Aerosal.it2

Rheology of Cystic Fibrosis Sputum after in vitro
Treatment with Hypertonic Saline Alone and in
Combination with Recombinant Human
Deoxyribonuclease I
MALCOLM KING, BONNIE DASGUPTA, ROBERT P. TOMKIEWICZ*, NEIL E. BROWN
Pulmonary Research Group, University of Alberta, Edmonton, Alberta, Canada Treatment with recombinant human deoxyribonuclease I (rhDNase) is currently used as therapy for
cystic fibrosis (CF) lung disease. Hypertonic saline (HS) acts as an expectorant promoting mucus se-
cretion and augmenting the volume of sputum. We evaluated the individual and combined effects of
HS and rhDNase in vitro on the viscoelasticity of CF sputum. Sputum samples were collected from
nine CF patients to use for in vitro testing. Aliquots of CF sputum (0.20 to 0.40 g) were subjected to
the following protocols: (1) negative control sample without any treatment; (2) positive control sam-
ple, adding 10% volume of normal saline (0.9% NaCl); (3) application of hypertonic saline (HS-3%
NaCl); (4) combining approximately 100 nM concentration of rhDNase with protocols 2 and 3. The
samples in protocols 2 through 4 were incubated for 30 min at 37Њ C. For each protocol, CF sputum
was analyzed at baseline and at 30 min for spinnability by filancemeter and viscoelasticity by mag-
netic microrheometry. Spinnability decreased for the sputum samples that were treated with rhD-
Nase, in combination with either HS or normal saline. Treatment with HS alone and combined treat-
ment with rhDNase and HS decreased log G* (the principal viscoelasticity index) to the same degree.
Saline alone and rhDNase in normal saline both increased the predicted cough clearability of the spu-
tum; however, the combined treatment with rhDNase and hypertonic saline had the best overall ef-
fect on cough clearability. The change in predicted mucociliary clearability, although greatest after
HS, was not significant. These in vitro results suggest that combined treatment with rhDNase and HS
should be evaluated further as a potential mucotropic approach to augment the clearance of puru-
lent sputum in CF lung disease. King M, Dasgupta B, Tomkiewicz RP, Brown NE. Rheology of
cystic fibrosis sputum after in vitro treatment with hypertonic saline alone and in combina-
tion with recombinant human deoxyribonuclease I.
AM J RESPIR CRIT CARE MED 1997;156:173–177.
Defects in the cystic fibrosis (CF) transmembrane regulator way secretions (5). Mucus rheology can be influenced by the protein are responsible for abnormal ion transport in airway concentration of ions within the mucous gel (6). One of the epithelial cells leading to complex pathophysiology of muco- pathologic mechanisms in CF is thought to be due to de- ciliary function with impaired mucus clearance in CF lung dis- creased ionic content in respiratory mucus that is believed to ease (1). Impaired lung clearance in CF has been correlated increase repulsive interactions within the mucin macromole- with deteriorating lung function (2), and severely impaired cules, increasing molecular dimensions and thereby augment- lung function has been associated with poor short-term sur- ing the rigidity of the mucous gel network (7).
Since accumulation of thickened secretions plugging the The rheological properties of the mucus, along with the ef- airways is considered a major contributing factor to the de- ficiency of ciliary action, are both responsible for the effective- creased lung function, the current treatment of CF lung dis- ness of mucociliary clearance (4). Two groups of macromole- ease includes strategies aimed at changing the physical prop- cules, DNA and mucous glycoproteins, have been reported to erties of respiratory mucus (1). One such strategy deals with be the major contributors of the physical properties of CF air- another mechanism responsible for increased rigidity of mu-cus, namely the increased concentration of high molecularneutrophilic DNA found in inflammatory secretions. Secre- (Received in original form December 18, 1995 and in revised form October 7, 1996 ) tions from patients with CF have been shown to have DNA *Current address: Pediatric Research Institute, St. Louis University.
contents up to 10.2% of the dry weight (8), and DNA has been Study supported by the Canadian Cystic Fibrosis Foundation.
