Table 3.
Association of DCM severity with the most deleterious DCM-related rare variant found.
|
Presence of DCM-related rare variants |
LVAD or heart transplant vs. Neither OR (95% CI) |
With ICD vs. Neither OR (95% CI) |
|---|---|---|
| Model 1, Crude | ||
| P/LP | 1.9 (1.2, 2.8) | 1.0 (0.6, 1.5) |
| VUS only | 0.9 (0.6, 1.2) | 0.8 (0.6, 1.1) |
| Negative | Reference | Reference |
| Model 2 1 . Controlled for demographic and social determinants | ||
| P/LP | 2.2 (1.4, 3.3) | 1.0 (0.7, 1.5) |
| VUS only | 0.9 (0.6, 1.2) | 0.8 (0.6, 1.1) |
| Negative | Reference | Reference |
| Model 3 2 . Additionally controlled for factors affecting disease severity | ||
| P/LP | 2.3 (1.5, 3.6) | 1.0 (0.6, 1.5) |
| VUS only | 0.8 (0.6, 1.2) | 0.8 (0.6, 1.1) |
| Negative | Reference | Reference |
Abbreviations: ICD=Implantable cardioverter-defibrillator. LVAD=Left ventricular assist device. P/LP=Pathogenic/Likely pathogenic. VUS=Variant of uncertain significance.
Note: Estimates are based on multinomial models with generalized logit link. Response variable is unordered and “Neither” ((i.e., with no LVAD, heart transplant or ICD)) is the reference group.
Model 2 controlled for genomic ancestry (AA, EA, or NA), US region of study sites, and tobacco use (ever or not). Adjustment for sex, education (<=12 or >12 years), and health insurance coverage (yes or no) did not alter the estimated ORs for the DCM-related rare variant groups.
Model 3 controlled for DCM duration (<5 or >=5 years) and comorbidities (diabetes, hypertension) in addition to socio-demographic variables controlled in model 2. There was no statistical interaction between ancestry (African ancestry and European ancestry) and DCM-related rare variant group (p=0.15) when patients with Native American ancestry were excluded.