- Email: [email protected]
Progression of structural lung disease on CT scans in children with cystic fibrosis related diabetes
Journal of Cystic Fibrosis 12 (2013) 216 – 221 www.elsevier.com/locate/jcf
Original Article
Progression of structural lung disease on CT scans in children with cystic fibrosis related diabetes John Widger a,⁎, Sarath Ranganathan
a, b, c
, Philip J. Robinson
a, b, c
a
c
Department of Respiratory Medicine, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia b The University of Melbourne, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia Murdoch Children's Research Institute, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia Received 22 April 2012; receive in revised 6 September 2012; accepted 21 September 2012 Available online 12 October 2012
Abstract Background: Diabetes has a deleterious effect on clinical status in children with Cystic Fibrosis (CF). We hypothesized that children with CF Related Diabetes (CFRD) or Impaired Glucose Tolerance (IGT) would have more rapidly progressive lung disease based on chest computed tomography (CT) than those with normal glucose tolerance (NGT). Methods: In a retrospective study we compared lung structure changes over time, as assessed by CT, in 34 CF children with CFRD, IGT or NGT. We then compared CT findings with changes in lung function. Results: Percentage forced expiratory volume in 1 s (%FEV1) remained stable over time with a mean (± SD) yearly change of − 0.5% (± 3.9), − 0.4% (± 2.3) and − 0.85% (± 2.8) (p = 0.92) for the CFRD, IGT and NGT groups respectively. However, there was a mean (95%CI) increase in % CT score of 3.86%/year (1.77–5.95%), 1.59%/year (0.6–2.58%) and 1.09%/year (0.07–2.11%) (p = 0.023). Conclusion: In patients with CFRD, there was a more rapid progression of structural lung disease, compared to those who had NGT that was not reflected by change in lung function. © 2012 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keywords: Chronic respiratory disease; Paediatrics; Radiology
1. Introduction It is recognised that cystic fibrosis related diabetes (CFRD) and impaired glucose tolerance (IGT) have a negative impact on lung function in children with CF [1]. Lung function may decline many years prior to CFRD diagnosis [2]. Although percent predicted forced expired volume in 1 s (%FEV1) remains the most widely used marker of CF lung disease, advances in CF care in recent years have lead to a slower decline in lung function with some studies reporting a yearly decline of less than 1% [3,4]. One consequence of this ⁎ Corresponding author at: Department of Respiratory and Sleep Medicine, Monash Children's, 246 Clayton Road, Clayton, Melbourne 3168, Australia. Tel.: + 61 3 9594 2900. E-mail address: [email protected] (J. Widger).
reduction in lung function decline is that %FEV1 has become a less sensitive marker of disease progression in CF, especially in the paediatric age group [5]. Early detection of lung disease is an imperative in the management of CF and so better tools are required to allow early and aggressive treatment as well as providing sensitive outcome measures for clinical trials. Chest computed tomography (CT) provides detailed information about structural lung disease in CF and has been shown in previous studies to be more sensitive than lung function in detecting disease progression in the CF lung [6]. CT is widely available and relatively easy to perform in older children and adolescents. Despite legitimate concerns regarding the long-term effects of radiation exposure, there has been much interest in this modality as a monitoring tool for disease progression in CF [7]. A number of CT scoring systems have been developed and validated to quantify CF structural lung
1569-1993/$ -see front matter © 2012 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jcf.2012.09.005
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
disease [8–11]. Studies have demonstrated a relationship between CT scores and clinical end points in CF such as pulmonary exacerbations [12] and Pseudomonas aeruginosa (P. aeruginosa) infection [13]. Further studies have demonstrated a significant difference in CT scores before and after an intervention in CF [14,15]. Finally, recent studies have shown that CT can predict future deterioration of CF lung disease and mortality [16,17]. Despite this, we are unaware of any published data on associated lung structure changes found on CT scans specifically in patients with CFRD. We hypothesized that patients with CFRD or IGT would have more rapidly progressive lung disease based on CT than those with normal glucose tolerance (NGT). We tested this hypothesis by comparing lung structure changes over time, as assessed by CT, in CF children with CFRD, IGT or NGT. We then compared disease progression found on CT with changes in lung function. Some of the results of this study have been previously reported in the form of an abstract [18]. 2. Methods 2.1. Subjects Subjects were attending the CF unit at the Royal Children's Hospital (RCH), Melbourne, Australia. Diagnosis of CF was based on newborn screening that was introduced across the state of Victoria in 1989. This program detects about 95% of CF affected infants born in Victoria each year with the remainder detected following clinical presentations including meconium ileus, failure to thrive, suppurative chest disease, or with a CF sibling. A retrospective review was performed of the RCH CF patient database as of 1st November 2009. Patients from 10 years of age who had had an oral glucose tolerance test (OGTT) and 2 CT scans of the chest were included in the study. CT scans were performed on alternate years in a subgroup of patients with unstable baseline clinical status (e.g. increased antibiotic use, poor nutritional status). CT scans were not performed during pulmonary exacerbations or at times of worse clinical symptoms. For subjects who had 3 CT scan available the first and third scans were compared. 2.2. Computed tomography High Resolution CT scans of the chest were obtained on a Siemens 16 slice CT scanner with a window width of 1500 HU to − 600 HU. Slices were obtained at 1 mm thickness, using a 1-second scan time. Inspiratory scans were obtained every 10 mm with three additional slices taken at maximal expiration. The CT scans, which were deidentified, were then scored by the primary author (JW), who was trained in the scoring system using standardised reference images (19). A subset of 15 scans was scored independently by a second investigator (PR) with extensive experience in scoring chest CT [17,20]. CTs were scored using a modified version of the method described by Brody et al. [9,19]. Each lobe (including lingula) was scored according to the presence, extent and severity of bronchiectasis, airway thickening, mucous plugging, air trapping and collapse
