Table 2:
Predictors of death/cardiac transplant >90 days post-1st M-MVR
| Adjusted Hazards Ratio | 95% CI | Significant p-value | |
|---|---|---|---|
| Male vs female | 1.5 | 1.0 – 2.3 | 0.05 |
| Sz/wt >2 vs ≤ 2 | |||
| 1 year post-M-MVR | 1.5 | 0.5–4.3 | |
| 10 years post-M-MVR | 0.3 | 0.1–0.8 | 0.02 |
| Age <2 years vs. 12 - <21 years | |||
| 1 year post-M-MVR | 1.8 | 0.4–7.0 | |
| 10 years post-M-MVR | 4.3 | 1.2–15.1 | 0.02 |
| Age 2 - <6 years vs. 12 - <21 years | |||
| 1 year post-M-MVR | 1.7 | 0.6 –4.8 | |
| 10 years post-M-MVR | 1.7 | 0.6–4.3 | |
| Age 6 - <12 years vs. 12 - <21 years | |||
| 1 year post-M-MVR | 2.0 | 0.7–5.3 | |
| 10 years post-M-MVR | 1.9 | 0.8–4.8 | |
| Concurrent other valve replacement vs. none | 1.8 | 0.9–3.5 | |
| Non-bi-leaflet vs. bi-leaflet valve | 2.4 | 1.3 –4.3 | <0.01 |
| Pacemaker ≤90 days following M-MVR vs. none | |||
| 1 year post-M-MVR | 1.7 | 0.9–3.0 | |
| 10 years post-M-MVR | 1.1 | 0.6–2.1 | |
Time of hazard ratio assessment for variables violating the proportional hazards assumption. Age and Sz/Wt were modeled using Heaviside functions; other syndromes and pacemaker implantation were modeled using a continuous term for interaction with the natural log of time.
Adjusted model includes all variables listed in the table and a random effect for M-MVR treatment center. Era of M-MVR, MV regurgitation and/or stenosis, and primary diagnoses evaluated with these covariates had p>0.2 and did not meaningfully contribute to the model.