Lung Clearance Index Is a Repeatable and Sensitive ... - ATS Journals

5 downloads 7374 Views 608KB Size Report
Mar 1, 2014 - performed with custom software. Two ... software were compared and the same ..... A, Price J, Carr SB, Shankar A, et al.; London Cystic Fibrosis.
ORIGINAL ARTICLE Lung Clearance Index Is a Repeatable and Sensitive Indicator of Radiological Changes in Bronchiectasis Stephen A. Rowan1, Judy M. Bradley2, Ian Bradbury2,3, John Lawson4, Tom Lynch4, Per Gustafsson5, Alex Horsley6, Katherine O’Neill1, Madeleine Ennis1, and J. Stuart Elborn1,4 1

Centre for Infection and Immunity, Queen’s University of Belfast, Belfast, United Kingdom; 2Centre for Health and Rehabilitation Technologies, Institute of Nursing and Health Research, University of Ulster, Newtownabbey, United Kingdom; 3Frontier Science Scotland, Kincraig, United Kingdom; 4Belfast City Hospital, Belfast, United Kingdom; 5Department of Pediatrics, Central Hospital, Skovde, ¨ Sweden; and 6Manchester Adult Cystic Fibrosis Centre, University Hospitals of South Manchester, Manchester, United Kingdom

Abstract Rationale: In bronchiectasis there is a need for improved markers

little evidence of an effect of FEV1. There were no significant associations between the lung clearance index or FEV1 and healthrelated quality of life.

of lung function to determine disease severity and response to therapy.

Conclusions: The lung clearance index is repeatable and a more

Objectives: To assess whether the lung clearance index is

a repeatable and more sensitive indicator of computed tomography (CT) scan abnormalities than spirometry in bronchiectasis. Methods: Thirty patients with stable bronchiectasis were recruited

and lung clearance index, spirometry, and health-related quality of life measures were assessed on two occasions, 2 weeks apart when stable (study 1). A separate group of 60 patients with stable bronchiectasis was studied on a single visit with the same measurements and a CT scan (study 2). Measurements and Main Results: In study 1, the intervisit intraclass correlation coefficient for the lung clearance index was 0.94 (95% confidence interval, 0.89 to 0.97; P , 0.001). In study 2, the mean age was 62 (10) years, FEV1 76.5% predicted (18.9), lung clearance index 9.1 (2.0), and total CT score 14.1 (10.2)%. The lung clearance index was abnormal in 53 of 60 patients (88%) and FEV1 was abnormal in 37 of 60 patients (62%). FEV1 negatively correlated with the lung clearance index (r = –0.51, P , 0.0001). Across CT scores, there was a relationship with the lung clearance index, with

sensitive measure than FEV1 in the detection of abnormalities demonstrated on CT scan. The lung clearance index has the potential to be a useful clinical and research tool in patients with bronchiectasis. Keywords: bronchiectasis; lung clearance index; multiple breath

washout

At a Glance Commentary Scientific Knowledge on the Subject: The lung clearance index has been shown to be a more sensitive marker of lung function, particularly in early disease, in cystic fibrosis. There are no published data on the use of the lung clearance index (LCI) in bronchiectasis. What This Study Adds to the Field: LCI offers the potential to be a more useful measure of lung function than spirometry to monitor disease and responsiveness to interventions in clinical trials.

( Received in original form October 1, 2013; accepted in final form December 30, 2013 ) Supported by a fellowship grant from the Health and Social Care Research and Development, Public Health Agency, Northern Ireland (S.A.R.). Author Contributions: S.A.R. performed data collection and manuscript preparation. This study was designed by J.S.E., J.M.B., and S.A.R. I.B. performed statistical analysis and review of data for manuscript preparation. J.L. and T.L. scored CT scans and assisted in manuscript preparation. J.M.B. and J.S.E. assisted with data interpretation and preparation of manuscript. P.G., A.H., K.O’N., and M.E. assisted in manuscript preparation. Correspondence and requests for reprints should be addressed to J. Stuart Elborn, M.D., Centre for Infection and Immunity, Queen’s University Belfast, 97 Lisburn Road, Belfast BT9 7BL, United Kingdom. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents online at www.atsjournals.org Am J Respir Crit Care Med Vol 189, Iss 5, pp 586–592, Mar 1, 2014 Copyright © 2014 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201310-1747OC on January 15, 2014 Internet address: www.atsjournals.org

