Abstract
Rationale: Unrecognized airway infection and inflammation in young children with cystic fibrosis (CF) may lead to irreversible lung disease; therefore early detection and treatment is highly desirable.
Objectives: To determine whether the lung clearance index (LCI) is a sensitive and repeatable noninvasive measure of airway infection and inflammation in newborn-screened children with CF.
Methods: Forty-seven well children with CF (mean age, 1.55 yr) and 25 healthy children (mean age, 1.26 yr) underwent multiple-breath washout testing. LCI within and between-test variability was assessed. Children with CF also had surveillance bronchoalveolar lavage performed.
Measurements and Main Results: The mean (SD) LCI in healthy children was 6.45 (0.49). The LCI was higher in children with CF (7.21 [0.81]; P < 0.001). The upper limit of normal for the LCI was 7.41. Fifteen (32%) children with CF had an elevated LCI. LCI measurements were repeatable and reproducible. Airway infection was present in 17 (36%) children with CF, including 7 (15%) with Pseudomonas aeruginosa. Polymicrobial growth was associated with worse inflammation. The LCI was higher in children with Pseudomonas (7.92 [1.16]) than in children without Pseudomonas (7.02 [0.56]) (P = 0.038). The LCI correlated with bronchoalveolar lavage IL-8 (R2 = 0.20, P = 0.004) and neutrophil count (R2 = 0.21, P = 0.001). An LCI below the upper limit of normality had a high negative predictive value (93%) in excluding Pseudomonas.
Conclusions: The LCI is elevated early in CF, especially in the presence of Pseudomonas and airway inflammation. The LCI is a feasible, repeatable, and sensitive noninvasive marker of lung disease in young children with CF.
Airway infection, inflammation, and structural change such as bronchiectasis are present within the first few months of life even in newborn-screened infants with cystic fibrosis (CF). A decline in lung function has also been reported particularly in infants infected with Pseudomonas aeruginosa. As these early pathological events occur mostly in the absence of symptoms, the roles of bronchoalveolar lavage, infant lung function testing, and high-resolution computed tomography for surveillance of early CF lung disease are under evaluation. The lung clearance index (LCI), a simple measure of ventilation inhomogeneity reflecting small airway disease, has been shown to sensitively identify abnormal lung function in preschool and older children with cystic fibrosis.
The LCI is a repeatable measure of small airway function in healthy infants and young children and similarly aged children with CF. The LCI is elevated in stable, well-nourished newborn-screened infants and young children with CF compared with their healthy peers. An abnormal LCI is associated with airway inflammation and Pseudomonas aeruginosa. Our findings suggest that the LCI may be a useful marker of early CF lung disease. Furthermore, our LCI repeatability data highlight the potential role of the LCI as an outcome measure for future intervention trials.
Cystic fibrosis (CF) lung disease is characterized by unremitting cycles of airway infection and inflammation that begin early in life and are often clinically inapparent (1). During infancy airway inflammation may be associated with significant structural change including bronchiectasis (2). As early childhood is a crucial period of rapid lung development, unrecognized airway disease may have irrevocable consequences for later respiratory health (3). Therefore investigations that can sensitively identify early disease and that are also noninvasive, repeatable, and simple to apply clinically may have the greatest potential to improve outcomes (4).
Oropharyngeal cultures (5, 6), chest X-rays (7), and serum antibodies (8) are insensitive to detect early CF lung disease. Therefore the indications for bronchoalveolar lavage (BAL), high-resolution computed tomography (HRCT), and infant lung function testing (ILFT) are currently being determined. In addition, clinically relevant outcome measures are needed to evaluate new therapies in trials involving younger children with CF (9–11). Bronchoalveolar lavage is the present standard for diagnosing lower airway infection and inflammation in young children (12, 13). However, BAL remains invasive and limited sampling may underestimate infection or inflammation because of regional variability (14, 15). Few centers perform surveillance BAL in young children with CF (16, 17).
Given the inherent invasiveness of BAL, the optimal interval between bronchoscopic sampling to monitor early disease is unknown. Similarly HRCT, which has sensitively revealed structural abnormalities in infants as young as 3 months (2), has cumulative radiation exposure concerns that may hamper its role as a serial outcome measure of early CF lung disease (18).
In contrast, ILFT has demonstrated both an ability to noninvasively identify early airway disease in cross-sectional studies (19–21) and to define its evolution over time (16, 22). Both the raised-volume rapid thoracoabdominal compression technique (21) and the multiple-breath washout (MBW) method (20) have detected abnormal lung function in infants with CF. However, a multicenter study using the raised-volume rapid thoracoabdominal compression technique identified “poor feasibility, low repeatability and the need for large sample sizes to detect reasonable treatment effects” as important constraints with this technique in clinical trials involving infants with CF (23).
The lung clearance index (LCI), a commonly reported MBW outcome measure, may more sensitively detect early functional pulmonary impairment, as it reflects ventilation inhomogeneity due to small airway pathology, the hallmark of early CF lung disease. The LCI has identified early disease more sensitively than spirometry in preschool children with CF (24). However, the ability of the LCI to detect presymptomatic disease in newborn-screened infants and young children with CF, as well as its variability in this age group, remains unknown.
The aims of the current research, therefore, were to evaluate the feasibility and repeatability of the LCI in infants and young children with and without CF and to determine the association between the LCI and airway inflammation and infection. Hence our objective was to determine the usefulness and sensitivity of the LCI as a noninvasive measure of early lung disease in infants and young children with CF. We hypothesized that the LCI would be elevated in presymptomatic/minimally symptomatic newborn-screened infants and young children with CF when compared with their healthy peers, reflecting early subclinical lung disease. Some of the results of this study have been previously reported in the form of abstracts (25, 26).
Children with CF less than 3 years of age, admitted for an annual BAL to Sydney Children's Hospital (Randwick, NSW, Australia) between June 2004 and August 2009 as part of an early disease surveillance program, were recruited for MBW testing. Infants were identified through newborn screening or by meconium ileus presentation and the diagnosis was confirmed by sweat chloride concentration greater than 60 mmol/L or by CF genetic mutation analysis. Exclusion criteria were (1) respiratory infection within 3 weeks and (2) coexisting cardiac, renal, or neuromuscular conditions or lung disease of prematurity. Parents completed a detailed symptom and history questionnaire (see the questionnaire in the online supplement).
Control subjects were infants and young children attending Sydney Children's Hospital, Randwick, between April 2005 and April 2009, either for a sedated echocardiograph, where normal cardiac structure and function were found, or a dimercaptosuccinic acid scan for previous urinary tract infection. Parental consent was obtained for MBW testing as an add-on procedure. Exclusion criteria were (1) the presence of cardiac, respiratory, or neuromuscular disease; (2) prematurity; (3) respiratory hospitalization; (4) history of asthma, wheezing, breathlessness, chronic cough, or use of antiasthma medication; and (5) respiratory infection within 3 weeks.
MBW testing was performed at the bedside in the pediatric medical procedures unit (medical day unit), using a commercially available mainstream ultrasonic flowmeter (Exhalyzer D; Eco Medics AG, Duernten, Switzerland) with sulfur hexafluoride as the tracer gas. The LCI was determined by dedicated data acquisition and analysis software (WBeath, version 3.19.8.0; ndd Medical Technologies, Zurich, Switzerland). Children were examined, weighed, and measured and then sedated with oral chloral hydrate as per guidelines (27, 28).
All children (CF and non-CF) were tested using the same equipment and technique. However, the dose of chloral hydrate in the non-CF children scheduled for a dimercaptosuccinic acid scan was lower (30–50 mg/kg compared with 50–80 mg/kg) according to the hospital's sedation protocol for that procedure. The MBW equipment was leak tested and calibrated before each patient assessment.
MBW was performed with the child in the supine position during quiet sleep after regular tidal breathing was established (usually 1–2 min). The wash-in was initiated if there was no evidence of mask leak or an unstable end-expiratory level. A minimum of two (ideally three to five) complete wash-in/wash-out curves were obtained for each child without adjusting the mask or body position. This formed the first set of measurements and was usually complete within 10–15 minutes. After a 5- to 10-minute interval and mask repositioning a second set of curves was obtained to allow assessment of between-test LCI reproducibility.
All wash-in/wash-out curves were saved, but only recordings that met acceptability criteria were used to derive the LCI (29). Hence the LCI was determined from wash-out curves in which there was no evidence of leak, sighs, hiccoughing, swallowing, or arousal and in which the functional residual capacity measurements differed less than 10% in relation to the lower value of the other curves within the set (29). The mean LCI was determined from three (minimum, two) acceptable wash-out curves within each set (29).
A chest X-ray was performed after MBW testing once the child was awake. Children remained overnight (for 24-h pH monitoring) and underwent BAL on the following day or were readmitted 48 hours later to the day surgery unit for this. They were subsequently discharged home after a 4-hour period of observation postbronchoscopy.
Two CF-specific scores, the Brasfield score (BS) (30) and the modified Chrispin–Norman score (MCNS) (31) were used to assess structural lung disease and were scored by a single pediatric radiologist blinded to the child's clinical status. Severity cutoff values were used to assess “irreversible” lung disease (32, 33). For the BS this was less than 21 and for the MCNS this was greater than 5.
Bronchoalveolar lavage was performed under general anesthesia within 72 hours of MBW testing. Suctioning through the bronchoscope (Olympus models BF-3C40, BF-3C160, and BF-XP16F; Olympus Medical Systems Corporation, Tokyo, Japan) was avoided until the tip had passed beyond the carina.
The bronchoscope was sequentially wedged into the right upper lobe, right middle lobe, and lingula. A single aliquot (1 ml/kg; minimum, 10 ml; maximum, 20 ml) of warmed nonbacteriostatic sterile saline was instilled into each lobe and BAL fluid was immediately aspirated. Three-lobe lavage was performed to optimize detection of airway infection (14) and inflammation (34). Similarly, topical anesthesia was applied only after BAL samples were collected to prevent bacterial growth inhibition (35, 36).
Pooled BAL fluid samples were processed for cell count and differential, IL-8 (37), and the complex of neutrophil elastase with α1-protease inhibitor (NE/α1-PI complex). In addition, quantitative bacterial microbiology and viral immunofluorescence and culture were performed (37). Airway infection was defined as pathogen growth equal to or greater than 105 colony-forming units per milliliter (cfu/ml) of BAL fluid, or a positive viral immunofluorescence or culture (1).
Statistical analyses were performed with SPSS version 15.0 (SPSS Inc., Chicago, IL). The LCI was the main outcome measure. The mean LCI value from set 1 in each child was used for all analyses. On the basis of available LCI data in 43 children with CF and 28 healthy control subjects, we estimated that 25 children per group (healthy and CF) were required to detect a 1–standard deviation (SD) difference in LCI with 80% power at the 5% significance level (38). In addition, an interim analysis of our own data demonstrated that a sample size of 16 per group would be sufficient to detect a 1-SD difference in LCI between infected and noninfected children with CF with 80% power and a 5% significance level.
Lung clearance index within-test repeatability was assessed by the coefficient of variation (CV) of acceptable curves within set 1. The between-test LCI reproducibility was determined by Bland-Altman analysis (39), using the mean LCI values from sets 1 and 2.
An LCI greater than the upper limit of normality (ULN), defined as the mean LCI + 1.96 SD value in healthy children, was classified as abnormal. Categorical data were compared by chi square or Fisher exact tests and continuous data by t tests or Mann-Whitney U tests as appropriate. Associations between continuous data such as age, inflammatory markers, chest X-ray scores, pathogen density, and LCI were assessed by scatter plot and linear regression. Variables were log transformed if required. Negative BAL cultures were assigned a value of 1 (100) cfu/ml to allow log transformation of pathogen density and because the lower level of sensitivity of quantitative culture was 101 cfu/ml. Each child contributed only one set of data. Receiver operating characteristic (ROC) curve analysis was used to assess the discriminative ability of the LCI to detect infection and inflammation. A P value less than 0.05 was considered statistically significant. The study was approved by the South Eastern Sydney Area Health Service Human Research Ethics Committee and registered at the Australian and New Zealand Clinical Trial Register (ACTRN: ACTRN12611000945921).
Sedated MBW testing was attempted in 50 of 55 eligible children with CF. Parents of five children declined lung function. All 50 children with CF were adequately sedated for the duration of MBW testing. However, in two children, quality criteria for LCI were not met. In one sedated child MBW testing was not performed as this child developed intermittent oxygen desaturation (88–94%) due to upper airway obstruction and required low-flow oxygen. Hence 47 children with CF, mean age (range) 1.55 (0.36–3.10) years, had technically acceptable LCI measurements. The mean age (range) of the eight eligible children who did not provide technically acceptable data was 1.96 (0.13–2.98) years.
Thirty-six healthy non-CF children were given chloral hydrate. Of these children 25, mean age (range) 1.26 (0.32–3.24) years, were able to be sedated and had technically acceptable LCI data. Eleven non-CF children, mean age 1.62 (0.50–3.50) years, did not sedate in time or woke up prematurely before test completion. Therefore overall 72 of 86 (84%) sedated children provided technically acceptable LCI data.
No major adverse events, defined as termination of procedure, need for resuscitation, intubation for suspected aspiration, need for supplemental oxygen greater than 1 hour, need for positive expiratory pressure, fever of at least 38.5°C, or unexpected admission, were related to bronchoscopy/BAL.
All children with CF had a normal clinical examination and 45 (96%) were receiving antistaphylococcal prophylaxis. There was no difference in age, sex, nutritional status, tobacco smoke exposure, family history of asthma, or breastfeeding duration between healthy children and children with CF (Table 1).
