Pulmonary Disease in Cystic Fibrosis: Assessment with Chest CT at Chest Radiography Dose Levels

Radiology: Volume 273: Number 2—November 2014 n radiology.rsna.org 597

Original research n

Thoracic imaging

1 From the Department of Radiology (C.W.E., I.A.B., B.I., N.B., G.V.G., K.H.N., F.V., M.D.M., J.d.M.), Department of Medical Imaging and Physical Sciences (N.B.), Department of Pediatric Pneumology (E.D.W., A.M.), Cystic Fibrosis Clinic (A.M.), Department of Medicine (D.C., J.D.M.), and Department of Biomedical Statistics and Informatics (D.C.), Universitair Ziekenhuis Brussel, Vrije Universiteit Brussel, Laarbeeklaan 101, 1090 Brussels, Belgium. Received September 19, 2013; revision requested October 25; revision received January 14, 2014; accepted March 21; final version accepted May 28. Address correspondence to C.W.E. (e-mail: [email protected]).

q RSNA, 2014

Purpose: To investigate a computed tomographic (CT) protocol with iterative reconstruction at conventional radiography dose levels for the assessment of structural lung abnor-malities in patients with cystic fibrosis (CF).

Materials and Methods:

In this institutional review board–approved study, 38 pa-tients with CF (age range, 6–58 years; 21 patients ,18 years and 17 patients .18 years) underwent investigative CT (at minimal exposure settings combined with itera-tive reconstruction) as a replacement of yearly follow-up posteroanterior chest radiography. Verbal informed con-sent was obtained from all patients or their parents. CT images were randomized and rated independently by two radiologists with use of the Bhalla scoring system. In ad-dition, mosaic perfusion was evaluated. As reference, the previous available conventional chest CT scan was used. Differences in Bhalla scores were assessed with the x2 test and intraclass correlation coefficients (ICCs). Radiation doses for CT and radiography were assessed for adults (.18 years) and children (,18 years) separately by using technical dose descriptors and estimated effective dose. Differences in dose were assessed with the Mann-Whit-ney U test.

Results: The median effective dose for the investigative protocol was 0.04 mSv (95% confidence interval [CI]: 0.034 mSv, 0.10 mSv) for children and 0.05 mSv (95% CI: 0.04 mSv, 0.08 mSv) for adults. These doses were much lower than those with conventional CT (median: 0.52 mSv [95% CI: 0.31 mSv, 3.90 mSv] for children and 1.12 mSv [95% CI: 0.57 mSv, 3.15 mSv] for adults) and of the same order of magnitude as those for conventional radiography (me-dian: 0.012 mSv [95% CI: 0.006 mSv, 0.022 mSv] for children and 0.012 mSv [95% CI: 0.005 mSv, 0.031 mSv] for adults). All images were rated at least as diagnosti-cally acceptable. Very good agreement was found in over-all Bhalla score (ICC, 0.96) with regard to the severity of bronchiectasis (ICC, 0.87) and sacculations and abscesses (ICC, 0.84). Interobserver agreement was excellent (ICC, 0.86–1).

Conclusion: For patients with CF, a dedicated chest CT protocol can replace the two yearly follow-up chest radiographic ex-aminations without major dose penalty and with similar diagnostic quality compared with conventional CT.

q RSNA, 2014

Caroline W. Ernst, MDInes A. Basten, MDBart Ilsen, MDNico Buls, PhDGert Van Gompel, PhDElke De Wachter, MDKoenraad H. Nieboer, MDFilip Verhelle, MScAnne Malfroot, PhD, MDDanny Coomans, PhDMichel De Maeseneer, PhD, MDJohan de Mey, PhD, MD

Pulmonary Disease in cystic Fibrosis: Assessment with Chest CT at Chest Radiography Dose Levels1

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

THORACIC IMAGING: Pulmonary Disease in Cystic Fibrosis Ernst et al

598 radiology.rsna.org n Radiology: Volume 273: Number 2—November 2014

Materials and Methods

Study PopulationThis study was approved by the institu-tional review board. Verbal informed consent was obtained from all patients or their parents in the case of minors. At our institution, all patients with CF un-dergo a routine yearly assessment that in-cludes biannual chest CT alternated with biannual chest radiography and annual PFT with a Jaeger whole-body plethys-mograph (Viasys Healthcare, Höchberg, Germany). Annual assessment is always performed when the condition of the pa-tient is stable—never during an exacer-bation. During an 8-month period (April to December 2011), consecutive patients who presented for their yearly evaluation and who had undergone routine chest CT in one of the previous 3 years were in-cluded in this single-center study.

