External Prognostic Validations and Comparisons of Age- and Gender-Adjusted Exercise Capacity Predictions
Esther S.H. Kim, MD, MPH*,
Hemant Ishwaran, PhD ,
Eugene Blackstone, MD, FACC, and
Michael S. Lauer, MD, FACC, FAHA*,, ,*
* Cardiovascular Medicine, Cleveland, Ohio
Quantitative Health Sciences, Cleveland, Ohio
Cardiothoracic Surgery, Cleveland Clinic, Cleveland, Ohio
Division of Prevention and Population Science, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland
View larger version (6K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1 Actual Versus Predicted 10-Year Survival Rates in Men
For this calibration plot based on the Veterans Affairs referral nomogram of predicted exercise capacity, patients were divided into quintiles of predicted risk, and 100 bootstrap resamplings were performed. The difference between actual and predicted Kaplan-Meier 10-year death rates was small across all levels of risk.
|
|
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2 Mortality in Men by Ability to Achieve  85% Predicted METs
Failure to achieve 85% of predicted exercise capacity with the VA referral nomogram in men predicted substantially higher death rates. MET = metabolic equivalent.
|
|
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3 Mortality in Women by Ability to Achieve 85% Predicted METs
Failure to achieve 85% of predicted exercise capacity with the St. James Take Heart Project nomogram in women predicted substantially higher death rates. MET = metabolic equivalent.
|
|
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4 Multivariable Adjusted 10-Year Survival Probability in Men on the Basis of Percent Pr METs and Age
All variables listed in Table 1 were used as covariates to predict survival probability. Continuous variables are considered as medians, whereas categorical variables are considered as modes. Dotted lines refer to 95% confidence intervals. There is a significant age interaction where percent predicted (Pr) metabolic equivalents (METs) achieved during stress testing predicts mortality in a different manner on the basis of age with the VA referral nomogram in men (p < 0.001).
|
|
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5 Multivariable Adjusted 10-Year Survival Probability in Women on the Basis of Percent Pr METs and Age
All variables listed in Table 1 were used as covariates to predict survival probability. Continuous variables are considered as medians, whereas categorical variables are considered as modes. Dotted lines refer to 95% confidence intervals. There is a significant age interaction where percent Pr METs achieved during stress testing predicts mortality in a different manner on the basis of age with the Gulati nomogram in women (p = 0.003). Abbreviations as in Figure 4.
|
|
|