586

American Journal of Respiratory and Critical Care Medicine Volume 189 Number 5 | March 1 2014

ORIGINAL ARTICLE Bronchiectasis (BE) not caused by cystic fibrosis (CF) is a debilitating illness with symptoms of recurrent cough, daily sputum production, recurrent chest infections, and a poor health-related quality of life (HRQoL) (1). The estimated prevalence of BE in the United States ranges from 4 per 100,000 young adults to almost 300 per 100,000 in those more than 75 years old (2–5). British Thoracic Society guidelines indicate that management should be focused on improving or maintaining lung function, reducing exacerbations, and improving HRQoL (1, 6). FEV1 is commonly used as the primary method of monitoring patients with BE. However, there is an increasing body of evidence that FEV1 is insensitive to early lung disease, and the majority of clinical trials in BE have been unable to demonstrate a treatment effect using FEV1 (7–9). This highlights the need for other, sensitive and responsive markers of lung function in BE (7, 10). The lung clearance index (LCI) measured by multiple breath washout (MBW) is a measure of ventilation inhomogeneity and has been shown to be a sensitive lung function test in early lung disease. Its usefulness has been demonstrated in CF, particularly in children and adults with mild disease (11–14). There are no studies published assessing LCI in BE. In this study we tested the hypothesis that LCI measured by MBW is a better predictor of structural lung changes seen on computed tomography (CT) scan compared with FEV1. The primary aim of this study was to assess whether LCI in patients with BE was (1) repeatable and (2) a more sensitive indicator of CT scan abnormalities than spirometry. Secondary aims were to explore the relationships of LCI with highresolution CT (HRCT) changes and HRQoL, and to assess the practicalities of MBW testing and patient comfort.

Methods This project consisted of two studies. Study 1 included 30 patients with BE who attended for two visits, 2 weeks apart, to assess the repeatability of LCI. The second study was a single visit including 60 patients with BE to compare the LCI measurement with CT scores performed at that visit.

Recruitment

Adults with a radiological diagnosis of BE by HRCT were recruited from the regional BE clinic at Belfast City Hospital, Belfast Health and Social Care Trust. Written informed consent was obtained. Patients were included if (1) they were 18–80 years of age; (2) a diagnosis of BE was confirmed, based on a CT report in the clinical notes, which described any evidence of BE. BE only was included. If only bronchial wall thickening was present, the patient was excluded. Traction BE was not included; (3) they were nonsmokers or ex-smokers with a less than 10–pack year smoking history; and (4) they were free of exacerbation symptoms (as defined by British Thoracic Society guidelines) for at least 4 weeks before taking part in the study (1). Patients were excluded if significant comorbidity was present (e.g., congestive cardiac failure, neoplasm). Full exclusion/inclusion criteria are given in the online supplement. In study 2, patients performed all investigations at a single visit, with the HRCT scan performed within 24 hours of each study visit. The study was approved by the Office of Research Ethics Committee for Northern Ireland (ORECNI-10/ NIR03/44).

LCI

MBW tests were performed in the seated position with a nose clip applied, using the Innocor gas analyzer (Innovision A/S, Odense, Denmark), modified as previously described to use an open circuit wash-in protocol (14). LCI was performed before spirometry in all patient visits. Three MBW tests were performed with the mean result of at least two acceptable tests recorded. A test was excluded if functional residual capacity differed by more than 10% from the median of recordings or there was evidence of a leak or irregular breathing pattern. Analysis of washout data was performed with custom software. Two trained personnel only were responsible for the accrual of data and calibration (S.A.R. and K.O’N.). Detailed methods are described further in the online supplement. After washout testing was completed, a visual analog score (VAS) was recorded to assess ease and comfort (0 representing very difficult to perform/very uncomfortable and 10 very easy to perform/very comfortable). The normal range of LCI mean (SD), 6.5 (0.5), was taken from a cohort of

Rowan, Bradley, Bradbury, et al.: Lung Clearance Index and Bronchiectasis

healthy normal control subjects (n = 30; age, 21–44 yr). LCI represents the number of lung turnovers required to wash out the inert gas from an initial to final value. The normal range in this study was taken to be 2 SD above the mean. An abnormal value was therefore greater than 7.5. Published control data collected using identical apparatus, procedures, and software were compared and the same normal range was achieved (14). Spirometry

