The multiple linear regression model: to predict peak metabolic equivalents and peak oxygen pulse in patients with coronary artery disease after percutaneous coronary intervention.

Background: The clinical indicators of patients with coronary artery disease (CAD) often affect their prognosis. Cardiopulmonary Exercise Testing (CPET) can effectively evaluate the cardiopulmonary ability of CAD patients. The objective of this research was to explore the correlation between some clinical indicators and peak metabolic equivalents (peak METs) and peak oxygen pulse (OP) in patients with CAD. Regression equations were further constructed for indicators with significant correlations to predict peak METs and OP.

Methods: 152 CAD patients were recruited (M: F = 109:43, age = 64.47 ± 7.80 years, including 32 patients with chronic myocardial infarction, 46 with frailty, 93 with hypertension, and 48 with diabetes). All participants had blood biochemistry analysis, cardiac ultrasound, CPET and five time sit-to-stand (FTSTS) test. CPET was tested according to an incremental loading scheme of 10-15 w/min and peak METs, OP were recorded. Stepwise multifactorial linear regression was used to determine which clinical variables should be adjusted to improve peak METs and OP.

Results: Results of multifactorial linear regression showed 2 equations: peak METs = 6.768-0.116*BMI + 0.018*Hgb-0.026*age-0.005*Gensini score (Adjusted R = 0.301, F = 17.239,  < 0.001); OP = -1.066 + 0.264*BMI + 0.049*Hgb-0.035*age (Adjusted R = 0.382, F = 32.106,  < 0.001).

Conclusion: BMI, Hgb, age and Gensini score can be used to predict peak METs and BMI, Hgb and age can be used to predict OP in patients with CAD clinically. Thus, tailored exercise program should be prescribed for individual CAD patient undergoing cardiac rehabilitation and modifying clinical factors such as BMI, Hgb and Gensini score will help to improve their cardiorespiratory fitness and quality of life.