| Healthy Children (n = 25) | Cystic Fibrosis (n = 47) | P Value | |
| Male, n (%) | 10 (40) | 22 (47) | 0.580 |
| Age, yr | 0.113 | ||
| Mean (SD) | 1.26 (0.69) | 1.55 (0.76) | |
| Range | 0.32–3.24 | 0.36–3.10 | |
| Weight (kg), mean (SD) | 10.1 (2.4) | 10.8 (2.2) | 0.264 |
| Weight < 10th centile, n (%) | 3 (12) | 6 (13) | 1.0 |
| Height (cm), mean (SD) | 77.6 (9.5) | 80.9 (9.1) | 0.152 |
| Height < 10th centile, n (%) | 0 | 4 (9) | 0.219 |
| Antenatal smoking, n (%)* | 1 (5) | 4 (9) | 1.0 |
| Household smokers, n (%)* | 1 (5) | 9 (19) | 0.152 |
| Breastfeeding, mo | |||
| Mean (SD) | 6.1 (4.8) | 4.3 (4.6) | 0.158 |
| Range | 0–14 | 0–15 | |
| Asthma/atopy first-degree relatives, n (%)* | 13 (59) | 31 (66) | 0.580 |
| Cough† last year, n (%) | 0 | 17 (36) | <0.001 |
| Cough last month, n (%) | 0 | 23 (49) | <0.001 |
| Wheeze last year, n (%) | 0 | 19 (40) | <0.001 |
| Wheeze last month, n (%) | 0 | 10 (21) | <0.001 |
| Respiratory admission, n (%) | 0 | 10 (21) | <0.001 |
| IV antibiotics last year, n (%) | 0 | 9 (20) | <0.001 |
| Previous P. aeruginosa (BAL), n % | 0 | 1 (2) | <0.001 |
The mean (SD) CVLCI was almost identical for children with CF and healthy children at 3.9 (2.4)%, range 0–9.9% and 3.8 (2.5)%, range 0.5–11.2%, respectively. There was no relationship between CVLCI and age in either group (P = 0.16 and P = 0.83, respectively).
The LCI between-occasion reproducibility (between the first and second set of wash-in/wash-out curves for each child measured 5–10 min apart) is shown in the mean-versus-difference (Bland–Altman) plot of Figure 1. The mean (SD) difference in LCI between the two sets in healthy children was –0.07 (0.31) and the 95% limits of agreement were –0.69; 0.54. In children with CF this was 0.11 (0.30) and –0.48; 0.70, respectively, demonstrating good and comparable reproducibility with healthy control subjects.
Figure 1. Bland-Altman analysis of lung clearance index (LCI) measured 5–10 minutes apart in all children. The thick dashed line indicates the mean difference in LCI measured 5–10 minutes apart from all children and the thinner dashed lines represent the 95% limits of agreement of this difference (mean ± 1.96 SD).
[More] [Minimize]The mean (SD) LCI for the 25 healthy children was 6.45 (0.49), range 5.42–7.37 with a ULN of 7.41. In comparison, the LCI was higher in the 47 children with CF (7.21 [0.81]; range, 6.19–11.04; P < 0.001). Importantly, 15 (32%) children with CF, age range 0.36–2.67 years, had an elevated LCI (Figure 2).
Figure 2. Lung clearance index (LCI) and age for all children. The dashed lines represent the mean ± 1.96 SD (95% limits of normality) for LCI for healthy children. The upper limit of normality is 7.41. There was no relationship between age and the LCI (R2 = 0.028, P = 0.164) when all children were combined. Fifteen children with CF had an abnormal LCI (>7.41).
[More] [Minimize]A difference in LCI between CF and non-CF children was apparent even in infants less than 12 months of age. The mean (SD) LCI in 14 infants with CF was 7.53 (1.10) compared with 6.73 (0.37), in healthy infants (P = 0.022). Six of 14 (43%) infants with CF had an LCI above the ULN.
In healthy children the LCI was negatively related to age (P = 0.026) but not sex, weight, or height. There was no correlation between LCI and age, sex, height, or weight in children with CF. Similarly, symptoms such as cough or wheeze or risk factors such as parental smoking were not associated with an abnormal LCI (Table E1).
Airway infection (≥105 cfu/ml BAL fluid) was present in 17 (36%) children with CF. In 11 (23.5%) children a respiratory pathogen was present but in low colony densities (101–105 cfu/ml). In 19 (40.5%) children no pathogens were detected (Table E2). Seven (15%) children had Pseudomonas aeruginosa infection. A further three had P. aeruginosa isolated in low colony densities.
P. aeruginosa and Haemophilus influenzae were the organisms most commonly detected (Table E2). Staphylococcus aureus was uncommon (2%). Fifteen children (32%) had 1 organism (of any colony density) isolated. Thirteen children (28%) had polymicrobial growth. In 10 children, 2 organisms were detected and in 3 children, 3 organisms were detected. There was no relationship between increasing age and number of pathogens (P = 0.190). No viruses were identified by immunofluorescence or culture, although two children were positive for cytomegalovirus by polymerase chain reaction.
Clinical characteristics including genotype, nutritional status, and respiratory symptoms did not distinguish children with or without infection (Table E3). Similarly, previous respiratory admission and intravenous antibiotic use in the last year were not associated with infection, although numbers were small (n = 10 and n = 9, respectively) (Table E3).
Inflammatory markers were higher in infected BAL samples (Table E4). There were significant correlations between inflammatory markers and actual pathogen load. Pathogen density explained 56% of the variability in BAL neutrophil percentage (P < 0.001) and 44% of the variability in IL-8 levels (P < 0.001) (Figures E1 and E2). However, there was no association between NE/α1-PI complex and infection.
Airway inflammation was also related to the number of pathogens in BAL fluid, indicating that polymicrobial growth was associated with worse inflammation in children with CF (Table 2).
| Not More Than One Pathogen (n = 34) | At Least Two Pathogens (n = 13) | P Value | |
| Neutrophils %, median (IQR) | 25 (7–42) | 75 (62–91) | <0.001 |
| Neutrophil count,* median (IQR) | 32 (10–115) | 402 (95–2,465) | 0.001 |
| Total cell count,* median (IQR) | 213 (125–391) | 682 (162–2,716) | 0.039 |
| IL-8 (pg/ml), median (IQR) | 1,144 (387–3,048) | 4,085 (2,340–6,638) | 0.003 |
| NE/α1-PI complex,† mean (SD) | 126 (114) | 146 (121) | 0.734 |
The BS ranged from 19 to 25 and the MCNS ranged from 0 to 8, suggesting mild structural lung disease. There was no association between either score and airway infection (P = 0.138 and P = 0.806, BS and MCNS, respectively). Furthermore, proposed cutoff scores indicating progression to “irreversible” CF lung disease did not relate to airway infection (BS < 21, P = 0.237 and MCNS > 5, P = 0.417).
The mean (SD) LCI was not different between children with and without airway infection: LCI = 7.54 (1.10) versus 7.02 (0.51), P = 0.083 (Figure 3). However, there was a positive correlation between LCI and pathogen load in children with CF (R2 = 0.10, P = 0.031) (Figure 4). In addition, there was a progressive increase in mean LCI when the three groups of children were compared, that is, healthy children (6.45 [0.49]) versus uninfected children with CF (7.02 [0.51]) versus infected children with CF (7.54 [1.10]) (Figure 5). Compared with healthy control subjects, the LCI was significantly elevated in both infected children with CF (P < 0.001) and noninfected children with CF (P < 0.001).
Figure 3. Lung clearance index (LCI) and airway infection in children with cystic fibrosis (CF). There was no association between LCI and airway infection in CF children (P = 0.083). The dashed line represents the upper limit of normality for healthy control subjects (LCI = 7.41). Bars represent the mean (noninfected, 7.02 vs. infected, 7.54).
[More] [Minimize]
Figure 4. Association between lung clearance index (LCI) and bronchoalveolar lavage (BAL) pathogen density in children with cystic fibrosis (CF). There was a weak relationship between LCI and pathogen load in young children with CF (R2 = 0.10, P = 0.031).
