Abstract
Background: As cardio-vascular diseases are the leading cause of death worldwide, establishing measures to improve cardiovascular health is of crucial importance. Exercise plays an essential role in cardiac rehabilitation of patients with coronary artery disease (CAD), in whom an evaluation of the cardiorespiratory fitness (CRF) is necessary. CRF of CAD patients could be assessed using 6-min walk test (6MWT), and the results interpreted by using Enright-Sherill prediction equation which has mainly been designed and evaluated for a healthy population. Hypothesizing that the Enright-Sherill prediction equation might not be best suited for CAD patients, our aim was to reevaluate this equation in CAD patients, and potentially establish a more accurate 6MWD prediction equation to be applied in these patients. Methods: 6MWD was measured in a cross-sectional study in 67 CAD patients (44 women) who were members of the Coronary club Ljubljana, Slovenia. In addition, the predicted 6MWD was calculated for men and women using Enright-Sherill gender specific regression equation. Multivariate regression analysis was used to obtain a new prediction equation, and the agreement between the measured and the predicted 6MWD analyzed using the repeated measures ANOVA. Results: Men achieved 451 ± 122 m and women 485 ± 69 m without significant differences between sexes (F = 0.022, p = 0.882) when adjusted for age, height, body mass, and waist circumference. When comparing the measured (473 ± 91 m) and the predicted (422 ± 57 m) values of 6MWD in CAD patients we found that the Enright-Sherill prediction equation significantly (F = 27.734, p < 0.001) underestimated the 6MWD by 52 ± 81 m. A significant regression equation was established [F (3,63) = 44.663, p < 0.001], with a R (2) of 0.680 where 6MWD equals 1,057 m-4.966 x age (years)-0.614 x WC (cm)-68.629 x NYHA class. Conclusion: The results of this study stress the importance of regular and actual walking ability testing in patients with stable CAD to obtain their CRF, rather than simply predicting it from regression equations obtained from non-representative or non-comparable samples. Our developed prediction equation warrants additional validation and may represent a good substitute for currently used predictions obtained from a healthy population.