Routine and Investigative Chest CT ProtocolsFor all patients, tidal breathing CT was performed from lung apex to lung base without the use of intravenous contrast material.

damage (12,13). Multiple investigators have shown that CT is more accurate than FEV1 (14–20) and chest radiogra-phy (21) in the detection and follow-up of early lung disease. CT enables early detection of clinically relevant patho-logic changes (eg, bronchiectasis, peri-bronchial thickening, mucus plugging, and emphysema), which can be rated with a scoring system (eg, Bhalla et al [22] and Brody [23]). These find-ings are used as the basis for clini-cal decision making in early stages of disease and in patients with newly diag-nosed CF (14–20,22).

Patients with CF undergo chest CT from a young age and are consequently exposed to substantial radiation doses. Owing to the increased radiation sen-sitivity of children, numerous follow-up images, and increased life expectancy, the chance of a radiation-induced ab-normality in patients with CF is of con-cern (24).

Hence, cumulative radiation dose is an important consideration and re-strictive factor in the use of CT for this patient group (24–30). Multiple proto-cols have been proposed to reduce this radiation dose. For example, Jiménez et al (31) and O’Connor et al (32) used a limited number of thin CT sections.

At our specialized CF center, the current follow-up strategy consists of two yearly chest CT examinations alter-nated with conventional chest radiogra-phy the other year.

The purpose of this study was to investigate a CT protocol with iterative reconstruction at conventional radi-ography dose level for the assessment of structural lung abnormalities in pa-tients with CF.

Published online before print10.1148/radiol.14132201 Content code:

Radiology 2014; 273:597–605

Abbreviations:CF = cystic fibrosisCI = confidence intervalCTDIvol = volume CT dose indexDLP = dose-length productFEV1 = forced expiratory volume in 1 secondICC = intraclass correlation coefficientPFT = pulmonary function testingSSDE = size-specific dose estimate

Author contributions:Guarantors of integrity of entire study, C.W.E., I.A.B., B.I., E.D.W., F.V., A.M., D.C., J.d.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, C.W.E., I.A.B., B.I., D.C., M.D.M., J.d.M.; clinical studies, C.W.E., I.A.B., B.I., E.D.W., K.H.N., F.V., A.M., D.C., J.d.M.; statistical analysis, I.A.B., B.I., N.B., D.C., M.D.M.; and manuscript editing, C.W.E., I.A.B., B.I., N.B., G.V.G., E.D.W., K.H.N., F.V., D.C., M.D.M., J.d.M.

Conflicts of interest are listed at the end of this article.

Advances in Knowledge

n A scanning protocol with 80 kVp, 4 mAs, and a volume CT dose index (CTDIvol) of 0.09 mGy in children and 0.11 mGy in adults was applied in consecutive patients with cystic fibrosis (CF) who presented for their annual check-up, resulting in patient doses in the range of 0.04 mSv for children and 0.05 mSv for adults.

n No difference was reported in overall Bhalla score between reduced-dose CT (CTDIvol: 0.09 mGy) and a conventional chest CT protocol (CTDIvol: 1.13 mGy).

Implications for Patient Care

n Radiation doses in chest CT can be reduced dramatically.

n Findings suggest that, in CF, fol-low-up chest radiography might be replaced by chest CT without important dose penalty.

n Diagnostic quality for patients with CF can be maintained with use of a lower dose protocol than that used with conventional CT.

Cystic fibrosis (CF) is one of the most common genetic disorders in the white population, with a

lethal outcome at a young age. Mor-bidity and mortality in patients with CF are essentially related to pulmo-nary disease (1–3). Early detection and follow-up of lung disease in CF is crucial to allow prompt treatment adaptation (4). Early changes in lung disease can be detected with various methods, such as pulmonary function testing (PFT), radiography, and com-puted tomography (CT) (4–7). PFT, by providing measurements of the forced expiratory volume in 1 second (FEV1), can adequately reflect the de-gree of lung damage in moderate and severe lung disease. However, in mild and early stage lung disease, FEV1 is usually normal and of limited use. Moreover, PFT is heavily influenced by patient cooperation, yielding unre-liable results for children younger than 6 years. Advancements in treatment currently allow for a reduction in the decrease in lung function to less than 1% a year. These subtle decreases in lung function cannot be assessed with PFT (8,9).

The current North American Cystic Fibrosis Foundation guidelines (10) and the European consensus on CF stan-dards of care (11) recommend, in ad-dition to PFT, a yearly follow-up chest radiographic examination to evaluate various morphologic parameters of lung



used for all analyses. Patient charac-teristics were described according to mean age and range. The difference in age distribution between male and fe-male patients was assessed with a non-parametric Mann-Whitney U test. For each category of morphologic change, the difference between protocols A and B was assessed by comparison of proportions with a x2 test. If small numbers invalidated the x2 approxima-tion, the Fisher exact test was used. The difference in FEV1 was assessed with a Wilcoxon signed rank test. P , .05 was considered indicative of a sig-nificant difference.

