Stone, P. H., Thompson, B., Zaret, B. L., Chaitman, B., Gibson, R. S., Schweiger, M. J., Steingart, R., Kirshenbaum, J., Thompson, C., Fung, A., McCabe, C. H., Knatterud, G. L., Braunwald, E. Factors associated with failure of medical therapy in patients with unstable angina and non-Q wave myocardial infarction. A TIMI-IIIB database study European heart journal. 1999;20(15):1084-93.

CONTEXT: Current management of patients with unstable angina and non-Q wave myocardial infarction generally consists of intensive medical therapy, with angiography and revascularization sometimes limited to those who fail such therapy. AIM: To determine if certain baseline characteristics are predictive of patients who fail medical therapy, since such patients could then be expeditiously directed to a more invasive strategy in a cost-effective manner.

METHODS: The study cohort consisted of the 733 patients in the Thrombolysis in Myocardial Ischemia (TIMI) IIIB study who were randomized to conservative strategy. Patients were to be treated with bedrest, anti-ischaemic medications, aspirin, and heparin, and were to undergo risk-stratifying tests, consisting of an exercise test with ECG and thallium scintigraphy, scheduled to be performed within 3 days prior to, or 5 days after, hospital discharge and 24 h Holter monitoring scheduled to begin 2-5 days after randomization. Baseline clinical and ECG characteristics were compared between patients who 'failed' medical therapy and those who did not 'fail'. Failure was defined using clinical end-points (death, myocardial infarction, or spontaneous ischaemia by 6 weeks after randomization) or a strongly positive risk-stratifying test. For each test an ordered failure profile of results was calculated and consisted of death, myocardial infarction, or rest ischaemia occurring prior to performance of the test, a markedly abnormal test result, and no abnormality.

RESULTS: Clinical end-points occurred in 241 (33%) patients and were more likely to occur in patients who at presentation were older, had ST-segment depression on the qualifying ECG, or were being treated with heparin or aspirin. Characteristics independently predictive of developing a clinical event or an abnormal exercise treadmill test included: ST-segment depression on the qualifying ECG, history of prior angina, family history of premature coronary disease (i.e. onset <55 years of age), prior use of heparin or aspirin, and increasing age. By combining these baseline risk characteristics for each outcome the incidence of developing a clinical event ranged from 8% if none was present to 63% if all six were present, and of developing a markedly abnormal risk stratifying test from 8-21% if none were present to approximately 90% if all six were present.

CONCLUSIONS: Baseline characteristics associated with developing a clinical event or a markedly abnormal risk stratifying test were similar: rest anginal episode accompanied by ST-segment depression and occurring despite treatment with aspirin and heparin, a history of angina, older age, and family history of coronary disease. Patients with these characteristics are appropriate candidates for expeditious cardiac catheterization and consideration for revascularization, while patients without them may be suitable for medical management alone.

Trial: TIMI 3B