PROGNosis in Obstructive Sleep Apnea after Myocardial Infarction - the PROGNOSA-MI study

Introduction Obstructive sleep apnea (OSA) is associated with increased risk for recurrent cardiovascular (CV) events and mortality in patients with established CV disease. As many as two-thirds of patients with acute myocardial infarction (MI) may at least have mild OSA. However, if OSA is associated with poor prognosis after an acute MI is not well established. Purpose To investigate the registry reported prevalence of OSA and association with adverse prognosis in patients after first MI. Methods We performed a retrospective cohort study with patients registered in the SWEDEHEART MI registry for first MI 2005-2017. Patients with an ICD-10 OSA diagnosis prior to discharge after first MI were identified by linkage to the Swedish National Patient Register. In adjusted Cox proportional hazards models, the association between OSA diagnosis, mortality, ischemic events, and hospitalization for heart failure were assessed. The no OSA diagnosis subset served as reference. Results The registry reported prevalence of OSA diagnosis prior to discharge after first MI was 1.37 % (2,129 out of 155,332). OSA patients were more likely to be younger (mean age 66.2 years versus [vs] 69.3 years), be male (81.6 % vs 64.1 %), former smokers (42.4 % vs 30.3 %), and more often had prescribed primary prevention therapies: acetylsalicylic acid (28.6 % vs 23.1 %), angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker (48.6 % vs 27.9 %), and statins (32.8 % vs 18.1 %). Further, OSA patients more often presented with non-ST-elevation myocardial infarction (NSTEMI; 65.0 % vs 46.5 %) and had more comorbidities: chronic obstructive pulmonary disease (11.4 % vs 6.1 %); congestive heart failure (11.8 % vs 6.3 %); diabetes mellitus (39.3 % vs 20.4 %); hypertension (68.6 % vs 46.4 %); and obesity (55.6 % vs 20.5 %). During a median follow-up of 4.67 (interquartile range 1.95-7.96) years, those with an OSA diagnosis had an increased risk of all-cause mortality (hazard ratio [HR], 95% confidence interval [CI]; 1.15, 1.04-1.27), non-CV mortality (1.28, 1.13-1.45), and hospitalization for heart failure (1.16, 1.01-1.34). No other associations with the studied outcomes were observed. Conclusions In patients with acute MI, registry reported prevalence of OSA was low, but OSA diagnosis remained independently associated with increased risk of all-cause mortality, non-CV mortality, and hospitalization for heart failure. Further, patients with OSA had higher prevalence of CV comorbidities and more often presented with NSTEMI. Consequently, OSA may constitute an important and potentially underdiagnosed component of the residual risk not affected by the intensive intervention of established risk factors applied after MI.