Spirometry was performed in accordance with American Thoracic Society/European Respiratory Society guidelines on a MicroLab 3500 spirometer (CareFusion, Basingstoke, UK) (15). Percent predicted values were calculated from the all ages reference data published by Stanojevic and colleagues (16). HRCT

Expiratory and inspiratory HRCT scans of the chest were performed on a 64-slice CT scanner (Siemens AG, Munich, Germany) in the supine position. Two thoracic radiologists (T.L. and J.L.) reviewed each scan independently, using a standardized scoring method (17). This result gives a global CT score of BE and, in addition, scores each lobe on type of BE, extent, mucus plugging, air trapping, and emphysema. The complete scoring system can be seen in the online supplement. A score for each lobe was recorded, which was then converted to a total score for each of the parameters assessed: total BE score, average bronchiectatic bronchus size, degree of mucus plugging, degree of peribronchial thickening, parenchymal score, and degree of air trapping (hyperinflation score). The higher the number, the more severe the abnormality. Scores can also be converted into a percent score for comparison and to account for those who have had previous surgery. Further information on the HRCT scoring methods is found in the online supplement. HRQoL

HRQoL was assessed with the St. George’s Respiratory Questionnaire (SGRQ) (18, 19). Questionnaires were completed independently by the patient at each study visit, with an investigator present for any queries. 587

ORIGINAL ARTICLE Inflammatory Markers, Sputum, and Clinical Data

A blood sample was collected and processed in the routine hospital laboratory for white cell count and C-reactive protein. A 24-hour sputum sample was collected in a dedicated preweighed collection tube from which volume (ml) and weight (g) were recorded. Frequency of pulmonary exacerbations requiring intravenous antibiotics over the last 12 months was determined from the clinical record.

Statistics

Statistical analysis was performed with Prism (version 5.01; GraphPad Software Inc., La Jolla, CA) and SPSS (version 18 PASW Statistics 18; IBM, Armonk, NY). The definition of repeatability was taken from previously published guidance on the investigation and interpretation of repeatability (20). To assess repeatability, the Lin concordance correlation coefficient (CCC) was used to calculate sample size. To obtain 95% confidence limits for the CCC with a lower limit no smaller than 0.65, 28 patients were required. Thirty patients were recruited to allow for dropouts or technical errors. The sample size calculation to determine the relationship between HRCT changes and LCI used the correlation coefficient. For the lower limit of the 95% confidence interval (CI) of the true correlation to be no lower than 0.7, the required sample size was 54. Sixty patients were recruited to allow for dropouts or technical errors. Descriptive statistics were used to describe the population characteristics (mean [SD]). Subject demographics and characteristics were summarized using mean (SD) and frequencies. Intervisit repeatability of LCI was assessed by two methods: the intraclass correlation coefficient (ICC) and Bland–Altman test. The ICC was interpreted as follows: 0–0.2, poor agreement; 0.2–0.4, fair agreement; 0.4–0.6, moderate agreement; 0.6–0.8, strong agreement; 0.8–1.0, almost perfect agreement. Pearson’s r correlation coefficient was used for the assessment of reliability of LCI. Intravisit repeatability of triplicate washout repeats performed at each visit was expressed as the coefficient of variation of those measurements. Subjects with only two washout repeats were excluded from this analysis. 588

Associations were assessed using Pearson’s correlation coefficient between variables of interest. Sensitivity and specificity were assessed by calculating the percentage of subjects with less than the 80% predicted value for spirometry and a score for LCI greater than 11.96 SD score (i.e., .7.5). Receiver-operator characteristic (ROC) curves were used to assess sensitivity and specificity across all measures of interest. Regression models were selected on the basis of the model with the smallest area under the curve (AUC) value as this has the best predictive performance. All models included age and sex in the analysis. A paired t test was used to compare results between visits. For tertile groups, analysis of variance, using the Dunnett correction for multiple comparisons, was used. A two-tailed P value less than 0.05 was considered statistically significant.

