The purpose of this study is to validate two models which categorize patients with syncope into high and low risk for either sudden death or diagnostic arrhythmias based on data available from the initial history, physical examination, and EKG. The models to be validated include: 1) Predictors of Sudden Death. A model for prediction of sudden death was developed in patients in whom a cause of syncope was not established by initial history and physical examination. The predictors included a history of diabetes mellitus, renal insufficiency (creatinine greater than 2), left ventricular hypertrophy, left bundle branch block and left axis deviation. 2) Predictors of Diagnostic Arrhythmias on Monitoring. Predictors of diagnostic arrhythmias were developed in patients in whom a cause of syncope was not established from initial history, physical examination, or EKG and include a history of ventricular tachycardia and an abnormal EKG by specific criteria. Approximately 500 patients with syncope will be accrued from the emergency room, the inpatient services and the ambulatory clinics of the Presbyterian University Hospital of Pittsburgh. All patients will undergo a basic standardized evaluation consisting of a history, physical examination, baseline laboratory tests, electrocardiogram, prolonged electrocardiographic monitoring, and special diagnostic tests as necessary. Diagnosis of a cause of syncope will be assigned by standardized criteria. Follow-up information regarding sudden death and mortality will be obtained at three-month intervals. Causes of death will be assigned in a standardized manner. The performance of the models will be determined by 1) estimation of the sensitivity and specificity of the previously derived models on the new group of patients, 2) a comparison of the area under the ROC curve derived from the derivation set to the area from the validation set, and 3) calculation of the likelihood ratios associated with each risk score in order to convert pretest probabilities to post-test probabilities. These models are important because they identify high risk patients by using data available at initial presentation. The identification of these high risk patients at presentation will lead to more rational decisions regarding their hospitalization and diagnostic evaluation. Furthermore, these models will lead to future studies investigating the impact of various treatment interventions in these high risk groups.
| Oh, J H; Hanusa, B H; Kapoor, W N (1999) Do symptoms predict cardiac arrhythmias and mortality in patients with syncope? Arch Intern Med 159:375-80 |
| Martin, T P; Hanusa, B H; Kapoor, W N (1997) Risk stratification of patients with syncope. Ann Emerg Med 29:459-66 |
| Kapoor, W N; Fortunato, M; Hanusa, B H et al. (1995) Psychiatric illnesses in patients with syncope. Am J Med 99:505-12 |
| Kapoor, W N; Smith, M A; Miller, N L (1994) Upright tilt testing in evaluating syncope: a comprehensive literature review. Am J Med 97:78-88 |
| Kapoor, W N (1994) Syncope in older persons. J Am Geriatr Soc 42:426-36 |
| Levy, M A; Arnold, R M; Fine, M J et al. (1993) Professional courtesy--current practices and attitudes. N Engl J Med 329:1627-31 |
| Kapoor, W N (1992) Evaluation and management of the patient with syncope. JAMA 268:2553-60 |
| Kapoor, W N; Brant, N (1992) Evaluation of syncope by upright tilt testing with isoproterenol. A nonspecific test. Ann Intern Med 116:358-63 |
| Kapoor, W N (1991) Diagnostic evaluation of syncope. Am J Med 90:91-106 |
| Atkins, D; Hanusa, B; Sefcik, T et al. (1991) Syncope and orthostatic hypotension. Am J Med 91:179-85 |
Showing the most recent 10 out of 16 publications
