OBJECTIVES: Previous community-based studies have failed to demonstrate an independent association between OSA and circulating cardiac troponin concentrations, a marker of myocardial injury. However, these studies have used troponin assays with modest analytic sensitivity to detect low-level, chronic increments in troponin levels. Using a highly sensitive troponin I (hs-TnI) assay, we tested the hypothesis that the severity of OSA is associated with myocardial injury independently of comorbidities. DESIGN: Cross-sectional study. SETTING: Community-based. PARTICIPANTS: 514 subjects (54% men, age 48 ± 11 y [mean ± SD]). INTERVENTIONS: N/A. MEASUREMENTS AND RESULTS: hs-TnI concentrations were measured in fasting morning blood samples and 318 participants (62%) had hs-TnI concentration above the limit of detection ([LoD] 1.2 ng/L). The severity of OSA, expressed as the apnea-hypopnea index (AHI) and nocturnal hypoxemia, was assessed by in-hospital polysomnography. After adjustment for age, gender, estimated creatinine clearance, history of coronary artery disease and hypertension, smoking, diabetes mellitus, systolic blood pressure, heart rate, body mass index, left ventricular hypertrophy, and cholesterol ratio in multivariate linear regression models, higher AHI (standardized β = 0.12, P = 0.006), lower mean SpO2 (β = -0.13, P = 0.012) and higher percentage of total sleep time with SpO2 < 90% (β = 0.12, P = 0.011) were all associated with higher hs-TnI levels in separate models. Additional analyses with hs-TnI categorized in tertiles or using a different strategy for persons with hs-TnI levels below the LoD did not change the results. CONCLUSION: Increased obstructive sleep apnea (OSA) severity is independently associated with higher concentrations of hs-TnI, suggesting that frequent apneas or hypoxemia in OSA may cause low-grade myocardial injury.