Abstract

SATUT, ZARAYA et al. Valor pronóstico del test de ejercicio. Rev.Urug.Cardiol. [online]. 2006, vol.21, n.1, pp.23-30. ISSN 1688-0420.

SUMMARY Background: the prognostic value of stress tests is not well established for every population. Objectives: to establish the capacity of stress tests in the prediction of cardiovascular events in a population without any limitation in its characteristics. Methods: from a total of 1904 patients who underwent a stress test in a institution during 1997, the clinical evolution at 7 seven years of 524 patients (28%) was assessed by using a structured telephonic questionnaire. The representativity of this group was validated by comparison of the characteristics of contacted and no-contacted patients. Bivariate and multivariate analysis of the variables that could have a predictive value was performed. By using the independent multivariate predictive variables the probability of cardiovascular events was calculated through a logistic function and a table of estimated risks was constructed. Results: there were no important differences between contacted and no-contacted patients, so the representativity of the group of contacted patients was accepted. Concerning the prediction of events, there were significant differences in gender, age, history of hypertension and dyslipidemia, angina and/or arrhythmia during the test, pre and posttest probability of coronary heart disease, ST segment deviation and maximal heart rate attained. By multivariate analysis, pretest probability, maximal heart rate and the magnitude of ST segment deviation during the test, were independent predictors of events. Using these variables, through logistic regression, the probability of events was calculated. The area under the ROC curve was 0.801. With the same variables a table of estimated risks was constructed. Conclusions: pretest probability and maximal heart rate and ST deviation during the test were identified as independent predictors of cardiovascular events; with these variables the probability of events was calculated; this results, especially the table of estimated risks, should be prospectively validated in a bigger cohort of patients.

Keywords : ERGOMETRY; EXERCISE TEST; PREDICTIVE VALUE OF TESTS.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )