Analysis of Sequential Aliquots of Hypertonic Saline Solution-Induced Sputum From Clinically Stable Patients With Cystic Fibrosis* Moira L. Aitken, MD, FCCP; Kelly E. Greene, MD; Mark R. Tonelli, MD, MA, FCCP; Jane L. Burns, MD; Julia C. Emerson, MD, MPH; Christopher H. Goss, MD, MSc; and Ronald L. Gibson, MD, PhD
Study objectives: Sputum induction (SI) is a noninvasive tool for sampling inflamed airways. The purpose of this study was to determine the optimal duration of collection in patients with cystic fibrosis (CF). The hypothesis was that the duration of SI collection would quantitatively and qualitatively alter the content of the induced sputum. Methods: In 10 clinically stable patients with CF (mean ⴞ SD age, 28 ⴞ 7 years; mean FEV1, 2.6 ⴞ 0.7 L), SI was performed with 3% hypertonic saline solution at five time points over 20 min. Results: SI was well tolerated, with an average maximum fall in FEV1 of 7 ⴞ 7%. The sample volumes, urea concentrations, interleukin-8 concentrations, total cell counts, and nonsquamous cell counts remained constant (p > 0.05). The percentage of neutrophils decreased from 89 ⴞ 5% to 86 ⴞ 4% (p ⴝ 0.03), and the percentage of alveolar macrophages increased 5 ⴞ 2% to 8 ⴞ 4% (p < 0.01). The mean quantitative microbiological counts of nonmucoid Pseudomonas aeruginosa and Staphylococcus aureus decreased over the 20-min time period each by half a log (p ⴝ 0.05 and p < 0.01, respectively). Surfactant protein-A concentration increased from 1.6 ⴞ 0.3 to 2.4 ⴞ 0.4 ng/mL (log10; p < 0.001). Conclusions: We conclude that aliquots of induced sputum are similar in clinically stable patients with CF during 4-min intervals, although there is more alveolar sampling after 20 min. When induced-sputum samples are fractionated for research monitoring of inflammatory or microbiologic indexes, power calculations accounting for these variations over time are required. (CHEST 2003; 123:792–799) Key words: cystic fibrosis; hypertonic saline; sputum Abbreviations: CF ⫽ cystic fibrosis; CI ⫽ confidence interval; ES ⫽ expectorated sputum; IL ⫽ interleukin; PMN ⫽ polymorphonuclear leukocyte; SI ⫽ sputum induction; SP ⫽ surfactant protein
saline has recently been used as a H ypertonic form of airway clearance in patients with cystic fibrosis (CF),1,2 and as a means to obtain a lower airway sample for clinical use from patients who are *From the Department of Medicine (Drs. Aitken, Tonelli, and Goss), Division of Pulmonary and Critical Care, University of Washington, Seattle, WA; Department of Pediatrics (Drs. Emerson, Burns, and Gibson), Children’s Hospital and Regional Medical Center, Seattle, WA; and University of Colorado (Dr. Greene), Denver, CO. Supported by a grant from the Firland’s Foundation, the CFF Regional Development Program, CFFTI, and the CFF Therapeutic Development Center. Manuscript received June 13, 2002; revision accepted August 16, 2002. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [emailprotected]). Correspondence to: Moira L. Aitken, MD, FCCP, Division of Pulmonary and Critical Care Medicine, Box 356522, 1959 NE Pacific St, Room BB1253, Seattle, WA 98195 792
unable to expectorate spontaneously.3 However, in high concentration (10%), hypertonic saline solution has been shown to cause bronchospasm in some patients with CF.4 Although sputum induction (SI) has been recognized as a research tool for over a decade with other airway disease such as asthma5–10 and COPD,11–13 only very recently has SI has been introduced as an outcome measure in CF.14 –16 BAL has been held as the “gold standard” for airway sampling in patients with CF.17,18 BAL has been used to follow serial microbiology cultures,18 –22 in therapeutic trials,23–25 and to monitor inflammatory mediators and other markers of the underlying pathophysiologic mechanisms of CF.26,27 However, there are problems with the BAL sampling technique in patients with CF. Foremost is the safety issue: young children may require general anesthesia for bronchoscopy and adults require conscious sedaClinical Investigations
tion. In addition, the CF lung compartments are not uniform. Inflammation is higher in the upper lobes than in the lower lobes.28 Some investigators compromise by sampling the lingula.29,30 Thirdly, the volume of saline solution instilled during BAL and whether to separate the first aliquot from later samples have not been standardized.31 The first aliquot from a CF BAL has a much higher percentage neutrophil count than subsequent samples.31 Finally, the BAL itself may cause an inflammatory response even in normal subjects.32 Thus, repeated BAL sampling to examine an airway inflammation outcome measure may lead to information reflecting adverse events from the bronchoscopy with BAL procedure rather than adverse events from the study drug.25 We have previously shown that SI may provide a more representative airway sample than BAL in patients with CF.14 When SI is compared with BAL and expectorated sputum (ES), SI had a lower percentage of squamous cells than ES, all three techniques had a similar nonsquamous cell differential count, quantitative microbiology correlated best between ES and SI, and cytokine measurements by all three techniques were similar once corrected for the dilution of BAL. The current study sought to determine whether a sputum sample obtained after a short time period (4 min) may reflect that of a longer time period, to examine whether 20 min of SI collection would sample the alveolar space as manifest by a change in WBC differential and surfactant concentration, and to determine if aliquots of induced-sputum samples showed changes in microbiologic and inflammatory indexes over time. Materials and Methods Subjects Subjects were recruited from the CF clinic at the University of Washington, Seattle, WA. All subjects had a proven diagnosis of CF. All subjects were ⱖ 18 years old and gave written informed consent for this study, which was approved by the Human Subjects Committee of the University of Washington. All subjects had an initial FEV1 ⬎ 40% predicted and were in clinically stable condition. Subjects remained receiving all long-term medications, including antibiotics. Study Design On the day of SI, spirometry was performed at baseline, 10 min after inhalation of 180 g of albuterol via metered-dose inhaler prior to SI, and 5 min following the completion of SI. Peak flow measurements were obtained every 4 min of the SI procedure to ensure that significant bronchospasm (20% fall) was not occurring. Spirometry was measured in all patients with a Cybermedic spirometer (Cybermedic; Boulder, CO) according to American Thoracic Society guidelines.33 www.chestjournal.org
SI Using a modification of the protocol described by Fahy et al,34 modified by the Cystic Fibrosis Foundation Therapeutics Development Network, subjects inhaled nebulized sterile hypertonic saline (3%) solution for 20 min from an Ultra-Neb 99 ultrasonic nebulizer (DeVilbiss; Somerset, PA). This nebulizer generates particles of a mean mass median diameter of 4.5 m and has an output of 2.4 mL/min. After expelling saliva, subjects were encouraged to cough, and sputum was collected at 4-min intervals. Sample Processing Samples were processed as previously described.14 All specimens were processed within 30 min of collection. The samples were weighed, and the volume was calculated from the weight. The whole sample, including plugs, was diluted 1:8 0.1% dithiothreitol (Sputolysin; Behring Diagnostics; Somerville, NJ) in phosphate-buffered saline solution. The sample was then mixed gently by vortex mixer and placed in a shaking water bath at 37°C for 15 min. Periodically, the samples were