[More] [Minimize]
Figure 5. Lung clearance index (LCI) in healthy versus noninfected and infected children with cystic fibrosis (CF). For healthy children versus noninfected children with CF, LCI = 6.45 (0.49) versus 7.02 (0.51) (P < 0.001). For noninfected children with CF versus infected children with CF, LCI = 7.02 (0.51) versus 7.54 (1.10) (P = 0.083). For healthy children versus infected children with CF, LCI = 6.45 (0.49) versus 7.54 (1.10) (P = 0.001). The dashed line represents the upper limit of normality for LCI = 7.41. Horizontal bars represent the mean. Data based on Kruskal-Wallis analysis.
[More] [Minimize]Figures 6 and 7 demonstrate the significant relationships between LCI, IL-8, and airway neutrophils. Bronchoalveolar lavage IL-8 levels explained 20% of the variability in LCI (P = 0.004). The absolute neutrophil count explained 21% of the variability in LCI (P = 0.001).
Figure 6. Association between bronchoalveolar lavage (BAL) IL-8 and lung clearance index (LCI) in children with cystic fibrosis (CF). There was a significant relationship between LCI and IL-8 in young children with CF (R2 = 0.20, P = 0.004).
[More] [Minimize]
Figure 7. Association between bronchoalveolar lavage (BAL) absolute neutrophil count and lung clearance index (LCI) in children with cystic fibrosis (CF). There was a significant relationship between LCI and absolute neutrophil count in infants and young children with CF (R2 = 0.21, P = 0.001).
[More] [Minimize]P. aeruginosa of any colony density was isolated from 10 children, of whom 7 (15%) had growths of at least 105 cfu/ml. In the three children with low P. aeruginosa colony counts (101–105 cfu/ml), BAL neutrophil percentage counts were 42, 43, and 81%, respectively, indicating a vigorous neutrophilic response even with low bacterial loads of this organism. There was no age difference between children with P. aeruginosa (mean [SD] age, 1.75 [0.84] yr) and children without P. aeruginosa (mean age, 1.50 [0.74] yr) (P = 0.369). The youngest child with P. aeruginosa was 4 months old and one child had the mucoid phenotype.
As we would treat the first or early isolation of any growth of P. aeruginosa to prevent chronic infection, we examined the effect of P. aeruginosa in any colony count on airway inflammation. Children with P. aeruginosa had significantly increased inflammatory markers compared with children with other, nonpseudomonal pathogens (Table 3).
| No PsA Isolated (n = 37) | PsA ≥ 101 cfu/ml (n = 10) | P Value | |
| Age (yr), mean (SD) | 1.50 (0.74) | 1.75 (0.84) | 0.369 |
| Airway inflammation (BAL) | |||
| Neutrophils %, median (IQR) | 25 (8–47) | 77 (45–91) | <0.001 |
| Neutrophil count,* median (IQR) | 38 (9–144) | 375 (45–2,160) | 0.009 |
| Total cell count,* median (IQR) | 218 (134–490) | 470 (92–2,366) | 0.219 |
| IL-8 (pg/ml), median (IQR) | 1,435 (387–3,671) | 3,940 (2,234–7,026) | 0.012 |
| NE/α1-PI complex,† mean (SD) | 125 (100) | 174 (158) | 0.368 |
| Lung clearance index | |||
| Mean (SD)‡ | 7.02 (0.56) | 7.92 (1.16) | 0.038 |
| Median (IQR) | 6.94 (6.53–7.38) | 7.63 (7.24–7.93) | 0.002 |
| AUC | 95% CI | P Value | |
| BAL neutrophils % | 0.874 | 0.766-0.982 | 0.001 |
| LCI | 0.819 | 0.686-0.951 | 0.004 |
| BAL IL-8 | 0.774 | 0.609-0.940 | 0.014 |
Furthermore, in children with CF, the LCI was significantly higher in those with P. aeruginosa than in subjects without this pathogen: 7.92 (1.16) versus 7.02 (0.56) (P = 0.038) (Figure 8). ROC curve analysis compared the discriminative ability of the LCI and BAL markers of inflammation to detect any growth of P. aeruginosa (Table 4 and Figure 9). LCI had similar ability to detect P. aeruginosa as increased BAL neutrophils and IL-8 levels.
Figure 8. Lung clearance index (LCI) in children with cystic fibrosis (CF) with and without P. aeruginosa (any growth). For children without P. aeruginosa, median LCI = 6.94 (interquartile range, 6.53–7.38) versus children with P. aeruginosa, LCI = 7.63 (7.24–7.93) (P = 0.002). This relationship remained significant when the outlier was removed (P = 0.006). The dashed line represents the upper limit of normality for LCI, 7.41.
[More] [Minimize]
Figure 9. Receiver operating characteristic (ROC) curves for lung clearance index (LCI), bronchoalveolar lavage (BAL) neutrophils (%), and BAL IL-8 for the detection of P. aeruginosa (any growth) in children with cystic fibrosis (CF).
[More] [Minimize]On the basis of ROC curve analysis, a BAL IL-8 level of 672 pg/ml had the best combination of sensitivity and specificity to detect the presence of any colony count of P. aeruginosa (sensitivity, 100%; specificity, 40%). Similarly, the value of 40% neutrophils had the best combination of sensitivity (100%) and specificity (74%) for detecting any growth of P. aeruginosa. An LCI of at least 7.41 (the upper limit of normality) had a sensitivity of 67%, a specificity of 80%, and a positive predictive value of 47% to detect P. aeruginosa whereas an LCI less than 7.41 had a negative predictive value of 93%.
To test the hypothesis that the ability of the LCI to detect P. aeruginosa was related to this organism's greater potential to induce inflammation, children with CF were assigned to three groups on the basis of their culture results (Table 5). Group 1 was composed of children with “nil growth/growth of commensals”; group 2 consisted of children from whom a pathogen was isolated, multiple or otherwise, in any colony count but excluding P. aeruginosa; and group 3 was composed of children from whom P. aeruginosa was isolated (Table 5). Neutrophil percentage (72 [22]% vs. 42 [29]%; P < 0.05) and LCI (8.0 [1.1] vs. 7.0 [0.6]; P < 0.01) were significantly higher in infants and children with P. aeruginosa compared with infected children without P. aeruginosa, suggesting that isolation of P. aeruginosa is associated with worse inflammation and greater ventilation inhomogeneity.
| Group 1 (n = 18): BAL Negative | Group 2 (n = 17): BAL Pathogens, PsA Negative | Group 3 (n = 10): BAL PsA Positive | P Value | |
| IL-8 (pg/ml), median (IQR) | 598 (291–1,519) | 2,427 (851–4,959)* | 3,940 (2,234–7,026)†‡ | 0.003 |
| Neutrophils % | 14 (12) | 42 (29)§ | 72 (22)§‖ | <0.001 |
| ANC (× 103/ml), median (IQR) | 13 (6–35) | 159 (28–628)§ | 375 (45–2,160)§‡ | <0.001 |
| TCC (× 103/ml), median (IQR) | 0.2 (0.1–0.3) | 0.4 (0.2–1.3)¶ | 0.5 (0.1–2.4)¶‡ | 0.079 |
| NE/α1-PI | 74 (78) | 148 (103)¶ | 174 (158)¶‡ | 0.703 |
| LCI | 6.9 (0.6) | 7.0 (0.6)¶ | 8.0 (1.1)†** | 0.004 |
The results of the present study demonstrate important relationships between airway infection, inflammation, and ventilation inhomogeneity assessed by BAL and LCI in young children with CF. The LCI was elevated in presymptomatic/minimally symptomatic newborn-screened infants and young children with CF, especially in the presence of airway inflammation and P. aeruginosa. LCI measurements were repeatable both in healthy non-CF infants and young children and in similar-aged children with CF, using a portable MBW system in a clinical setting. This study highlights the feasibility, reproducibility, and sensitivity of the LCI as a noninvasive measure of small airway function and marker of early lung disease in children with CF.
