Table 2.
Univariate and multivariate Cox regression models for AF detection.
| Variables | Crude HR | 95% CI | p-value | Adjusted HR* | 95% CI | p-value |
|---|---|---|---|---|---|---|
| Age (/1 Y) | 1.03 | 1.01–1.06 | <0.01 | 1.01 | 0.99–1.03 | 0.48 |
| Male | 0.91 | 0.59–1.40 | 0.66 | — | — | — |
| Hypertension | 1.18 | 0.77–1.82 | 0.45 | — | — | — |
| Diabetes mellitus | 1.09 | 0.64–1.85 | 0.76 | — | — | — |
| History of CHF | 0.85 | 0.27–2.70 | 0.79 | — | — | — |
| CHADS2 score (/1 point) | 1.17 | 0.95–1.43 | 0.13 | — | — | — |
| 1st degree atrioventricular block (n = 264) | 1.05 | 0.61–1.81 | 0.86 | — | — | — |
| PAC ≥200/day (n = 243) | 2.82 | 1.81–4.40 | <0.01 | 2.04 | 1.23–3.38 | <0.01 |
| LAD (/1 mm) (n = 264) | 1.04 | 1.01–1.08 | 0.022 | 1.02 | 0.98–1.06 | 0.35 |
| LVEF (/1%) (n = 264) | 1.00 | 0.97–1.03 | 0.99 | — | — | — |
| Large vessel occlusion | 1.30 | 0.77–2.18 | 0.33 | — | — | — |
| Index stroke to BNP measure (/1 day) | 1.00 | 0.97–1.04 | 0.85 | — | — | — |
| BNP measure to ICM implantation (/1 day) | 1.01 | 0.99–1.00 | 0.086 | — | — | — |
| Index stroke to ICM implantation (/1 day) | 1.00 | 0.99–1.00 | 0.088 | — | — | — |
| BNP tertile | ||||||
| Low-BNP (≤19.0) | 1 (reference) | — | — | 1 (reference) | — | — |
| Mid-BNP (19.1–48.4) | 1.01 | 0.53–1.92 | 0.98 | 0.91 | 0.46–1.78 | 0.77 |
| High-BNP (≥48.5) | 3.43 | 1.99–5.90 | <0.01 | 2.17 | 1.14–4.13 | 0.019 |
*Multivariate Cox regression model was developed adjusting for the variables associated with AF detection in the univariate analysis (n = 241). BNP, B-type natriuretic peptide; CHF, chronic heart failure; CI, confidence interval; HR, hazard ratio; ICM, insertable cardiac monitor; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; PAC, premature atrial contraction.