Objective: To evaluate various methods for studying the perception of dyspnea in chronic obstructive pulmonary disease (COPD) using a new parameter, the change in Borg scale rating, and others already in use: the linear regression slope and the application of Stevens' law to the response perception curve--ie change in forced expiratory volume in 1 second (delta FEV1)--change in dyspnea (delta dyspnea).
Patients and methods: A bronchial challenge test was performed on 70 patients with stable COPD and no contraindications for performing the test (European Respiratory Society criteria), during which dyspnea was measured (Borg scale) after each nebulization. Perception was analyzed using: a) the linear regression slope of delta FEV1 plotted against (delta dyspnea); b) the exponent n of Stevens' law (psi=k phi n, in which psi is delta dyspnea and phi is delta FEV1, with perception being poor when n<1 and good when n>1), and c) change in Borg: difference between dyspnea when FEV1 has fallen 20% and dyspnea after saline inhalation. Subjects were classified according to the slope and change in Borg as hypoperceivers, normal perceivers, or hyperperceivers. These 2 methods of classification were compared using the kappa statistic.
Results: According to the exponent n, all patients were hypoperceivers (n<1). According to the slope, there were 33 hypoperceivers, 28 normal perceivers, and 9 hyperperceivers. The change in Borg classified 37 subjects as hypoperceivers, 23 as normal perceivers, and 10 as hyperperceivers. All except 5 subjects were classified in the same way by the slope and the change in Borg (kappa=0.88). In most of the 5 cases of discrepancy, the slope classified subjects as better perceivers.
Conclusions: The n exponent is not valid for evaluating the perception of dyspnea induced by a bronchial challenge test in COPD. Change in Borg is at least as useful as the slope for evaluating perception of dyspnea. The percentage of patients with this disease who are hyperperceivers is high.