217
and/or consolidation. The scores for each of the 6 lobes were added to give a maximum composite score of 246 for each patient (see online supplement). 2.3. Oral glucose tolerance test Oral glucose tolerance tests (OGTT) were performed in patients who were 10 years and older during a period of clinical stability at the hospital day medical unit using a standardised protocol. Fasting patients were given a glucose drink to the equivalent of 1.75 g/kg glucose to a maximum of 75 g. Blood glucose measurements were obtained prior to glucose ingestion as well as 60 and 120 min post ingestion. Glucose tolerance status was defined according to WHO guidelines based on the 2 hour glucose level (b 7.8 mmol/L–NGT; ≥ 7.8 mmol/L and b 11.1 mmol–IGT; ≥ 11.1 mmol/L–CFRD). 2.4. Data collection Spirometry testing was carried out according to ATS/ERS standards [21]. In general results for spirometry closest to the time of the CT scan were used. In patients where there was more than one spirometry result within 3 months the best effort was used (to represent the baseline). CF genotype, sputum microbiology, nutritional data and presence of liver disease were also recorded. P. aeruginosa infection was considered chronic if infection persisted in sputum despite an appropriate eradication regime. The study was approved by the RCH human ethics review board as a clinical audit. 2.5. Statistics Results are presented as mean and standard deviation (SD). All data were examined for normality prior to analysis. Continuous variables were compared using ANOVA and the Tukey post hoc test. Categorical variables were compared using the Fisher's exact test. Interobserver agreement of CT scores was calculated using intraclass correlation coefficients. Raw composite and component CT scores were converted to percent of composite scores. Multiple regression was used to test the influence of gender and P. aeruginosa on CT scores. The relationship between composite CT scores and FEV1 at baseline was tested with Pearson correlation. A p b 0.05 was considered statistically significant. Statistical analysis was performed using Stata Version 11.0 (Stata Corporation, College Station, Texas. USA). 3. Results 3.1. Baseline data One hundred and ten patients aged 10–19 years on 1-11-2009 attended the RCH CF clinic at the period of the review. Fifty-seven of these (51%) had been tested with an OGTT. Fifteen patients had been classified as CFRD, 20 with IGT and 22 with NGT. Two or more CT scans were available for scoring in 9 CFRD, 13 IGT and 12 NGT patients.
218
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
Characteristics for these patients at the time of the review are shown in Table 1. Mean age of CFRD diagnosis was 13.6 (1.2) years and mean HbA1c was 6.83%. All patients with CFRD had been commenced on insulin with a mean time of treatment of 1.9 (0.9) years and a mean dose of 0.8 (0.6) units/kg/day. Nutritional status and genotype was similar across the three groups. Patients with CFRD had worse lung function at the time of the review, a female predominance, a higher rate of chronic infection with P. aeruginosa and a higher prevalence of liver disease than those with either IGT or NGT (Table 1). 3.2. CT data Mean age of first scan was similar across the three groups (125 months CFRD vs. 130 IGT vs. 133 NGT, Table 1). In the CFRD group, the first CT scan was performed at a mean of 22 months prior to diagnosis of diabetes. Intraclass correlation coefficient for composite CT scores between observers was considered good at 0.87. A mean (95%CI) increase in % composite CT score of 3.86%/year (1.77–5.95%), 1.59%/year (0.6–2.58%) and 1.09%/year (0.07–2.11%) (p = 0.023) was found in those with CFRD, IGT and NGT respectively (Fig. 1). This was significantly greater in the CFRD compared with the NGT group (p = 0.03). Chronic infection with P. aeruginosa was not found to be associated with change in CT scores when adjusted for glucose tolerance (p = 0.65). There were more females in the CFRD group however gender did not have a statistical association with CT score (p = 0.16). When component CT scores were analysed, subjects with CFRD had significantly more deterioration for bronchiectasis, airway thickening and parenchymal changes than did subjects with NGT (Table 2). 3.3. Lung function At the time of the review, mean FEV1 was better in the NGT group than the IGT and CFRD groups (Table 1). However,
Table 1 Characteristics of patients with 2 CT scans available.
Female n(n%) Age 1st scan (months) Months between scans (SD) BMI (kg/m2) Chronic Psa n(n%) CFLD n(n%) %FEV1 (SD) deltaF508 Homozygous n(%) Heterozygous n(%)
NGT (n = 12)
IGT (n = 13)
CFRD (n = 9)
p value
4(33%) 133 30 (9.6) 19.1 2(16) 2(16) 93.8(9)
5(38%) 130 40(8.8) 19.5 2(15) 2(15) 85.8(13)
8(89%) 125 30(8.4) 20.1 4(44) 6(66) 76.8(17)
0.028 a 0.79 0.8 0.26 0.6 0.025 a 0.02 b
5(42) 4(33)
9(69) 3(23)
5(55) 2(22)
0.38 0.69
NGT = Normal glucose tolerance, IGT = Impaired, CFRD = Cystic fibrosis related diabetes. Psa = Pseudomonas aeruginosa, CFLD = Cystic Fibrosis Liver Disease. a Fisher exact — significant difference between NGT vs. CFRD. b Bonferroni post hoc test- significant difference between NGT vs. CFRD.
FEV1 remained relatively stable over time with a mean yearly change in %FEV1 (± SD) of − 0.5% (± 3.9), − 0.4% (± 2.3) and − 0.85% (± 2.8) (p = 0.92) for the CFRD, IGT and NGT groups respectively. There was a negative correlation between lung function and CT score at base line (r = 0.69, p b 0.0001) (Fig. 2). 4. Discussion We identified more rapid deterioration in lung structural disease, as reflected by CT scores, in patients with CFRD compared with those with NGT. FEV1 remained relatively stable over time in all three groups. CT scores were worse to start within the CFRD group and this fact may have contributed to the more rapid progression of structural lung disease seen in this group. The female predominance in the CFRD group is noteworthy as it is known that females have an increased clinical decline compared to males, especially during puberty. Although chronic P. aeruginosa infection and female gender did not significantly contribute to differences in CT scores after adjustment for glucose tolerance status, the numbers in this analysis were small. We cannot out rule that the worse CT scores and more rapid progression of structural lung disease seen in the CFRD group was in part due to the female majority in the group. The results of this study are consistent with literature suggesting that CT may be justified as a more sensitive method than lung function to monitor disease progression in CF. In their study of 48 patients (mean age 11 ± 3.3 years), De Jong and colleagues found a mean improvement in %FEV1 of 1.24%/year, while the mean Brody composite CT score increased by 2.01%/ year (p b 0.05) [6]. Other studies have found that composite CT/ lung function score was at least 2.7 times more sensitive at detecting disease progression than FEV1 alone and that using CT as an outcome marker can significantly reduce the sample size needed for clinical trials (14–15). Trials of insulin treatment in patients with CFRD and IGT have appropriately focused on lung function and nutritional parameters as their main outcome measures [22,23]. However, our data suggest that lung function may be relatively insensitive in these patients, particularly over a duration of 1 to 2 years (the likely treatment period for clinical trials). This insensitivity may lead to a lack of treatment effect that may otherwise have been shown by a more sensitive test. Controversy persists regarding the use of CT in clinical practice due to concerns over the long-term effects of exposure to ionizing radiation. However, in recent years, the radiation dose associated with CT scans has been significantly reduced with further improvements likely in the future [24]. Thus, in future clinical trials studying the benefit of insulin in CF patients with IGT, it may be appropriate to adopt CT score as an outcome measure. The most significant lung structural changes specifically identified in this study were bronchiectasis, airway wall thickening and parenchymal changes. The relationship between the development of CFRD and deterioration of CF lung disease has been well established [1,2]. The pathophysiology of this relationship is still to be clearly defined. One postulated mechanism is the effect of insulin deficiency and protein catabolism leading to nutritional decline. Poor nutritional status and diminished lung function are linked in CF and abnormal
CFRD
100 90 80 70 60 50 40 30 20 10 0
100
15
20
5
10
15
Age (years)
Age (years)
IGT
IGT
100 90 80 70 60 50 40 30 20 10 0
20
100
%FEV1
Composite CTscore %
10
50
0 5
10
15
20
5
10
15
Age (years)
Age (years)
NGT
NGT
20
100
%FEV1
Composite CTscore %
50
0 5
100 90 80 70 60 50 40 30 20 10 0
219
CFRD
%FEV1
Composite CTscore %
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
5
10
15
50
0
20
5
10
Age (years)
15
20
Age (years)
Fig. 1. Composite CT score and lung function (%FEV1) in patients with normal glucose tolerance (NGT), impaired glucose tolerance (IGT), and cystic fibrosis related diabetes. (CFRD).
Bronchiectasis Airway wall thickening Parenchyma changes Mucous plugging Gas trapping Composite score
IGT
CFRD
p value
0.97 ± 2.17 1.1 ± 2.1 − 0.34 ± 0.82 0.45 ± 2.6 3.54 ± 8.41 1.09 ± 1.8
2.49 ± 3.34 1.06 ± 2.25 0.31 ± 0.70 0.75 ± 2.54 3.11 ± 4.86 1.59 ± 1.8
3.36 ± 3 4.35 ± 4.1 1.36 ± 1.47 1.42 ± 5.6 11.5 ± 11.7 3.8 ± 3.2
0.05 0.02 0.005 0.44 0.07 0.02
NGT = Normal glucose tolerance. IGT = Impaired glucose tolerance. CFRD = Cystic fibrosis related diabetes.
50 40 30 20 10
NGT
0
Table 2 CT scores showing mean change in % score per year ± SD.
[28]. P. aeruginosa is associated with worsening lung function in CF and its increased prevalence in those with diabetes might explain the poorer lung function seen in these patients.
Composite CTscore (%)
glucose tolerance in CF is associated with poor nutritional outcome [25,26]. However insulin appears to restore nutritional status more effectively than lung function so that at least part of the clinical decline in those with abnormal glucose tolerance appears to be independent of insulin deficiency [22]. Abnormal glucose tolerance may also have a more direct effect on lung structural disease. It is known that airway glucose concentrations are elevated in CF when blood glucose concentrations acutely rise [27], and this has been shown to promote bacterial growth especially Staphylococcus aureus and P. aeruginosa
60
70
80
90
100
110
FEV1 (%) Fig. 2. Scatter plot showing relationship between CT scores and FEV1 at baseline. (r = 0.69, p b 0.0001).
220
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
There are several limitations to the present study. Only 34 of the possible 110 patients aged 10 years and above were included in this study. These patients were further divided into three subgroups according to glucose status so that our group sizes were small. Glucose tolerance screening was performed only for symptomatic patients, which may have led to a selection bias in the study i.e. sicker patients selected for diabetic screening. Furthermore, CT scans were performed in only a subgroup of patients because of a relatively unstable clinical baseline. We cannot rule out the possibility that there might have been patients with ‘asymptomatic’ CFRD amongst those not studied and that their CTs may have shown less progressive lung disease. Despite these limitations, progression of lung disease was demonstrated by CT in subjects with CFRD compared with those undergoing OGTT for similar clinical indications but in whom the result suggested NGT. In conclusion, in this pilot study, patients with CFRD had more rapid progression of structural lung disease compared to those who had NGT that was not reflected by deterioration in lung function. Longitudinal data are required to confirm these findings and to show the temporal relationship between the development of abnormal glucose tolerance and lung structural damage. This study indicates the need for earlier diagnosis and aggressive treatment of CFRD in order to limit the associated decline in structural lung disease. Given the stability of lung function in these patients, CT may be a more appropriate outcome marker in future trials of insulin treatment in CFRD. Conflict of interest statement None of the authors have any conflict in interest to declare. Acknowledgements The authors would like to acknowledge Prof. Susan Donath for her help with the statistical analysis. This study was funded by the Royal Children's Cystic Fibrosis Research Trust. Appendix A. Supplementary material Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2012.09.005. References [1] Lanng S, Thorsteinsson B, Nerup J, Koch C. Influence of the development of diabetes mellitus on clinical status in patients with cystic fibrosis. Eur J Pediatr 1992 Sep;151(9):684-7. [2] Milla CE, Warwick WJ, Moran A. Trends in pulmonary function in patients with cystic fibrosis correlate with the degree of glucose intolerance at baseline. Am J Respir Crit Care Med 2000 Sep;162(3 Pt 1):891-5. [3] Que C, Cullinan P, Geddes D. Improving rate of decline of FEV1 in young adults with cystic fibrosis. Thorax 2006 Feb;61(2):155-7. [4] McPhail GL, Acton JD, Fenchel MC, Amin RS, Seid M. Improvements in lung function outcomes in children with cystic fibrosis are associated with better nutrition, fewer chronic pseudomonas aeruginosa infections, and dornase alfa use. J Pediatr 2008 Dec;153(6):752-7.
