Blood Pressure Response during the Exercise Treadmill Test and the Risk of Future Hypertension and Cardiovascular Disease

BACKGROUND AND OBJECTIVES: Many studies had established the risk factors for cardiovascular disease. The Duke treadmill score has gained widespread acceptance for making the prognosis and diagnosis for patients with cardiac disease. Recently, the changes in blood pressure during and after exercise have also been studied to predict the prognosis of cardiac disease. We examined the relationship between the incidence of hypertension or cardiovascular disease and the changes of blood pressure during a routine exercise treadmill test. SUBJECTS AND METHODS: 256 men were screened, and they performed exercise treadmill tests from March to May, 2000. Those subjects with histories of hypertension and ischemic heart disease or who were newly diagnosed with ischemic heart disease were excluded. 109 subjects were selected for the final analysis. The follow up period was 78 months. Review of medical records and telephone interviews were used for follow up. We defined clinical events as new onset hypertension, ischemic heart disease, congestive heart failure, cerebrovascular accident, diabetes and atrial fibrillation. The peak systolic blood pressure of 182.5mmHg had the highest specificity and sensitivity on the receiver operating characteristic (ROC) curve of the systolic blood pressure for prediction of clinical events. We defined a hypertensive response as a peak systolic blood pressure over 180 mmHg. RESULTS: 43 (39.4%) of the subjects had a hypertensive response on their exercise treadmill test. The mean exercise capacity was higher in the hypertensive response group. No significant differences were found between the hypertensive and non-hypertensive response groups, in terms of age, gender, body weight, height, body mass index and resting blood pressure. 18 (41.8%) of the hypertensive response subjects had clinical events, while only 11 (16.6%) of the non-hypertensive response subjects had clinical events. The hypertensive response group had more clinical events (p=0.006). 14 (32.5%) of the hypertensive response subjects had hypertension, while only 10 (15.1%) of the non-hypertensive response group had hypertension. The hypertensive response group had more hypertension (p=0.044). On the multivariate analysis, the hypertensive response on the exercise treadmill test was an independent risk factor for hypertension and clinical events (odds ratio=3.990, 95% confidence interval; 1.473-10.808, p=0.006). CONSLUSION: These results indicate that the exercise blood pressure response seems to be a risk factor for hypertension and clinical events. Careful medical care and close follow up may be needed for subjects with a hypertensive blood pressure response on the exercise treadmill test. Further study is needed to understand the significance of an exaggerated blood pressure response on the exercise treadmill test.

Influences of Perfusion Defect on the Measurement of Left Ventricular Ejection Fraction and Volumes in Gated Myocardial Perfusion SPECT

BACKGROUND AND OBJECTIVES: The left ventricular ejection fraction (LVEF) and volume (LVV) are important variables in patients with coronary artery disease. Quantitative gated myocardial SPECT (QGS) permits the simultaneous assessment of perfusion, LVEF and LVV. However, the presence of a perfusion defect may influence the LVEF and LVV measured by QGS. SUBJECTS AND METHODS: 67 subjects (M/F=47/20; mean age: 60.2+/-12.4 years) underwent both QGS with Tc-99m MIBI and 2-D echocardiography (Echo) at less than 7 days apart. The LVEF and LVV were measured by Echo, using the modified Simpson's method, and by QGS, using the automatic software, AutoQUANT(TM). The QGS rest images were used to compare with the Echo. RESULTS: The correlations between the QGS and Echo for LVEF, LVEDV and LVESV were good in all 67 subjects (r=0.781, 0.754 and 0.906, respectively, p<0.0001). In patients with no perfusion defect (n=34), the correlations between the QGS and Echo for LVEF, LVEDV and LVESV were good (r=0.689, 0.593 and 0.586, p<0.0001). In patients with a perfusion defect (n=33), the LVEF between the QGS and Echo was well correlated (r=0.777, p<0.0001), but the LVEF was higher by 7.1+/-8.7% from the Echo results. The LVEDV and LVESV by both QGS and Echo were also well correlated (r=0.804 and 0.929, respectively, p<0.0001), but the LVEDV and LVESV were higher from QGS by 17.9+/-34 and 16.9+/-25 mL, respectively. A Bland-Altman analysis showed the agreement between the QGS and Echo in patients without perfusion defect was better than for those with a perfusion defect. CONCLUSION: The perfusion defect from QGS might affect the measurements of the LVEF and LVV; therefore, the QGS and Echo values are not interchangeable.