Author + information
- Received September 20, 1983
- Revision received January 9, 1984
- Accepted January 19, 1984
- Published online July 1, 1984.
- Joseph Hung, MB*,1,
- Bernard R. Chaitman, MD‡,1,
- Jules Lam, MD1,
- Jacques Lesperance, MD1,
- Georges Dupras, MD1,
- Phillippe Fines, MSc1 and
- Martial G. Bourassa, MD, FACC1
- ↵†Address for reprints: Bernard R. Chaitman, MD, St. Louis University Medical Center, Division of Cardiology, 1325 South Grand Boulevard, St. Louis, Missouri 63104.
Several diagnostic noninvasive tests to detect coronary and multivessel coronary disease are available for women. However, all are imperfect and it is not yet clear whether one particular test provides substantially more information than others. The aim of this study was to evaluate clinical findings, exercise electrocardiography, exercise thallium myocardial scintigraphy and cardiac fluoroscopy in 92 symptomatic women without previous infarction and determine which tests were most useful in determining the presence of coronary disease and its severity. Univariate analysis revealed two clinical, eight exercise electrocardiographic, seven myocardial scintigraphic and seven fluoroscopic variables predictive of coronary or multivessel disease with 70% or greater stenosis. The multivariate discriminant function analysis selected a reversible thallium defect, coronary calcification and character of chest pain syndrome (p ¼ 0.05) as the variables most predictive of presence or absence of coronary disease. The ranked order of variables most predictive of multivessel disease were cardiac fluoroscopy score, thallium score and extent of ST segment depression in 14 electrocardiographic leads. Each provided statistically significant information to the model. The estimate of predictive accuracy was 89% for coronary disease and 97% for multivessel coronary disease.
The results suggest that cardiac fluoroscopy or thallium scintigraphy provide significantly more diagnostic information than exercise electrocardiography in women over a wide range of clinical patient subsets.
- Received September 20, 1983.
- Revision received January 9, 1984.
- Accepted January 19, 1984.
- American College of Cardiology Foundation