Results Repeatability and Intravisit Reproducibility

Thirty patients were recruited (15 male) with a mean (SD) age of 56.7 (14.1) years, FEV1 (% predicted) 84.8 (20.7), and LCI 9.2 (1.8). A full outline of demographics and methods and further results for this are included in the online supplement. The intervisit ICC for LCI was 0.94 (95% CI, 0.89 to 0.97; P , 0.001). ICC between visits for FEV1 (% predicted) was 0.99 (95% CI, 0.97 to 0.99; P , 0.001) and for forced expiratory flow between 25 and 75% of FVC (FEF25–75%) (% predicted) it was 0.95 (95% CI, 0.90 to 0.98; P , 0.001). In subjects with a change in values between visits, there was no difference in bronchodilator therapy, time of airway clearance, or breathing pattern during the test to explain the variability. The coefficient of variation between LCI measurements for visit 1 was 4.4% and for visit 2 it was 4.8%. LCI versus Spirometry as a Sensitive Marker of CT Scan Abnormalities in BE

Sixty patients were recruited and all patients completed this study with no dropouts. A summary of demographic data, lung function, and HRQoL measures is given in Table 1. The majority of patients were female and the mean FEV1 (% predicted)

Table 1: Baseline Demographics and Characteristics of Study Participants in Sensitivity Study Characteristic Age, yr Male/female, no./no. Etiology of bronchiectasis ABPA CTD Immune defect IBD None Postinfectious Other FEV1, % FEV1, z-score FVC, % FVC, z-score Ratio Ratio, z-score FEF25–75% LCI, lung turnovers LCI, CV % WCC, 3 109/L CRP, mg/L 24-h sputum volume, ml 24-h sputum weight, g SGRQ Symptoms Activity Impact Total

Result 62.4 (10.7) 18/42 1.7% 5% 3.3% 5% 43.3% 38.3% 3.3% 76.5 (18.9) 21.7 (1.4) 82.3 (16.6) 21.3 (1.2) 71.5 (9.3) 20.7 (1.0) 67.9 (34.4) 9.1 (2.0) 4.0 (2.9) 7.0 (1.9) 4.5 (6.7) 11.4 (8.0) 10.8 (8.2) 57.6 45.9 27.6 38.2

(22.8) (29.1) (17.1) (19.2)

Definition of abbreviations: ABPA = allergic bronchopulmonary aspergillosis; CRP = C-reactive protein; CTD = connective tissue disorder; CV = coefficient of variation; FEF25–75% = forced expiratory flow between 25 and 75% of FVC; IBD = inflammatory bowel disease; LCI = lung clearance index; SGRQ = St. George’s Respiratory Questionnaire; WCC = white cell count. n = 60 patients. Results are presented as mean (SD), number, or %.

was 76.5, reflecting overall mild to moderate airflow obstruction. The most common etiologies for BE were idiopathic (43.3%) and postinfectious (38.3%). There was no evidence of a reader effect between the two independent CT scorers. LCI was abnormal in 53 of 60 patients (88%). On the basis of conventional upper limits of normality (.80% predicted), FEV1 was abnormal in 37 (62%) and FEF25–75% in 41 (68%). Using z-scores, abnormal FEV1 and FEF25–75% values were 26 (43%) and 15 (25%), respectively. LCI increased with declining lung function (Figure 1). An ROC analysis confirmed that LCI is more sensitive than spirometric measures when compared between patients with

American Journal of Respiratory and Critical Care Medicine Volume 189 Number 5 | March 1 2014

ORIGINAL ARTICLE were 9.3 (0.8) and 8.9 (1.2), respectively, out of a maximal score of 10. The mean time for a single LCI test (inclusive of wash-in and washout) was 7.4 minutes. There was no significant difference between visits for ease, comfort, or time of test. It was a safe procedure with no patients reporting adverse events.

Figure 1. Plot of FEV1 and forced expiratory flow between 25 and 75% of FVC (FEF25–75%) predicted scores against lung clearance index (LCI) value for each participant. Dashed lines represent limits of normality (FEV1 and FEF25–75% . 80% and LCI , 7.5). LCI versus FEV1, r = –0.51, P , 0.0001; LCI versus FEF25–75%, r = –0.58, P , 0.0001.