[5] Ranganathan SC, Davis SD, Rosenfeld M. Monitoring of Structure and Function in Early Cystic Fibrosis Lung Disease. Pediatr Allergy Immunol Pulmonol 2011 Sept;24(3):133-7. [6] de Jong PA, Nakano Y, Lequin MH, Mayo JR, Woods R, Pare PD. Progressive damage on high resolution computed tomography despite stable lung function in cystic fibrosis. Eur Respir J 2004 Jan;23(1):93-7l. [7] Tiddens HA. Chest computed tomography scans should be considered as a routine investigation in cystic fibrosis. Paediatr Respir Rev 2006 Sep;7(3): 202-8. [8] Bhalla M, Turcios N, Aponte V, Jenkins M, Leitman BS, McCauley DI, et al. Cystic fibrosis: scoring system with thin-section CT. Radiology 1991 Jun;179(3):783-8. [9] Brody AS, Kosorok MR, Li Z, Broderick LS, Foster JL, Laxova A, et al. Reproducibility of a scoring system for computed tomography scanning in cystic fibrosis. J Thorac Imaging 2006 Mar;21(1):14-21. [10] Helbich TH, Heinz-Peer G, Fleischmann D, Wojnarowski C, Wunderbaldinger P, Huber S, et al. Evolution of CT findings in patients with cystic fibrosis. Am J Roentgenol 1999 Jul;173(1):81-8. [11] Santamaria F, Grillo G, Guidi G, Rotondo A, Raia V, de Ritis G, et al. Cystic fibrosis: when should high-resolution computed tomography of the chest Be obtained? Pediatrics 1998 May;101(5):908-13. [12] Loeve M, Gerbrands K, Hop WC, Rosenfeld M, Hartmann IC, Tiddens HA. Bronchiectasis and pulmonary exacerbations in children and young adults with CF. Chest 2011 Jul;140(1):178-85. [13] Farrell PM, Collins J, Broderick LS, Rock MJ, Li Z, Kosorok MR, et al. Association between mucoid Pseudomonas infection and bronchiectasis in children with cystic fibrosis. Radiology 2009 Aug;252(2):534-43. [14] Nasr SZ, Gordon D, Sakmar E, Yu X, Christodoulou E, Eckhardt BP, et al. High resolution computerized tomography of the chest and pulmonary function testing in evaluating the effect of tobramycin solution for inhalation in cystic fibrosis patients. Pediatr Pulmonol 2006 Dec;41(12):1129-37. [15] Robinson TE, Leung AN, Northway WH, Blankenberg FG, Chan FP, Bloch DA, et al. Composite spirometric-computed tomography outcome measure in early cystic fibrosis lung disease. Am J Respir Crit Care Med 2003 Sep 1;168(5):588-93. [16] Sanders DB, Li Z, Brody AS, Farrell PM. Chest computed tomography scores of severity are associated with future lung disease progression in children with cystic fibrosis. Am J Respir Crit Care Med 2011 Oct 1;184(7):816-21. [17] Loeve M, Hop WC, de Bruijne M, van Hal PT, Robinson P, Aitken ML, et al. Chest computed tomography scores are predictive of survival in patients with cystic fibrosis awaiting lung transplantation. Am J Respir Crit Care Med 2012 May 15;185(10):1096-103. [18] Widger J, Ranganathan S, Robinson P. Progression of structural lung disease on CT scans in children with Cystic Fibrosis Related Diabetes. Am J Respir Crit Care Med 2010;181:A1834. [19] de Jong PA, Tiddens HA. Cystic fibrosis specific computed tomography scoring. Proc Am Thorac Soc 2007 Aug 1;4(4):338-42. [20] Loeve M, van Hal PT, Robinson P, de Jong PA, Lequin MH, Hop WC, et al. The spectrum of structural abnormalities on CT scans from patients with CF with severe advanced lung disease. Thorax 2009 Oct;64(10):876-82. [21] Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardisation of spirometry. Eur Respir J 2005 Aug;26(2):319-38. [22] Mohan K, Israel KL, Miller H, Grainger R, Ledson MJ, Walshaw MJ. Long-term effect of insulin treatment in cystic fibrosis-related diabetes. Respiration 2008;76(2):181-6. [23] Moran A, Pekow P, Grover P, Zorn M, Slovis B, Pilewski J, et al. Insulin therapy to improve BMI in cystic fibrosis-related diabetes without fasting hyperglycemia: results of the cystic fibrosis related diabetes therapy trial. Diabetes Care 2009 Oct;32(10):1783-8. [24] O'Connor OJ, Vandeleur M, McGarrigle AM, Moore N, McWilliams SR, McSweeney SE, et al. Development of low-dose protocols for thin-section CT assessment of cystic fibrosis in pediatric patients. Radiology 2010 Dec;257(3):820-9. [25] Corey M, McLaughlin FJ, Williams M, et al. A comparison of survival, growth, and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol 1988;41:583-91.
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221 [26] White H, Pollard K, Etherington C, Clifton I, Morton AM, et al. Nutritional decline in cystic fibrosis related diabetes: the effect of intensive nutritional intervention. J Cyst Fibros 2009 May;8(3):179-85. [27] Baker EH, Clark N, Brennan AL, Fisher DA, Gyi KM, Hodson ME, et al. Hyperglycemia and cystic fibrosis alter respiratory fluid glucose concentrations
221
estimated by breath condensate analysis. J Appl Physiol 2007 May;102(5): 1969-75. [28] Brennan AL, Gyi KM, Wood DM, Johnson J, Holliman R, Baines DL, et al. Airway glucose concentrations and effect on growth of respiratory pathogens in cystic fibrosis. J Cyst Fibros 2007 Apr;6(2):101-9.
