Colistin

Study objective: Inhaled colistin is used for the treatment of Pseudomonas aeruginosa infection in cystic fibrosis (CF) patients despite reports of chest tightness and bronchospasm. The main objective of the study was to assess whether bronchospasm occurred in pediatric CF patients with or without clinical evidence of airway hyperreactivity.

Design and methods: A prospective placebo-controlled clinical trial with crossover design was devised using challenge tests with 75 mg colistin in 4 mL saline solution and a placebo solution of the same osmolarity using a breath-enhanced nebulizer for administration. Subjects were recruited as follows: high risk (HR) for bronchospasm due to a personal history of recurrent wheezing, a family history of asthma and/or atopy, or bronchial lability, as demonstrated in pulmonary function tests; or low risk (LR) without these characteristics.

Results: The mean [FEV.sub.1] (expressed as the mean [[+ or -] SD] fall from baseline) of the HR group (n = 12) fell 12 [+ or -] 9% after placebo was administered, and fell 17 [+ or -] 10% after colistin was administered. For the LR group (n = 8), the mean [FEV.sub.1] fell 9 [+ or -] 4% following placebo administration and 13 [+ or -] 8% following colistin administration. There was a greater number of subjects in the HR group compared to the LR group, which had a mean fall in [FEV.sub.1] of [greater than or equal to] 15% (p < 0.01) after inhaling colistin The differences between placebo and colistin. therapy in the LR group were not significant.

Conclusion: The results demonstrated that colistin can cause bronchospasm, particularly in those patients with coexisting CF and asthma.

Key words: bronchial spasm; colistin; cystic fibrosis; inhaled antibiotics

Abbreviations: CF = cystic fibrosis; HR = high risk; HSC = Hospital for Sick Children; LR = low risk; PA = Pseudomonas aeruginosa; PF = pulmonary function

**********

Chronic endobronchial infection is a primary feature of the pulmonary disease in cystic fibrosis (CF) patients. The acquisition of Pseudomonas aeruginosa (PA) is associated with a mild but statistically demonstrable worsening in pulmonary function (PF), (1) resulting in aerosolized antibiotic therapy becoming a mainstay of ambulatory treatment for chronic endobronchial infection. Therapy with inhaled tobramycin, in particular, has gained popularity among many CF centers due to its ease of nebulization and lack of systemic toxicity in comparison to parenteral tobramycin administration. (2) When given by the aerosol route compared to the parenteral route, there appear to be fewer problems with bacterial resistance against tobramycin due to the high concentrations that are achievable in the airway. Some patients with advanced disease harbor tobramycin-resistant strains of PA in their respiratory tracts. (3) This has prompted the search for other antibiotics that have antimicrobial activity against PA and can be aerosolized. (4,5)

Colistin sulfomethate, a polymyxin E antibiotic, has the potential to be a suitable agent for aerosolization in CF patients due to its high antimicrobial activity against PA, including multiresistant strains. (6,7) Although its use systemically was largely curtailed in the early 1970s because of nephrotoxicity and neurotoxicity, (7,8) there has been increasing interest in aerosol administration. Presently, nebulized colistin is used in many European centers treating CF patients, (7) with some studies, (4,9,10) but not all, (11) showing positive results. One study (3) in adult CF patients with panresistant PA who were awaiting lung transplantation suggested the return of sensitivity to tobramycin in PA strains in 100% of patients compared to 30% in control subjects.

Nebulized colistin causes chest tightness and bronchospasm in some patients. This side effect has been demonstrated largely in adult studies of CF patients. (12,13) A bronchospasm reaction to nebulized solutions of colistin has been attributed to hypertonicity and/or to the drug itself. Hypertonic solutions have been shown to produce bronchoconstriction, particularly in patients with hyperreactive airways, (14,15) and patients receiving hypertonic antibiotic therapy have been shown to have more bronchospasm than those patients receiving antibiotics with lower tonicity. (16) The proposed mechanism involves the release of histamine through changes in osmotic loads around mast cells. (12,15) During the nebulization of an isotonic solution, the tonicity becomes progressively higher as the evaporation of water from nebulized droplets occurs. In a study with adults, (12) nebulized colistin produced equal symptoms of chest tightness and magnitude of fall in [FEV.sub.1] irrespective of the tonicity of the solution. These effects were evident earlier with the use of the hypertonic solution (7.8 min after the onset of nebulization) and were delayed with the use the hypotonic solution (34.2 min after the onset of nebulization). In one study, Cunningham and colleagues (17) found a significant fall in [FEV.sub.1] in response to the inhalation of colistin, which was not either related to IgE levels or associated with the chest discomfort experienced by some patients. These results are hard to interpret in that different nebulizers with different levels of efficiency (18) were used, and there was no comparison with placebo inhalation. Furthermore, 95% of their subjects had been regularly receiving inhaled colistin and had clearly demonstrated previously that they had not experienced any major adverse effects. In a study comparing inhaled tobramycin to colistin, Hodson et al (11) found a fall in [FEV.sub.1] in patients receiving both preparations. These results are also difficult to interpret since 85% of their subjects were receiving [beta]-adrenergic receptor agonists, and no information about the time of administration of the bronchodilators was given. While bronchoconstriction may be blunted by the coadministration of salbutamol, the preservative benzalkonium chloride in the salbutamol (Ventolin; GlaxoSmithKline; Research Triangle Park, NC) respiratory solution interacts with colistin to cause precipitation, (19) further complicating the issue. The occurrence of bronchospasm with therapy using an isotonic nebulized colistin solution suggests the possibility of airway reactivity to the colistin molecule itself. This has been supported by an in vitro demonstration of mast cell degranulation caused by colistin. (20) If the illustrated benefits outweigh the risks of bronchospasm, aerosolized colistin may be an effective therapeutic agent in the treatment of multiresistant PA infection in children with CF. To date, the magnitude of bronchospasm in children with CF is unknown. In addition, the bronchospasm that may develop in children may be related to the dose of colistin deposited in the lungs or to possible inherent airway hyperreactivity in selected CF patients.

The hypothesis of this study was that bronchospasm occurs following a colistin inhalation challenge in pediatric CF patients compared to challenge with placebo and that the likelihood of bronchospasm is greater in subjects with coexistent indicators of asthma compared to those without a previous personal or family history, or a response to therapy with bronchodilators.

MATERIALS AND METHODS

This study was a prospective, placebo-controlled clinical trial with a crossover design. The aim of the study was to recruit 24 subjects from the Cystic Fibrosis Clinic at the Hospital for Sick Children (HSC). The medical records of CF patients between 5 and 18 years of age who were able to perform technically acceptable spirometry (21) and could use a nebulizer with a mouthpiece were reviewed. The selection criteria were based on a previously published study (22) in which the subjects were separated into those who were judged to be at high risk (HR) and low risk (LR) for bronchoconstriction following the inhalation of antibiotics. For this study, the first group deemed to be at HR for bronchospasm possessed characteristics of possible inherent airway lability, and the second group considered to be at LR for bronchospasm did not possess any characteristics of airway lability. This subdivision of the study population helped to determine whether bronchoconstriction was part of an asthmatic component coexisting with CF or was independent of inherent airway lability and likely then to be an irritant side effect of colistin. All subjects had positive culture results for PA within the past 12 months and were receiving therapy with inhaled tobramycin. None had ever received colistin in the past. Details of the selection have already been published (22) and are summarized briefly in Table 1. Allocation to the two groups was not randomized. Recruitment involved a letter to the parents and/or the patient, depending on age, requesting their participation in this study. On entering the study, written informed consent was obtained, and the patient was enrolled into the appropriate group (ie, HR or LR). The project was approved by the Research Ethics Board of the HSC.

