Patient-Reported Dyspnea Correlates Poorly with Aerobic Exercise Capacity Measured During Cardiopulmonary Exercise Testing

Patient-reported dyspnea plays a central role in assessing cardiopulmonary disease. There is little evidence, however, that dyspnea correlates with objective exercise capacity measurements. If the correlation is poor, dyspnea as a proxy for objective assessment may be misleading.

To compare patient’s perception of dyspnea with maximum oxygen uptake (MaxVO2) during cardiopulmonary exercise testing (CPET).

Fifty patients undergoing CPET for dyspnea evaluation were studied prospectively. Dyspnea assessment was measured by a metabolic equivalent of task (METs) table, Mahler Dyspnea Index, Borg Index, number of blocks walked, and flights of stairs climbed before stopping due to dyspnea. These descriptors were compared to MaxVO2.

MaxVO2 showed low correlation with METs table (r = 0.388, p = 0.005) and no correlation with Mahler Index (r = 0.24, p = 0.093), Borg Index (r = −0.017, p = 0.905), number of blocks walked (r = 0.266, p = 0.077) or flights of stairs climbed (r = 0.188, p = 0.217). When adjusted for weight (maxVO2/kg), there was significant correlation between MaxVO2 and METs table (r = 0.711, p < 0.001), moderate correlation with blocks walked (r = 0.614, p < 0.001), and low correlation with Mahler Index (r = 0.488 p = 0.001), Borg Index (r = −0.333 p = 0.036), and flights of stairs (r = 0.457 p = 0.004). Subgroup analysis showed worse correlation when patients with normal CPET were excluded (12/50 excluded). Patients with BMI < 30 had no correlation between Max VO2 and the assessment methods, while patients with BMI > 30 had moderate correlation between MaxVO2 and METs table (r = 0.568, p = 0.002).

Patient-reported dyspnea correlates poorly with MaxVO2 and fails to predict exercise capacity. Reliance on reported dyspnea may result in suboptimal categorization of cardiopulmonary disease severity.