Intraclass correlation coefficients (ICCs) and Bland-Altman plots were used to evaluate the agreement be-tween both Bhalla scores as well as to evaluate interobserver agreement. ICC values of 0.41–0.60 were indicative of moderate agreement; 0.61–0.80, good agreement; and 0.81 or more, very good agreement (38). For the dosime-try data, normality was investigated by using a Shapiro-Wilk test at a signifi-cance level of = 0.05. In case of non-normality, the nonparametric Mann-Whitney U test was used.

Results

Thirty-eight patients (22 male patients aged 6–50 years [mean age 6 standard deviation, 22 years 6 13]; 16 female patients aged 6–58 years [mean age, 19 years 6 15]) were included in the study. There was no difference in age distribution between male and female patients (P = .39). Twenty-three pa-tients had mild disease, nine had mod-erate disease, and six were severely affected by CF. FEV1 did not change significantly (P = .18) in the time be-tween protocol A (median: 90%; 95% confidence interval [CI]: 75%, 91%) and protocol B (median: 91%; 95% CI: 73%, 89%). The median effective diameters for both protocols and for children (age ,18 years) and adults (age .18 years) are shown in Table 1. Eleven patients had chronic Pseudo-monas aeruginosa colonization. Chest CT was performed in 2010–2011 in 27

bronchiectasis, peribronchial thicken-ing, mucus plugging, sacculations and abscesses, bullae, emphysema, col-lapse, and consolidation. A score of zero indicates that no abnormalities are depicted. The sum of the subscores of each category provides a total score, with the highest possible score being 25. The presence or absence of mosaic perfusion was also evaluated.

Dose EvaluationBoth standard technical dose descrip-tors and estimated effective dose were assessed in this study. To quantify the radiation dose difference between pro-tocols A and B, the DLP and CTDIvol were recorded from the dose reports of all patients. The estimated effective dose was calculated by CT dosimetry software (ImPACT, version 1.0.4; Im-PACT, St George’s Healthcare NHS, London, England) by using the organ-weighing scheme of the International Commission on Radiological Protec-tion (33). For patients younger than 18 years, a radiation dose correction fac-tor was used to obtain normalized pe-diatric doses (34). In addition, the size-specific dose estimate (SSDE), which is an estimate of the individual absorbed dose, was computed for both the CTDI-

vol (SSDE CTDIvol) and DLP (SSDE DLP) on the basis of patient size (anteropos-terior and lateral dimensions) (35). Linear regression models were used to assess the dependence of CTDIvol and SSDE on patient size (35–37).

The effective doses from chest radi-ography performed at follow-up in the year 2009 were estimated by using x-ray dosimetry software (PCXMC, ver-sion 2.0; Stuk, Helsinki, Finland) and based on the recorded dose-area prod-uct, patient size, and technical and geo-metric exposure parameters.

Although assessment and interpre-tation of effective doses from medical exposures can be problematic, it can be of value for comparing the relative doses from different procedures for the same patient groups (33).

Statistical AnalysisStatistical software (SPSS, version 19.0 for Windows; SPSS, Chicago, Ill) was

Routine CT data acquisition (pro-tocol A) was performed with one of two 64–detector row CT scanners (Brilliance CT 64 [Philips, Best, the Netherlands] and Discovery 750 HD [GE Healthcare; Milwaukee, Wis]). For the Philips system, scans were ob-tained with the following parameters: rotation time, 0.5 second; tube volt-age, 80 kVp; automatic tube current modulation; and section thickness, 0.67 mm. For the GE system, scans were obtained with the following pa-rameters: rotation time, 0.4 second; tube voltage, 80 kVp; automatic tube current modulation; and section thick-ness, 0.625 mm.

Data acquisition for investigative protocol B was performed with the Dis-covery 750HD unit using the following parameters: rotation time, 0.4 second; tube voltage, 80 kVp; minimal fixed tube current, 10 mA; and section thickness, 0.625 mm. CT scans acquired for pro-tocol B were reconstructed by means of a model-based iterative reconstruction technique (VEO, GE Healthcare). Vol-ume CT dose index (CTDIvol) and dose-length product (DLP) were recorded from the dose report of the patient af-ter the examination.

Image AnalysisThe chest CT reconstructions were anonymized and reviewed with a pic-ture archiving and communications system (Impax 6.4.0; Agfa Healthcare, Mortsel, Belgium) on 3-megapixel viewing stations. Images were evalu-ated at lung window settings (window width, 1600 HU; window level, 2600 HU). Two radiologists independently read the studies (B.I. and C.W.E., with 8 and 14 years of experience, respectively, in chest CT of CF). The readings of CT scans from protocol A and protocol B were separated by at least 4 weeks to obviate recall effects. The readers were blinded to all clinical information.