We established a mean (SD) LCI value of 6.45 (0.49) with an ULN of 7.41 for healthy children aged 0.32 to 3.24 years, using a mainstream ultrasonic flowmeter. This is consistent with the literature reporting LCI for healthy preschool children (19) and healthy school-aged children assessed by mass spectrometry (40); healthy children assessed with a sidestream ultrasonic flowmeter (41); and healthy children and adults assessed with a photoacoustic analyzer (42) (Table 6). As such it validates the use of our non-CF group as healthy control subjects and confirms the constancy of LCI values across the age spectrum in healthy individuals whether assessed by mass spectrometry or photoacoustic gas analysis, or with an ultrasonic flowmeter.
| Study | Reference No. | Subjects | Age | MBW Equipment | LCI: Mean (SD) | LCI Mean Difference (95% CI) | Children with CF with a High LCI (%) |
| Current study | 25 healthy | 0.32–3.24 yr | USFM mainstream | 6.45 (0.49) | 0.76 (0.40–1.11) | 32 | |
| 47 CF (screened) | 0.36–3.10 yr | 7.21 (0.81) | |||||
| Lum et al. (2007) | 20 | 21 healthy | 15.3–77.9 wk | MS | 7.20 (0.30) | 1.20 (0.70–1.70) | 56.4 |
| 39 CF (nonscreened) | 7.6–94.1 wk | 8.40 (1.50) | |||||
| Aurora et al. (2005) | 19 | 30 healthy | 4.31 (0.84) yr | MS | 6.89 (0.44) | 2.72 (1.90–3.54) | 73 |
| 30 CF (nonscreened) | 4.43 (0.77) yr | 9.61 (2.19) | |||||
| Gustafsson et al. (2003) | 38 | 28 healthy | 4.5–18.7 yr | MS | 6.33 (0.43) | 2.00 (1.06–2.95) | 63 |
| 43 CF (nonscreened) | 3.0–18.2 yr | 8.33 (2.48) | |||||
| Fuchs et al. (2008) | 46 | 22 healthy | 6.8–18.9 yr | USFM sidestream | 6.70 (0.50) | 3.5 (2.36–4.68) | 77 |
| 26 CF (nonscreened) | 4.7–17.6 yr | 10.20 (2.80) | |||||
| Aurora et al. (2004) | 40 | 33 healthy | 11.3 (3.1) yr | MS | 6.45 (0.49) | 5.08 (4.07–6.10) | 95 |
| 22 CF (nonscreened) | 11.5 (3.2) yr | 11.53 (2.86) | |||||
| Horsley et al. (2008) | 42 | 12 healthy children | 6–16 yr | P | 6.30 (0.50) | Not stated | Not stated |
| 48 healthy adults | 19–58 yr | 6.70 (0.40) | |||||
| 33 adult CF | 17–49 yr | 13.10 (3.80) |
We found a negative relationship between the LCI and age in healthy, non-CF infants and young children. A longitudinal study assessing ventilation inhomogeneity in preterm and term healthy control infants monitored from the newborn period to 15–18 months of age also found a significant decrease in LCI between these two time points (43). During early infancy rapid alveolarization is associated with dysanaptic growth, that is the growth and development of the alveoli is greater than the growth of the airway, and this physiological process may contribute to greater ventilation inhomogeneity during this period (43). The lack of a similar fall in LCI during infancy in children with CF may reflect persisting ventilation inhomogeneity due to evolving early airway disease. Larger studies of ventilation inhomogeneity in healthy infants and young children assessed longitudinally from birth through to 2–3 years may be needed to confirm our observations and precisely define the normal range for LCI in very young subjects, thereby clarifying the relationship between LCI and normal postnatal lung growth.
This study has also shown that the LCI is a highly repeatable measure of lung function in young children. We demonstrated a within-test CVLCI less than 5% in both healthy subjects and children with CF. This agrees with reports of LCI within-test repeatability in preschool children with CF (19), older children (40), and adults with CF (42) and studies using different methods of inert gas analysis (see Table E5). For example, the mean (SD) CVLCI was 7.8 (5.4)% in awake children with CF, aged 2–6 years measured by mass spectrometry (MS) (19); 6.2 (2.9)% in CF children aged 6–16 years, also measured by MS (40); and 4.4 (2.8)% in adult subjects with CF, aged 17–49 years, measured with a photoacoustic gas analyzer (42). We found no relationship between age and CVLCI, a finding also reported by Aurora and colleagues, who examined school-aged children by MS (40).
The short-term (over 5–10 min) reproducibility between mean LCI measurements in set 1 and set 2, reported in the current research, has not been previously determined in children with CF of this age group. This study found a 5- to 10-minute between-test reproducibility of approximately ±0.60 unit in both healthy children and children with CF. The latter is an important finding as the minimal clinically important difference for the LCI is not known for young children with mild disease, and therefore our data potentially define the physiological variation in LCI for this age group. We did not assess the day-to-day LCI reproducibility as we could not justify repeat sedation for multiple testing occasions over short time periods.
No major adverse events were related to either BAL or MBW testing. Safety of sedation for ILF testing has been long established (44). However, one child with unsuspected upper airway obstruction developed intermittent oxygen desaturation requiring low-flow oxygen. This child subsequently underwent adenotonsillectomy. Similarly, the procedure of bronchoscopy/BAL performed under general anesthesia is generally safe, well tolerated (45), and acceptable to parents.
Important strengths of this study were that children with CF were identified by newborn screening and had been segregated from birth into cohort groups according to infection status determined by annual BAL. This is the first study to demonstrate an elevated LCI in clinically well infants and young children with CF diagnosed early though newborn screening. Both BAL and LCI were performed as elective procedures when the child was well with no clinical evidence of a respiratory exacerbation. The LCI was elevated in 32% of children despite early diagnosis (mean age, 3.8 wk), normal nutrition, regular clinical assessment, and absence of respiratory symptoms, highlighting the subclinical onset of early CF lung disease. In the only other study that has assessed the LCI in infants with CF, subjects were diagnosed after symptomatic presentation and were lighter and shorter than their healthy peers, indicating clinically established disease (20).
In this study, proportionally fewer children with CF had an abnormal LCI and the difference in LCI between children with CF and healthy control subjects was modest compared with values reported in successively older CF cohorts (Table 6) (19, 38, 40, 42, 46). This suggests that the LCI reflects disease progression as well as sensitively identifying early lung disease, although longitudinal studies are required to confirm this.
A major strength of this study was the assessment of the LCI in young children with CF, in whom concurrent lower respiratory infection and inflammation were quantitatively determined. In addition, our three-lobe BAL method, which included the right upper lobe and preceded topical anesthesia, may have optimized the assessment of both airway infection and inflammation.
The results of the present study demonstrate a significant infection burden in non/minimally symptomatic, well-nourished, screened infants and young children with CF assessed electively. Airway infection was present in 36% of children with CF. P. aeruginosa infection was present in 15% and polymicrobial growth was present in 28%. S. aureus detection was infrequent, which, given the high adherence to antistaphylococcal prophylaxis, supports the preventive role of this strategy for early infection.