reported to accumulate at an average concentration of 59 mg/ Correspondence and requests for reprints should be addressed to Malcolm King, ml (9). Treatment of CF sputum with rhDNase decreases the Ph.D., 173 Heritage Medical Research Center, University of Alberta, Edmonton, molecular size of the DNA, thus reducing its contribution to AB, T6G 2S2 Canada. E-mail: [email protected] viscoelasticity (10). Clinical studies with rhDNase have shown Am J Respir Crit Care Med
Vol. 156. pp. 173–177, 1997
it to improve lung function in patients with CF and decrease AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE the number of exacerbations of infections necessitating the Viscoelasticity and Clearance Indices
The magnetic microrheometer technique was used to measure the vis- Administration of hypertonic saline (HS) has been re- cosity and elasticity of the sputum samples. A 100 ␮m steel ball was ported to increase whole lung clearance in CF patients (12), positioned in a 5–10 ␮l sample of sputum, and the motion of this and treatment of CF sputum in vitro has been demonstrated to sphere under the influence of an electromagnet was used to determine increase its clearability on ciliated epithelium (13). The mech- the theological properties of the sputum. The image of the steel ball anism for these effects is not fully understood. HS may be was projected via a microscope onto a pair of photocells, whose out- changing the physical and transport properties of CF secre- put was amplified and transmitted to an oscilloscope. By plotting thedisplacement of the ball against the magnetic driving force, the vis- tions by modifying the hydrogen and ionic bonding between coelastic properties of the mucus were ascertained (18).
The parameters of mucus viscoelasticity determined were the ri- Therefore the purpose of this study was to examine the ef- gidity index or mechanical impedance, i.e., G*, reported here on a log fects of administration of HS on the rheological properties of scale, expressing the vector sum of “viscosity ϩ elasticity”, and the CF sputum that could provide a mechanism for the reported loss tangent, i.e., tan ␦, reflecting the ratio of “viscosity/elasticity”, at clearance changes with HS treatment. Also the in vitro effect low (1 rad/s) and high (100 rad/s) frequency (18). Two derivative pa- of combined treatment with HS and DNase was evaluated.
rameters—mucociliary clearability index (MCI) and cough clearabil- The individual effects of hypertonic saline (HS 3% NaCl), as ity index (CCI)—were computed from in vitro relationships (19).
well as normotonic saline (NS 0.9% NaCl) were compared These two indices predict mucus clearability by ciliary and coughmechanisms, respectively, based on the measured theological proper- with the combined effects of DNase and HS or NS on the rhe- ties and observations of clearance from model studies. The respective ology and clearability of CF sputum.
MCI = 1.62 – (0.22 × log G* 1) – (0.77 × tan δ 1) Subjects
CCI = 3.44 – (1.07 × log G* 100) + (0.89 × tan δ 100) Sputum samples were collected from nine patients with CF by volun- Statistical Analysis
tary expectoration during a routine clinical visit. The patients (18 to 37 yr, mean age 27 yr) were all infected with Pseudomonas aeruginosa Data from each protocol are presented as mean SD of the mean. To and treated with steroids, antibiotics, and bronchodilators as required.
analyze the significance of changes in spinnability, log G* at 1 rad/s, None of the patients had been exposed to either rhDNase or HS aero- MCI, and CCI after administration of NS, rhDNase, and HS, the spu- sol up to the time of the sputum collection, and none were using any tum from each patient served as its own control. Equality of means other type of mucolytic preparation. Approval to collect and use spu- was tested by analysis of variance (ANOVA), and post hoc analysis of tum for this in vitro analysis was obtained from the University of Al- changes from baseline was determined by the two-tailed, paired t-test.
The paired t-test was also used to determine the differences betweenspinnability and viscoelasticity after different treatments. The Stat- Study Design
View statistical package (Abacus Concepts, Palo Alto, CA) was usedto carry out these analyses.
Aliquots of each sputum sample (0.2 to 0.4 g) were subjected to threedifferent treatment protocols: (1) baseline (negative control), without application of any treatment (e.g., rhDNase, HS); (2) positive control,to test the dilution effect, adding 10% (vol/wt) of 0.9% NaCl (NS); (3) Complete spinnability measurements pre- and post-treatment incubation with ten percent of 3% NaCl (HS). The samples in proto- were obtained on six samples of CF sputum; no spinnability cols 2 and 3 were incubated at 37Њ C for 30 min. Ten percent volume of data were obtained on three other samples because of inade- treatment solutions per weight of sputum (vol/wt) was chosen because quate sample size. Post-treatment viscoelasticity measure- by itself this vehicle treatment had no significant effect on viscoelastic- ments by magnetic rheometry were successful on eight of nine ity in previous in vitro experiments involving DNase (14, 16).
