Table 3.
Independent predictors of increased TRV by multiple linear regression analysis.
| N = 320 | Beta (95% CI) [Estimated increase in TRV in m/sec] | P |
|---|---|---|
| Hemolytic component (1 SD increase) | 0.09 (0.06, 0.12) | <0.0001 |
| LV lateral E/e′ ratio (natural log increase) | 0.19 (0.09, 0.29) | 0.0001 |
| BUN (natural log increase) | 0.10 (0.05, 0.16) | 0.0002 |
| Serum erythropoietin (natural log increase) | 0.07 (0.03, 0.11) | 0.002 |
Variables entered into model were: age, systolic blood pressure, pulse pressure, oxygen saturation, hemoglobin, erythropoietin, BUN, creatinine, hemolytic component, left atrial volume index, left ventricular mass index, LV lateral E/e′.
Eleven outliers were removed, R-square = 0.25
Each of the following variables could substitute for hemolytic component in the multiple linear regression model as follows (beta and 95% CI are given):
Total bilirubin (natural log): N=373, beta = 0.06, 95% CI = 0.02, 0.11.
Aspartate aminotransferase* (natural log): N=368, beta = 0.07, 95% CI = 0.01, 0.13.
Reticulocyte % (square root): N=350, beta = 0.05, 95% CI = 0.02, 0.09.
Lactate dehydrogenase* (natural log): N=348, beta = 0.10, 95% CI = 0.04, 0.17
site adjusted