removed from the water bath for further brief gentle mixing with a plastic transfer pipette to ensure visible liquefaction. Following liquefaction with dithiothreitol, 1 mL was removed for quantitative culture. The liquefied sputum sample was used to determine the total cell count, and cytospin samples were prepared and stained (Hema 3; Fisher Scientific; Santa Clara, CA). The remainder of liquefied sputum was centrifuged at 2,000 revolutions per minute for 5 min. The supernatants were aspirated and frozen at ⫺ 70°C for later analysis. Slides were batched, and a blinded observer counted 500 nonsquamous epithelial cells and cell differentials per sputum preparation. Interleukin (IL)-8 concentrations were measured using a quantitative “sandwich” enzyme immunoassay technique (Quantikine; R&D Systems; Minneapolis, MN). Samples were run in triplicate and compared to a standard curve. In order to ensure that dithiothreitol did not degrade IL-8, we added 100 ng and 500 ng IL-8 for 90 min to each sample from three patients (ie, total of 15 samples) and generated standard curves. Dithiothreitol did not degrade IL-8. Urea concentration was measured at the clinical laboratory of the University of Washington Medical Center. Surfactant protein (SP)-A and SP-D were measured with a standard enzyme-linked immunosorbent assay as reported previously.35 Human SP-A was measured with a kit using two monoclonal antibodies provided by the Teijin Institute of Bio-Medicine (Hino, Japan). Recombinant SP-D was used as the standard for SP-D. The enzyme-linked immunosorbent assay was based on a sandwich method, using two monoclonal antibodies against human SP-D, 6B2 and 7C6, which were prepared against human SP-D purified from BAL fluids of patients with pulmonary alveolar proteinosis. Quantitative Culture Quantitative microbiology was performed at Children’s Hospital and Regional Medical Center by methods previously described.14,36 The 1-mL aliquots removed for microbiologic cultures were plated within 2 h of the addition of dithiothreitol. Statistical Methods Descriptive statistics and graphical displays were used to evaluate results by patient and sampling time. Analyses to examine the effect of sampling time on induced-sputum outcome measures were performed using generalized estimating equation methods. These methods allowed the fitting of linear regression models that accounted for the correlated nature of within-patient CHEST / 123 / 3 / MARCH, 2003
793
observations. The regression model for each outcome measure included indicator variables for the effects of sampling time, and a test that the time variables were jointly equal to zero was performed. The null hypothesis for these tests was that sampling times, taken together, had no effect on the outcome of interest. A test for linear time trend was also performed for each outcome measure by fitting a regression in which time was modeled as a continuous variable. Finally, the mean change calculated as the 20-min value minus the 4-min value with 95% confidence intervals (CIs) was calculated. Sample Size Calculations This was a pilot study. There was no prior information from which a sample size calculation could be made. However, we performed a post hoc power analysis to determine what size effect could have been missed given the sample and results of our pilot study.