To observe the combined effects of HS or NS with rhDNase, pro- samples; one sample was rejected because the sputum was too tocols 2 and 3 were combined with rhDNase treatment (Pulmozyme; liquid for analysis by this rheological technique. Pretreatment Genentech Inc., South San Francisco, CA) to achieve a final concen- viscoelastic data were only obtained for six samples in order to tration of 2.5 ␮g/ml or approximately 100 nM and then incubated at preserve sufficient sample for the spinnability measurements.
37Њ C for 30 min. This concentration of DNase was chosen on the basisof previous experience in our laboratory (14–16), while 3% NaCl solu- Spinnability Measurements
tion has been commonly used for clinical sputum induction.
Compared with baseline, 30 min of single treatments with NS For each treatment protocol, spinnability and viscoelasticity were (0.9% NaCl) and HS (3% NaCl) resulted in a mean decrease measured where possible, and mucociliary clearability index (MCI) in spinnability of CF sputum from baseline of 16% and 26%, and cough clearability index (CCI) were calculated prior to any treat-ment (baseline), and then after 30 min of application of the treatment.
respectively. The mean decrease in spinnability for the com- Three spinnability readings per aliquot were taken, and the arithmetic bined treatment of NS and DNase was 37%, while after com- mean of the three readings were calculated. Viscoelasticity measure- bined treatment of HS and DNase the decrease was 40%.
When compared with NS as a control (i.e., at 30 min), HS ver-sus NS gave a 12% decrease in spinnability, and the combina- Rheological Measurements on CF Sputum
tion of NS and DNase treatment versus NS gave a 25% de- In this in vitro study, two techniques were used to measure the theo- crease, whereas the combination of HS and DNase versus NS logical properties of sputum: spinnability by filancemeter and vis- treatment resulted in a 29% decrease.
Individual effects of mucolysis. Incubation with HS over 30 Spinnability is the thread forming ability of mucus under the influ- min resulted a small decrease in spinnability in comparison ence of low amplitude elastic deformation. The spinnability of CF with incubation with NS (10.7 Ϯ 0.61 versus 12.18 Ϯ 0.79 mm, sputum samples was measured using a Filancemeter (SEFAM, Nancy, respectively; p ϭ 0.0085). Administration of NS itself resulted France) (17), in which a 20 to 30 ␮l mucus sample is stretched at a dis-traction velocity of 10 mm/s. An electric signal conducted through the in a small but highly significant decrease in spinnability com- mucus sample is interrupted at the point where the mucus thread is pared with the undiluted control, which had no treatment ap- broken. The length of this thread is known as the mucus spinnability plied over 30 min (12.18 Ϯ 0.79 versus 14.43 Ϯ 1.17 mm, re- spectively; p ϭ 0.0001) (Figure 1). King, Dasgupta, Tomkiewicz, et al.: Hypertonic Saline, DNase and CF Sputum Figure 1. Individual and combined effects of rhDNase (final con-
Figure 2. Individual and combined effects of rhDNase (final con-
centration 100 nM) and hypertonic saline (HS, 3% NaCl) on spin- centration 100 nM) and hypertonic saline (HS Ϫ ѨNaCl ϭ 38 mM) nability (mean Ϯ SEM) of six samples of CF sputum for different on viscoelasticity (log G* at 1 rad/s, mean Ϯ SEM) of eight sam- treatment protocols. Control ϭ pre-treatment; saline ϭ NS ϭ ples of CF sputum for different treatment protocols. Saline ϭ NS ϭ 0.9% NaCl. The estimated increase in sputum NaCl concentration 0.9% NaCl. For Control only (no treatment), n ϭ 6.
due to the incubation with HS (10% vol/wt) was 38 mM.
Combined effects of mucolysis. After a period of 30 min, HS ϩ DNase versus 1.57 Ϯ 0.46 for NS alone; p ϭ 0.050). The the combination of DNase and HS treatments decreased spin- differences in predicted mucociliary clearability index (MCI) nability considerably more than the administration of HS were smaller than the CCI changes. The greatest MCI oc- alone over the same period of time (8.67 Ϯ 1.03 versus 10.7 Ϯ curred for HS treatment, but the difference from NS control 0.61 mm, respectively; p ϭ 0.0009). Combined treatment with was nonsignificant (0.99 Ϯ 0.14 versus 0.92 Ϯ 0.09, respec- DNase and NS also decreased spinnability significantly more than administration of NS alone over a period of 30 min(9.10 Ϯ 1.04 versus 12.18 Ϯ 0.79 mm, respectively; p ϭ 0.0011).
