|
|
APPENDIX 1A | APPENDIX 1B | APPENDIX 2A | APPENDIX 2B | APPENDIX 3 |
APPENDIX 4 | APPENDIX 5 | APPENDIX 6 | APPENDIX 7 | APPENDIX 8 |
APPENDIX 3
Multivariable Logistic Regression
The following are the results of the multivariable logistic regression analyses with and without propensity adjustment.
Table Appendix 3. Results of Multivariable Logistic Regression for 5-Year Mortality in Study Patients with and without Propensity Adjustment*
|
|
Analysis in Presence of Covariates without Propensity Adjustment† |
Analysis in Presence of Covariates with Propensity Adjustment‡ |
Risk Factor |
Odds Ratio (95% CI) |
P Value |
Odds Ratio (95% CI) |
P Value |
Aprotinin vs. control |
1.51 (1.17 – 1.96) |
.002 |
1.48 (1.13 – 1.93) |
.005 |
Aminocaproic acid vs. control |
0.90 (0.67 – 1.21) |
.483 |
0.85 (0.63 – 1.15) |
.293 |
Tranexamic acid vs. control |
1.14 (0.80 – 1.60) |
.471 |
1.12 (0.79 – 1.59) |
.519 |
Propensity score, deciles |
― |
― |
1.03 (0.98 – 1.07) |
.222 |
Complex vs. primary surgery§ |
1.65 (1.34 – 2.04) |
<.001 |
1.62 (1.29 – 2.03) |
<.001 |
Age per 10 years or part thereof over
60 years |
1.62 (1.43 – 1.83) |
<.001 |
1.61 (1.42 – 1.82) |
<.001 |
African American, American Indian, or
Hispanic ethnicity║ |
1.79 (1.23 – 2.61) |
.002 |
1.64 (1.12 – 2.43) |
.012 |
History of smoking |
1.34 (1.06 – 1.69) |
.014 |
1.32 (1.04 – 1.67) |
.022 |
History of congestive heart failure with
hospitalization |
2.05 (1.55 – 2.72) |
<.001 |
2.03 (1.53 – 2.71) |
<.001 |
History of no angina¶ |
1.38 (1.02 – 1.87) |
.036 |
1.41 (1.04 – 1.92) |
.029 |
History of valve disease without prior or
current valve surgery |
1.46 (1.05 – 2.02) |
.025 |
1.38 (0.98 – 1.94) |
.065 |
History of diabetes mellitus |
1.47 (1.18 – 1.82) |
<.001 |
1.48 (1.19 – 1.84) |
<.001 |
History of syncope |
1.45 (1.04 – 2.03) |
.031 |
1.46 (1.04 – 2.05) |
.027 |
History of pulmonary disease |
1.39 (1.11 – 1.74) |
.004 |
1.37 (1.09 – 1.73) |
.007 |
History of peripheral vascular disease |
1.65 (1.31 – 2.07) |
<.001 |
1.69 (1.34 – 2.13) |
<.001 |
Medication of warfarin or coumadin in the
past week of admission |
1.88 (1.36 – 2.61) |
<.001 |
1.81 (1.29 – 2.54) |
<.001 |
Creatinine > 1.3 mg/dl on admission** |
1.93 (1.52 – 2.46) |
<.001 |
1.90 (1.49 – 2.42) |
<.001 |
Preoperative congestive heart failure |
1.27 (1.02 – 1.58) |
.032 |
1.30 (1.04 – 1.62) |
.022 |
Preoperative myocardial infarction |
2.24 (1.28 – 3.93) |
.005 |
2.04 (1.14 – 3.65) |
.017 |
* Cohort includes patients completing the entire 5-year study (n=3357; alive=2809; dead=548). CI denotes confidence interval.
† Excluded were 26 patients with missing values for at least one of the risk factors in the model, including the covariates. The Hosmer–
Lemeshow goodness-of-fit chi-square test statistic was 9.82 (P=.278). The C-index for the model was .763.
‡ Excluded were 97 patients with missing values for at least one of the risk factors in the model, including the covariates. The Hosmer–
Lemeshow goodness-of-fit chi-square test statistic was 12.45 (P=.132). The C-index for the model was .763.
§ Complex surgery was defined as surgery under any of the following conditions: a history of coronary-artery bypass grafting, valve surgery,
noncoronary angioplasty or stenting, or other cardiac or vascular noncardiac surgery, combined current heart surgery, or current surgery
in emergency status or urgent status with evidence of congestive heart failure preoperatively.
║Race or ethnic group was determined by the clinical investigator.
¶ No angina indicates admission to cardiac surgery without angina, and generally for other reasons, such as: heart failure, myocardial
infarction, sudden death, or failed percutaneous intervention.
**To convert the value for creatinine to micromoles per liter, multiply by 88.4.
|
