Prognostic Value of a Treadmill Exercise Score in Symptomatic Patients With Nonspecific ST-T Abnormalities on Resting ECG.

Kwok JMF, Miller TD, Christian TF, et al (Division of Cardiovascular Diseases and Departments of Internal Medicine and Biostatistics, Mayo Clinic and Mayo Foundation, Rochester, Minn; Medical Department, Princess Margaret Hospital, Kowloon, Hong Kong), JAMA. 1999;282:1047-1053.

This study examined the prognostic capability of the Duke Treadmill Score for comparing patients with symptoms of chest pain or dyspnea who had nonspecific ST-T abnormalities on a resting electrocardiogram (ECG) to those patients with the same symptoms who had normal resting ECGs. The authors stated that data from patients with nonspecific ST-T abnormalities was used in the formulation of the Duke Treadmill Score but that this subgroup had not been tested as a separate entity.

The researchers studied 2,405 patients with symptoms of chest pain or dyspnea who underwent exercise thallium testing between 1989 and 1991. Of these patients, 939 had nonspecific ST-T changes and formed the study group, and 1,466 had normal findings on the resting ECG and formed the control group. The patients excluded from the study were those with congenital, valvular, or cardiomyopathic heart disease; with prior coronary artery revascularization; with a resting ECG with secondary ST-T abnormalities; or with missing data. Bruce or Naughton protocols were used for the exercise test, and the endpoint of the test was determined by at least 1 of 3 criteria: severe fatigue, moderate angina, or a 2-mm or greater ST-segment depression. The data were placed in the following formula to determine the Duke Treadmill Score: exercise time -- (5 x maximum ST deviation) -- (4 x angina index).

The 7-year follow-up on each patient was carried out by mailed questionnaires, telephone interviews, medical record review, or physician contact to determine subsequent cardiac incidents. These incidents included death, nonfatal myocardial infarction, or revascularization procedures including percutaneous transluminal coronary angioplasty or coronary artery bypass graft.

The authors pointed out 2 limitations to their study: (1) the patients were symptomatic and referred for the exercise thallium imaging test and not the standard treadmill test, and (2) it was not a true, natural, history study because the patients were treated during the follow-up period, which probably lowered the incidence of cardiac events. Despite these limitations, they concluded that patients with resting nonspecific ST-T abnormalities could be successfully categorized into low-, intermediate-, and high-risk groups using the Duke Treadmill Score. Within these categories, the study group had a significantly lower 7-year survival rate, fewer low-risk patients, and more high-risk patients than the control group. The authors concluded that this successful risk stratification demonstrates the importance of the exercise ECG and that the Duke Treadmill Score and can lead to appropriate follow-up care.

Charles P Gray, PT University of South Alabama Mobile, Ala