Results Demographic and Safety Profile Ten patients with CF (eight male) entered and completed the study. Three patients were homozygous for DF508, six patients were heterozygous for DF508, and one patient’s alleles were G551D/ N1303K. Mean age was 28 ⫾ 7 years (range, 18 to 38 years). Three subjects were prescribed inhaled tobramycin and one patient was prescribed oral ciprofloxacin at the time of the collection. Four adverse events occurred in three subjects, including throat irritation in three subjects and headache in one subject. The baseline prealbuterol FEV1 was 2.58 ⫾ 0.73 L/min (66 ⫾ 16% predicted; range, 48 to 97%). The postalbuterol FEV1 was 2.74 ⫾ 0.74 L/min (71 ⫾ 16% predicted; range, 51 to 98%). The FEV1 fell an average of 7 ⫾ 7% predicted, with a range of zero to 17.6% predicted. Two patients had a ⬎ 15% fall in FEV1. This fall in FEV1 was seen at the end of the 20-min collection, as no subject had to prematurely stop the collection because of a 20% fall in peak flow measurement at any time point. Sample Volumes The sample volumes determined by weight were 3.0 ⫾ 1.8 mL, 2.8 ⫾ 1.5 mL, 4.6 ⫾ 6.6 mL, 4.8 ⫾ 5.1 mL, and 3.8 ⫾ 2.1 mL, collected at 4, 8, 12, 16, and 20 min, respectively. The mean change from the 0- to 4-min sample to the 16- to 20-min sample was 0.81 mL (95% CI, ⫺ 0.43 to 2.05). There was no significant difference in volume of inducedsputum aliquots over time (p ⫽ 0.12). Urea Urea was detected in all subjects and in all samples. The mean urea concentration was 4 ⫾ 1 mg/dL, 794
3 ⫾ 1 mg/dL, 3 ⫾ 1 mg/dL, 4 ⫾ 1 mg/dL, and 3 ⫾ 1 mg/dL at the five time points. The mean change from the 0- to 4-min sample to the 16- to 20-min sample was 0.20 mg/dL (95% CI, ⫺ 0.65 to 0.25). There was no linear trend in urea concentration with time (p ⫽ 0.99). Total Cell Counts and Cell Differentials The total cell counts of samples collected at 4, 8, 12, 16, and 20 min (mean ⫾ SEM) were 1.8 ⫾ 1.6 ⫻ 107/mL, 1.8 ⫾ 1.8 ⫻ 107/mL, 1.8 ⫾ 1.7 ⫻ 107/ mL, 1.6 ⫾ 1.7 ⫻ 107/mL, and 1.8 ⫾ 1.8 ⫻ 107/mL, respectively (Fig 1). The mean change was 0.016 log10 cells per milliliter (95% CI, ⫺ 0.193 to 0.225). The nonsquamous cell counts per sample were 1.8 ⫾ 1.7 ⫻ 107/mL, 1.7 ⫾ 1.8 ⫻ 107/mL, 1.8 ⫾ 1.7 ⫻ 107/mL, 1.6 ⫾ 1.7 ⫻ 107/mL, and 1.9 ⫾ 1.8 ⫻ 107/ mL, respectively. There was no sample at the 20-min time point in one patient. There was no difference in nonsquamous cell count over time (p ⫽ 0.99). The percentage of squamous cells were 5.9 ⫾ 6.6%, 6.8 ⫾ 10.1%, 4.7 ⫾ 5.2%, 5.1 ⫾ 4.7%, and 3.4 ⫾ 2.6% for each time point (p ⫽ 0.83). The percentage of neutrophils (mean ⫾ SD) were 89 ⫾ 5%, 88 ⫾ 4%, 87 ⫾ 4%, 84 ⫾ 6%, and 86 ⫾ 4%, sequentially (Fig 1), thus slightly decreasing over time (p ⫽ 0.03). The percentage macrophage count was 5 ⫾ 2%, 7 ⫾ 3%, 7 ⫾ 3%, 10 ⫾ 5%, and 8 ⫾ 4%, thus mildly increasing over time (p ⬍ 0.01; Fig 1). Microbiology The number of organisms in study subjects ranged from two to seven. In addition to the presence, the quantitation of a specific set of organisms was examined. Mucoid and nonmucoid Pseudomonas aeruginosa (mucoid n ⫽ 9, nonmucoid n ⫽ 5), Staphylococcus aureus (n ⫽ 6), Burkholderia cepacia (n ⫽ 1), Stenotrophomonas maltophilia (n ⫽ 4), Achromobacter (formerly Alcaligenes) xylosoxidans (n ⫽ 2), Haemophilus influenzae (n ⫽ 0), Enterococus (n ⫽ 1), Acinobacter (n ⫽ 1) and Ralstonia pickettii (n ⫽ 6) were selected. Quantitation of each organisms in each sample was also examined (Fig 2). There was a mean half-log decrease in both nonmucoid P aeruginosa and S aureus with time (p ⫽ 0.05 and p ⬍ 0.01, respectively). The mean changes in nonmucoid P aeruginosa density and S aureus were 0.0.54 log10 cfu/mL (95% CI, ⫺ 1.60 to 0.52) and ⫺ 0.55 log10 cfu/mL (95% CI, ⫺ 1.42 to 0.32), respectively. There was no change in bacterial density of mucoid P aeruginosa (p ⫽ 0.12). R pickettii was seen in 60% of subjects but at low colony counts up to 7,600 colonies per milliliter. Clinical Investigations
Figure 1. The total number of cells and the differential WBC count from induced-sputum aliquots. The total number of cells from the induced-sputum aliquots remained constant over time (p ⫽ 0.99). The percentage of PMNs decreased slightly and the percentage alveolar macrophages increased slightly over time (p ⫽ 0.03 and p ⬍ 0.01, respectively).Macs ⫽ alveolar macrophages.
