Inherent variability of pulmonary function tests in infants with bronchiolitis.

During the last decade, many new methods of assessing pulmonary function in infants have been introduced. However, considerable controversy has arisen when these different techniques are used to assess the efficacy of various inhaled medications. The currently employed measures of pulmonary function have been reported to have coefficients of variation of less than 10%. However, these figures refer to repeated measurements made in individual babies over a short period of time, with the infant in a constant position. They may not be applicable to studies on the effects of drugs when a significant time interval elapses between measurements, particularly if the infant is repositioned during this time. In the present study, we found that the variability of measurements of thoracic gas volume (TGV), passive respiratory mechanics, and forced expiration approximately doubled if the infant was taken out of the plethysmograph, or repositioned within the plethysmograph, between sets of measurements. The between-test coefficients of variation ranged from 4.8% to 18.9% for TGV, 7.8% to 23.5% for respiratory system resistance, from 5.0% to 24.1% for respiratory system compliance, and from 10.8% to 36.1% for maximal flow at functional residual capacity. These results demonstrate that the commonly employed infant pulmonary function tests have wide variability in infants with acute wheezing illnesses. Before studying the efficacy of therapeutic interventions in such infants, it is necessary to establish the individual variability of the tests in the infants under the conditions of the study. Failure to do so may result in erroneous conclusions and recommendations.