Table 2.
Multivariable Cox regression
| ICD implantation | Sustained VA event | |||
|---|---|---|---|---|
| Multivariable Cox HR (95% CI) | P value | Multivariable Cox HR (95% CI) | P value | |
| North American location | 3.084 (2.521; 3.774) | <.0001 | 1.355 (1.046; 1.754) | .021 |
| Age (years) | 0.999 (0.993; 1.005) | .69 | 0.987 (0.979; 0.995) | .0015 |
| Male sex | 1.215 (0.997; 1.479) | .053 | 1.840 (1.429; 2.369) | <.0001 |
| # ECG leads w/ TWI | 1.052 (1.002; 1.105) | .041 | 1.046 (0.987; 1.108) | .13 |
| log(PVC count) | 1.039 (0.989; 1.092) | .13 | 1.157 (1.073; 1.248) | .0002 |
| History of syncope | 2.272 (1.728; 2.988) | <.0001 | 1.679 (1.234; 2.283) | .0010 |
| History of NSVT | 1.726 (1.387; 2.149) | <.0001 | 2.325 (1.741; 3.106) | <.0001 |
| RVEF (%) | 0.984 (0.975; 0.993) | .0007 | 0.981 (0.970; 0.992) | .0008 |
| Presence of ICD | NA | NA | 3.283 (2.497; 4.317) | <.0001 |
Multivariable Cox regression assessing associations between geographic location (North American vs. European centres) and both ICD implantation and sustained VA event. In assessment of association with sustained VA events, presence of ICD was modelled as a time-dependent variable. Hazard ratios (HR) are presented along with 95% confidence intervals (CI).