All spirometry was performed at the PF laboratory at HSC according to American Thoracic Society guidelines, (21) and the predicted values were derived from a large series of healthy children who were studied in this laboratory and used as control subjects. (23) Patients were instructed to abstain from using any bronchodilators for 12 h prior to the inhalation of the challenge drug, and no subject at the time of the study was using long-acting bronchodilators, leukotriene inhibitors, or cromolyn. For subjects receiving therapy with inhaled corticosteroids, their therapy was continued without interruption. Each subject visited the PF laboratory on two separate occasions within a 2-week interval. The subjects were randomized to receive either the preparation of colistin or placebo on the first visit and received the remaining preparation on the second visit (Fig 1). At study end, each patient had received both colistin and placebo. There were various factors, such as taste, noise generated during nebulization, and the foaming that colistin generates around the baffle system of the nebulizer, that made blinding the patient to the preparation difficult. The investigator and technologist performing the spirometry were blinded to the type of solution being administered, and the nebulizer was enclosed in an opaque sleeve that concealed any foaming.

[FIGURE 1 OMITTED]

Colistin and placebo were delivered by a nebulizer (LC Star; Pari Respiratory Equipment; Midlothian, VA) that was driven by a compressor (Proneb Turbo; Pari Respiratory Equipment). During operation, the system generates a flow of 6 L/min at the output of the nebulizer. The nebulizer-compressor combination is the standard equipment for the home delivery of inhaled antibiotics in the clinic at HSC. (18) Each 75-mg vial of powdered colistin (Parke-Davis Canada; Toronto, ON, Canada) was diluted with 4 mL saline solution, in keeping with the recommended dose (4) and the previously characterized nebulization output for this device. (24) The osmolarity of this solution was 367 mmol/L. (24) The placebo solution was formulated such that its osmolarity matched that of the colistin solution by adding 1.55 mL 3% NaCl solution to 2.45 mL distilled water. This gave an osmolarity of 368 mmol/L, with a concentration of 11.625 mg/mL NaCl. Each patient inhaled the preparations for a selected total duration of 20 min, to prevent the osmolarity of the colistin solution from exceeding 500 mmol/L (24) at the end of the session. Postinhalation spirometry was completed immediately after treatment.

All episodes of coughing or wheezing were graded based on a scale of no symptoms to 2+, and oxygen saturation was monitored during nebulization to ensure that it remained at > 93%. In the event of any respiratory embarrassment, the test was to be terminated and the symptoms recorded. Coughing by itself was not a reason for termination, unless the patient was coughing to the extent that nebulization could not be continued without multiple pauses. If the test was terminated prematurely, spirometry was performed as soon as possible after termination to assess the [FEV.sub.1], and then the patient was treated with salbutamol.

For each patient, the total drug output of the nebulizer was calculated by taking concentrating effects due to evaporative losses into account. The details of these calculation have been described elsewhere. (25-28) In order to estimate the expected deposition, device efficiency was estimated from the flow tracings of four CF patients (18) breathing through a nebulizer using the output algorithms based on the work by Katz et al, (24) and using the method described by Ho and colleagues. (18) The mean in vivo efficiency, which was defined as the amount of the drug that is likely to enter the airway and deposit below the cords per breath in relation to the total amount of output of drug per breath, was found to be 59.7%. For the sake of simplicity, the deposition was estimated as the amount that left the nebulizer during each session multiplied by 0.6.

Data analysis of the primary outcome evaluated the number of patients with a significant fall in [FEV.sub.1], (defined as a decrease of [FEV.sub.1] from baseline by [greater than or equal to] 15%) in each group and for each solution using paired t tests and [chi square] analysis. Values are expressed as the mean and SD.

RESULTS

Twenty-three patients were enrolled into the study, and 20 patients completed it (HR group, 12 patients 8 to 18 years of age; LR group, 8 patients 8 to 17 years of age). The LR group lost three patients due to incomplete visits. One patient did not return for scheduled visit 2, and there was difficulty in rebooking during the appropriate time period. Two patients had exacerbations between visits with a baseline [FEV.sub.1] > 15% lower on visit 2, which excluded them from continuing. None of the three patients who were lost had experienced an excessive fall in [FEV.sub.1] during visit 1 or residual symptoms of bronchospasm that would have prevented completion of the study. The demographic details, personal history, and family history of the completed subjects are shown in Table 1. The HR group had a slightly lower mean baseline [FEV.sub.1] than the LR group (67 [+ or -] 17% vs 74 [+ or -] 13% predicted, respectively) [Fig 2], and these differences were not clinically or statistically significant.

