Classification system for cardiorespiratory fitness based on a sample of the Brazilian population

Background: Peak oxygen consumption (VO2peak) is an important prognostic marker and its classification helps the cardiologist in the therapeutic decision-making process. The most commonly used cardiorespiratory fitness (CRF) classification has not been validated for the Brazilian population. Objective: To elaborate a CRF classification using a Brazilian sample and to compare it with the American Heart Association (AHA), Cooper and UNIFESP classifications. Methods: A total of 6,568 healthy subjects were analyzed through cardiopulmonary exercise testing (CPET). They were distributed by sex and the following age groups (years): 7-12, 13-19, 20-79 (per decades) and > 80 years. After measurement of the VO2peak, participants were distributed into quintiles of CRF in very poor, poor, moderate, high and very high (AEMA Table). The CRF classifications by AEMA, AHA, Cooper, and UNIFESP were compared using the Wilcoxon, Kappa and concordance percentages. Results: VO2peak presented an inverse and moderate correlation with age considering both sexes (R = -0.488, p < 0.001). All paired comparisons between CRF classification systems showed differences (p < 0.001) and disagreement percentage - AEMA versus AHA (k = 0.291, 56.7%), AEMA versus Cooper (k = 0.220, 62.4%) and AEMA versus UNIFESP (k = 0.201, 63.9 %). Conclusion: The AEMA table showed important discrepancies in the classification of CRF when compared to other tables widely used in our setting. Because it was obtained from a large sample of the Brazilian population, the AEMA table should be preferred over other classification systems in our population