Clinical parameters such as symptoms of cough and wheeze in the month before bronchoscopy or previous admission for a respiratory exacerbation were not associated with infection or an abnormal LCI. Conversely, 24% of children with demonstrated airway infection and 60% of children with an abnormal LCI had no cough or wheeze in the 12 months before their BAL, underscoring the lack of sensitivity of symptoms in the detection of early CF lung disease.
The strong association between airway infection and inflammation has been previously reported (1, 37) and is confirmed in this study. However, a novel finding was that undiagnosed polymicrobial growth, present in almost one-third of children, was associated with significantly greater airway inflammation.
In this study the LCI was not higher in children with CF and airway infection. However, the LCI was positively correlated with pathogen density, suggesting that the LCI does reflect the impact of early infection.
Perhaps the most important finding was the strong association between the LCI and the presence of P. aeruginosa in the lower airways, with a mean difference of 1.03 units between young children with and without this organism. Early P. aeruginosa isolation may be asymptomatic yet associated with airway inflammation and structural disease: its initial low colony density and nonmucoid status may present a window of opportunity for eradication. Therefore its isolation in any amount has clinical relevance, as an eradication protocol would most likely be initiated (47). Hence we evaluated the results of all 10 children with P. aeruginosa irrespective of pathogen load.
Although the sensitivity of the LCI to detect P. aeruginosa was modest, its high negative predictive value (93%) suggests that the LCI has the potential to rule out P. aeruginosa in the majority of well young children with CF. In comparison, a study using the raised-volume rapid thoracoabdominal compression technique was unable to detect lower baseline lung function in the presence of P. aeruginosa despite a more rapid decline in FEV0.5 (forced expiratory volume in 0.5 second) at follow-up (16).
We found that 68% of well young children with CF at elective BAL had airway neutrophil levels that exceeded values reported in healthy young children (48, 49). Airway neutrophils were significantly higher in infected children, in children with two or more pathogens, and in children with P. aeruginosa. The latter finding suggests that this organism has greater pathogenic potential for inducing airway inflammation.
This study demonstrated significant relationships between the LCI and airway inflammation. Higher LCI values were associated with increased airway neutrophils and IL-8 levels, a finding that has not previously been reported. There are inconsistent reports in the literature regarding the association between markers of airway inflammation and lung function abnormalities in infants with CF. Brennan and colleagues (50) demonstrated significant relationships between parenchymal hysteresivity and tissue damping and neutrophilic inflammation, and Dakin and colleagues (37) reported a significant association between specific respiratory system compliance and both IL-8 and percentage neutrophils. Pillarisetti and colleagues (16) reported an association between neutrophil elastase and FVC and FEV0.5, but not FEF75 (forced expiratory flow after expiration of 75% of vital capacity), and no association between any of these lung function parameters and IL-8, IL-1, or total cell count. Similarly, both Nixon and colleagues (51) and Linnane and colleagues (52) were unable to demonstrate an association between measures of forced expiration and airway inflammation.
Our data suggest that the LCI may be more sensitive than timed expiratory flows to detect early inflammatory disease. It is also possible that our three-lobe lavage technique may have enhanced our ability to detect inflammation. In addition, sampling each lobe once and then pooling the resultant lavage fluid is likely to produce a more bronchial sample with higher neutrophil counts (13). This practice is used by some (37, 51, 53) but not all centers (16) performing surveillance BAL, making comparisons challenging.
We recognize that this study has limitations, which include the lack of a robust measure of structural lung disease. The two commonly used CF-specific chest X-ray scoring systems did not distinguish between children with infection and those without infection. High-resolution CT sensitively detects early structural lung change in infants with CF (2, 17). It has been suggested that the LCI and HRCT have comparable sensitivity to detect lung disease in older (6–10 yr), nonscreened children with CF (18). However, Hall and colleagues (54) reported that the LCI did not relate to the presence of bronchiectasis in infants with CF, suggesting that the LCI may detect early infection and inflammation sensitively but not the onset of structural lung disease.
We acknowledge that our cutoff for airway infection (≥105 cfu/ml BAL fluid) is contentious and that other BAL studies have used lower diagnostic thresholds including 104 cfu/ml or more (17), and 103 cfu/ml or more (53). We chose this threshold as it is commonly used (1, 37, 55), consensus endorsed (12, 13), and is based on the marked increase in BAL IL-8 concentrations seen at this pathogen load, indicating a significant host response (5, 55).
Furthermore, the concept of lower airway sterility in healthy individuals has been questioned with the detection, using sensitive molecular techniques, of a lung microbiome that is indistinguishable from upper respiratory flora except in biomass (lower) (56). Whether this is a transient colonization or a normal microbiological population of the lower airways is unknown. In addition, it is recognized that the procedure of BAL in itself is not free of contamination risk as the bronchoscope passes through the upper airway, with its myriad of organisms, including common CF respiratory pathogens such as H. influenzae and S. aureus. Therefore defining a “threshold” colony count that constitutes lower airway infection or clinically relevant disease in CF remains problematic, especially in the current era of early disease surveillance. For this reason we considered the effect of any growth of P. aeruginosa on airway inflammation and ventilation inhomogeneity.
The role of BAL in young children with CF unable to expectorate has been questioned as a result of the findings of a randomized controlled trial of Australian newborn-screened infants in whom BAL-directed therapy did not improve medium-term outcomes (53). In this multicenter study BAL was performed when the child was unwell or had identified P. aeruginosa from oropharyngeal cultures. This design was different from our own study and the results cannot be extrapolated to surveillance BAL performed electively. We acknowledge that, to date, there are no data to demonstrate that our approach of annual surveillance BAL leads to improved outcomes. However, although surveillance BAL, particularly three-lobe sampling, has not been shown to alter outcomes, we believe that this technique is valuable for identifying clinically inapparent infection and assessing occult lower airway inflammation.
We have provided data demonstrating that the LCI is a feasible, repeatable, and sensitive noninvasive marker of early lung disease in well newborn-screened young children with CF. Reproducible measurements of the LCI were achievable in sedated infants and young children, using a portable MBW system for measurement at the bedside in a pediatric medical procedures unit. We acknowledge that our results represent a single time point in the clinical course of each child and that a longitudinal assessment of the LCI with respect to changes in airway infection and inflammation is required to confirm the role of the LCI as an early marker of CF lung disease. However, our study emphasizes that subclinical infection, inflammation, and functional lung disease can be evaluated through sensitive techniques such as BAL and MBW in infants and young children with CF, and our repeatability data highlight the potential role of the LCI as an outcome measure for future intervention trials involving infants and young children with CF.