BE and healthy control subjects (Figure 2 and Table 2). This showed a differentiation between health and disease, and assumes that healthy control subjects have no abnormal CT findings, although this was not specifically addressed in this study. Table 3 shows the relationship between LCI and spirometric/lung function measures with CT abnormalities. With the exception of airway thickening, which was consistent across all subscales, there is clear evidence of a relationship with LCI. The addition of FEV1 to the model was nonsignificant. There was no relationship between any

measure of lung function to airway thickening; however, LCI had a stronger relationship with all scores in comparison with FEV1 and FEF25–75%. In Figure 3, the data were divided into tertiles based on the severity of BE on CT scan. This shows that LCI has a clearer gradient and less variation in comparison with FEV1. Feasibility and Acceptability

LCI by MBW has not been used in this population, and therefore we assessed the impact of this test on patient experience. VAS scores for ease and comfort of the test

Figure 2. Receiver-operator characteristic (ROC) analysis curve showing sensitivity of lung clearance index (LCI) between healthy control subjects and those with bronchiectasis. As the LCI rises with abnormality and spirometric measures fall, FEV1 and forced expiratory flow between 25 and 75% of FVC (FEF25–75%) (% predicted) were inverted to be comparable. Areas under the curve: LCI, 0.96 (P , 0.0001); FEV1, 0.82 (P , 0.0001); FEF25–75%, 0.76 (P , 0.0001).

Rowan, Bradley, Bradbury, et al.: Lung Clearance Index and Bronchiectasis

Relationship between HRQoL, Lung Function, and LCI

There was evidence of a relationship between LCI and the symptom domain of SGRQ, showing an increase in symptoms with increasing LCI (R2 = 0.18, P = 0.03). This effect appeared to level off at higher LCI values. There was no evidence of a relationship with the subscales of activity, impact, and total score.

Discussion This study is the first report of LCI in adults with BE and provides evidence that LCI is a useful measure of lung function in this common condition. LCI in this study has been shown to be repeatable between visits, and has good inter- and intravisit reproducibility. Repeatability was not affected by disease severity and compares well with data seen in CF (14). Many studies in patients with CF have shown that LCI by MBW is a more sensitive indicator of impaired lung function than FEV1 (21). In this study LCI also demonstrated improved sensitivity to abnormalities demonstrated on HRCT scan compared with FEV1 and FEF25–75% (% predicted). Although the patients studied here exhibited a wide range of airflow obstruction, 38% of them had an FEV1 within the normal range. Despite this, LCI was markedly abnormal in the majority of patients with BE (88%). The AUC data from ROC analysis confirm the improved sensitivity of LCI over FEV1. The data on repeatability of this measure compare well with the ranges of coefficient of variation (3–9%) in the European Cystic Fibrosis Society consensus document on the use of LCI in CF (22). The correlation of structural abnormalities on CT and pulmonary function tests in BE has been linked primarily to evidence of intrinsic disease of small and medium-sized airways (23). The dominant finding in patients with BE at presentation is mild or no airflow 589

ORIGINAL ARTICLE Table 2: Summary of Receiver-Operator Characteristic Statistics for Lung Clearance Index and Spirometry

Variable LCI Inverse FEV1 (100 – value) Inverse FEF25–75% (100 2 value)

Area

SE

P Value

0.96 0.82 0.76

0.02 0.04 0.05

,0.0001 ,0.0001 ,0.0001

95% Confidence Interval Lower Upper 0.93 0.74 0.67

0.99 0.90 0.86

Definition of abbreviations: FEF25–75% = forced expiratory flow between 25 and 75% of FVC; LCI = lung clearance index.

obstruction (24). In this study, LCI correlates more strongly with quantitative radiological assessment of structural changes on CT scan compared with FEV1. The possibility of using LCI as a surrogate marker of lung damage is promising, based on these data. In CF, this apparent normal lung function based on spirometry in the early stages of the disease process has been termed the “silent zone” or “silent years” of lung function decline (25, 26). Monitoring disease progression and response to interventions in these subjects requires measurements that are both sensitive and repeatable. In this regard, the enhanced sensitivity of LCI assessing ventilation inhomogeneity over FEV1 represents a significant advantage. The box plots in Figure 3 indicate that LCI is a better discriminator of severity of structural disease on CT than conventional spirometry. Further studies are needed to determine the usefulness of LCI in the diagnosis and management of people with BE. In a study assessing the role of LCI in a small group of patients (n = 21) with primary ciliary dyskinesia (PCD) and CF, the authors hypothesized that LCI and FEV1 were not related and that there was a poor relationship between LCI and

