Obstructive sleep apnea (OSA) and central sleep apnea (CSA) are common among patients with heart failure, and both forms of sleep disordered breathing are associated with worse overall function and poor prognosis.
Given this association, it stands to reason that aggressive treatment of sleep disordered breathing would help people with heart failure live longer. But the reality is far more complicated and the optimal treatment strategy, especially for patients with CSA, remains unclear.
The challenges and uncertainties regarding the treatment of OSA and CSA in the setting of heart failure were discussed at CHEST 2018, the annual meeting of the American College of Chest Physicians.
Experts in the field of sleep medicine reviewed the research, and discussed the known and unknown about the benefits and potential harms of continuous positive airway pressure (CPAP), targeted adaptive servo-ventilation (ASV), and alternative treatments, including phrenic nerve stimulation and supplemental nasal cannula oxygen therapy.
While some trials on the impact of OSA and CSA on death risk in heart failure patients found no association, signals suggest that sleep disordered breathing disorders are tied to an increased risk of death, said Neomi Shah, MD, of Mount Sinai Health System in New York City.
CPAP for OSA in Heart Failure
The data on the impact of treating OSA on death risk in the heart failure setting are inconclusive, Shah said.
"Short-term trials examining the treatment of OSA in heart failure show overall improvement in intermediate endpoints such as sleepiness, left ventricular ejection fraction [LVEF], and symptoms," she said. "But does removing OSA from the picture for those with heart failure improve survival? I don't think we have the answer to that yet."
In the 2016 international Sleep Apnea Cardiovascular Endpoints () trial, CPAP failed to improve cardiovascular outcomes in patients with moderate-to-severe OSA, and cardiovascular or cerebrovascular disease.
The study involved 89 centers in seven countries, with patients ages 45-75, who had coronary artery disease or cerebrovascular disease, and moderate-to-severe OSA. They were followed for 3.7 years. The study's primary composite endpoint included heart failure. The trial excluded patients with excessive daytime sleepiness and patients with severe hypoxemia.
Adherence to CPAP in SAVE was low at a mean of 3.3 hours per night.
Shah noted that hospitalization for heart failure was comparable in patients who were, and were not treated, with CPAP. Those randomized to CPAP had more coronary revascularization procedures, although the numbers were small.
Shah said her own earlier work, and that of others, has raised the possibility that sleep apnea in the setting of acute MI could potentially be cardioprotective.
"I think it's hard to know what OSA treatment is doing to incident heart failure or survival. I think we need to understand the large epidemiological studies better in order to characterize which patients are going to benefit, not benefit, or potentially be harmed by treatment," she stated.
CSA Trials to Date
CSA, characterized by cyclic hyperventilation and falls in partial pressure of arterial carbon dioxide below the apnea index, has been reported to be present in 25% to 40% of chronic heart failure patients.
To date, two prospective trials of CSA in heart failure have been completed -- the (Canadian Continuous Positive Airway Pressure for Patients With Central Sleep Apnea) study in 2005, and the SERVE-HF (Treatment of Predominant Central Sleep Apnea by Adaptive Servo Ventilation in Patients With Heart Failure) trial from 2015.
If findings from CANPAP could be characterized as disappointing, SERVE-HF findings were considered cause for alarm, prompting warnings to clinicians to in patients with heart failure and LVEF <45%.
CANPAP lead researcher T. Douglas Bradley, MD, of the University of Toronto, discussed the lessons from that study, and other major CNS trials in the heart failure setting, at CHEST 2018.
The randomized trial found that CPAP attenuated CSA, improved nocturnal oxygenation, increased EF, lowered norephinephine levels, and improved endurance, as measured by distance walked in the 6-minute walk test. However, it did not impact heart transplant-free survival.
Bradley noted that in the first 18 months of the study, the CPAP group tended to do worse than the control group. After that there was a crossover, and the CPAP group did better.
"This suggested to us that there may be two groups of patients -- a group that responds to CPAP and a group that doesn't," he said.
To test the hypothesis, the researchers did a post-hoc analysis comparing CPAP responders with an Apnea-Hypopnea Index (AHI) <15 at 3 months to non-responders.
The observation that a decline in AHI to <15 within 3 months of CPAP initiation was associated with higher transplant-free survival in the CPAP group compared with the control group suggested that marked lowering of the AHI may be a key therapeutic goal, Bradley said.
He suggested that based on the observation, "it may be reasonable to provide a trial of CPAP, and to reassess CSA 1 to 3 months later." If the AHI does not decrease, then CPAP should be stopped, he said, adding that "further trials are needed to test this hypothesis."
A decade after CANPAP, the SERVE-HF trial examined the use of the synchronized positive pressure ventilation ASV, which has been shown to be more effective against CSA than CPAP. The trial included 1,325 ambulatory patients with CSA and heart failure with LVEF <40%.
The study showed ASV therapy to be associated with significant increases in all-cause and cardiovascular death versus the control group. Post-hoc analysis suggested that sudden cardiac death was the mode of increased mortality, and that the risk was greater with lower EF.
SERVE-HF researcher Virend K. Somers, MD, PhD, of Mayo Clinic in Rochester, Minnesota, discussed the study at CHEST 2018. He offered a potential explanation for the increased mortality reported with the treatment.
In patients with heart failure, having the nocturnal condition known as periodic limb movement in sleep (PLMS) is associated with worse survival.
In a separate retrospective study of patients treated with ASV, Somers's group found a very high prevalence of PLMS in heart failure patients. They also found that the treatment increased PLMS episodes in these patients. The likelihood of an increase in PLMS was also higher in patients with low EF.
"In the analysis we just completed of SERVE-HF, we found a significant increase in PLMS in the ASV-treated group," he said.
This could potentially be the mechanism to explain the worse survival in these patients, he said.
Bradley is currently conducting the , examining whether ASV improves cardiovascular outcomes in heart failure patients with OSA (71%) and CSA (29%). The trial is being conducted in 49 centers in nine countries, and 603 patients have been randomized.
The trial was conceived before the SERVE-HF findings were reported, and Bradley reported early findings, noting that compliance is much better than previous trials. In addition, there have been no safety concerns, and no signals of early death, in either the OSA or CSA patients treated with ASV.
"Stay tuned for the results. Hopefully this trial will be finished next year," Bradley said.
Another trial examining ASV ( in hospitalized patients with heart failure and OSA or CSA was terminated during recruitment based on the findings from SERVE-HF.
Disclosures
Bradley disclosed support from Philips Respironics.