Original Article
Progression of structural lung disease on CT scans in children with cystic fibrosis related diabetes John Widger a,⁎, Sarath Ranganathan
a, b, c
, Philip J. Robinson
a, b, c
a
c
Department of Respiratory Medicine, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia b The University of Melbourne, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia Murdoch Children's Research Institute, The Royal Children's Hospital, 50 Flemington Road, Parkville, VIC 3052 Australia Received 22 April 2012; receive in revised 6 September 2012; accepted 21 September 2012 Available online 12 October 2012
Abstract Background: Diabetes has a deleterious effect on clinical status in children with Cystic Fibrosis (CF). We hypothesized that children with CF Related Diabetes (CFRD) or Impaired Glucose Tolerance (IGT) would have more rapidly progressive lung disease based on chest computed tomography (CT) than those with normal glucose tolerance (NGT). Methods: In a retrospective study we compared lung structure changes over time, as assessed by CT, in 34 CF children with CFRD, IGT or NGT. We then compared CT findings with changes in lung function. Results: Percentage forced expiratory volume in 1 s (%FEV1) remained stable over time with a mean (± SD) yearly change of − 0.5% (± 3.9), − 0.4% (± 2.3) and − 0.85% (± 2.8) (p = 0.92) for the CFRD, IGT and NGT groups respectively. However, there was a mean (95%CI) increase in % CT score of 3.86%/year (1.77–5.95%), 1.59%/year (0.6–2.58%) and 1.09%/year (0.07–2.11%) (p = 0.023). Conclusion: In patients with CFRD, there was a more rapid progression of structural lung disease, compared to those who had NGT that was not reflected by change in lung function. © 2012 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keywords: Chronic respiratory disease; Paediatrics; Radiology
1. Introduction It is recognised that cystic fibrosis related diabetes (CFRD) and impaired glucose tolerance (IGT) have a negative impact on lung function in children with CF [1]. Lung function may decline many years prior to CFRD diagnosis [2]. Although percent predicted forced expired volume in 1 s (%FEV1) remains the most widely used marker of CF lung disease, advances in CF care in recent years have lead to a slower decline in lung function with some studies reporting a yearly decline of less than 1% [3,4]. One consequence of this ⁎ Corresponding author at: Department of Respiratory and Sleep Medicine, Monash Children's, 246 Clayton Road, Clayton, Melbourne 3168, Australia. Tel.: + 61 3 9594 2900. E-mail address: [email protected] (J. Widger).
reduction in lung function decline is that %FEV1 has become a less sensitive marker of disease progression in CF, especially in the paediatric age group [5]. Early detection of lung disease is an imperative in the management of CF and so better tools are required to allow early and aggressive treatment as well as providing sensitive outcome measures for clinical trials. Chest computed tomography (CT) provides detailed information about structural lung disease in CF and has been shown in previous studies to be more sensitive than lung function in detecting disease progression in the CF lung [6]. CT is widely available and relatively easy to perform in older children and adolescents. Despite legitimate concerns regarding the long-term effects of radiation exposure, there has been much interest in this modality as a monitoring tool for disease progression in CF [7]. A number of CT scoring systems have been developed and validated to quantify CF structural lung
1569-1993/$ -see front matter © 2012 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jcf.2012.09.005
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
disease [8–11]. Studies have demonstrated a relationship between CT scores and clinical end points in CF such as pulmonary exacerbations [12] and Pseudomonas aeruginosa (P. aeruginosa) infection [13]. Further studies have demonstrated a significant difference in CT scores before and after an intervention in CF [14,15]. Finally, recent studies have shown that CT can predict future deterioration of CF lung disease and mortality [16,17]. Despite this, we are unaware of any published data on associated lung structure changes found on CT scans specifically in patients with CFRD. We hypothesized that patients with CFRD or IGT would have more rapidly progressive lung disease based on CT than those with normal glucose tolerance (NGT). We tested this hypothesis by comparing lung structure changes over time, as assessed by CT, in CF children with CFRD, IGT or NGT. We then compared disease progression found on CT with changes in lung function. Some of the results of this study have been previously reported in the form of an abstract [18]. 2. Methods 2.1. Subjects Subjects were attending the CF unit at the Royal Children's Hospital (RCH), Melbourne, Australia. Diagnosis of CF was based on newborn screening that was introduced across the state of Victoria in 1989. This program detects about 95% of CF affected infants born in Victoria each year with the remainder detected following clinical presentations including meconium ileus, failure to thrive, suppurative chest disease, or with a CF sibling. A retrospective review was performed of the RCH CF patient database as of 1st November 2009. Patients from 10 years of age who had had an oral glucose tolerance test (OGTT) and 2 CT scans of the chest were included in the study. CT scans were performed on alternate years in a subgroup of patients with unstable baseline clinical status (e.g. increased antibiotic use, poor nutritional status). CT scans were not performed during pulmonary exacerbations or at times of worse clinical symptoms. For subjects who had 3 CT scan available the first and third scans were compared. 2.2. Computed tomography High Resolution CT scans of the chest were obtained on a Siemens 16 slice CT scanner with a window width of 1500 HU to − 600 HU. Slices were obtained at 1 mm thickness, using a 1-second scan time. Inspiratory scans were obtained every 10 mm with three additional slices taken at maximal expiration. The CT scans, which were deidentified, were then scored by the primary author (JW), who was trained in the scoring system using standardised reference images (19). A subset of 15 scans was scored independently by a second investigator (PR) with extensive experience in scoring chest CT [17,20]. CTs were scored using a modified version of the method described by Brody et al. [9,19]. Each lobe (including lingula) was scored according to the presence, extent and severity of bronchiectasis, airway thickening, mucous plugging, air trapping and collapse