disease severity on HRCT scan, with a particular lack of sensitivity in advanced PCD (27). In contrast to CF, HRCT imaging studies have shown that a decreased attenuation pattern on expiratory CT scan is common in severe BE, and may also be detected in lobes without bronchiectatic airways visible on HRCT (28). The lack of a relationship between LCI, FEV1, and HRCT scores in patients with PCD may be explained by the population size studied. In this study, 21 patients with PCD were included, which is less than the number required to power a study for an association. In addition, when comparing the LCI values, the CF group and our population of BE had an LCI greater than 12 in about 10% of the participants. The PCD group had about 30% with an LCI greater than 12. In patients with this level of LCI, there is a loss of the linear relationship between FEV1 and LCI. This supports our observation that LCI is a more discriminatory measurement in mild to moderate disease compared with those with more severe disease. In addition, the assessment of acceptability and feasibility of LCI shows that this measure of lung function is simple to perform, with a high degree of

comfort and no major adverse effects seen in this group. Feasibility has been reported in numerous studies in CF; however, as patients with BE are older, it is important to establish feasibility and acceptability in this population (14, 29). Despite taking a longer time to perform than spirometry, ease of use and comfort scores are good. Ease of use and comfort of the test for spirometry were not measured in this study or compared with MBW measurements. It is an inexpensive test, with the consumables costing less than €10 per triplicate of tests per patient. The limitations of this study are that these data are cross-sectional taken during clinical stability, with no longitudinal data comparing HRCT findings in stable disease. To further validate the use of LCI, it will be important to measure the response of LCI during an intervention such as intravenous antibiotics, antiinflammatory therapy, or physiotherapy. In addition, serial CT scans would be ideal during these measures to correlate lung function parameters with structural changes on CT. This, however, is limited from an ethical perspective because of the exposure to radiation that would be required. Also, this study included patients with relatively good lung function measured by FEV1 and excluded those with an FEV1 less than 40% predicted. In advanced disease the benefit of LCI over FEV1 in terms of sensitivity may be reduced and the practicalities of LCI testing are complicated by the longer wash-in and washout times seen with higher LCI. A further limitation is that MBW methods used in this study are not commercially available, and sulfur hexafluoride is a greenhouse gas that is not approved for medical use in the United States and France. A more recently available nitrogen washout system can be

Table 3: Correlations (R Values) between Lung Clearance Index and Spirometric Measures with Computed Tomography Scan Abnormalities

FEV1 FEF25–75% LCI

Percent Bronchiectasis

Percent Airway Thickening

Percent Mucus Plugging

Percent Parenchymal

Percent Air Trapping

Percent Total

20.19 20.27† 0.41*

20.25 20.24 0.21

20.39* 20.44‡ 0.49‡

20.35* 20.41* 0.56‡

20.37* 20.35* 0.36*

20.37* 20.44* 0.55‡

Definition of abbreviations: FEF25–75% = forced expiratory flow between 25 and 75% of FVC; LCI = lung clearance index. *P , 0.01. † P , 0.05. ‡ P , 0.001.

590

American Journal of Respiratory and Critical Care Medicine Volume 189 Number 5 | March 1 2014

ORIGINAL ARTICLE

Figure 3. Data arranged into tertiles according to bronchiectasis computed tomography (CT) score, with low score to high representing increasing disease severity. Each tertile group contained 20 patients and with each successive tertile reflects increasing disease severity based on CT score. (a) Tertiles for CT score values, P , 0.0001 (analysis of variance [ANOVA]) between tertiles; all P , 0.0001. (b) Lung clearance index (LCI) scores in CT tertiles; P = 0.0005 (ANOVA) adjusted for comparison of tertile 1 with tertile 2 (P = 0.06), and of tertile 2 with tertile 3 (P = 0.0002). (c) FEV1 (% predicted) in CT tertiles; P = 0.03 (ANOVA) adjusted for comparison of tertile 1 with tertile 2 (P = 0.16) and of tertile 2 with tertile 3 (P = 0.015).

used without further modification and does not require an inert gas (30). A study in this area has shown a small increase in LCI with age of 0.22 unit/decade (31). The healthy control data used to define a normal LCI were not age-matched and represent a limitation of this study. Although this data set involves an older population than in many other LCI studies, this should not impact the interpretation of the data. LCI by MBW has been identified as a key clinical and research tool in children with CF and has been used to be a primary outcome measure in clinical

trials (32, 33). It has also been proposed to be a predictor of subsequent future lung function in preschool children (34). In BE, LCI has the potential to be used as an outcome measure either in association with, or as an alternative to, conventional spirometry. It may be most valuable in patients with preserved lung function and normal FEV1 , where it is a better predictor of structural changes on CT. In many studies in BE the lack of responsiveness of FEV1 to therapy makes LCI attractive as an outcome measure in clinical trials.