Structural lung abnormalities on CT scans were assessed by using the Bhalla scoring system (31,32). This scoring system includes evaluation of pattern and severity of the following categories of morphologic changes:



patients, in 2009 in eight patients, and in 2008 in three patients.

Figures 1 and 2 provide an illus-tration of image and diagnostic quality with the two protocols. The mean Bhal-la score was 8.06 6 5.70 (range, 1–18) for protocol A and 8.63 6 4.91 (range, 1–18) for protocol B.

Bland-Altman plots for the total Bhalla score (Fig 3) showed that the differences in scores between proto-cols A and B were independent of the magnitude of the scores. No signifi-cant differences were found between protocol A and protocol B with regard to the detection of severity of bronchi-ectasis (P = .83), peribronchial thick-ening (P = .91), extent of bronchiecta-sis (P = .74), extent of mucus plugging (P = .18), sacculations or abscesses (P = .55), generations of bronchial divisions involved (P = .19), number of bullae (P = .74), and collapse or

Figure 1

Figure 1: CT scans obtained with (a, b) protocol A and (c, d) protocol B in 50-year-old woman with CF show small saccular bronchiectasis in both upper lobes and in apical segment of lower lobes. Mucus plugging is present in apical segment of both lower lobes. Bhalla score was 14 with protocol A and 15 with protocol B.

Table 1

Dose Results and Effective Diameters for CT Protocols A and B and Chest Radiography in Children and Adults

Parameter Children (,18 y) Adults (.18 y)

Protocol A CTDIvol (mGy) 1.13 (0.62, 7.60) 2.25 (1.17, 5.05) DLP (mGy · cm) 33.52 (15.56, 275.37) 80.35 (40.39, 225.09) SSDE CTDIvol (mGy) 1.74 (1.19, 10.89) 3.12 (1.71, 6.42) SSDE DLP (mGy · cm) 55.58 (29.88, 394.49) 115.03 (59.22, 285.81) Estimated effective dose (mSv) 0.52 (0.31, 3.90) 1.12 (0.57, 3.15) Effective diameter (cm) 22.1 (18.55, 26.87) 27.1 (22.90, 29.44)Protocol B CTDIvol (mGy) 0.09 (0.09, 0.22) 0.11 (0.10, 0.14) DLP (mGy · cm) 2.69 (2.01, 4.96) 3.52 (3.07, 4.83) SSDE CTDIvol (mGy) 0.15 (0.12, 0.42) 0.14 (0.13, 0.21) SSDE DLP (mGy · cm) 3.93 (3.42, 9.34) 4.76 (4.36, 7.06) Estimated effective dose (mSv) 0.04 (0.03, 0.10) 0.05 (0.04, 0.08) Effective diameter (cm) 22.38 (19.04, 28.08) 26.85 (22.35, 29.33)Estimated effective dose at chest radiography (mSv) 0.012 (0.006, 0.022) 0.012 (0.005, 0.031)

Note.—Data are medians, with 95% CIs in parentheses.



CTDIvol, SSDE CTDIvol, SSDE DLP, and estimated effective doses, along with the 95% CIs, are listed for protocols A and B and for the two age groups (chil-dren and adults).

The median CTDIvol for children was 0.09 mGy (95% CI: 0.09 mGy, 0.22 mGy) for protocol B and 1.13 mGy (95% CI: 0.62 mGy, 7.60 mGy) for protocol A, resulting in a radiation dose reduction of 92%. A similar re-duction (95%) in mean CTDIvol was found for adults.

The median DLP for protocol B was significantly lower (P , .01) than that for protocol A (2.69 mGy · cm

consolidation. Moderate agreement (0.4 , ICC , 0.6) was observed for peribronchial thickening. There was excellent interobserver agreement, in which CT protocol A performed slightly better than CT protocol B (Table 3).

In 34 of the 38 patients (89%), mo-saic perfusion was evident with both protocol A and protocol B (Fig 4). Mo-saic perfusion was present only with protocol B in two patients (5%) and was absent with both protocols A and B in two other patients (5%).

The dosimetry data were not nor-mally distributed (Shapiro-Wilk, P , .05). In Table 1, the median DLP,

consolidation (P = .31). There was no evidence of emphysema in our patient group.

