Table 5.
Overall and Landmark Analysis for the Adverse Events in Patients With OSA Versus Those Without OSA
| Variables | Unadjusted HR (95% CI) | P Value | Adjusted HR (95% CI)* | P Value |
|---|---|---|---|---|
| Overall analysis | ||||
| MACCE | 1.59 (1.01–2.50) | 0.043 | 1.55 (0.94–2.57) | 0.085 |
| Cardiovascular death† | 0.80 (0.25–2.63) | 0.716 | ··· | ··· |
| Myocardial infarction† | 0.54 (0.16–1.85) | 0.327 | ··· | ··· |
| Stroke† | 1.93 (0.48–7.71) | 0.353 | ··· | ··· |
| Ischemia‐driven revascularization | 1.57 (0.72–3.42) | 0.261 | 1.52 (0.65–3.56) | 0.334 |
| Hospitalization for unstable angina | 1.89 (1.04–3.44) | 0.036 | 2.10 (1.09–4.05) | 0.027 |
| Hospitalization for heart failure† | 0.97 (0.20–4.81) | 0.972 | ··· | ··· |
| All repeat revascularization | 1.32 (0.75–2.35) | 0.340 | 1.51 (0.81–2.83) | 0.195 |
| Composite of all events | 1.48 (0.99–2.21) | 0.057 | 1.54 (0.98–2.40) | 0.059 |
| Landmark analysis (≤1 y) | ||||
| MACCE | 1.27 (0.76–2.12) | 0.353 | 1.18 (0.67–2.09) | 0.575 |
| Hospitalization for unstable angina | 1.62 (0.79–3.31) | 0.187 | 1.84 (0.84–4.03) | 0.130 |
| Ischemic‐driven revascularization | 1.21 (0.48–3.07) | 0.688 | 1.27 (0.46–3.50) | 0.646 |
| All repeat revascularization | 1.14 (0.61–2.14) | 0.682 | 1.41 (0.71–2.82) | 0.328 |
| Composite of all events | 1.26 (0.81–1.96) | 0.310 | 1.28 (0.78–2.09) | 0.322 |
| Landmark analysis (>1 y) | ||||
| MACCE | 3.55 (1.20–10.56) | 0.023 | 3.87 (1.20–12.46) | 0.023 |
| Hospitalization for unstable angina | 2.68 (0.87–8.21) | 0.085 | 2.82 (0.84–9.51) | 0.095 |
| Ischemic‐driven revascularization | 2.92 (0.61–14.04) | 0.182 | 2.46 (0.46–13.26) | 0.295 |
| All repeat revascularization | 2.85 (0.59–13.71) | 0.192 | 2.54 (0.47–13.73) | 0.278 |
| Composite of all events | 3.30 (1.10–9.86) | 0.033 | 3.67 (1.13–11.95) | 0.031 |
HR indicates hazard ratio; MACCE, major adverse cardiovascular and cerebrovascular event; OSA, obstructive sleep apnea; PCI, percutaneous coronary intervention.
*
Model adjusted for age, sex, body mass index, hypertension, and diabetes mellitus, clinical presentation (acute myocardial infarction vs unstable angina), PCI procedure, and minimum SaO2.
†
Multivariate Cox regression and landmark analysis was not done because of too few events.