[FIGURE 2 OMITTED]

After inhaling the placebo, there was a small but significant fall in [FEV.sub.1] (expressed as a percent fall from baseline) in both the HR group (12 [+ or -] 9%; p < 0.001), with 4 of 12 patients having a fall in [FEV.sub.1] of [greater than or equal to] 15%, and in the LR group (9 [+ or -] 4%; p < 0.001), with 1 of 8 patients having a fall in [FEV.sub.1] of [greater than or equal to] 15% (Fig 2). The mean fall in [FEV.sub.1] after the inhalation of colistin was 17 [+ or -] 10% for the HR group, with 7 of 12 patients having a fall of [greater than or equal to] 15%, and 13 [+ or -] 8% for the LR group, with 4 of 8 having a fall of [greater than or equal to] 15%. The changes in [FEV.sub.1] postinhalation of placebo vs colistin were significant for the HR group (p < 0.002), but not for the LR group. There was a statistical difference in the response between the HR and LR groups for a fall in [FEV.sub.1] of [greater than or equal to] 15% post-colistin inhalation (p < 0.01), but not post-placebo inhalation. The largest fall in [FEV.sub.1] was 29% in the HR group following the inhalation of colistin, which resulted in an [FEV.sub.1] of 36% predicted. For the HR group, 3 of 12 patients had an [FEV.sub.1] of < 40% predicted following colistin inhalation, but this did not occur in any of the HR patients after placebo inhalation. In all three patients, the baseline [FEV.sub.1] was between 45% and 50% predicted. The largest fall in [FEV.sub.1] for the LR group was 27% following placebo inhalation. In no case did the [FEV.sub.1] fall to < 40% predicted in the LR group, irrespective of the agent inhaled. The lowest baseline [FEV.sub.1] in the LR group was 54% predicted, and all others were > 60%.

No serious adverse events occurred during the study, and the test was never terminated before completion. No patient in either group experienced any desaturations following either placebo or colistin inhalation. Wheezing that had not been present preinhalation occurred in four patients postinhalation in the HR group (three recorded as mild, one as 2+), and two in the LR group (recorded as mild and 1+). The wheezing was not related to whether or not the inhalation was of placebo or colistin, nor was it related to the degree of fall in [FEV.sub.1]. In the HR group, four patients experienced 2+ coughing post-placebo inhalation, and two patients were recorded as having 2+ coughing and one as having mild coughing post-colistin inhalation. For the LR group, two patients experienced coughing (1+ and 2+) post-placebo inhalation, and two experienced it post-colistin inhalation (both 1+). None of the coughing episodes were related to the degree of fall in [FEV.sub.1]. Three of four patients in the HR group who complained of mild chest tightness did so after inhaling colistin, whereas with the two patients in the LR group (both of whom experienced mild tightness), one episode occurred after placebo inhalation and the other after colistin inhalation. While 8 of 12 subjects in the HR group met the criteria for salbutamol administration post-colistin inhalation, this was not different from the situation following placebo inhalation ([chi square] analysis) in which half of the patients met the criteria. Whether or not the bronchoconstriction was caused by placebo or colistin, it was completely reversed by the inhalation of salbutamol. For the LR group, five of eight patients received salbutamol after both placebo and colistin inhalation. Although the fall in [FEV.sub.1] was less in the LR group, interestingly, it was still slightly but significantly lower than baseline following salbutamol administration in those who had inhaled colistin (72 [+ or -] 12% vs 65 [+ or -] 13%, respectively; p < 0.05) compared to those who had inhaled placebo (75 [+ or -] 12% vs 71 [+ or -] 13%, respectively; p < 0.05) [paired t test for both].