References 1. Pasteur MC, Bilton D, Hill AT; British Thoracic Society Bronchiectasis Non-CF Guideline Group. British Thoracic Society guideline for nonCF bronchiectasis. Thorax 2010;65:i1–i58. 2. Weycker D, Edelsberg J, Oster G, Tino G. Prevalence and economic burden of bronchiectasis. Clin Pulm Med 2005;12:205–209. 3. Bilton D, Jones AL. Chapter 1: Bronchiectasis: epidemiology and causes. In: Floto RA, Haworth CS, editors. Bronchiectasis. Lausanne, Switzerland: European Respiratory Society; 2011. 4. Wynn-Williams N. Bronchiectasis: a study centred on Bedford and its environs. BMJ 1953;1:1194–1199. 5. Bronchiectasis. The European Lung White Book, Charlesworth Press. Available from: http://www.Erswhitebook.Org/chapters/ bronchiectasis/ 6. McShane PJ, Naureckas ET, Tino G, Strek ME. Concise clinical review: non–cystic fibrosis bronchiectasis. Am J Respir Crit Care Med 2013; 188:647–656. 7. Murray MP, Govan JR, Doherty CJ, Simpson AJ, Wilkinson TS, Chalmers JD, Greening AP, Haslett C, Hill AT. A randomized controlled trial of nebulized gentamicin in non–cystic fibrosis bronchiectasis. Am J Respir Crit Care Med 2011;183:491–499. 8. Bilton D, Daviskas E, Anderson SD, Kolbe J, King G, Stirling RG, Thompson BR, Milne D, Charlton B; B301 Investigators. Phase 3 randomized study of the efficacy and safety of inhaled dry powder

In conclusion, LCI is a repeatable and highly sensitive measure of lung function in BE, particularly in those with preserved spirometric lung function, and has the potential to play a key role in future assessment of lung function in BE and in clinical trials as an outcome measure in clinical trials. n Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: Permission to use custom software used to analyze LCI measurements was kindly given by Dr. Nicolas Bell.

mannitol for the symptomatic treatment of non–cystic fibrosis bronchiectasis. Chest 2013;144:215–225. 9. Hill SL, Morrison HM, Burnett D, Stockley RA. Short term response of patients with bronchiectasis to treatment with amoxicillin given in standard or high doses orally or by inhalation. Thorax 1986;41:559–565. 10. Bilton D, Henig N, Morrissey B, Gotfried M. Addition of inhaled tobramycin to ciprofloxacin for acute exacerbations of Pseudomonas aeruginosa infection in adult bronchiectasis. Chest 2006;130:1503–1510. 11. Owens CM, Aurora P, Stanojevic S, Bush A, Wade A, Oliver C, Calder A, Price J, Carr SB, Shankar A, et al.; London Cystic Fibrosis Collaboration. Lung clearance index and HRCT are complementary markers of lung abnormalities in young children with CF. Thorax 2011;66:481–488. 12. 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. 13. 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. 14. 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.