Table 2 lists the ICCs between the Bhalla scores from protocol B versus protocol A. Very good ICC agreement (ICC . 0.8) was found between pro-tocol A and protocol B for the overall Bhalla score as well as for the detec-tion of severity of bronchiectasis, ex-tent of bronchiectasis, sacculations or abscesses, and generations of bronchial divisions involved. Good agreement (0.6 , ICC , 0.8) was found for the detection of number of bullae, extent of mucus plugging, and collapse or

Figure 2

Figure 2: CT scans obtained with (a, b) protocol A and (c, d) protocol B in 28-year-old woman with CF. Mucus plugging is predominantly present in posterior segment of left upper lobe in d. Middle lobe collapse with cystic bronchiectasis involving the whole bronchial tree is seen in a–c. Note important central and peripheral mucus plugging in both lower lobes in c. Saccular bronchiectasis is seen in upper and lower lobes and cystic bronchiectasis in middle and lower lobes in b–d. Mucus plugging is predominantly present in posterior segment of left upper lobe in d. Bhalla score was 17 for protocol A and 18 for protocol B.



mGy, 23.06 mGy) with protocol A and 0.15 mGy (95% CI: 0.12 mGy, 0.42 mGy) with protocol B. The me-dian SSDE DLP was 55.58 mGy · cm (95% CI: 29.88 mGy · cm, 349.49 mGy · cm) with protocol A and 3.93 mGy · cm (95% CI: 3.42 mGy · cm, 9.3 mGy · cm) for protocol B.

For adults, the median SSDE CT-DIvol was 42.14 mGy (95% CI: 28.39 mGy, 79.85 mGy) for protocol A and 0.14 mGy (95% CI: 0.13 mGy, 0.21 mGy) for protocol B. The SSDE DLP was 115.03 mGy · cm (95% CI: 59.22 mGy · cm, 285.81 mGy · cm) for pro-tocol A and 4.76 mGy · cm (95% CI: 4.36 mGy · cm, 7.06 mGy · cm) for protocol B.

Discussion

The qualitative study results indicate that an almost-perfect agreement in terms of ICC is found between protocol A and protocol B for the detection of bronchiectasis, generations of bronchial divisions involved, extent of bronchi-ectasis, and sacculations or abscesses in patients with a wide range of lung disease, as shown by their PFT (FEV1) results. However, the mean FEV1 of the study group was normal, which reflects its inability to reveal subtle or mild lung disease at PFT, as stressed by other in-vestigators (8,9).

The extent of mucus plugging and collapse or consolidation and the number of bullae showed good agreement with the previous CT scan. Peribronchial thickening showed mod-erate agreement with the previous CT scan. According to a recent study, however, the latter is a reversible CT feature caused by fibroinflammatory infiltrate in and around the bronchi. Healing is accompanied by scarring of the bronchial wall and surrounding al-veoli, resulting in an increase in mural thickness because of the accumulation of collagen in the adventitia (39). This cause could, however, not be con-firmed in our study because no addi-tional acquisition with protocol A was performed at the same time as pro-tocol B owing to ethical reasons. An-nual assessment is always performed

0.005 mSv, 0.031 mSv) for adults. These doses are of the same order of magnitude and much lower than those with protocol A (median, 1.13 mSv [95% CI: 0.62 mSv, 7.60 mSv] for children and 2.25 mSv [95% CI: 1.17 mSv, 5.05 mSv] for adults). In cor-respondence with the results for the mean CTDIvol, this reflects a reduction in estimated effective dose of protocol B compared with protocol A of 92% for children and 95% for adults.

For children, the median SSDE CTDIvol was 18.2 mGy (95% CI: 9.8

and 33.52 mGy · cm, respectively, in children and 3.52 mGy · cm and 80.35 mGy · cm in adults). Figure 5 shows the SSDE DLP distribution in children and adults with both protocols.

The median estimated effective dose with protocol B was 0.04 mSv (95% CI: 0.03 mSv, 0.1 mSv) for chil-dren and 0.05 mSv (95% CI: 0.04 mSv, 0.08 mSv) for adults, and the mean estimated effective dose for conventional radiography was 0.012 mSv (95% CI: 0.006 mSv, 0.022 mSv) for children and 0.012 mSv (95% CI:

Figure 3

Figure 3: Agreement of Bhalla score between protocols B and A at expert analysis. Bland-Altman plot shows difference in total Bhalla score between protocols B and A as a function of average Bhalla score. Dotted lines represent average difference 6 95% CI.

Table 2

Agreement between Bhalla Scores with Protocol B versus Protocol A

Category ICC

Overall score 0.961 (0.926, 0.980)Severity of bronchiectasis 0.870 (0.803, 0.916)Peribronchial thickening 0.620 (0.459, 0.741)Extent of bronchiectasis 0.882 (0.820, 0.924)Extent of mucus plugging 0.801 (0.704, 0.869)Sacculations or abscesses 0.840 (0.758, 0.895)Generations of bronchial divisions involved 0.812 (0.719, 0.877)No. of bullae 0.807 (0.711, 0.873)Collapse or consolidation 0.635 (0.479, 0.752)

Note.—Numbers in parentheses are 95% CIs.



