Limitations of applicability of current calibration procedures for respiratory inductive plethysmography

Repeatability of calibration of the respiratory inductive plethysmograph and its stability with time were investigated in young healthy subjects. Eight were submitted to sequential runs of calibration-validation by simultaneous equation (SEM) and least squares methods (LSM) (30 in all): in five subjects, after an initial acceptable calibration, accuracy and precision were repeatedly checked over three hours (a total of 50 trials). Although mean errors for the whole sample seemed acceptably low (7.6%, SD 7.0 for SEM; 7.3%, SD 7.3 for LSM), a marked inconsistency of apparently comparable calibrations and a wide variability of results along but independently of the time in steady conditions (ranging up to 30%) were discovered. An analysis of factors interfering with the response of the equipment was made. The unreliability of the current calibration procedure was related to the fact that in most cases the calibration line derived by the change-in-posture technique, as currently applied, may not be adequately representative of the full range of relationships between the participant variables, as they occur in the selected posture. To overcome this difficulty, two suggestions are made: a) prolong the calibration procedure so as to use more data points, and b) investigate alternative procedures of calibration in any one position.