Isolated Small Airway Dysfunction and Ventilatory Response to Cardiopulmonary Exercise Testing

Abstract

BACKGROUND:

The effect of isolated small airway dysfunction (SAD) on exercise remains incompletely characterized. We sought to quantify the relationship between isolated SAD, identified with lung testing, and the respiratory response to exercise.

METHODS:

We conducted a prospective evaluation of service members with new-onset dyspnea. All subjects underwent plethysmography, diffusing capacity of the lung for carbon monoxide (D_LCO), impulse oscillometry, high-resolution computed tomography (HRCT), and cardiopulmonary exercise testing (CPET). In subjects with normal basic spirometry, D_LCO, and HRCT, SAD measures were analyzed for associations with ventilatory parameters at submaximal exercise and at maximal exercise during CPET.

RESULTS:

We enrolled 121 subjects with normal basic spirometry (ie, FEV₁, FVC, and FEV₁/FVC), D_LCO, and HRCT. Mean age and body mass index were 37.4 ± 8.8 y and 28.4 ± 3.8 kg/m², respectively, and 110 (90.9%) subjects were male. The prevalence of SAD varied from 2.5% to 28.8% depending on whether FEV₃/FVC, FEF_25-75%, residual volume/total lung capacity, and R5-R20 were used to identify SAD. Agreement on abnormal SAD across tests was poor (kappa = −0.03 to 0.07). R5-R20 abnormalities were related to higher minute ventilation ( ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{E}}}_{}}$ ) and higher ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{E}}}_{}}$ /maximum voluntary ventilation (MVV) during submaximal exercise and to lower ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{O}}}_{\mathbf{\text{2}}}}$ during maximal exercise. After adjustment for differences at baseline, there remained a trend toward a relationship between R5-R20 and an elevated ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{E}}}_{}}$ /MVV during submaximal exercise (β = 0.04, 95% CI −0.01 to 0.09, P = .10), but there was no significant association with ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{E}}}_{}}$ during submaximal exercise or with ${\dot{\mathbf{\text{V}}}}_{{\mathbf{\text{O}}}_{\mathbf{\text{2}}}}$ during maximal exercise. No other SAD measures showed a relationship with ventilatory parameters.

CONCLUSIONS:

In 121 subjects with normal basic spirometry, D_LCO, and HRCT, we found poor agreement across tests used to detect SAD. Among young, healthy service members with postdeployment dyspnea, SAD as identified by lung function testing does not predict changes in the ventilatory response to exercise.