217
and/or consolidation. The scores for each of the 6 lobes were added to give a maximum composite score of 246 for each patient (see online supplement). 2.3. Oral glucose tolerance test Oral glucose tolerance tests (OGTT) were performed in patients who were 10 years and older during a period of clinical stability at the hospital day medical unit using a standardised protocol. Fasting patients were given a glucose drink to the equivalent of 1.75 g/kg glucose to a maximum of 75 g. Blood glucose measurements were obtained prior to glucose ingestion as well as 60 and 120 min post ingestion. Glucose tolerance status was defined according to WHO guidelines based on the 2 hour glucose level (b 7.8 mmol/L–NGT; ≥ 7.8 mmol/L and b 11.1 mmol–IGT; ≥ 11.1 mmol/L–CFRD). 2.4. Data collection Spirometry testing was carried out according to ATS/ERS standards [21]. In general results for spirometry closest to the time of the CT scan were used. In patients where there was more than one spirometry result within 3 months the best effort was used (to represent the baseline). CF genotype, sputum microbiology, nutritional data and presence of liver disease were also recorded. P. aeruginosa infection was considered chronic if infection persisted in sputum despite an appropriate eradication regime. The study was approved by the RCH human ethics review board as a clinical audit. 2.5. Statistics Results are presented as mean and standard deviation (SD). All data were examined for normality prior to analysis. Continuous variables were compared using ANOVA and the Tukey post hoc test. Categorical variables were compared using the Fisher's exact test. Interobserver agreement of CT scores was calculated using intraclass correlation coefficients. Raw composite and component CT scores were converted to percent of composite scores. Multiple regression was used to test the influence of gender and P. aeruginosa on CT scores. The relationship between composite CT scores and FEV1 at baseline was tested with Pearson correlation. A p b 0.05 was considered statistically significant. Statistical analysis was performed using Stata Version 11.0 (Stata Corporation, College Station, Texas. USA). 3. Results 3.1. Baseline data One hundred and ten patients aged 10–19 years on 1-11-2009 attended the RCH CF clinic at the period of the review. Fifty-seven of these (51%) had been tested with an OGTT. Fifteen patients had been classified as CFRD, 20 with IGT and 22 with NGT. Two or more CT scans were available for scoring in 9 CFRD, 13 IGT and 12 NGT patients.
218
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
Characteristics for these patients at the time of the review are shown in Table 1. Mean age of CFRD diagnosis was 13.6 (1.2) years and mean HbA1c was 6.83%. All patients with CFRD had been commenced on insulin with a mean time of treatment of 1.9 (0.9) years and a mean dose of 0.8 (0.6) units/kg/day. Nutritional status and genotype was similar across the three groups. Patients with CFRD had worse lung function at the time of the review, a female predominance, a higher rate of chronic infection with P. aeruginosa and a higher prevalence of liver disease than those with either IGT or NGT (Table 1). 3.2. CT data Mean age of first scan was similar across the three groups (125 months CFRD vs. 130 IGT vs. 133 NGT, Table 1). In the CFRD group, the first CT scan was performed at a mean of 22 months prior to diagnosis of diabetes. Intraclass correlation coefficient for composite CT scores between observers was considered good at 0.87. A mean (95%CI) increase in % composite CT score of 3.86%/year (1.77–5.95%), 1.59%/year (0.6–2.58%) and 1.09%/year (0.07–2.11%) (p = 0.023) was found in those with CFRD, IGT and NGT respectively (Fig. 1). This was significantly greater in the CFRD compared with the NGT group (p = 0.03). Chronic infection with P. aeruginosa was not found to be associated with change in CT scores when adjusted for glucose tolerance (p = 0.65). There were more females in the CFRD group however gender did not have a statistical association with CT score (p = 0.16). When component CT scores were analysed, subjects with CFRD had significantly more deterioration for bronchiectasis, airway thickening and parenchymal changes than did subjects with NGT (Table 2). 3.3. Lung function At the time of the review, mean FEV1 was better in the NGT group than the IGT and CFRD groups (Table 1). However,
Table 1 Characteristics of patients with 2 CT scans available.
Female n(n%) Age 1st scan (months) Months between scans (SD) BMI (kg/m2) Chronic Psa n(n%) CFLD n(n%) %FEV1 (SD) deltaF508 Homozygous n(%) Heterozygous n(%)
NGT (n = 12)
IGT (n = 13)
CFRD (n = 9)
p value
4(33%) 133 30 (9.6) 19.1 2(16) 2(16) 93.8(9)
5(38%) 130 40(8.8) 19.5 2(15) 2(15) 85.8(13)
8(89%) 125 30(8.4) 20.1 4(44) 6(66) 76.8(17)
0.028 a 0.79 0.8 0.26 0.6 0.025 a 0.02 b
5(42) 4(33)
9(69) 3(23)
5(55) 2(22)
0.38 0.69
NGT = Normal glucose tolerance, IGT = Impaired, CFRD = Cystic fibrosis related diabetes. Psa = Pseudomonas aeruginosa, CFLD = Cystic Fibrosis Liver Disease. a Fisher exact — significant difference between NGT vs. CFRD. b Bonferroni post hoc test- significant difference between NGT vs. CFRD.
