Advertisement

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Electronic Letters to:

Pediatric Imaging: Pim A. de Jong, Mark D. Ottink, Simon G. F. Robben, Maarten H. Lequin, Wim C. J. Hop, Johan J. E. Hendriks, Peter D. Paré, and Harm A. W. M. Tiddens Pulmonary Disease Assessment in Cystic Fibrosis: Comparison of CT Scoring Systems and Value of Bronchial and Arterial Dimension Measurements Radiology 2004; 231: 434-439 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Noninvasive assessment of pulmonary involvement in children with cystic fibrosis

Georgia Papaioannou, Dr Catherine Owens, Dr Øystein E. Olsen, Dr Adam Jaffe   (13 June 2005)

Noninvasive assessment of pulmonary involvement in children with cystic fibrosis 13 June 2005
Georgia Papaioannou, MD Radiological Department, Great Ormond Street Hospital for Children, London UK, Dr Catherine Owens, Dr Øystein E. Olsen, Dr Adam Jaffe Send letter to journal: Re: Noninvasive assessment of pulmonary involvement in children with cystic fibrosis gpapaio{at}hotmail.com Georgia Papaioannou, et al.	Editor: We applaud the article by de Jong et al (1) for addressing the need for a noninvasive and sensitive test for the assessment of lung morphology in children with cystic fibrosis, which ideally should be applicable to infants with clinically silent disease. The authors retrospectively compared five thin-section CT scoring systems in children with CF and evaluated the diagnostic helpfulness of bronchial and arterial diameter measurements. We have, however, four major and several minor reservations concerning this study: 1. This was not a “randomly selected” patient population; all the children had established and regularly monitored cystic fibrosis. There were no controls and no children under 5 years of age. Accordingly, the results are not valid in “young children,” as concluded. 2. The study proved the reproducibility and the relatively equal performance of the scoring systems. This is not surprising, as some are modified versions of others. However, it is unclear what is meant by the “robustness” of a diagnostic test. There is obviously an attempt to test the predictability of an abnormal pulmonary function test (PFT) based on thin-section CT findings, which may not have been justified in this study. The authors base their conclusions on a battery of correlation statistics. Although such analyses are undoubtedly useful for generating hypotheses, they may hinder useful evaluation of diagnostic tests. One may, as an example, wonder if patient age correlates with PFTs; we would assume so, but it does not follow that age is a useful clinical variable for assessment of an individual child. The crucial question is what the additional value of a diagnostic test may be. With reference to the authors’ Figure 1, it appears that a Bhalla score in the range 8-10 would predict FEV1% in the range from less than 40 to above 100. In a clinical setting, these numbers may not seem very promising. It is also of importance that such data cannot be extrapolated to patient groups with subclinical disease or to the group of infants or toddlers in whom PFTs cannot easily be obtained. The relatively poor interobserver agreement for scans with mild change is particularly discouraging. 3. The radiation burden from the applied examination protocol is high compared with the well-established standards. More sections are obtained with higher milliampere-second values and hence an inappropriately high radiation dose. Studies have highlighted that use of >50 mAs for thin-section CT in children is not justified (2). 4. We are not convinced that the study justifies the rejection of bronchial and arterial measurements. Scan time was 2.5 seconds per section; how reliably can one detect early change in the presence of even mild motion artifact? Additionally, the assumption that there is no change in arterial caliber as cystic fibrosis progresses is not valid (3). As for the minor points, we assume that the reported lung window of –600/1500 HU (width/level) represents a typographical error. The lack of expiratory scans may well explain the surprisingly low prevalence of small airways disease in this population (4). However, if the prevalence of emphysema, air trapping, and hyperinflation was really 1%, as stated in Table 2, that would suggest such involvement in 0.25 child! In conclusion, we question the authors’ statement that their findings “support the use of thin-section CT to monitor progression of structural abnormality in cystic fibrosis”. In our opinion, the validity of the scoring systems as outcome surrogates has not been proven. References 1. de Jong PA, Ottink MD, Robben SGF, et al. Pulmonary disease assessment in cystic fibrosis: comparison of CT scoring systems and value of bronchial and arterial dimension measurements. Radiology 2004; 231:434-439. 2. Lucaya J, Piqueras J, Garcia-Pena P, et al. Low-dose high-resolution CT of the chest in children and young adults: dose, cooperation, artifact incidence, and image quality. AJR Am J Roentgenol 2000; 175:985-992. 3. Long FR, Williams RS, Castile RG. Structural airway abnormalities in infants and young children with cystic fibrosis. J Pediatr 2004; 144:154-161. 4. Bonnel AS, Song SM, Kesavarju K, et al. Quantitative air-trapping analysis in children with mild cystic fibrosis lung disease. Pediatr Pulmonol 2004; 38:396-405. Dr de Jong responds: We appreciate the interest from Dr Papaioannou and colleagues in our article (1). In addition, we appreciate that they share our view that a noninvasive and sensitive test is needed for the assessment of lung morphology in children with cystic fibrosis. The article discussed by Papaioannou and colleagues, published in July 2004, is one of a series on this topic. Some of our more recently published studies answer some of the reservations raised. 1. Dr Papaioannou questions the random selection of our study population. As described (1), all children with cystic fibrosis at our institution undergo biennial CT scans together with pulmonary function tests (PFTs). We start CT scanning in patients 4-5 years old and PFTs in patients 5-6 years old. For the purpose of our study, 25 patients were randomly selected from among the approximately 100 children who had undergone CT scans. As described in the Methods section, our study included only children between the ages of 5 and 18 years who had a diagnosis of cystic fibrosis and were followed regularly in the clinic. Clearly this is not a random population sample but rather a random sample from among those who fulfilled these criteria. 