Inflammatory Mediators IL-8 was detected in all subjects and in all samples. The mean ⫾ SD concentration (log10) was 1.8 ⫾ 1.4 ng/mL, 1.8 ⫾ 1.5 ng/mL, 1.9 ⫾ 1.6 ng/mL, 1.8 ⫾ 1.6 ng/mL, and 1.8 ⫾ 1.5 ng/mL at the five time points, respectively. The mean change in IL-8 (log10) was ⫺ 0.026 ng/mL (95% CI, ⫺ 0.103 to 0.051). There was no linear time trend in IL-8 concentrations (p ⫽ 0.99; Fig 3). SPs SP-D was not found in any sample. SP-A was detected in all subjects and in all samples. The mean ⫾ SD concentration (log10) was 1.62 ⫾ 0.29 ng/mL, 1.95 ⫾ 0.62 ng/mL, 2.23 ⫾ 0.58 ng/mL, 2.30 ⫾ 0.59 ng/mL, and 2.40 ⫾ 0.44 ng/mL at the five time points, respectively. The mean change in SP-A (log10) was 0.79 ng/mL (95% CI, 0.46 to 1.11). There was a linear time trend in SP-A concentrations (p ⬍ 0.001; Fig 4). www.chestjournal.org
Interpretation of Results in Light of Multiple Comparisons As 10 different statistical outcome measures were performed, a value of p ⱕ 0.005 is required to maintain the overall risk of a type I error equal to 0.05. We can reasonably conclude that only SP-A is statistically remarkable (p ⬍ 0.001). Discussion Serial aliquots of induced sputum in clinically stable adult CF patients with mild-to-moderate obstructive lung disease show significant changes over time in SP-A concentration and a 3% increase in alveolar macrophages. This provides evidence for increased distal airway sampling over time with the SI procedure in patients with CF. In addition, there was a 3% decreased in polymorphonuclear leukocytes (PMNs), and a half-log decrease in P aeruginosa and S aureus density between the 0- to 4-min and 16- to 20-min collections. CHEST / 123 / 3 / MARCH, 2003
795
Figure 2. The microbiological counts from induced-sputum aliquots. Nine subjects had mucoid P aeruginosa, five subjects had nonmucoid P aeruginosa, and six subjects had S aureus. The bacteriological counts of nonmucoid P aeruginosa and S aureus changed over time (p ⫽ 0.05 and p ⬍ 0.01, respectively).
