Cardiopulmonary exercise testing is a common tool used to differentiate the numerous causes of dyspnea. Exceeding 70% of the maximal voluntary ventilation (MVV) during maximal exercise is usually viewed as a sign of ventilatory limitation to exercise. Traditionally, the MVV is estimated by multiplying the FEV1 by 40 rather than by direct measurement. Our goal was to determine if including other spirometric variables in a mathematical model could more accurately predict the MVV. We also sought to determine if different models were appropriate for different pathophysiologic subgroups of patients.
Retrospective review of spirometries and MVVs was performed. Data collected included patient demographics, spirometric values, and measured MVVs. Stepwise regression was performed to build mathematical models to correlate the measured MVV with spirometric data of all individuals. Furthermore, additional models were built for patients with normal, obstructive, and restrictive spirometry.
585 records were reviewed. One third were normal spirometries; 39% had obstructive physiology; 24% had restrictive physiology. A mathematical model including multiple spirometric variables such as the peak expiratory flow rate (PEFR) and the maximal inspiratory flow rate (MIFR) more accurately predicted the measured MVV than did the traditinal formula. Furthermore even better correlation was obtained by constructing different formulas specific for the physiologic subgroups of normal, obstructive, and restrictive.CONCLUSIONS: Including of multiple spirometric variables in a mathematical model improved upon the estimation of MVV compared to the traditional method of multiplying the FEV1 by 40. Specific formulas for different spirometric patterns further improved the accuracy of the model.
Specific prediction equations for MVV based on the underlying spirometric physiology lead to a more accurate estimation, thus allowing improved interpretation of cardiopulmonary exercise testing.
S.S. Razi, None.