2. Dr Papaioannou questions the added value of CT scanning as a diagnostic test in children with cystic fibrosis. We share this critical view and remained cautious in our report. We were not suggesting that CT is a useful predictor of lung function tests, but the results did suggest that CT may be helpful in detecting abnormalities when lung function is normal. For example, we stated: “Despite these excellent correlations between PFT results and the scores derived with the thin-section CT scoring systems, abnormalities may be observed on CT scans in patients with normal PFT results. These observations suggest that at least in some instances thin-section CT may be more sensitive to early lung damage in patients with cystic fibrosis” (1). Recently, we published longitudinal data that provide more convincing evidence for the added value of CT scanning in children with cystic fibrosis (2). We demonstrated that in these children, CT scoring systems and bronchial and arterial measurements (3) are more sensitive in the detection of the onset and progression of cystic fibrosis-related lung disease than are PFTs. In our children with cystic fibrosis, PFTs remained stable while CT scoring systems (2) and quantitatively measured airway wall thickness (3) worsened. In addition, a poor correlation was demonstrated between structural information obtained with CT and functional data obtained with PFTs. Recently, a paper was submitted confirming our observation in adults. Hence, structural abnormalities may not be predicted adequately with use of PFTs both in early and more advanced cystic fibrosis. 3. Dr Papaioannou raises concerns about the radiation dose used in this study, which was conducted between 1996 and 2001. We share these concerns and therefore discussed this issue in our manuscript (1) and previously in correspondence with Dr Rawlings and colleagues (4,5). The reference mentioned in Dr Papaioannou’s letter (6) was included in the Discussion section of our manuscript. In January 2002, we changed our protocol to select the CT parameters, and thus radiation dose, on the basis of patient weight. This change has resulted in a more than twofold reduction in the radiation dose per CT scan. We are investigating options to further reduce the dose per scan and have recently modeled the risks associated with repeat CT scanning (7). On the basis of this model, we feel strongly that the risks associated with our current CT protocol are acceptable in the light of the severity of cystic fibrosis-related lung disease and the potential benefits offered by early detection. 4. Dr Papaioannou suggests that we reject bronchial and arterial measurements in cystic fibrosis. This is incorrect; we stated in our report (1) that “further research is necessary to determine the role of bronchial measurements….” The protocol used to quantify bronchial and arterial dimensions included a scanning time of 1 second as stated in the paragraph “Thin-section CT and Evaluation” (1). Nevertheless, we were unable to demonstrate a significant correlation between lung function abnormalities and the ratio of bronchial diameter to the accompanying pulmonary artery diameter or of the bronchial wall thickness to the accompanying pulmonary artery diameter. In a more recent study (3), we demonstrated that the increase in airway wall thickness over 2 years correlated with the decrease in forced expiratory flow of between 25% and 75% of vital capacity predicted (FEF25-75). In that article (3), we discuss that the possibility that pulmonary arterial caliber in CF patients may be less than normal (8, 9). 5. Dr Papaioannou wonders how 1% of the patients could only have emphysema, air trapping or hyperinflation. This can be explained by the fact that this abnormality was never scored by two observers; it was seen by only one observer in a single patient. he reasons we feel that this abnormality is rarely detected is discussed in our article (1). Dr Papaioannou is correct that there was a typographical error. The window level in our study was –600 HU, and the width was 1500 HU. In conclusion, we clearly stated that our data support the use of CT and that CT can be important for therapeutic studies. We did not state that the use of scoring systems is a validated outcome measurement, as suggested by the last sentence in Dr Papaioannou’s letter. The requirements for a useful alternative outcome surrogate in cystic fibrosis were summarized by Brody et al in 1999 (10). Our findings of good interobserver and intraobserver agreement are an important condition for the use of CT scoring as an outcome surrogate. The current insensitivity of PFTs to monitor disease progression in cystic fibrosis demands the development of more sensitive outcome surrogates. On the basis of the results of our study and of more recent studies, we believe that CT scoring systems are promising for use as an outcome measure, but we agree that further research is necessary. References 1. de Jong PA, Ottink MD, Robben SG, et al. Pulmonary disease assessment in cystic fibrosis: comparison of CT scoring systems and value of bronchial and arterial dimension measurements. Radiology 2004; 231:434-439. 2. de Jong PA, Nakano Y, Lequin MH, et al. Progressive damage on high resolution computed tomography despite stable lung function in cystic fibrosis. Eur Respir J 2004; 23:93-97. 3. De Jong PA, Nakano Y, Hop WC, et al. Changes in airway dimensions on computed tomography scans of children with cystic fibrosis. Am J Respir Crit Care Med 2005 April 14; [Epub ahead of print]. 4. de Jong PA, Lequin MH, Mayo JR, Pare PD, Tiddens H. Re: Progressive damage on high-resolution computed tomography [letter]. Eur Respir J 2004; 24:1071-1072. 5. Rawlings D, Tennant D, Furness J. Progressive damage on high-resolution computed tomography [letter]. Eur Respir J 2004; 24:1071. 6. Lucaya J, Piqueras J, Garcia-Pena P, Enriquez G, Garcia-Macias M, Sotil J. Low-dose high-resolution CT of the chest in children and young adults: dose, cooperation, artifact incidence, and image quality. AJR Am J Roentgenol 2000; 175:985-992. 7. de Jong PA, Golmohammadi K, Nakano Y, et al. Modelling the mortality risk associated with radiation exposure from cumulative computed tomography in cystic fibrosis. Am J Respir Crit Care Med 2005; 2:A 576. 8. Ryland D, Reid L. The pulmonary circulation in cystic fibrosis. Thorax 1975; 30:285-292. 9. Long FR, Williams RS, Castile RG. Structural airway abnormalities in infants and young children with cystic fibrosis. J Pediatr 2004; 144:154-161. 10. 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.