Rowan, Bradley, Bradbury, et al.: Lung Clearance Index and Bronchiectasis

591

ORIGINAL ARTICLE 15. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, Crapo R, Enright P, van der Grinten CP, Gustafsson P, et al.; ATS/ ERS Task Force. Standardisation of spirometry. Eur Respir J 2005; 26:319–338. 16. Stanojevic S, Wade A, Stocks J, Hankinson J, Coates AL, Pan H, Rosenthal M, Corey M, Lebecque P, Cole TJ. Reference ranges for spirometry across all ages: a new approach. Am J Respir Crit Care Med 2008;177:253–260. 17. Brody AS, Klein JS, Molina PL, Quan J, Bean JA, Wilmott RW. Highresolution computed tomography in young patients with cystic fibrosis: distribution of abnormalities and correlation with pulmonary function tests. J Pediatr 2004;145:32–38. 18. Chan SL, Chan-Yeung MM, Ooi GC, Lam CL, Cheung TF, Lam WK, Tsang KW. Validation of the Hong Kong Chinese version of the St. George Respiratory Questionnaire in patients with bronchiectasis. Chest 2002;122:2030–2037. 19. Jones PW, Quirk FH, Baveystock CM. The St. George’s Respiratory Questionnaire. Respir Med 1991;85(Suppl B):25–31; discussion 33–37. 20. Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. Statistician 1983;32:307–317. 21. Gustafsson PM, De Jong PA, Tiddens HAWM, Lindblad A. Multiplebreath inert gas washout and spirometry versus structural lung disease in cystic fibrosis. Thorax 2008;63:129–134. 22. Kent L, Reix P, Innes JA, Zielen S, Le Bourgeois M, Braggion C, Lever S, Arets HG, Brownlee K, Bradley JM, et al.; European Cystic Fibrosis Society Clinical Trial Network (ECFS-DTN) Standardisation Committee. Lung clearance index: evidence for use in clinical trials in cystic fibrosis. J Cyst Fibros 2014;13:123–138. 23. Roberts HR, Wells AU, Milne DG, Rubens MB, Kolbe J, Cole PJ, Hansell DM. Airflow obstruction in bronchiectasis: correlation between computed tomography features and pulmonary function tests. Thorax 2000;55:198–204. 24. King PT, Holdsworth SR, Freezer NJ, Villanueva E, Holmes PW. Characterisation of the onset and presenting clinical features of adult bronchiectasis. Respir Med 2006;100:2183–2189.

592

25. Horsley A. Lung clearance index in the assessment of airways disease. Respir Med 2009;103:793–799. 26. Macklem PT. The physiology of small airways. Am J Respir Crit Care Med 1998;157:S181–S183. 27. Irving SJ, Ives A, Davies G, Donovan J, Edey AJ, Gill SS, Nair A, Saunders C, Wijesekera NT, Alton EW, et al. Lung clearance index and high-resolution computed tomography scores in primary ciliary dyskinesia. Am J Respir Crit Care Med 2013;188:545–549. 28. Hansell DM, Wells AU, Rubens MB, Cole PJ. Bronchiectasis: functional significance of areas of decreased attenuation at expiratory CT. Radiology 1994;193:369–374. 29. Fuchs SI, Ellemunter H, Eder J, Mellies U, Grosse-Onnebrink J, Tummler B, Staab D, Jobst A, Griese M, Ripper J, et al. Feasibility and variability of measuring the lung clearance index in a multicenter setting. Pediatr Pulmonol 2012;47:649–657. 30. Robinson PD, Latzin P, Verbanck S, Hall GL, Horsley A, Gappa M, Thamrin C, Arets HG, Aurora P, Fuchs SI, et al. Consensus statement for inert gas washout measurement using multiple- and singlebreath tests. Eur Respir J 2013;41:507–522. 31. Verbanck S, Thompson BR, Schuermans D, Kalsi H, Biddiscombe M, Stuart-Andrews C, Hanon S, Van Muylem A, Paiva M, Vincken W, et al. Ventilation heterogeneity in the acinar and conductive zones of the normal ageing lung. Thorax 2012;67:789–795. 32. Fuchs SI, Gappa M. Lung clearance index: clinical and research applications in children. Paediatr Respir Rev 2011;12:264–270. 33. Subbarao P, Stanojevic S, Brown M, Jensen R, Rosenfeld M, Davis S, Brumback L, Gustafsson P, Ratjen F. Lung clearance index as an outcome measure for clinical trials in young children with cystic fibrosis. A pilot study using inhaled hypertonic saline. Am J Respir Crit Care Med 2013;188:456–460. 34. Aurora P, Stanojevic S, Wade A, Oliver C, Kozlowska W, Lum S, Bush A, Price J, Carr SB, Shankar A, et al.; London Cystic Fibrosis Collaboration. Lung clearance index at 4 years predicts subsequent lung function in children with cystic fibrosis. Am J Respir Crit Care Med 2011;183:752–758.

American Journal of Respiratory and Critical Care Medicine Volume 189 Number 5 | March 1 2014