The median estimated effective dos-es for 38 patients with CF who under-went chest CT protocol B as a replace-ment of follow-up chest radiography were as low as 0.04 mSv for children and 0.05 mSv for adults, which is of the same order of magnitude as the dose of conventional posteroanterior chest ra-diography in this study (0.012 mSv for both children and adults). The chest ra-diography doses in our center are in the lower range of internationally reported effective doses for conventional chest radiography, ranging from 0.02 to 0.299 mSv (29,41). Although not included in the standard follow-up protocol, pa-tients often undergo an additional lat-eral chest radiographic examination for diagnostic purposes—which increases the chest radiography dose.

The dose levels of protocol A in our study are relatively low compared with reported diagnostic reference levels of pediatric chest CT from national mul-ticenter studies (CTDIvol of 10 mGy in United Kingdom, 4.3 mGy in Germany, 5 mGy in Switzerland, and 5.5 mGy in France) (42).

Furthermore, the effective dose of protocol B is well below that of the dedicated CF low-dose protocols pro-posed in the literature (31,32), which used doses of 0.19 mSv and 4 mSv, respectively.

The main limitation of this study is the absence of evaluation of centrilobu-lar nodules and ground glass opacities because these features are reversible

protocol B and protocol A. Further-more, these results suggest that the Bhalla score may reveal abnormalities in patients with normal findings at PFT and that mosaic perfusion is often pre-sent in patients with CF and normal findings at PFT.

Previous studies demonstrated the superiority of chest CT to conventional chest radiography in terms of improved and additional diagnostic information such as bronchiectasis and mucus plugging (22,30). Although the diag-nostic qualities of protocol B and chest radiography were not compared in this study, we assume that this superiority to conventional chest radiography is still valid for protocol B because pro-tocol B offers similar diagnostic quality for CF follow-up compared with stan-dard chest CT protocol A.

when the condition of the patient is stable—never during an exacerbation. With use of this approach, transient lung changes between the time of per-forming protocol A and protocol B may have been underestimated. Because emphysema was not present in any of our patients, no information concern-ing this feature can be provided.

Interobserver agreement calcu-lations in our study showed that the Bhalla scoring system is highly repro-ducible, supporting the results of pre-vious studies (22,40). Mean Bhalla scores for protocol B and protocol A matched closely (8.63 and 8.06, re-spectively). Mosaic perfusion pattern detection was also very similar with the two techniques. These results suggest that there is good agreement in terms of assessing disease severity between

Table 3

Interobserver Agreement for Bhalla Scores with Protocols B and A

Category Protocol B Protocol A

Overall score 0.987 (0.975, 0.993) 0.994 (0.988, 0.997)Severity of bronchiectasis 0.968 (0.940, 0.983) 0.986 (0.973, 0.993)Peribronchial thickening 0.861 (0.749, 0.925) 0.956 (0.917, 0.977)Extent of bronchiectasis 0.982 (0.966, 0.991) 0.995 (0.990, 0.997)Extent of mucus plugging 0.980 (0.962, 0.989) 0.972 (0.948, 0.985)Sacculations or abscesses 0.941 (0.889, 0,969) 0.972 (0.947, 0.986)Generations of bronchial divisions involved 0.951 (0.908, 0.974) 0.945 (0.897, 0.971)No. of bullae 1.000 (1, 1) 1.000 (1, 1)Collapse or consolidation 1.000 (1, 1) 1.000 (1, 1)

Note.—Data are ICCs. Numbers in parentheses are 95% CIs.

Figure 4

Figure 4: CT scans obtained with (a) protocol A and (b) protocol B in 10-year-old girl with CF show presence of zones of varying lung attenu-ation representing mosaic perfusion. Bhalla score was 11 for both protocols.



8. Döring G, Elborn JS, Johannesson M, et al. Clinical trials in cystic fibrosis. J Cyst Fibros 2007;6(2):85–99.

9. Que C, Cullinan P, Geddes D. Improving rate of decline of FEV1 in young adults with cystic fibrosis. Thorax 2006;61(2):155–157.

10. Cystic Fibrosis Foundation guidelines for patient services, evaluation, and monitor-ing in cystic fibrosis centers. The Cystic Fibrosis Foundation Center Committee and Guidelines Subcommittee. Am J Dis Child 1990;144(12):1311–1312.

11. Kerem E, Conway S, Elborn S, Heijerman H; Consensus Committee. Standards of care for patients with cystic fibrosis: a European consensus. J Cyst Fibros 2005;4(1):7–26.

