Table 6.
Association of preoperative echocardiography with 1 year mortality within subgroups
| Cohort/subgroup | 1 year mortality | |
|---|---|---|
| Relative risk (95% CI) | Interaction* | |
| Entire matched cohort | 1.07 (1.01 to 1.12) | NA |
| Subgroups | ||
| Hospital teaching status: | ||
| Teaching | 1.02 (0.94 to 1.12) | P=0.30 |
| Non-teaching | 1.09 (1.01 to 1.16) | |
| Hospital procedure quarter: | ||
| First (lowest) | 1.13 (1.01 to 1.28) | P=0.36 |
| Second | 1.01 (0.90 to 1.12) | |
| Third | 1.10 (1.00 to 1.22) | |
| Fourth (highest) | 1.02 (0.93 to 1.13) | |
| Age: | ||
| <70 years | 1.07 (1.01 to 1.14) | P=0.87 |
| ≥70 years | 1.06 (0.96 to 1.18) | |
| Coronary artery disease: | ||
| Present | 1.05 (0.95 to 1.15) | P=0.70 |
| Absent | 1.07 (1.01 to 1.14) | |
| Congestive heart failure: | ||
| Present | 1.04 (0.90 to 1.21) | P=0.80 |
| Absent | 1.07 (1.01 to 1.13) | |
| Pulmonary disease: | ||
| Present | 1.07 (0.96 to 1.20) | P=0.96 |
| Absent | 1.07 (1.01 to 1.14) | |
| Concurrent cardiac stress test and RCRI†: | ||
| Concurrent stress test‡ | 1.01 (0.92 to 1.11) | |
| No concurrent stress test: | ||
| RCRI: 0 points | 1.44 (1.14 to 1.82) | P=0.02 |
| RCRI: 1-2 points | 1.10 (1.02 to 1.18) | |
| RCRI: ≥3 points | 1.00 (0.87 to 1.13) | |
| Procedure: | ||
| Vascular | 1.00 (0.90 to 1.12) | P=0.16 |
| Abdominal or thoracic | 1.06 (1.00 to 1.13) | |
| Orthopaedic | 1.20 (1.04 to 1.38) | |
RCRI=revised cardiac risk index.
*Tests whether treatment effects differed significantly between subgroups.
†Given previously described interaction between stress testing and RCRI,6 this subgroup analysis involved exact matching on both characteristics.
‡No significant interaction (P=0.26) between the RCRI score and preoperative echocardiography among patients who had cardiac stress testing.