SI was relatively well tolerated by all subjects. However, two subjects had a ⬎ 15% fall in FEV1 with this procedure even when premedicated with albuterol. However, the adverse effects associated with the sedation and the procedure of bronchoscopy with BAL on pulmonary function and oxygenation in patients with CF is not well reported but would appear greater than those of SI.14,25,37 Throat irritation was felt by three subjects. Induction of sputum with 3% hypertonic saline solution has been shown to be relatively safe in patients with moderate-to-severe asthma,9,34,38 and was associated with no serious adverse events in our 10 subjects with CF. The majority of CF patients have reactive airway disease, but the degree of airway reactivity to the challenge of hypertonic saline solution in the limited number of CF subjects studied to date appears to be less than that observed among asthmatic subjects.14 –16 Only 1 of our 10 subjects was not able to produce 1 mL of sample volume at each time point, the minimum volume often necessary to perform routine assays and reproducible cell counts. In the inducedsputum protocol of asthmatic subjects that we adopted, the methodologies call for a collection time of 12 min,9,34,38 although a 1-mL sample after a minimum time period of 4 min is deemed accept796
able. In asthmatic subjects, fractional analysis of sequential induced sputum at 4-min intervals over a 20-min interval showed a decrease in percentage of PMNs and an increase in percentage of macrophages after the 0- to 4-min aliquot. In the CF population, the total cell counts are higher than those seen in asthmatics, reflecting the chronic infected airways of CF patients. The change in aliquots appeared to be less pronounced than those of asthmatics, so that a 4-min collection may be more representative of a longer time sample in CF subjects. Thus, shorter sputum collection times may be adequate for CF subjects. Microbiology results suggested SI may be a useful tool in the multicentered research setting. Aliquots of induced sputum appear to provide a representative sample as quantitative bacterial cultures only varied by approximately a half-log count over time. In standard induced-sputum sample processing, dithiothreitol has to be added to the sputum sample to homogenize it. However, the addition of dithiothreitol decreases microbiological counts after 2 h. Thus, for multicenter studies, when microbiological samples may be sent to a centralized laboratory, it appears that a sample from the first 4-min aliquot may vary from the last aliquot by a half-log count. Inflammatory mediators were found in concentraClinical Investigations
Figure 3. Concentrations of IL-8 from induced-sputum aliquots. IL-8 was detected in all subjects and in all samples. There was no linear time trend in IL-8 concentrations (p ⫽ 0.99).
tions equivalent to or greater than those previously reported in patients with CF.14,30 The presence of hypertonic saline solution may increase the
amount of Na⫹ and Cl⫺ in airway surface liquid and rehydrate the periciliary fluid.39 Hyperosmolarity has been shown to stimulate IL-8 production
Figure 4. Concentrations of SP-A from induced-sputum aliquots. SP-A was detected in all subjects and in all samples. There was a linear time trend in SP-A concentrations (p ⬍ 0.001). HSPA ⫽ human surfactant protein-A. www.chestjournal.org
CHEST / 123 / 3 / MARCH, 2003
797
in human bronchial epithelial cells in vitro,39 but we doubt that a 20-min exposure to hypertonic saline solution in vivo would allow adequate time to increase intracellular IL-8 production. Type II alveolar epithelial cells are responsible for producing and secreting surfactant and most of the surfactant-associated proteins (SP-A, SP-B, SP-C, and SP-D) into the alveolar lining fluid. Deficiencies in SP-A and SP-D in lavage fluid have been documented in various human inflammatory lung diseases, including pneumonia, ARDS,35 and CF.40 Greene et al35 demonstrated that reduced levels of SP-D in BAL fluid from patients with ARDS are associated with a worse outcome. Greene et al35 hypothesize that the low surfactant levels are not just a biomarker of acute lung injury, but that their absence exacerbates the dysfunctional inflammation