focusing on hard clinical end points as to the benefit of obstructive sleep apnea treatment for cardiovascular disease prevention is lacking," McEvoy told a press conference here.

The SAVE trial was a multicenter, randomized, parallel-group but open-label trial with blinded end-point assessment. After a 1-week run-in period where patients used sham CPAP to establish whether or not they could adhere to therapy at least 3 hours per night, 2717 eligible participants with a diagnosis of coronary or cerebrovascular disease and moderate to severe sleep apnea were randomized to receive CPAP treatment plus usual care or usual care alone.

Moderate to severe apnea was defined as an oxygen desaturation index—that is, the number of times per hour during oximetry that the blood oxygen level dropped 4 or more points below baseline—of 12 using a home sleep study screening device (ApneaLink, ResMed).

Patients were excluded if they had severe daytime sleepiness (Epworth Sleepiness Scale score >15), were considered at increased risk of an accident from falling asleep, if they had very severe hypoxemia, or showed a pattern of Cheyne-Stokes respiration on the ApneaLink nasal pressure recording.

Mean duration of adherence to CPAP therapy in the treated group was 3.3 hours per night. The mean apnea-hypopnea index, the number of apnea or hypopnea events per hour of recording, decreased from 29.0 events per hour at baseline to 3.7 events per hour during follow-up.

However, after a mean follow-up of 3.7 years, there was no significant difference in the occurrence of the primary composite end point when CPAP therapy was added to usual care.

SAVE: Primary End Point

End Point CPAP plus usual care, n (%) Usual care, n (%) Hazard ratio (95% CI) P
Death from cardiovascular causes, MI, stroke, or hospitalization for unstable angina, heart failure or TIA 229 (17.0) 207 (15.4) 1.10 (0.91–1.32) 0.34

One-to-one propensity score matching comparing 561 patients in the treatment group who were more adherent to CPAP therapy with 561 usual-care patients showed a nonsignificant reduction in events, 15.3% in the treatment group vs 17.5% in the usual-care group (HR 0.80, 95% CI 0.60–1.07; P=0.13).

No significant differences were seen between groups in any cause-specific or composite secondary end points in the primary or subsidiary analyses, the authors note, "except for a higher rate of total hospital admissions for transient ischemic attack among the patients in the CPAP group (relative risk 2.29, 95% CI 1.05–4.99, P=0.05).

The propensity score-matched analyses also showed that patients who were adherent to therapy had a lower risk of stroke (HR 0.56; P=0.05) and lower risk of the "nonprespecified" composite end point of cerebrovascular events, but they note, "These results were not adjusted for multiple testing."

Reductions from baseline in sleepiness and other symptoms of sleep apnea were greater in the CPAP group, with an estimated mean between-group difference in the change from baseline in Epworth Sleepiness Scale score of -2.5 (P<0.001).

Greater reductions from baseline were also seen in the anxiety and depression scores on the Hospital Anxiety and Depression Scale with CPAP treatment.

During the press conference here, moderator Dr Stephan Gielen (Klinikum Lippe Kardiologie, Angiologie, und Intensivmedizin Sekretariat, Detmold, Germany) said he would like to ask a question "from my perspective as a cardiologist."

"In the past, we have always seen obstructive sleep apnea as a risk factor for hypertension and developing heart disease and stroke. So now you say that reducing obstructive sleep apnea by effective treatment does not change the end points," Gielen said. "Does this mean that it's more, from a prognostic perspective, an innocent bystander of the same shared risk profile, namely obesity, smoking, and others, alcohol for example, that have the same effect on disease progression, both with regard to OSA and heart disease?"

"That is a distinct possibility," McEvoy replied. For example, he said, "one possibility is that the observational studies overestimate the risk because of difficulty controlling for confounders you don't know about and for confounders like visceral obesity, which is not measured accurately.

"However," he added, "the strength of the association in the epidemiologic studies is substantial, and it is still possible that we have not determined the amount of treatment necessary to modify the risk associated with obstructive sleep apnea.

Next steps may be to look at the effects of treating sleep apnea on individual diseases, rather than the composite of different cardiovascular diseases they used in this study, he said.


"Our study would suggest that stroke may be a very good target to look at, and certainly atrial fibrillation," McEvoy told heartwire . "We really need a randomized, controlled trial of sleep apnea treatment for atrial-fibrillation reduction—reducing the burden of atrial fibrillation, as well as the burden of disease and hospitalization in heart failure. Our study was not powered to look at those specific cardiac events."

In an accompanying editorial[2], Dr Babak Mokhlesi (University of Chicago, IL) and Dr Najib T Ayas (University of British Columbia, Vancouver), consider why the results of the trial were negative.

More likely than sleep apnea not actually having clinically significant adverse cardiovascular effects is the possibility that patients didn't use CPAP long enough each night to derive cardiovascular benefits, they note.

Mean duration of CPAP adherence in SAVE was 3.3 hours per night, while a propensity analysis of those treated more than 4 hours per night suggested a reduction in the primary end point among CPAP-treated patients, as well as a significant reduction in cerebrovascular events, although these data should be interpreted "cautiously," the editorialists write.

Other issues may have been the timing of CPAP therapy later in the night, the "pragmatic" use of a home sleep study device rather than formal polysomnography, and the possibility that in a population with established cardiovascular disease, CPAP may have a limited effect.

"We believe that for symptomatic patients with obstructive sleep apnea, a trial of CPAP should be offered. It would also be prudent to offer CPAP to patients with obstructive sleep apnea and severe hypoxemia during sleep regardless of symptoms—these patients were excluded from the SAVE trial," Mokhlesi and Ayas write.

"However, on the basis of the results of the SAVE trial, prescribing CPAP with the sole purpose of reducing future cardiovascular events in asymptomatic patients with obstructive sleep apnea and established coronary disease cannot be recommended."

Ongoing trials such as the ISAACC study will provide more information on the effect of CPAP in nonsleepy patients with obstructive sleep apnea and acute coronary syndromes, they conclude. "Furthermore, although improving CPAP technology to maximize adherence is important, we believe that there is also a need for novel treatment options that allow for better adherence."

The study was funded by the National Health and Medical Research Council of Australia and Philips Respironics. McEvoy reports receiving research study equipment from Air Liquide; disclosures for coauthors are listed in the article. Ayas reports being on the scientific advisory board of BresoTec and personal fees from RHS Medical outside the submitted work. Mokhlesi reports personal fees from Itamar Medical and Zephyr Medical Technologies, grants from Philips/Respironics and the National Institutes of Health, and personal fees for expert testimony in medical-malpractice cases outside the submitted work.

For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.