ASSESSMENT OF RESPIRATORY SYSTEM MECHANICS IN ADULTS: EFFECT OF WEIGHT LOSS, POSTURE, BRONCHODILATION AND ARTEFACTS ON RESPIRATORY IMPEDANCE AND ITS REPEATABILITY

Abstract

Obesity is associated with respiratory symptoms that often improve with weight loss. However, traditional methods of testing pulmonary function such as spirometry and plethysmography show little or no changes with weight loss. Oscillometry which measures respiratory system mechanics could potentially fill this knowledge gap, but it has never been used to measure weight-induced changes in respiratory system mechanics in the supine position, even though such assessment would be more relevant to the mechanics of breathing during sleep. We evaluated 19 severely obese female subjects using spirometry, plethysmography, Pittsburgh Sleep Quality Index, and hand-held oscillometry to obtain respiratory system resistance (Rrs) and reactance from 6-19Hz before, and at five weeks and six months after weight loss surgery. These assessments were performed in both upright and supine positions, and pre- and post-bronchodilation with 200µg of salbutamol. An average weight loss of 11.9±2.7kg at 5 weeks was not associated with changes in upright respiratory mechanics, but Rrs at 19Hz (Rrs,19) was reduced by 13.1±3.8% in the supine position and this correlated with improvements in sleep quality. Weight loss also increased bronchodilator responsiveness, perhaps indicating improvements in airway-parenchymal tethering. At six months, greater mean weight-loss of 21.4±7.1kg caused significant changes in respiratory mechanics in both upright and supine positions. However, weight-loss induced a greater reduction in supine Rrs,19. Together with the early changes at 5 weeks, these results demonstrate the importance of breathing mechanics to sleep quality. By comparison, no significant changes were detected with spirometry. We also characterized the biasing effects of inappropriate positioning of head-and-neck during oscillometry and compared these to the established effects of the upper airway shunt artefact. Impedance values were not significantly affected by 20° neck flexion, and although significant, changes in Rrs during 10° neck extension were small. Oscillometry outcomes were highly repeatable since within-test coefficient of variation (COV) was < 8%, and day-to-day COV was < 6% in all subjects. Furthermore, the test-retest reliability of oscillometry was also very high with Pearson’s correlation coefficient of 0.99. Taken together, hand-held oscillometry was very repeatable, reliable and more sensitive than spirometry at detecting changes in lung mechanics with weight-loss.