For the HR group, the mean postnebulization osmolarity of colistin was 475 [+ or -] 24 mosm/L, and of placebo was 511 [+ or -] 52 mosm/L. The degree of fall in [FEV.sub.1] was not related to the increase in osmolarity, nor were these differences significant (p > 0.05). The relatively low rate of output of colistin, (24) compared to other medications, (18) and the restriction of nebulization to 20 min resulted in a significantly larger mean residual volume in the nebulizer compared to that for placebo (1.59 [+ or -] 0.29 vs 0.95 [+ or -] 0.16 mL, respectively) with similar results found in the LR group. For the HR group, 48 [+ or -] 8% of the colistin left the nebulizer and either entered the patient or impacted on the expiratory filter vs 67 [+ or -] 5% for placebo, differences that were highly significant (p < 0.001 [paired t test]). The results were similar in the LR group, 55 [+ or -] 13% vs 67 [+ or -] 6%, respectively; p < 0.04). In terms of dose of colistin, the estimated mean pulmonary deposition in the HR group was 30.2 [+ or -] 2.2% of the charge dose of 75 mg, which did not differ from a mean of 29.9 [+ or -] 2.9% seen in the LR group. The calculated expected deposition of colistin did not correlate with the fall in [FEV.sub.1].

DISCUSSION

The results indicate that a certain degree of bronchoconstriction following the inhalation of an aerosol is common, irrespective of the agent inhaled. However, for those patients with demonstrated airway lability (ie, the HR group), the fall in [FEV.sub.1] was greater if the agent being inhaled was colistin. These differences were not apparent in the LR group. The degree of fall in [FEV.sub.1] was higher than that found in the study by Cunningham et al, (17) where 95% of the subjects were regularly receiving colistin. The estimated amount of colistin deposited in the lung agreed closely with that estimated by Katz et al (24) using the same nebulizer and compressor system. The estimated deposition was much higher than that shown by Chua et al, (29) reflecting the greater efficiency of modern breath-enhanced nebulizers. (24,25)

The design of this study was intended to give insight into the risk of bronchospasm in two extremes of patients in the CF population as opposed to the risk for the general population as a whole. The number of patients who complained of chest tightness was too small to be able to decide whether this was due to the inhalation of colistin, as it was also observed in both groups following the inhalation of placebo. Similarly, the presence of wheezing postinhalation that was not observed preinhalation occurred in insufficient numbers of patients to allow any conclusion as to whether or not this was related to colistin. In most cases, the degree of bronchoconstriction, even in the HR group, was relatively mild and easily reversed with bronchodilator treatment, but in the ease of the LR group bronchoconstriction was not completely reversed. It is possible that the dose of the bronchodilator salbutamol (400 [micro]g) given by metered-dose inhaler and valved holding chamber was too low. On the other hand, the failure to completely reverse the fall in [FEV.sub.1] occurred in patients in the LR group who had never demonstrated a response to bronchodilators during routine testing. This leads to speculation that the fall in [FEV.sub.1] in this group could be due in part to factors other than a spasm of the smooth muscle in the airway, such as local irritation, which could conceivably give rise to edema that was unaffected by bronchodilators. For those patients whose prenebulization [FEV.sub.1] was [less than or equal to] 50% predicted, a case could be made for pretreatment with bronchodilators prior to the inhalation of colistin.

One limitation of this study is that the general CF population, which frequently includes patients who have at one point demonstrated a response to bronchodilators but have no other markers for increased airway lability, made up only a small portion of the groups studied. Subjects 10 and 12 (Table 2) did not have a personal or family history of asthma or atopy, but subject 10 had multiple responses to bronchodilators of > 8% (with the largest being 32%), and subject 12 had two of four responses > 8% (with the largest being 15%). These two subjects met the HR inclusion criteria based on their responses to bronchodilators. Subject 10 had a 34% fall in [FEV.sub.1] in response to colistin inhalation, while subject 12 had only a 5% fall in [FEV.sub.1]. From the data in the LR group, it could be inferred that many of these patients would be at risk for some degree of bronchospasm following inhalational therapy.