The authors thank the children and their families for their participation in the study, Dr. Linda Xu for cytokine analysis, and Ms. Rhonda Bell for patient recruitment. The authors also thank Professor Janet Stocks, Dr. Sooky Lum, Dr. Jane Pillow, Dr. Andreas Schibler, and especially A/Professor Graham Hall for their support in establishing ILFT at Sydney Children's Hospital, Randwick and the Sydney Children's Hospital Foundation whose generosity made this study possible.
| 1. | Armstrong DS, Grimwood K, Carzino R, Carlin JB, Olinsky A, Phelan PD. Lower respiratory infection and inflammation in infants with newly diagnosed cystic fibrosis. BMJ 1995;310:1571–1572. |
| 2. | Stick SM, Brennan S, Murray C, Douglas T, Ungern-Sternberg BS, Garratt LW, Gangell CL, de Klerk N, Linnane B, Ranganathan S, Robinson P, Robertson C, Sly PD, and Australian Respiratory Early Surveillance Team for Cystic Fibrosis(AREST CF). Bronchiectasis in infants and preschool children diagnosed with cystic fibrosis after newborn screening. J Pediatr 2009;155:623–628. |
| 3. | Sly PD, Brennan S. Detecting early lung disease in cystic fibrosis: are current techniques sufficient? Thorax 2004;59:1008–1010. |
| 4. | Stick SM, Sly PD. Exciting new clinical trials in cystic fibrosis: infants need not apply. Am J Respir Crit Care Med 2011;183:1577–1578. |
| 5. | Armstrong DS, Grimwood K, Carlin JB, Carzino R, Olinsky A, Phelan PD. Bronchoalveolar lavage or oropharyngeal cultures to identify lower respiratory pathogens in infants with cystic fibrosis. Pediatr Pulmonol 1996;21:267–275. |
| 6. | Rosenfeld M, Emerson J, Accurso F, Armstrong D, Castile R, Grimwood K, Hiatt P, McCoy K, McNamara S, Ramsey B, et al.. Diagnostic accuracy of oropharyngeal cultures in infants and young children with cystic fibrosis. Pediatr Pulmonol 1999;28:321–328. |
| 7. | Brody AS, Molina PL, Klein JS, Rothman BS, Ramagopal M, Swartz DR. High-resolution computed tomography of the chest in children with cystic fibrosis: support for use as an outcome surrogate. Pediatr Radiol 1999;29:731–735. |
| 8. | Douglas TA, Brennan S, Berry L, Winfield K, Wainwright CE, Grimwood K, Stick SM, Sly PD; on behalf of the members of AREST CF and the ACFBAL Trial. Value of serology in predicting Pseudomonas aeruginosa infection in young children with cystic fibrosis. Thorax 2010;65:985–990. |
| 9. | Davis SD, Brody AS, Emond MJ, Brumback LC, Rosenfeld M. Endpoints for clinical trials in young children with cystic fibrosis. Proc Am Thorac Soc 2007;4:418–430. |
| 10. | Hall GL, Stocks J. Intervention trials and ventilation distribution in mild cystic fibrosis lung disease: will it all come out in the wash? Eur Respir J 2011;37:757–759. |
| 11. | Sly PD. Early detection of lung disease in CF: do we have the necessary techniques? Paediatr Respir Rev 2008;9:149–150. |
| 12. | Brennan S, Gangell C, Wainwright C, Sly PD. Disease surveillance using bronchoalveolar lavage [review]. Paediatr Respir Rev 2008;9:151–159. |
| 13. | de Blic J, Midulla F, Barbato A, Clement A, Dab I, Eber E, Green C, Grigg J, Kotecha S, Kurland G, et al..; ERS Task Force on bronchoalveolar lavage in children. European Respiratory Society. Bronchoalveolar lavage in children [review]. Eur Respir J 2000;15:217–231. |
| 14. | Gilchrist FJ, Salamat S, Clayton S, Peach J, Alexander J, Lenney W. Bronchoalveolar lavage in children with cystic fibrosis: how many lobes should be sampled? Arch Dis Child 2011;96:215–217. |
| 15. | Meyer KC, Sharma A. Regional variability of lung inflammation in cystic fibrosis. Am J Respir Crit Care Med 1997;156:1536–1540. |
| 16. | Pillarisetti N, Williamson E, Linnane B, Skoric B, Robertson CF, Robinson P, Massie J, Hall GL, Sly P, Stick S, et al.. Infection, inflammation, and lung function decline in infants with cystic fibrosis. Am J Respir Crit Care Med 2011;184:75–81. |
| 17. | Sly PD, Brennan S, Gangell C, de Klerk N, Murray C, Mott L, Stick SM, Robinson PJ, Robertson CF, Ranganathan SC; Australian Respiratory Early Surveillance Team for Cystic Fibrosis (AREST-CF). Lung disease at diagnosis in infants with cystic fibrosis detected by newborn screening. Am J Respir Crit Care Med 2009;180:146–152. |
| 18. | Owens CM, Aurora P, Stanojevic S, Bush A, Wade A, Oliver C, Calder A, Price J, Carr SB, Shankar A, et al.. Lung clearance index and HRCT are complementary markers of lung abnormalities in young children with CF. Thorax 2011;66:481–488. |
| 19. | Aurora P, Bush A, Gustafsson P, Oliver C, Wallis C, Price J, Stroobant J, Carr S, Stocks J, London Cystic FC. Multiple-breath washout as a marker of lung disease in preschool children with cystic fibrosis. Am J Respir Crit Care Med 2005;171:249–256. |
| 20. | Lum S, Gustafsson P, Ljungberg H, Hulskamp G, Bush A, Carr SB, Castle R, Hoo AF, Price J, Ranganathan S, et al.. Early detection of cystic fibrosis lung disease: multiple-breath washout versus raised volume tests. Thorax 2007;62:341–347. |
| 21. | Ranganathan SC, Dezateux C, Bush A, Carr SB, Castle RA, Madge S, Price J, Stroobant J, Wade A, Wallis C, et al..; London Collaborative Cystic Fibrosis Group. Airway function in infants newly diagnosed with cystic fibrosis. Lancet 2001;358:1964–1965. |
| 22. | Ranganathan SC, Parsons F, Gangell C, Brennan S, Stick SM, Sly PD. Evolution of pulmonary inflammation and nutritional status in infants and young children with cystic fibrosis. Thorax 2011;66:408–413. |
| 23. | Davis SD, Rosenfeld M, Kerby GS, Brumback L, Kloster MH, Acton JD, Colin AA, Conrad CK, Hart MA, Hiatt PW, et al.. Multicenter evaluation of infant lung function tests as cystic fibrosis clinical trial endpoints. Am J Respir Crit Care Med 2010;182:1387–1397. |
| 24. | Aurora P. Multiple-breath washout in preschool children—FRC and ventilation inhomogeneity. Paediatr Respir Rev 2006;7:S14–S16. |
| 25. | Belessis Y, O'Donovan B, Numa AH, Morton JR, Lui K, Jaffe A. The lung clearance index measured by an ultrasonic flowmeter detects early lung disease in infants and young children with cystic fibrosis identified by newborn screening [abstract]. Pediatr Pulmonol Suppl 2006;29:331. |
| 26. | Belessis Y, O'Donovan B, Numa AH, Hawkins G, Xu L, Lui K, Morton JR, Jaffe A. Early detection of cystic fibrosis lung disease: lung clearance index predicts bronchoalveolar lavage evidence of infection and inflammation [abstract]. Respirology 2009;14:A15. |
| 27. | American Academy of Pediatrics Committee on Drugs and Committee on Environmental Health. Use of chloral hydrate for sedation in children. Pediatrics 1993;92:471–473. |