characteristic of acute lung injury. As anticipated, the concentration of SP-A increased in the later time samples in our study, reflecting that the content of these later samples contained more alveolar fluid. Sputum induction has become a well-validated research tool for evaluating a variety of indexes of inflammation, both cellular and humoral, in non-CF diseases of the airways such as asthma. This study demonstrates that SI is well tolerated and provides valid information regarding the markers of inflammation and the microbiology of the airways in subjects with CF. If patients or research subjects are able to expectorate at least 1 mL of sample using hypertonic saline solution, then this can provide an adequate sample for clinical information. However, further studies are needed to examine modest, but significant changes in CF pathogen density and PMN count for use of serial aliquots as outcome measurements in clinical trials. In summary, fractional analysis of induced sputum from clinically stable CF subjects shows that aliquots of induced sputum provide representative samples for microbiologic and inflammatory indexes, and this suggests it should be possible to apply the varied processing methodologies necessary for different outcome measurements in multicenter studies. References 1 Robinson M, Regnis J, Bailey D, et al. Effect of hypertonic saline, amiloride, and cough on mucociliary clearance in patients with cystic fibrosis. Am J Respir Crit Care Med 1996; 153:1503–1509 2 Eng P, Morton J, Douglass J, et al. Short-term efficacy of ultrasonically nebulized hypertonic saline in cystic fibrosis. Pediatr Pulmonol 1996; 21:77– 83 3 DeBoeck K, Alifier M, Vandeputte A. Sputum induction in young cystic fibrosis patients. Eur Respir J 2000; 16:91–94 4 Rodwell L, Anderson S. Airway responsiveness to hyperosmolar saline challenge in cystic fibrosis: a pilot study. Pediatr Pulmonol 1996; 21:282–289 798
5 Pin I, Gibson P, Kolendowicz R, et al. Use of induced sputum cell counts to investigate airway inflammation in asthma. Thorax. 1992; 47:25–29 6 Fahy J, Fleming H, Wong H, et al. The effect of an anti-IgE monoclonal antibody on the early- and late-phase responses to allergen inhalation in asthmatic subjects. Am J Respir Crit Care Med 1997; 155:1828 –1834 7 Pizzichini M, Popov T, Efthimiadis A, et al. Spontaneous and induced sputum to measure indices of airway inflammation in asthma. Am J Respir Crit Care Med 1996; 154:866 – 869 8 Kips J, Fahy J, Hargreave F, et al. Methods for sputum induction and analysis of induced sputum method for assessing airway inflammation in asthma. Eur Respir J 1998; 11:9S–12S 9 Fahy J, Liu J, Wong H, et al. Cellular and biochemical analysis of induced sputum from healthy and asthmatic subjects. Am Rev Respir Dis 1993; 147:1126 –1131 10 Pin I, Freitag AP, O’Byrne PM, et al. Changes in the cellular profile of induced sputum after allergen-induced asthmatic responses. Am Rev Respir Dis 1992; 145:1265–1269 11 Keatings V, Collins PD, Scott DM, et al. Differences in interleukin-8 and tumor necrosis factor-␣ in induced sputum from patients with chronic obstructive pulmonary disease or asthma. Am J Respir Crit Care Med 1996; 153:530 –534 12 Bhowmik A, Seemungal TA, Sapsford R, et al. Comparison of spontaneous and induced sputum for investigation of airway inflammation in chronic obstructive pulmonary disease. Thorax 1998; 53:953–956 13 Gibson P, Gabardo A, Morris M, et al. Cellular characteristics of sputum from patients with asthma and chronic bronchitis. Thorax 1989; 44:693– 699 14 Henig N, Tonelli M, Pier M, et al. Sputum induction as a research tool for sampling the airways of subjects with cystic fibrosis. Thorax 2001; 56:306 –311 15 Ordonez C, Stulbarg M, Grunland H, et al. The safety and reproducibility of sputum induction to measure airway inflammation in adults with cystic fibrosis [abstract]. Pediatr Pulmonol 2000; 20(suppl):A319 16 Ordonez C, Remold-O’donnell E, Wohl M. The feasibility, safety and repeatability of sputum induction to study infection and airway inflammation in children with cystic fibrosis [abstract]. Pediatr Pulmonol 2000; 20(Suppl):A320 17 Rosenfeld M, Emerson J, Accurso F, et al. Diagnostic accuracy of oropharyngeal cultures in infants and