DISCUSSION
The additional effect of HS in the presence of DNase, al- Previous in vitro studies report HS to be more effective in re- though very small, was consistent and statistically significant ducing mucoid sputum viscosity in comparison to water (21, (8.67 Ϯ 1.03 versus 9.10 Ϯ 1.04 mm, respectively; p ϭ 0.0071) 22). A decrease in viscoelasticity and an increase in sputum clearability was also demonstrated in this study. These in vitroresults provide support for the work done by Pavia and co- Viscoelasticity and Clearance Indices
workers (23), who observed enhanced mucociliary clearance Incubation of CF sputum samples with HS demonstrated a de- rates in patients with chronic bronchitis after inhalation with crease in the sputum rigidity index (log G* at 1 rad/s) com- 1.21 M NaCl, and for Robinson and colleagues (12), who re- pared with treatment with NS (1.41 Ϯ 049 for HS versus 1.97Ϯ0.35 for NS; p ϭ 0.041). Combined treatment with HS andDNase showed a comparable, significant decrease in log G* 1compared with CF sputum treated with NS and DNase (1.39Ϯ0.46 for HS ϩ DNase versus 1.91 Ϯ 0.81 for NS ϩ DNase; p ϭ0.049). However, singular treatment with DNase (in NS) re-sulted in only a small, nonsignificant decrease in rigidity com-pared with NS treatment itself. Administration of NS itself re-sulted in no significant change in viscoelasticity comparedwith the undiluted control. These viscoelastic data (log G* 1)are presented in Figure 2. The changes in rigidity determined at 100 rad/s were simi- lar to those seen at 1 rad/s, i.e., a decrease for HS treatment,with or without DNase, but only a small, nonsignificant de-crease in log G* 100 for DNase treatment itself. None of thechanges in tan ␦ by treatment group were statistically signifi-cant.
As illustrated in Figure 3, HS treatment of CF sputum also demonstrated significant changes in the predicted cough clear-ability index (CCI) in comparison to treatment with NS(2.31 Ϯ 0.80 for HS versus 1.57 Ϯ 0.46 for NS; p ϭ 0.046). The Figure 3. Individual and combined effects of rhDNase (final con-
mean CCI obtained for DNase treatment was numerically centration 100 nM) and hypertonic saline (HS Ϫ ѨNaCl ϭ 38 mM) similar (2.33 Ϯ 2.17), but the difference from NS control was on cough clearability index (CCI mean Ϯ SEM) of eight samples of nonsignificant. The largest change in CCI was seen for the CF sputum for different treatment protocols. Saline ϭ NS ϭ 0.9% combination of DNase and HS treatment (2.90 Ϯ 1.60 for NaCl. For Control only (no treatment), n ϭ 6.
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE cently found that inhalation of 7% NaCl (1.20 M) stimulated The combined treatments of DNase and HS and DNase pulmonary mucus clearance in cystic fibrosis patients better and NS showed a greater reduction in spinnability in compari- than either aerosolized amiloride or coughing. After observ- son to individual treatments of HS alone. The additive effects ing the results of our study and work done by others, it would of both HS and DNase and DNase and NS in reducing the seem that HS solutions have the potential to stimulate airway sputum spinnability is largely due to DNase. Treatment with mucus clearance in both chronic bronchitis and CF patients, rhDNase has been reported to cleave neutrophil-derived DNA by changing its ionic content and rheological properties.
present in CF infected lungs, reducing the adhesiveness and Addition of HS to CF sputum demonstrated a small reduc- viscoelasticity of CF sputum (26). By separating the DNA mol- tion in spinnability compared to treatment with NS. Treatment ecules from the mucoprotein, it was expected that HS would with rhDNase, as in our previous studies (14–16), resulted in a make the DNA more susceptible to enzyme digestion (27).
significant decrease in spinnablity. Although, combined treat- However, this was not the case, as indicated by the results.
ment of HS and DNase showed a greater reduction in spin- The combination of HS and DNase may have possibly had a nability than either treatment by itself, the total reduction was less than additive effect on spinnability due to a reduction in less than additive. In fact, in this study, the decrease in spin- activity of DNase at high salt concentrations, given the fact nability for combined in vitro treatment of rhDNase and HS that DNase and HS were combined prior to incubating the was similar to the decrease demonstrated after treatment with rhDNase and NS. Thus, as opposed to the co-treatment with In the previous studies (15, 16), as well as in the present in- DNase and gelsolin (15) or DNase and Nacystelyn (16), in this vestigation, the change in molecular weight of the DNA study, we saw no clear evidence for synergy of mucolytic ac- present in the sputum samples after administration of rh- DNase was reflected in the results obtained by the filanceme- The sputum samples in the study were treated with 3% ter, but not in the results from the magnetic microrheometer.