FEV1 remained relatively stable over time with a mean yearly change in %FEV1 (± SD) of − 0.5% (± 3.9), − 0.4% (± 2.3) and − 0.85% (± 2.8) (p = 0.92) for the CFRD, IGT and NGT groups respectively. There was a negative correlation between lung function and CT score at base line (r = 0.69, p b 0.0001) (Fig. 2). 4. Discussion We identified more rapid deterioration in lung structural disease, as reflected by CT scores, in patients with CFRD compared with those with NGT. FEV1 remained relatively stable over time in all three groups. CT scores were worse to start within the CFRD group and this fact may have contributed to the more rapid progression of structural lung disease seen in this group. The female predominance in the CFRD group is noteworthy as it is known that females have an increased clinical decline compared to males, especially during puberty. Although chronic P. aeruginosa infection and female gender did not significantly contribute to differences in CT scores after adjustment for glucose tolerance status, the numbers in this analysis were small. We cannot out rule that the worse CT scores and more rapid progression of structural lung disease seen in the CFRD group was in part due to the female majority in the group. The results of this study are consistent with literature suggesting that CT may be justified as a more sensitive method than lung function to monitor disease progression in CF. In their study of 48 patients (mean age 11 ± 3.3 years), De Jong and colleagues found a mean improvement in %FEV1 of 1.24%/year, while the mean Brody composite CT score increased by 2.01%/ year (p b 0.05) [6]. Other studies have found that composite CT/ lung function score was at least 2.7 times more sensitive at detecting disease progression than FEV1 alone and that using CT as an outcome marker can significantly reduce the sample size needed for clinical trials (14–15). Trials of insulin treatment in patients with CFRD and IGT have appropriately focused on lung function and nutritional parameters as their main outcome measures [22,23]. However, our data suggest that lung function may be relatively insensitive in these patients, particularly over a duration of 1 to 2 years (the likely treatment period for clinical trials). This insensitivity may lead to a lack of treatment effect that may otherwise have been shown by a more sensitive test. Controversy persists regarding the use of CT in clinical practice due to concerns over the long-term effects of exposure to ionizing radiation. However, in recent years, the radiation dose associated with CT scans has been significantly reduced with further improvements likely in the future [24]. Thus, in future clinical trials studying the benefit of insulin in CF patients with IGT, it may be appropriate to adopt CT score as an outcome measure. The most significant lung structural changes specifically identified in this study were bronchiectasis, airway wall thickening and parenchymal changes. The relationship between the development of CFRD and deterioration of CF lung disease has been well established [1,2]. The pathophysiology of this relationship is still to be clearly defined. One postulated mechanism is the effect of insulin deficiency and protein catabolism leading to nutritional decline. Poor nutritional status and diminished lung function are linked in CF and abnormal
CFRD
100 90 80 70 60 50 40 30 20 10 0
100
15
20
5
10
15
Age (years)
Age (years)
IGT
IGT
100 90 80 70 60 50 40 30 20 10 0
20
100
%FEV1
Composite CTscore %
10
50
0 5
10
15
20
5
10
15
Age (years)
Age (years)
NGT
NGT
20
100
%FEV1
Composite CTscore %
50
0 5
100 90 80 70 60 50 40 30 20 10 0
219
CFRD
%FEV1
Composite CTscore %
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
5
10
15
50
0
20
5
10
Age (years)
15
20
Age (years)
Fig. 1. Composite CT score and lung function (%FEV1) in patients with normal glucose tolerance (NGT), impaired glucose tolerance (IGT), and cystic fibrosis related diabetes. (CFRD).
Bronchiectasis Airway wall thickening Parenchyma changes Mucous plugging Gas trapping Composite score
IGT
CFRD
p value
0.97 ± 2.17 1.1 ± 2.1 − 0.34 ± 0.82 0.45 ± 2.6 3.54 ± 8.41 1.09 ± 1.8
2.49 ± 3.34 1.06 ± 2.25 0.31 ± 0.70 0.75 ± 2.54 3.11 ± 4.86 1.59 ± 1.8
3.36 ± 3 4.35 ± 4.1 1.36 ± 1.47 1.42 ± 5.6 11.5 ± 11.7 3.8 ± 3.2
0.05 0.02 0.005 0.44 0.07 0.02
NGT = Normal glucose tolerance. IGT = Impaired glucose tolerance. CFRD = Cystic fibrosis related diabetes.
50 40 30 20 10
NGT
0
Table 2 CT scores showing mean change in % score per year ± SD.
[28]. P. aeruginosa is associated with worsening lung function in CF and its increased prevalence in those with diabetes might explain the poorer lung function seen in these patients.
Composite CTscore (%)
glucose tolerance in CF is associated with poor nutritional outcome [25,26]. However insulin appears to restore nutritional status more effectively than lung function so that at least part of the clinical decline in those with abnormal glucose tolerance appears to be independent of insulin deficiency [22]. Abnormal glucose tolerance may also have a more direct effect on lung structural disease. It is known that airway glucose concentrations are elevated in CF when blood glucose concentrations acutely rise [27], and this has been shown to promote bacterial growth especially Staphylococcus aureus and P. aeruginosa
60
70
80
90
100
110
FEV1 (%) Fig. 2. Scatter plot showing relationship between CT scores and FEV1 at baseline. (r = 0.69, p b 0.0001).
220
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221
There are several limitations to the present study. Only 34 of the possible 110 patients aged 10 years and above were included in this study. These patients were further divided into three subgroups according to glucose status so that our group sizes were small. Glucose tolerance screening was performed only for symptomatic patients, which may have led to a selection bias in the study i.e. sicker patients selected for diabetic screening. Furthermore, CT scans were performed in only a subgroup of patients because of a relatively unstable clinical baseline. We cannot rule out the possibility that there might have been patients with ‘asymptomatic’ CFRD amongst those not studied and that their CTs may have shown less progressive lung disease. Despite these limitations, progression of lung disease was demonstrated by CT in subjects with CFRD compared with those undergoing OGTT for similar clinical indications but in whom the result suggested NGT. In conclusion, in this pilot study, patients with CFRD had more rapid progression of structural lung disease compared to those who had NGT that was not reflected by deterioration in lung function. Longitudinal data are required to confirm these findings and to show the temporal relationship between the development of abnormal glucose tolerance and lung structural damage. This study indicates the need for earlier diagnosis and aggressive treatment of CFRD in order to limit the associated decline in structural lung disease. Given the stability of lung function in these patients, CT may be a more appropriate outcome marker in future trials of insulin treatment in CFRD. Conflict of interest statement None of the authors have any conflict in interest to declare. Acknowledgements The authors would like to acknowledge Prof. Susan Donath for her help with the statistical analysis. This study was funded by the Royal Children's Cystic Fibrosis Research Trust. Appendix A. Supplementary material Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2012.09.005. References [1] Lanng S, Thorsteinsson B, Nerup J, Koch C. Influence of the development of diabetes mellitus on clinical status in patients with cystic fibrosis. Eur J Pediatr 1992 Sep;151(9):684-7. [2] Milla CE, Warwick WJ, Moran A. Trends in pulmonary function in patients with cystic fibrosis correlate with the degree of glucose intolerance at baseline. Am J Respir Crit Care Med 2000 Sep;162(3 Pt 1):891-5. [3] Que C, Cullinan P, Geddes D. Improving rate of decline of FEV1 in young adults with cystic fibrosis. Thorax 2006 Feb;61(2):155-7. [4] McPhail GL, Acton JD, Fenchel MC, Amin RS, Seid M. Improvements in lung function outcomes in children with cystic fibrosis are associated with better nutrition, fewer chronic pseudomonas aeruginosa infections, and dornase alfa use. J Pediatr 2008 Dec;153(6):752-7.