12. Shale DJ. Chest radiology in cystic fibrosis: is scoring useful? Thorax 1994;49(9):847.

13. Langton Hewer SC. Is limited computed to-mography the future for imaging the lungs of children with cystic fibrosis? Arch Dis Child 2006;91(5):377–378.

14. 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(1):93–97.

15. de Jong PA, Lindblad A, Rubin L, et al. Progression of lung disease on computed tomography and pulmonary function tests in children and adults with cystic fibrosis. Thorax 2006;61(1):80–85.

16. Brody AS, Klein JS, Molina PL, Quan J, Bean JA, Wilmott RW. High-resolution com-puted tomography in young patients with cystic fibrosis: distribution of abnormalities and correlation with pulmonary function tests. J Pediatr 2004;145(1):32–38.

17. Brody AS, Sucharew H, Campbell JD, et al. Computed tomography correlates with pulmonary exacerbations in children with cystic fibrosis. Am J Respir Crit Care Med 2005;172(9):1128–1132.

18. Judge EP, Dodd JD, Masterson JB, Galla-gher CG. Pulmonary abnormalities on high-resolution CT demonstrate more rapid de-cline than FEV1 in adults with cystic fibrosis. Chest 2006;130(5):1424–1432.

19. Maffessanti M, Candusso M, Brizzi F, Pio-vesana F. Cystic fibrosis in children: HRCT findings and distribution of disease. J Tho-rac Imaging 1996;11(1):27–38.

20. Cademartiri F, Luccichenti G, Palumbo AA, et al. Predictive value of chest CT in patients with cystic fibrosis: a single-center 10-year experience. AJR Am J Roentgenol 2008;190(6):1475–1480.

21. Jacobsen LE, Houston CS, Habbick BF, Genereux GP, Howie JL. Cystic fibrosis: a

Disclosures of Conflicts of Interest: C.W.E. disclosed no relevant relationships. I.A.B. dis-closed no relevant relationships. B.I. disclosed no relevant relationships. N.B. Activities re-lated to the present article: disclosed no rel-evant relationships. Activities not related to the present article: is paid to be on the med-ical advisory board of GE Healthcare. Other relationships: disclosed no relevant relation-ships. G.V.G. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: receives fees from GE Heathcare for presen-tation content. Other relationships: disclosed no relevant relationships. E.D.W. disclosed no relevant relationships. K.H.N. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is paid to be on the med-ical advisory board of GE Healthcare Medical Diagnostics. Other relationships: disclosed no relevant relationships. F.V. Activities related to the present article: disclosed no relevant rela-tionships. Activities not related to the present article: receives payment to be on medical advisory board of GE Healthcare. Other rela-tionships: disclosed no relevant relationships. A.M. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: received an educational grant from Novartis; institution received payment from Abbott for develop-ment of educational presentations; institution received travel grants from Novartis, Abbott, and Gilead. Other relationships: disclosed no relevant relationships. D.C. disclosed no rele-vant relationships. M.D.M. disclosed no rel-evant relationships. J.d.M. disclosed no rele-vant relationships.

References 1. Davis PB. Cystic fibrosis since 1938. Am J

Respir Crit Care Med 2006;173(5):475–482.

2. Wood BP. Cystic fibrosis: 1997. Radiology 1997;204(1):1–10.

3. Ruzal-Shapiro C. Cystic fibrosis: an overview. Radiol Clin North Am 1998;36(1):143–161.

4. Weiser G, Kerem E. Early intervention in CF: how to monitor the effect. Pediatr Pul-monol 2007;42(11):1002–1007.

5. Moskowitz SM, Gibson RL, Effmann EL. Cystic fibrosis lung disease: genetic in-fluences, microbial interactions, and ra-diological assessment. Pediatr Radiol 2005;35(8):739–757.

6. Gibson RL, Burns JL, Ramsey BW. Patho-physiology and management of pulmonary infections in cystic fibrosis. Am J Respir Crit Care Med 2003;168(8):918–951.

7. Ramsey BW, Boat TF. Outcome measures for clinical trials in cystic fibrosis: summary of a Cystic Fibrosis Foundation consensus conference. J Pediatr 1994;124(2):177–192.

findings mostly present during an acute exacerbation and annual assessment is never performed during an acute exac-erbation (39). In addition, air trapping, a CT feature typically depicted at expi-ration CT, could not be scored in this study because tidal breathing CT was performed in all patients for both pro-tocols. Furthermore, this study was per-formed with a relatively small sample size; hence, a larger scale study must be set up to confirm the presented results.

Protocol B is specifically designed for CF follow-up and only diagnostically evaluated for CF follow-up; hence, this protocol is not advised to replace the two yearly conventional chest CT exam-inations (out of precaution) or pediatric chest CT in general.