The study design called for a small number of subjects who would be studied fairly intensively. Given both the subject numbers and the low incidence of chest tightness seen in both groups and with both agents, it is unclear how this relates to reports of chest tightness following colistin inhalation in adults. (12,13) Given that chest tightness was reported with both agents, it is possible that high tonicity in the airways, not colistin, was the source of these concerns. (14,15)

Bronchoconstriction may be caused by direct chemical stimulation as occurs in a methacholine challenge test, allergy in the airway, irritation from chemicals or fumes, hyperosmolarity in the airway, and from nonspecific causes. Given the degree of response to placebo, the osmolarity of the aerosol toward the end of nebulization may well have played a role in causing bronchoconstriction. As a word of caution, the nebulizer was driven by a compressor and consequently a "wet" gas source. Had it been driven by dry, hospital gas (or tanks of compressed air), there would have been much more drying and a greater increase in osmolarity. (26) There is a suggestion that colistin may have acted as an irritant, causing increased bronchoconstriction in the HR group. It is unlikely that any of the subjects were allergic to colistin, since none had previously been exposed to the drug.

In conclusion, it is apparent that the inhalation of colistin, because of the drug itself, the tonicity of the airway, or a combination of the two, results in a certain degree of bronchospasm and a fall in [FEV.sub.1] postinhalation. This occurs whether or not the patient has a coexisting tendency for "asthma" or has no family history for asthma or atopy and no response to bronchodilators. However, the degree of fall in [FEV.sub.1] appears to be greater in the former, placing them at greater risk for respiratory embarrassment due to the nebulization of colistin. Bronchospasm does respond to therapy with bronchodilators, and pretreatment with these agents could be an option in those patients with advanced lung disease and low baseline spirometry. In other words, colistin-induced bronchospasm, while it does exist, appears to be manageable and does not constitute a reason not to administer colistin by inhalation when medically indicated.

* From the Division of Respiratory Medicine, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada.

REFERENCES

(1) Kerem E, Corey ML, Gold R, et al. Pulmonary function and clinical course in patients with cystic fibrosis after pulmonary colonization with Pseudomonas aeruginosa. J Pediatr 1990; 116:714-719

(2) Smith AL, Ramsey BW, Hedges DL, et al. Safety of aerosolized tobramycin administration for 3 months to patients with cystic fibrosis. Pediatr Pulmonol 1989; 7:265-271

(3) Bauldoff GS, Nunley DR, Manzetti JD, et al. Use of aerosolized colistin sodium in cystic fibrosis patients awaiting lung transplantation. Transplantation 1997; 64:748-752

(4) Littlewood JM, Miller MG, Ghoneim AT, et al. Nebulized colomycin for early Pseudomonas colonization in cystic fibrosis [letter]. Lancet 1985; I:865

(5) Diot P, Gagnadoux F, Martin C, et al. Nebulization and anti-Pseudomonas aeruginosa activity of colistin. Eur Respir J 1997; 10:1995-1998

(6) Catchpole CR, Andrews JM, Brenwald N, et al. A reassessment of the in vitro activity of colistin sulphomethate sodium. J Antimicrob Chemother 1997; 39:255-260

(7) Jensen T, Pedersen SS, Garne S, et al. Colistin inhalation therapy in cystic fibrosis patients with chronic Pseudomonas aerugmosa lung infection. J Antimicrob Chemother 1987; 19:831-838

(8) Bosso JA, Liptak CA, Seilheimer DK, et al. Toxicity of colistin in cystic fibrosis patients. DICP 1998; 25:1168-1170

(9) Frederiksen B, Koch C, Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997; 23:330-335

(10) Frederiksen B, Koch C, Hoiby N. Changing epidemiology of Pseudomonas aeruginosa infection in Danish cystic fibrosis patients (1974-1995). Pediatr Pulmonol 1999; 28:159-166

