The sleep apnea cardiovascular endpoints study (SAVE)

Presented at the ESC congress 2016 by: R. Doug McEvoy (Adelaide, Australia)

Background

Obstructive sleep apnea (OSA) occurs in about 40-60% of patients with CV disease. OSA is associated with increased blood pressure, insulin resistance, badly functioning endothelium and increased CV morbidity and mortality. Treatment with continuous positive air pressure (CPAP) during sleeping may have a preventive effect for these patients in terms of CV disease. Randomised controlled study data that show the advantage of CPAP in light of CV disease, is however lacking.
To answer this question, the Sleep Apnea Cardiovascular Endpoints (SAVE) study was conducted. 2717 patients from 89 clinical centers were enrolled. These patients all had moderate to severe sleep apnea, were mostly elder (range 45-75 years), overweight and who snored a lot, and with coronary or cerebrovascular disease. These patients were randomised to standard treatment (CV risk management, sleep advice and lifestyle improvement, n=1341) or standard treatment plus CPAP (n=1346).

Main results

• 42% of patients used CPAP on average 4 or more hours per night.
• After a follow-up of a mean of 3.7 years the apnea-hypapnea index (AHI) from 29 to 3.7 events per hour in patients using CPAP.
• No effect of CPAP treatment was seen on the primary (composite of mortality, myocardial infarction (MI), stroke, hospitalization for TIA, unstable angina or heart failure) or secondary CV endpoints. 17.0% of CPAP patients experienced a CV event, as compared with 15.4% of the patients receiving standard treatment (HR for primary endpoint: 1.10, 95% CI 0.91-1.32).
• A trend was seen towards fewer cerebrovascular events in patietns who used CPAP >4 hours per night.
• CPAP also improved the mood of patients: they snored less, and fell asleep less often during the day, were less depressed, had better quality of life and less sick leave.

Conclusion

Treatment with CPAP during sleep in persons with moderate or severe sleep apnea did not improve CV mortality. It is unclear why this is the case. During the press conference it was speculated that a mean of 3.3 hours per night might not suffice to provide a decline in CV mortality or it could be that the timing of treatment is important. Although no benefit was seen on CV mortality, it is important that patients’ mood was positively affected.


Our reporting is based on the information provided during the ESC congress –

On August 28, the accompanying article was published in NEJM