In conclusion, the results of this study show that a dedicated CF chest CT protocol can replace the two yearly CF follow-up chest radiographic exami-nations without major dose penalty and with similar diagnostic quality com-pared with conventional CT.

Figure 5

Figure 5: Box and whisker plots of SSDE DLP for both protocols in adults (.18 years, white boxes) and children (,18 years, gray boxes) with a loga-rithmic scale. Solid line in box represents median value, and upper and lower bars represent first and third quartiles, respectively. Whiskers represent 95% CIs. = outliers.



comparison of computed tomography and plain chest radiographs. Can Assoc Radiol J 1986;37(1):17–21.

22. Bhalla M, Turcios N, Aponte V, et al. Cys-tic fibrosis: scoring system with thin-section CT. Radiology 1991;179(3):783–788.

23. Brody AS. Scoring systems for CT in cystic fibrosis: who cares? Radiology 2004;231(2): 296–298.

24. Mathews JD, Forsythe AV, Brady Z, et al. Cancer risk in 680,000 people exposed to computed tomography scans in childhood or adolescence: data linkage study of 11 million Australians. BMJ 2013;346:f2360.

25. Brenner DJ, Hall EJ. Computed tomography: an increasing source of radiation exposure. N Engl J Med 2007;357(22):2277–2284.

26. de Jong PA, Mayo JR, Golmohammadi K, et al. Estimation of cancer mortality asso-ciated with repetitive computed tomogra-phy scanning. Am J Respir Crit Care Med 2006;173(2):199–203.

27. Frederiksen B, Lanng S, Koch C, Høiby N. Improved survival in the Danish center-treated cystic fibrosis patients: results of aggressive treatment. Pediatr Pulmonol 1996;21(3):153–158.

28. Donadieu J, Roudier C, Saguintaah M, Maccia C, Chiron R. Estimation of the radiation dose from thoracic CT scans in a cystic fibrosis population. Chest 2007;132(4):1233–1238.

29. Terheggen-Lagro SWJ, Arets HG, van der Laag J, van der Ent CK. Radiological and

functional changes over 3 years in young children with cystic fibrosis. Eur Respir J 2007;30(2):279–285.

30. Mettler FA Jr, Huda W, Yoshizumi TT, Ma-hesh M. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Ra-diology 2008;248(1):254–263.

31. Jiménez S, Jiménez JR, Crespo M, Sant-amarta E, Bousoño C, Rodríguez J. Com-puted tomography in children with cystic fi-brosis: a new way to reduce radiation dose. Arch Dis Child 2006;91(5):388–390.

32. O’Connor OJ, Vandeleur M, McGarrigle AM, et al. Development of low-dose proto-cols for thin-section CT assessment of cys-tic fibrosis in pediatric patients. Radiology 2010;257(3):820–829.

33. The 2007 Recommendations of the Inter-national Commission on Radiological Pro-tection. ICRP publication 103. Ann ICRP 2007;37(2-4):1–332.

34. Khursheed A, Hillier MC, Shrimpton PC, Wall BF. Influence of patient age on normal-ized effective doses calculated for CT exam-inations. Br J Radiol 2002;75(898):819–830.

35. American Association of Physicists in Medicine. Size-specific dose estimates in pediatric and adult body CT examina-tions. American Association of Physicists in Medicine Report 204. College Park, Md: American Association of Physicists in Medicine, 2011.

36. Brady SL, Kaufman RA. Investigation of American Association of Physicists in Medi-

cine Report 204 size-specific dose estimates for pediatric CT implementation. Radiology 2012;265(3):832–840.

37. Christner JA, Braun NN, Jacobsen MC, Carter RE, Kofler JM, McCollough CH. Size-specific dose estimates for adult patients at CT of the torso. Radiology 2012;265(3):841–847.

38. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1):159–174.

39. Shah RM, Sexauer W, Ostrum BJ, Fiel SB, Friedman AC. High-resolution CT in the acute exacerbation of cystic fibrosis: evalu-ation of acute findings, reversibility of those findings, and clinical correlation. AJR Am J Roentgenol 1997;169(2):375–380.

40. Helbich TH, Heinz-Peer G, Fleischmann D, et al. Evolution of CT findings in patients with cystic fibrosis. AJR Am J Roentgenol 1999;173(1):81–88.

41. Zenone F, Aimonetto S, Catuzzo P, et al. Effective dose delivered by conven-tional radiology to Aosta Valley popula-tion between 2002 and 2009. Br J Radiol 2012;85(1015):e330–e338.

42. Thomas KE. CT utilization—trends and devel-opments beyond the United States’ borders. Pediatr Radiol 2011;41(Suppl 2):562–566.

Documents

Pulmonary Disease in Cystic Fibrosis: Assessment with Chest CT at Chest Radiography Dose Levels