(11) Hodson ME, Gallagher CG, Govan JRW. A randomized clinical trial of nebulised tobramycin or colistin in cystic fibrosis. Eur Respir J 2003; 20:658-664

(12) Dodd M, Abbott J, Maddison J, et al. Effect of tonicity of nebulized colistin on chest tightness and pulmonary function in adults with cystic fibrosis. Thorax 1997; 52:656-658

(13) Maddison J, Dodd M, Webb AK. Nebulized colistin causes chest tightness in adults with cystic fibrosis. Respir Med 1994; 88:145-147

(14) Beasley R, Rafferty P, Holgate ST. Adverse reactions to the non-drug constituents of nebuliser solutions. Br J Clin Pharmacol 1988; 25:283-287

(15) Schoeffel RE, Anderson SD, Altounyan RE. Bronchial hyper-reactivity in response to inhalation of ultrasonically nebulised solutions of distilled water and saline. BMJ 1981; 283:1285-1287

(16) Chua HL, Collis GG, LeSouef PN. Bronchial response to nebulized antibiotics in children with cystic fibrosis. Eur Respir J 1990; 3:1114-1116

(17) Cunningham S, Prasad A, Collyer L, et al. Bronchoconstriction following nebulised colistin in cystic fibrosis. Arch Dis Child 2001; 84:432-433

(18) Ho SL, Kwong WTJ, Coates AL. Evaluation of four breath-enhanced nebulizers for home use. J Aerosol Med 2001; 14:467-475

(19) Roberts GW, Badock NR, Jarvinen AO. Cystic fibrosis inhalation therapy: stability of a combined salbutamol/colistin solution. Aust J Pharm 1992; 22:378-380

(20) Jasani B, Kreil G, Mackler BF, et al. Further studies on the structural requirements for polypeptide-mediated histamine release from rat mast cells. Biochem J 1979; 181:623-632

(21) American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med 1995; 152:1107-1136

(22) Alothman GA, Alsaadi MM, Ho BL, et al. Evaluation of bronchial constriction in cystic fibrosis children after inhaling two different preparations of tobramycin. Chest 2002; 122: 930-934

(23) Corey ML, Crozier DN, Levison H. Five to seven-year course of pulmonary function in cystic fibrosis. Am Rev Respir Dis 1976; 114:1085-1092

(24) Katz SL, Ho SL, Coates AL. Nebulizer choice for treatment of cystic fibrosis patients with inhaled colistin. Chest 2001; 119:250-255

(25) Coates AL, MacNeish CF, Lands LC, et al. A comparison of the availability of tobramycin for inhalation from vented versus unvented nebulizers. Chest 1998; 113:951-956

(26) Coates AL, MacNeish CF, Lands LC, et al. Factors influencing the rate of drug output during the course of wet nebulization. J Aerosol Med 1998; 11:101-111

(27) Kwong E, MacNeish CF, Meisner D, et el. The use of osmometry as a means of determining changes in drug concentration during jet nebulization. J Aerosol Med 1998; 11:89-100

(28) MacNeish CF, Meisner D, Thibert R, et el. A comparison of pulmonary availability between Ventolin (albuterol) nebules and Ventolin (albuterol) respirator solution. Chest 1997; 111:204-208

(29) Chua HL, Collis GG, Newbury AM, et al. The influence of age on aerosol deposition in children with cystic fibrosis. Eur Respir J 1994; 7:2185-2191

The nebulizers were generously provided by Pari Respiratory Equipment, and the colistin was provided by Parke Davis Canada. The study was supported in part by the Canadian Cystic Fibrosis Foundation, and no author had any financial relationship with either Pari Respiratory Equipment Inc. or Parke Davis Canada.

Manuscript received February 6, 2004; revision accepted September 15, 2004.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org).

Correspondence to Allan L. Coates, BEng, MDCM, Division of Respiratory Medicine, Hospital for Sick Children 555 University Ave, Toronto, ON, Canada M5G 1X8; e-mail: allan.coates@ sickkids.ca

COPYRIGHT 2005 American College of Chest Physicians
COPYRIGHT 2005 Gale Group