NaCl (0.512 M) (1 part to 10 parts of sputum). The initial spu- The reasons why spinnability by filancemeter appears to be a tum NaCl content was not measured because of insufficient more sensitive indicator for this type of mucolysis (i.e., molec- volume of sputum with four treatment protocols used; al- ular weight reduction) than viscoelasticity by magnetic rheom- though this is one of the limitations of the study, we can calcu- eter are likely related to the molecular weight dependence of the late the amount of NaCl added by HS and estimate the final two techniques. Viscosity (and probably viscoelasticity) classi- NaCl content in the sputum. Based on previous literature re- cally exhibits a 3.4-power dependence on molecular weight ports (5, 7), the initial NaCl concentration should have been (28). The power dependency for spinnability is less certain, about 88 mM (mean Naϩ ϭ 101 mM, mean ClϪ ϭ 76 mM).
but to the extent that it is related to the first normal stress dif- Thus, the final concentration of NaCl in the sputum treated ference, its power law dependency should be much higher, with 3% NaCl should have been about 126 mM, an increase of theoretically 6.8 power (29). This extra power dependency approximately 38 mM. These electrolyte concentrations are means that a mere 10% reduction in molecular weight would lower than serum levels and less than those reported for laryn- reduce spinnability to (0.9)6.8 or 49% of control, which is com- gectomized patients by Potter and coworkers (165 mM for parable to the reduction in spinnability observed for rhDNase treatment. At the same time, viscoelasticity should only de- Ten percent volume of treatment solutions per weight of crease to (0.9)3.4 or 70%. On the logarithmic scale used for vis- sputum (vol/wt) was chosen for this in vitro study because by coelasticity (log G*), this reduction would only amount to 0.16 itself this vehicle treatment had no significant effect on vis- log units, which is within the usual limit of detectability for coelasticity in previous experiments involving DNase (15, 16).
magnetic rheometry (0.2 log units). With hypertonic saline It is not known if the clinical use of HS or DNase results in de- treatment, it is assumed that there is no cleavage of intramo- livery of 1:10 of total volume of secretions in the airways; this lecular bonds and no reduction in molecular weight. However, ratio will most likely vary depending on the distribution of the treatment (shielding of ionic charge) should produce sub- aerosol deposition. Therefore it is difficult to make any pre- stantial reductions in intermolecular interaction and macro- dictions related to an optimal volume/dose of solutions deliv- molecular conformation, both of which will reduce the degree of entanglement coupling. For this type of alteration, vis- The 38 mM increase in sputum NaCl content by this in vitro coelasticity and spinnability changes should become compara- treatment resulted in a modest decrease in spinnability com- pared with NS, as well as a major decrease in viscoelasticity by In conclusion, this in vitro study demonstrated that treat- magnetic microrheometry. The increase in NaCl content was ment with HS alone, as well as combined treatment with HS comparable to that reported by Tomkiewicz and coworkers and DNase, but not DNase alone, decreased log G* (the prin- (7) for long-term amiloride treatment of CF patients, where cipal viscoelasticity index) to a similar degree, suggesting that the mean Naϩ content went from from 95 mM to 121 mM.
this effect was mainly due to HS and that DNase had smaller (Sputum chloride content increased from 64 to 77 mM with effect on viscoelasticity in these samples than HS. The clinical chronic amiloride treatment; further acute treatment in- relevance of our results is based mainly on established correla- creased the sodium content to 143 mM, and 95 mM for ClϪ).
tions between viscoelasticity and mucociliary clearance, whereas These changes in ion content were associated with a signifi- cough clearance also increases with decreased spinnability (31).
