Sleep medicine · Does CPAP prevent cardiovascular events?
Marin 2005 observational cohort
Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without CPAP
Marin 2005 trial design and results
Design
Prospective observational cohort study
Treatment
CPAP in those who accepted and tolerated it
Control
Untreated apnoea, simple snorers, and healthy men
Population
1651 men: 264 healthy, 377 simple snorers, 403 untreated mild-moderate apnoea, 235 untreated severe apnoea and 372 treated with CPAP
Follow-up
Mean 10.1 years
Primary endpoint
Fatal and non-fatal cardiovascular events
Result
Men with untreated severe disease had more fatal (1.06 per 100 person-years) and non-fatal (2.13) cardiovascular events than untreated mild-moderate disease (0.55, p=0.02; 0.89, p<0.0001), simple snorers (0.34, p=0.0006; 0.58, p<0.0001) or CPAP-treated patients.
In men, severe obstructive sleep apnoea-hypopnoea significantly increases the risk of fatal and non-fatal cardiovascular events. CPAP treatment reduces this risk.
The trial authors, in the published abstract
How it has aged
The study that made everyone expect SAVE to be positive, and a lesson in why observational comparisons of treated and untreated patients mislead. Men who accept and tolerate CPAP differ systematically from those who do not, in ways no adjustment fully captures. When the randomised trials came, the effect did not.
Publications
Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study
Marin JM, Carrizo SJ, Vicente E, et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005 Mar 19-25;365(9464):1046-53.
Untreated severe sleep apnoea associated with markedly higher cardiovascular event rates
CPAP-treated men had event rates similar to controls
Observational design, with treatment self-selected rather than randomised