Macrorheology of cystic fibrosis, chronic obstructive pulmonary disease & normal sputum

Abstract

Background: Prior microrheologic assessments of selected, microlitre plugs of cystic fibrosis (CF) sputum suggest no intrinsic rheologic abnormality. However, such analyses may not be representative of CF sputum as a whole. We therefore reassessed this question using whole sputum macrorheology. Additionally, we wished to further explore the relationships between sputum rheology, inflammation and infection.

Methods: Dynamic oscillatory macrorheometry was performed on whole expectorated sputum from stable adults with CF (n = 18) and COPD (n = 12) and induced sputum from normal controls (n = 7). Concomitant sputum inflammatory mediator levels were measured in CF and COPD samples. Sputum collected from CF subjects (n = 6) at commencement and completion of intravenous antibiotic therapy for an infective exacerbation was also assessed.

Results: CF sputum neutrophil elastase activity (NE) was significantly related to degree of sputum purulence (p = 0.049) and correlated significantly with measures of sputum viscoelasticity (r = 0.696, p = 0.008 for storage modulus G' at 9 Hz). There were significant differences in viscoelasticity between subject groups when samples were compared irrespective of appearance/degree of sputum purulence. However, the macrorheology of mucoid CF sputum did not differ from normal sputum (eg median (range) G' at 9 Hz 2.25 (0.79, 3.26) vs 2.04 (1.4,4.6) Pa, p = 1). In contrast, mucoid COPD samples demonstrated significantly greater viscoelasticity (G' at 9 Hz 4.5 (2.4, 23) Pa) than sputum from both CF (p = 0.048) & normal subjects (p = 0.009). Antibiotic therapy during exacerbations was associated with significant reductions in CF sputum viscoelasticity, with mean (SD) G' at 9 Hz decreasing from 28.5 (11.5) Pa at commencement to 6.4 (4.6) Pa on day 7 (p = 0.01).

Conclusion: The macrorheologic properties of whole, mucoid CF sputum are not different from normal, confirming the results of prior microrheologic studies. Instead, CF sputum viscoelasticity is related to secondary infection, decreases with intravenous antibiotic therapy and correlates with inflammation. In contrast, COPD sputum demonstrates inherently greater viscoelasticity, providing a novel target for potential therapeutic interventions.

Figure 1

Box and whisker plot comparing mucoid, mucopurulent and purulent sputum from CF subjects for storage modulus. (Illustrating the graded increase in elasticity measures at 9 Hz occurring with increasing macroscopic purulence of sputum from stable CF subjects. Mucoid n = 4, mucopurulent n = 10, purulent n = 8. * by Mann-Whitney U-test).

Figure 2

Box and whisker plot comparing rheology of sputum (irrespective of macroscopic appearance) from normal, COPD and CF subjects, for dynamic viscosity. (As indicated, significant differences were observed between all groups for dynamic viscosity η' at 0.3 Hz. Normal n = 7, COPD n = 12, CF n = 18. * by Mann-Whitney U Test).

Figure 3

Box and whisker plot comparing rheology of mucoid sputum from normal, COPD and CF subjects for storage modulus. (No significant differences were observed between normal and CF sputum for storage modulus G' at 9 Hz. Normal n = 7, CF n = 4, COPD n = 8. * by Mann-Whitney U Test).

Figure 4

Change in CF sputum rheology with therapy for an infective exacerbation (Change in mean storage modulus G' at 9 Hz is depicted. Error bars indicate SEM. * p = 0.01 for the comparison with day 0 by paired T-test; Using a Bonferroni adjustment, significance at the 5% level is indicated by a p value of 0.01).