cant decrease in viscoelasticity despite the absence of detect- Spinnability decreased for the sputum samples that were treated able change in sputum water content. It was suggested at the with rhDNase, in combination with either HS or normal sa- time that CF sputum rheology might be particularly suscepti- line. Hypertonic saline alone and rhDNase in normal saline ble to small changes in ion content (25), and the results of the both increased the predicted cough clearability of the sputum; current study would seem to confirm this. The interpatient however, the combined treatment with rhDNase and hyper- variation in rheological data for a log transformed mean Ϯ SD tonic saline had the best overall effect on cough clearability.
for G* 1 in the control group was 1.97 Ϯ 0.35 (i.e., 124% coef- Our results of combined treatment with HS and rhDNase ficient of variation), while the intrapatient variation was ap- seem encouraging since they demonstrate a change in the proximately half of that; this is comparable to what has been physical property of CF sputum from a rigid gel to a lower elasticity fluid, more readily clearable by airflow interaction.
King, Dasgupta, Tomkiewicz, et al.: Hypertonic Saline, DNase and CF Sputum Because of the possibility that co-administration of HS with 14. Dasgupta, B., R. P. Tomkiewicz, W. A. Boyd, N. E. Brown, and M. King.
rhDNase might reduce the effectiveness of rhDNase, the po- 1995. Effects of combined treatment with rhDNase and airflow oscil- tential for alternate or consecutive treatments with those two lations on spinnability of cystic fibrosis sputum in vitro. Pediatr. Pul-monol. 20:78–82.
forms of mucotropic agents should be considered. Further 15. Dasgupta, B., R. P. Tomkiewicz, G. T. De Sanctis, W. A. Boyd, and M.
studies are required to confirm these findings with a larger King. 1996. Rheological properties in cystic fibrosis airway secretions number of CF sputum samples and to carry out similar experi- with combined rhDNase and gelsolin treatment. In M. Singh and V. P.
ments with different ratios of HS and rhDNase.
Saxena, editors. Advances in Physiological Fluid Dynamics. Narosa,New Delhi. 74–78.
16. Dasgupta, B., and M. King. 1996. Reduction in viscoelasticity of cystic fi- References
brosis sputum in vitro with combined treatment by Nacystelyn and 1. Collins, F. S. 1992. Cystic fibrosis: molecular biology and therapeutic im- rhDNase. Pediatr. Pulmonol. 22:161–166.
plications. Science 256:774–779.
17. Puchelle, E., J. M. Zahm, and C. Duvivier. 1983. Spinnability of bron- 2. Regnis, J. A., M. Robinson, D. L. Bailey, P. Cook, P. Hooper, H. K.
chial mucus: relationship with viscoelasticity and mucus transport prop- Chan, I. Gonda, G. Bautovich, and P. T. P. Bye. 1994. Mucociliary erties. Biorheology 20:239–249.
clearance in patients with cystic fibrosis and in normal sublects. Am. J. 18. King, M. 1988. Magnetic microrheometer. In P. C. Braga and L. Allegra, Respir. Crit. Care Med. 150:66–71.
editors. Methods in Bronchial Mucology. Raven Press, New York. 73– 3. Kerem, E., J. Reisman, M. Corey, G. J. Canny, and H. Levison. Predic- tion of mortality in patients with cystic fibrosis. N. Engl. J. Med. 326: 19. King, M. 1987. Role of mucus viscoelasticity in cough clearance. Biorhe- 4. King, M. 1989. Mucus, mucociliary clearance and coughing. In D. V.
20. Zayas, J. G., G. C. W. Man, and M. King. 1990. Tracheal mucus rheology Bates, editors. Respiratory Function in Disease, 3rd ed. Saunders, Phila- in patients undergoing diagnostic bronchoscopy: interrelations with smoking and cancer. Am. Rev. Respir. Dis. 141:1107–1113.
5. Boat, T. F., and L. W. Matthews. 1973. Chemical composition of human 21. Ziment, I. 1989. Hypertonic solutions, urea, and ascorbic acid. In P. C.
tracheobronchial secretions. In M. J. Dulfano, editor. Sputum: Funda- Braga and L. Allegra, editors. Drugs in Bronchial Mucology. Raven mentals and Clinical Pathology. Charles C. Thomas, Springfield, IL.
22. Sheffner, A. L., E. M. Medler, L. W. Jacobs, and H. P. Sarett. 1964. The 6. Lutz, R. J., M. Litt, and L. W. Chakrin. 1973. Physical-chemical factors in in vitro reduction in viscosity of human tracheobronchial secretions by mucus rheology. In H. L. Gabelnick and M. Litt, editors. Rheology of acetylcysteine. Am. Rev. Respir. Dis. 90:721–729.