[5] Ranganathan SC, Davis SD, Rosenfeld M. Monitoring of Structure and Function in Early Cystic Fibrosis Lung Disease. Pediatr Allergy Immunol Pulmonol 2011 Sept;24(3):133-7. [6] de Jong PA, Nakano Y, Lequin MH, Mayo JR, Woods R, Pare PD. Progressive damage on high resolution computed tomography despite stable lung function in cystic fibrosis. Eur Respir J 2004 Jan;23(1):93-7l. [7] Tiddens HA. Chest computed tomography scans should be considered as a routine investigation in cystic fibrosis. Paediatr Respir Rev 2006 Sep;7(3): 202-8. [8] Bhalla M, Turcios N, Aponte V, Jenkins M, Leitman BS, McCauley DI, et al. Cystic fibrosis: scoring system with thin-section CT. Radiology 1991 Jun;179(3):783-8. [9] Brody AS, Kosorok MR, Li Z, Broderick LS, Foster JL, Laxova A, et al. Reproducibility of a scoring system for computed tomography scanning in cystic fibrosis. J Thorac Imaging 2006 Mar;21(1):14-21. [10] Helbich TH, Heinz-Peer G, Fleischmann D, Wojnarowski C, Wunderbaldinger P, Huber S, et al. Evolution of CT findings in patients with cystic fibrosis. Am J Roentgenol 1999 Jul;173(1):81-8. [11] Santamaria F, Grillo G, Guidi G, Rotondo A, Raia V, de Ritis G, et al. Cystic fibrosis: when should high-resolution computed tomography of the chest Be obtained? Pediatrics 1998 May;101(5):908-13. [12] Loeve M, Gerbrands K, Hop WC, Rosenfeld M, Hartmann IC, Tiddens HA. Bronchiectasis and pulmonary exacerbations in children and young adults with CF. Chest 2011 Jul;140(1):178-85. [13] Farrell PM, Collins J, Broderick LS, Rock MJ, Li Z, Kosorok MR, et al. Association between mucoid Pseudomonas infection and bronchiectasis in children with cystic fibrosis. Radiology 2009 Aug;252(2):534-43. [14] Nasr SZ, Gordon D, Sakmar E, Yu X, Christodoulou E, Eckhardt BP, et al. High resolution computerized tomography of the chest and pulmonary function testing in evaluating the effect of tobramycin solution for inhalation in cystic fibrosis patients. Pediatr Pulmonol 2006 Dec;41(12):1129-37. [15] Robinson TE, Leung AN, Northway WH, Blankenberg FG, Chan FP, Bloch DA, et al. Composite spirometric-computed tomography outcome measure in early cystic fibrosis lung disease. Am J Respir Crit Care Med 2003 Sep 1;168(5):588-93. [16] Sanders DB, Li Z, Brody AS, Farrell PM. Chest computed tomography scores of severity are associated with future lung disease progression in children with cystic fibrosis. Am J Respir Crit Care Med 2011 Oct 1;184(7):816-21. [17] Loeve M, Hop WC, de Bruijne M, van Hal PT, Robinson P, Aitken ML, et al. Chest computed tomography scores are predictive of survival in patients with cystic fibrosis awaiting lung transplantation. Am J Respir Crit Care Med 2012 May 15;185(10):1096-103. [18] Widger J, Ranganathan S, Robinson P. Progression of structural lung disease on CT scans in children with Cystic Fibrosis Related Diabetes. Am J Respir Crit Care Med 2010;181:A1834. [19] de Jong PA, Tiddens HA. Cystic fibrosis specific computed tomography scoring. Proc Am Thorac Soc 2007 Aug 1;4(4):338-42. [20] Loeve M, van Hal PT, Robinson P, de Jong PA, Lequin MH, Hop WC, et al. The spectrum of structural abnormalities on CT scans from patients with CF with severe advanced lung disease. Thorax 2009 Oct;64(10):876-82. [21] Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardisation of spirometry. Eur Respir J 2005 Aug;26(2):319-38. [22] Mohan K, Israel KL, Miller H, Grainger R, Ledson MJ, Walshaw MJ. Long-term effect of insulin treatment in cystic fibrosis-related diabetes. Respiration 2008;76(2):181-6. [23] Moran A, Pekow P, Grover P, Zorn M, Slovis B, Pilewski J, et al. Insulin therapy to improve BMI in cystic fibrosis-related diabetes without fasting hyperglycemia: results of the cystic fibrosis related diabetes therapy trial. Diabetes Care 2009 Oct;32(10):1783-8. [24] O'Connor OJ, Vandeleur M, McGarrigle AM, Moore N, McWilliams SR, McSweeney SE, et al. Development of low-dose protocols for thin-section CT assessment of cystic fibrosis in pediatric patients. Radiology 2010 Dec;257(3):820-9. [25] Corey M, McLaughlin FJ, Williams M, et al. A comparison of survival, growth, and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol 1988;41:583-91.
J Widger et al. / Journal of Cystic Fibrosis 12 (2013) 216–221 [26] White H, Pollard K, Etherington C, Clifton I, Morton AM, et al. Nutritional decline in cystic fibrosis related diabetes: the effect of intensive nutritional intervention. J Cyst Fibros 2009 May;8(3):179-85. [27] Baker EH, Clark N, Brennan AL, Fisher DA, Gyi KM, Hodson ME, et al. Hyperglycemia and cystic fibrosis alter respiratory fluid glucose concentrations
221
estimated by breath condensate analysis. J Appl Physiol 2007 May;102(5): 1969-75. [28] Brennan AL, Gyi KM, Wood DM, Johnson J, Holliman R, Baines DL, et al. Airway glucose concentrations and effect on growth of respiratory pathogens in cystic fibrosis. J Cyst Fibros 2007 Apr;6(2):101-9.