young children with cystic fibrosis. Pediatr Pulmonol 1999; 28:321– 328 18 Baughman R, Keeton D, Perez C, et al. Use of bronchoalveolar lavage semiquantitative cultures in cystic fibrosis. Am J Respir Crit Care Med 1997; 156:286 –291 19 Ramsey B, Wentz K, Smith A, et al. Predictive value of oropharyngeal cultures for identifying lower airway bacteria in cystic fibrosis patients. Am Rev Respir Dis 1991; 144:331– 337 20 Burns J, Gibson R, McNamara S, et al. Longitudinal assessment of Pseudomonas aeruginosa in young children with cystic fibrosis. J Infect Dis 2001; 183:444 – 452 21 Armstrong D, Grimwood K, Carlin J, et al. Bronchoalveolar lavage or oropharyngeal cultures to identify lower respiratory pathogens in infants with cystic fibrosis. Pediatr Pulmonol 1996; 21:267–275 22 Nixon G, Armstrong D, Carzino R, et al. Clinical outcome after early Pseudomonas aeruginosa infection in cystic fibrosis. J Pediatr 2001; 138:699 –704 23 McElvaney N, Hubbard R, Birrer P, et al. Aerosol ␣1antitrypsin treatment for cystic fibrosis. Lancet 1991; 337: 392–394 24 McElvaney N, Nakamura H, Birrer P, et al. Modulation of Clinical Investigations
25
26
27
28
29
30
31
32
airway inflammation in cystic fibrosis: in vivo suppression of interleukin-8 levels on the respiratory epithelial surface by aerosolization of recombinant secretory leukoprotease inhibitor. J Clin Invest 1992; 90:1296 –1301 Aitken M, Moss R, Waltz D, et al. A phase 1 study of aerosolized administration of tgAAVCF to CF patients with mild lung disease. Hum Gene Ther 2001; 12:1907–1916 Khan T, Wagener J, Bost T, et al. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995; 151:939 –941 Balough K, McCubbin M, Weinberger M, et al. The relationship between infection and inflammation in the early stages of lung disease from cystic fibrosis. Pediatr Pulmonol 1995; 20:63–70 Meyer K, Sharma A. Regional variability of lung inflammation in cystic fibrosis. Am J Respir Crit Care Med 1997; 156:1536 – 1540 Konstan M, Walenga R, Hilliard K, et al. Leukotriene B4 markedly elevated in the epithelial lining fluid of patients with cystic fibrosis. Am Rev Respir Dis 1993; 148:896 –901 Konstan M, Hilliard K, Norvell T, et al. Bronchoalveolar lavage findings in cystic fibrosis patients with stable, clinically mold lung disease suggest ongoing infection and inflammation. Am J Respir Crit Care Med 1994; 1994:448 – 454 Ratjen F, Rietschel E, Griese M, et al. Fractional analysis of bronchoalveolar lavage fluid cytology in cystic fibrosis patients with normal lung function; Bronchoalveolar Lavage for the Evaluation of Anti-inflammatory Treatment (BEAT) study group. Eur Respir J 2000; 151:141–145 Picard E, Schwartz S, Goldberg S, et al. A prospective study
www.chestjournal.org
33 34
35
36 37
38
39 40
of fever and bacteremia after flexible fiberoptic bronchoscopy in children. Chest 2000; 117:573–577 American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1995; 152:1107– 1136 Fahy J, Liu J, Wong H, et al. Analysis of cellular and biochemical constituents of induced sputum after allergen challenge: a method for studying allergic airway inflammation. J Allergy Clin Immunol 1994; 93:1031–1039 Greene K, Wright J, Steinberg K, et al. Serial changes in surfactant-associated proteins in lung and serum before and after onset of ARDS. Am J Respir Crit Care Med 1999; 160:1843–1850 Burns J, Emerson J, Stapp J, et al. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin Infect Dis 1998; 27:158 –163 Wagener J, Rock M, McCubbin M, et al. Aerosol delivery and safety of recombinant human deoxyribonuclease in young children with cystic fibrosis: a bronchoscopic study. J Pediatr 1998; 133:486 – 491 Gershman N, Liu H, Wong H, et al. Fractional analysis of sequential induced sputum samples during sputum induction: evidence that different lung compartments are sampled at different time points. J Allergy Clin Immunol 1999; 104:322– 328 Folkesson H, Kheradmand F, Mathay M. The effect of saltwater on alveolar epithelial barrier function. Am J Respir Crit Care Med 1994; 150:1555–1563 von Bredow C, Birrer P, Griese M. Surfactant protein A and other bronchoalveolar lavage fluid proteins are altered in cystic fibrosis. Eur Respir J 2001; 17:716 –722
CHEST / 123 / 3 / MARCH, 2003
799