Biological Systems. Charles C. Thomas, Springfield, IL. 158–194.
23. Pavia, D., D. L. Thompson, and S. W. Clarke. 1978. Enhanced clearance 7. Tomkiewicz, R. P., E. M. App, J. G. Zayas, O. Ramirez, N. Church, R.C.
of secretions from the human lung after administration of hypertonic Boucher, M. R. Knowles, and M. King. 1993. Amiloride inhalation saline aerosol. Am. Rev. Respir. Dis. 117:199–203.
therapy in cystic fibrosis: influence on ion content, hydration and rhe- 24. Potter, J. L., L. W. Matthews, S. Spector, and J. Lemm. 1967. Studies on ology of sputum. Am. Rev. Respir. Dis. 148:1002–1007.
pulmonary secretions. 2. Osmolality and the ionic enviroment of pul- 8. Chernick, W. S., and G. J. Barbero. 1959. Composition of tracheobron- monary secretions from patients with cystic fibrosis, bronchiectasis chial secretions in cystic fibrosis of the pancreas and bronchiectasis.
and laryngectomy. Am. Rev. Respir. Dis. 96:83–87.
25. Quinton, P. M. 1994. Viscosity versus composition in airway pathology 9. Smith, A. L., G. Redding, C. Doershuk, et al. 1988. Sputum changes as- [editorial]. Am. J. Respir. Crit. Care Med. 149:6–7.
sociated with therapy for endobronchial exacerbation in cystic fibro- 26. Zahm, J. M., S. Girod de Bentzmann, E. Deneuville, C. Perrot-Minnot, sis. J. Pediatr. 112:547–554.
E. Depret, F. Pennaforte, M. Roussey, and E. Puchelle. 1995. Dose- 10. Shak, S., D. J. Capon, R. Hellmiss, S. A. Marsters, and C. L. Baker. Re- dependent in vitro effect of recombinant human DNase on the trans- combinant human DNase I reduces the viscosity of cystic fibrosis spu- port properties of cystic fibrosis respiratory mucus. Eur. Respir. J. 8:381– tum. Proc. Natl. Acad. Sci. U.S.A. 87:9188–9192.
11. Fuchs, H. J., D. S. Borowitz, D. H. Christiansen, E. M. Morris, M. L.
27. Lieberman, J., and N. B. Kurnick. 1962. Influence of deoxyribonucleic Nash, B. W. Ramsey, B. J. Rosenstein, A. L. Smith, and M. E. Wohl.
acid content on the proteolysis of sputum and pus. Nature (London) 1994. Effect of aerosolized recombinant human DNase on exacerba- tions of respiratory symptoms and on pulmonary function in cystic fi- 28. Ferry, J. D. 1970. Molecular theory for undiluted polymers and concen- brosis. N. Engl. J. Med. 331:637–648.
trated solutions: networks and entanglements. Viscoelastic Properties 12. Robinson, M., J. A. Regnis, D. L. Bailey, M. King, G. J. Bautovich, and of Polymers, 2nd ed. Wiley, New York. 268–270.
P. T. P. Bye. 1996. Effect of hypertonic saline, amiloride, and cough 29. Ferry, J. D. 1970. The nature of viscoelastic behavior. Viscoelastic Prop- on mucociliary clearance in patients with cystic fibrosis. Am. J. Respir. erties of Polymers, 2nd ed. Wiley, NewYork. 27–29.
30. Ferry, J. D. 1970. Illustrations of viscoelastic behavior. Viscoelastic Prop- 13. Wills, P. J., R. L. Hall, W. M. Chan, and P. J. Cole. 1997. Sodium chlo- erties of Polymers, 2nd ed. Wiley, NewYork. 52–54.
ride increases the ciliary transportability of cystic fibrosis and bron- 31. King, M., J. M. Zahm, D. Pierrot, S. Vaquez-Girod, and E. Puchelle.
chiectasis sputum on the mucus-depleted bovine trachea. J. Clin. Invest. 1989. The role of mucus gel viscosity, spinnability, and adhesive prop- erties in clearance by simulated cough. Biorheology 26:737–745.

Source: http://aerosal.it/bibliografia/RCCM1561173.pdf