| 28. | Gaultier C, Fletcher ME, Beardsmore C, England S, Motoyama E; Working Group of the European Respiratory Society and the American Thoracic Society. Respiratory function measurements in infants: measurement conditions. Eur Respir J 1995;8:1057–1066. |
| 29. | Beydon N, Davis SD, Lombardi E, Allen JL, Arets HG, Aurora P, Bisgaard H, Davis GM, Ducharme FM, Eigen H, et al.. An official American Thoracic Society/European Respiratory Society statement: pulmonary function testing in preschool children. Am J Respir Crit Care Med 2007;175:1304–1345. |
| 30. | Brasfield D, Hicks G, Soong S, Peters J, Tiller R. Evaluation of scoring system of the chest radiograph in cystic fibrosis: a collaborative study. AJR Am J Roentgenol 1980;134:1195–1198. |
| 31. | Chrispin AR, Norman AP. The systematic evaluation of the chest radiograph in cystic fibrosis. Pediatr Radiol 1974;2:101–105. |
| 32. | Farrell PM, Li Z, Kosorok MR, Laxova A, Green CG, Collins J, Lai HC, Makholm LM, Rock MJ, Splaingard ML. Longitudinal evaluation of bronchopulmonary disease in children with cystic fibrosis. Pediatr Pulmonol 2003;36:230–240. |
| 33. | Terheggen-Lagro SWJ, Arets HGM, van der Laag J, van der Ent CK. Radiological and functional changes over 3 years in young children with cystic fibrosis. Eur Respir J 2007;30:279–285. |
| 34. | Davis SD, Fordham LA, Brody AS, Noah TL, Retsch-Bogart GZ, Qaqish BF, Yankaskas BC, Johnson RC, Leigh MW. Computed tomography reflects lower airway inflammation and tracks changes in early cystic fibrosis. Am J Respir Crit Care Med 2007;175:943–950. |
| 35. | Chandan SS, Faoagali J, Wainwright CE. Sensitivity of respiratory bacteria to lignocaine. Pathology 2005;37:305–307. |
| 36. | Olsen KM, Peddicord TE, Campbell GD, Rupp ME. Antimicrobial effects of lidocaine in bronchoalveolar lavage fluid. J Antimicrob Chemother 2000;45:217–219. |
| 37. | Dakin CJ, Numa AH, Wang H, Morton JR, Vertzyas CC, Henry RL. Inflammation, infection, and pulmonary function in infants and young children with cystic fibrosis. Am J Respir Crit Care Med 2002;165:904–910. |
| 38. | Gustafsson PM, Aurora P, Lindblad A. Evaluation of ventilation maldistribution as an early indicator of lung disease in children with cystic fibrosis. Eur Respir J 2003;22:972–979. |
| 39. | Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–310. |
| 40. | Aurora P, Gustafsson P, Bush A, Lindblad A, Oliver C, Wallis CE, Stocks J. Multiple breath inert gas washout as a measure of ventilation distribution in children with cystic fibrosis. Thorax 2004;59:1068–1073. |
| 41. | Fuchs SI, Eder J, Ellemunter H, Gappa M. Lung clearance index: normal values, repeatability, and reproducibility in healthy children and adolescents. Pediatr Pulmonol 2009;44:1180–1185. |
| 42. | Horsley AR, Gustafsson PM, Macleod KA, Saunders C, Greening AP, Porteous DJ, Davies JC, Cunningham S, Alton EW, Innes JA. Lung clearance index is a sensitive, repeatable and practical measure of airways disease in adults with cystic fibrosis. Thorax 2008;63:135–140. |
| 43. | Schulzke SM, Hall GL, Nathan EA, Simmer K, Nolan G, Pillow JJ. Lung volume and ventilation inhomogeneity in preterm infants at 15–18 months corrected age. J Pediatr 2009;156:542–549.e2. |
| 44. | Stocks J, Sly P, Tepper RS, Morgan W. Infant respiratory function testing. New York: Wiley-Liss, 1996. |
| 45. | Wainwright CE, Grimwood K, Carlin JB, Vidmar S, Cooper PJ, Francis PW, Byrnes CA, Whitehead BF, Martin AJ, Robertson IF, et al.. Safety of bronchoalveolar lavage in young children with cystic fibrosis. Pediatr Pulmonol 2008;43:965–972. |
| 46. | Fuchs SI, Sturz J, Junge S, Ballmann M, Gappa M. A novel sidestream ultrasonic flow sensor for multiple breath washout in children. Pediatr Pulmonol 2008;43:731–738. |
| 47. | 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. |
| 48. | Midulla F, Villani A, Merolla R, Bjermer L, Sandstrom T, Ronchetti R. Bronchoalveolar lavage studies in children without parenchymal lung disease: cellular constituents and protein levels. Pediatr Pulmonol 1995;20:112–118. |
| 49. | Ratjen F, Bredendiek M, Brendel M, Meltzer J, Costabel U. Differential cytology of bronchoalveolar lavage fluid in normal children. Eur Respir J 1994;7:1865–1870. |
| 50. | Brennan S, Hall GL, Horak F, Moeller A, Pitrez PM, Franzmann A, Turner S, de Klerk N, Franklin P, Winfield KR, et al.. Correlation of forced oscillation technique in preschool children with cystic fibrosis with pulmonary inflammation. Thorax 2005;60:159–163. |
| 51. | Nixon GM, Armstrong DS, Carzino R, Carlin JB, Olinsky A, Robertson CF, Grimwood K, Wainwright C. Early airway infection, inflammation, and lung function in cystic fibrosis. Arch Dis Child 2002;87:306–311. |
| 52. | Linnane BM, Hall GL, Nolan G, Brennan S, Stick SM, Sly PD, Robertson CF, Robinson PJ, Franklin PJ, Turner SW, Ranganathan SC; AREST-CF. Lung function in infants with cystic fibrosis diagnosed by newborn screening. Am J Respir Crit Care Med 2008;178:1238–1244. |
| 53. | Wainwright CE, Vidmar S, Armstrong DS, Byrnes CA, Carlin JB, Cheney J, Cooper PJ, Grimwood K, Moodie M, Robertson CF, et al.. Effect of bronchoalveolar lavage directed therapy on Pseudomonas aeruginosa infection and structural lung injury in children with cystic fibrosis. JAMA 2011;306:163–171. |
| 54. | Hall GL, Logie KM, Parsons F, Schulzke SM, Nolan G, Murray C, Ranganathan SC, Robinson P, Sly P, Stick S. Air trapping on chest CT is associated with worse ventilation distribution in infants with cystic fibrosis diagnosed following newborn screening. PLoS One 2011;6:e23932. |
| 55. | Khan TZ, Wagener JS, Bost T, Martinez J, Accurso FJ, Riches DW. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995;151:1075–1082. |
| 56. | Charlson ES, Bittinger K, Haas AR, Fitzgerald AS, Frank I, Yadav A, Bushman FD, Collman RG. Topographical continuity of bacterial populations in the healthy human respiratory tract. Am J Respir Crit Care Med 2011;184:957–963. |
Author Contributions: Conception and design—Y.B., J.M., K.L., A.N.; performance of bronchoscopies—Y.B.; performance of multiple-breath washout testing—Y.B., B.D., R.Mac.D., A.N.; data collection and interpretation—Y.B., B.D., R.Mac.D., G.H., P.F., J.P., J.P., A.N., A.J., K.L.; drafting the manuscript—Y.B.; contribution to final draft and important intellectual content—Y.B., A.J., P.F., K.L.
This article has an online supplement, which is available from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201109-1631OC on February 9, 2012
Author disclosures
