乌鸦传媒

ROME -- Treating obstructive sleep apnea (OSA) with continuous positive airway pressure (CPAP) did not reduce risk of recurrent coronary or cerebrovascular events, the randomized SAVE trial showed.

For people with coronary or cerebrovascular disease, treatment of moderate-to-severe OSA showed no advantage over usual care for the composite of cardiovascular death, myocardial infarction (MI), stroke, or hospitalization for unstable angina, heart failure, or transient ischemic attack.

The rate over 3.7 years was 17.0% with CPAP used for a mean of 3.3 hours per night compared with 15.4% with usual care alone (HR 1.10, 95% CI 0.91-1.32), , of Australia's Adelaide Institute for Sleep Health, and colleagues found.

No individual component of that primary composite endpoint showed a benefit either, the group reported here at the European Society of Cardiology annual meeting and simultaneously online in the .

"This is a really important study. People have been waiting for some answer to this question for a long time," , of the University of Minnesota School of Public Health in Minneapolis, told 乌鸦传媒. "At least in the U.S., this has become an enormous industry ... Part of the leading thinking of this was that CPAP indeed would reduce cardiovascular disease outcomes."

"Clinically, we prescribe CPAP for our patients for symptoms -- commonly excessive daytime sleepiness, but also mood changes or neurocognitive deficits -- and also for cardiovascular benefit," , of the Sleep Disorders Center Research at the Cleveland Clinic, told 乌鸦传媒.

While she cited more than 15 clinical trials that "have consistently demonstrated improvement in blood pressure and other studies that have pointed towards benefit of other cardiovascular endpoints" with apnea treatment, , of the Duke Clinical Research Institute in Durham, N.C., noted that only SAVE and SERVE-HF have looked at hard clinical outcomes -- and both went at least numerically in the wrong direction.

The finding of mortality risk in heart failure when central sleep apnea was treated with adaptive servo-ventilation in SERVE-HF led to a big drop in use, Ohman noted. "Now we have two trials overall neutral but possibly with one trial with some drift toward harm."

However, SAVE did still show the expected benefits for snoring, daytime sleepiness, and quality of life.

So a trial of CPAP should still be offered for OSA patients with heart disease or prior stroke, albeit not with the sole purpose of reducing future cardiovascular events, concluded an accompanying editorial by , of the University of Chicago, and , of the University of British Columbia in Vancouver.

"That'll be the clinical dilemma to sort out are people having symptoms," commented , of California's Stanford University and an American Heart Association spokesperson. "My guess is a trial of CPAP might be instituted to see if it makes people feel better."

But it's not clear whether the trial had disappointing results because OSA doesn't have a clinically significant impact on the heart, which would render any treatment ineffective, or because patients didn't get a big enough dose of CPAP, they wrote. "Given the substantial human and animal data that have consistently documented links between obstructive sleep apnea and cardiovascular health, we suspect it may be the latter."

They pointed to the suggestive, but nonsignificant, trend for a slightly lower risk of the primary composite endpoint in the CPAP group among the 561 of the total 2,717 participants who used the CPAP machine more than 4 hours per night (HR 0.80, 95% CI 0.60-1.07) and the significantly lower cerebrovascular event risk in the same group (HR 0.52, 95% CI 0.30-0.90).

McEvoy, speaking at a press conference where the findings were presented, agreed, citing the strength of the association in prior observational studies. "It's still possible we have not hit the target in terms of the amount of treatment necessary to modify the risk associated with sleep apnea."

The situation might have been worsened in the trial because of a high proportion of patients being enrolled in countries with few sleep medicine experts (60% were from China) that can help motivate patients and address issues like mask fit and humidity and heated air settings that boost adherence, suggested , of the University of California San Diego and a past president of the American College of Cardiology.

But realistically, no trial has been able to target a population adherent to CPAP for the entire the night, he noted.

"CPAP in it's current formulation appears to be able to used by most people for a period of only about 50% of their sleep," he told 乌鸦传媒. "At the moment, it's a therapy that has adherence that is limited."

Other possible explanations suggested included:

• The timing of CPAP was off, with few patients using during the REM sleep predominant in the early morning hours when apnea or hypopnea events last longer and reduce oxygen saturation more.
• Inclusion only of patients with established cardiovascular or cerebrovascular disease, for whom CPAP might have a limited effect.
• Use of a pragmatic approach to OSA diagnosis with the home ApneaLink screening device, rather than a full polysomnography test because many participants were from geographic locations with limited resources, although McEvoy told 乌鸦传媒 that validation studies showed this device to be "very effective and accurate."

The trial included 2,717 adults, ages 45 to 75, with an oxygen saturation index of at least 12 on screening with the ApneaLink device, but minimal daytime sleepiness (Epworth Sleepiness Scale score of 15 or less). Those with Cheyne-Stokes respiration, which had been earlier linked to mortality risk in heart failure when treated with adaptive servo-ventilation, were excluded.

Participants were randomized open-label to usual care with advice on healthful sleep habits and lifestyle changes to minimize OSA or to the addition of mask-delivered CPAP, set to automatic mode for the first week then fixed to the 90th percentile of pressure calculated by the device.

The study also had the sample size revised after enrollment problems were encountered, and recalculation was based mostly on primary prevention studies. "Although it seems doubtful that the recruitment of more patients would have changed the results of the primary analysis, it may have made the results for the patients who were adherent to CPAP therapy clearer," the editorialists wrote.

Luepker noted that a trial in primary prevention patients would be difficult because of the large size and long follow-up time that would be required with the lower event rate expected in this group.

Whatever the reason for it's failure, "I think payers are going to look at this because it's expensive treatment," Luepker concluded. "One of the things they very well may say is 'If you're prescribing for cardiovascular benefits, that's not adequate justification.'"

But "once these things become common it's hard to go back," he acknowledged.

Comment