Pediatr Radiol DOI 10.1007/s00247-014-3112-z
ORIGINAL ARTICLE
Pulmonary radiographic findings in young children with cystic fibrosis Jesiana Ferreira Pedrosa & Cássio da Cunha Ibiapina & Cristina Gonçalves Alvim & Paulo Augusto Moreira Camargos & Fabiana Paiva Martins & Elizabet Vilar Guimarães & Renata Marcos Bedran
Received: 7 February 2014 / Revised: 19 May 2014 / Accepted: 29 June 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Pulmonary evaluation is one of the greatest challenges in children with cystic fibrosis who are younger than 6 years. Although chest CT can provide the most information for initial and progressive changes in cystic fibrosis, the radiation exposures can lead to significant cumulative exposure in children as they are followed with serial scanning to monitor early and progressive disease. Therefore the systematized study of chest radiographs using the Brasfield score has been used scientifically with the aim of evaluating the evolution of pulmonary abnormalities in children with cystic fibrosis. Objective This study was performed to assess the radiologic findings in children younger than 6 years with cystic fibrosis. We used the Brasfield score to compare radiographs performed in patients with Pseudomonas aeruginosa vs. oxacillin-sensitive Staphylococcus aureus and to compare radiographs in children with early vs. delayed diagnosis. Materials and methods A total of 254 chest radiographs from 67 children who had undergone material cultures of the airways as part of routine care were evaluated in this crosssectional study. The statistical analysis was performed by the Kruskal–Wallis test, with a significance level of 5%. Results Approximately 35.8% of chest radiographs had a Brasfield score lower than 21 points, which is compatible with potentially irreversible pulmonary disease. Brasfield scores decreased (this score decreases with increasing disease severity) in older children, and both bacterial colonization with Pseudomonas and later diagnosis were associated with lower (greater disease) scores. J. F. Pedrosa (*) : C. da Cunha Ibiapina : C. G. Alvim : P. A. M. Camargos : F. P. Martins : E. V. Guimarães : R. M. Bedran Departamento de Anatomia e Imagem (IMA), Faculdade de Medicina, Universidade Federal de Minas Gerais, Av. Prof. Alfredo Balena, 190, Belo Horizonte/MG, Minas Gerais 30.310-100, Brazil e-mail: [email protected]
Conclusion The evaluation of radiographs using the Brasfield score demonstrated the most important pulmonary findings in cystic fibrosis and identified the age group when these alterations began to appear more pronounced. Keywords Cystic fibrosis . Chest radiography . Brasfield score . Bacterial infection . Children
Introduction Cystic fibrosis (CF) lung disease begins in early childhood. Evidence of the presence of potentially irreversible pulmonary findings is detected in the chest radiographs of up to 50% of children by the age of 2 years [1–3]. The pulmonary evaluation of children younger than 6 years is limited, particularly in low–middle income settings, because of the unavailability of spirometers and chest CT, the difficulty in gaining the cooperation of young children when performing spirometry and the reluctance to expose young children to the higher dose of ionizing radiation attributed to chest CT. Chest radiography has been the most commonly used tool for evaluating pulmonary disease in this age group, together with the microbiological monitoring of the airway [1–3]. The systematized study of chest radiographs by radiographic scoring has allowed the objective analyses of pulmonary lesions and the comparison of disease severity among patients. Authors have recently shown that quantitative chest radiology is the best procedure for frequent assessment of bronchopulmonary disease in CF [4]. The Brasfield score was created in 1979 and has been extensively used scientifically with the aim of evaluating the evolution of pulmonary abnormalities in children with CF. The Brasfield score is simple, has a high degree of intra- and inter-observer reproducibility, and correlates with the child’s clinical status and with pulmonary function tests [5]. Greater
Pediatr Radiol
knowledge of the disease, early diagnosis, follow-up of patients with the aim of controlling pulmonary infections, and appropriate weight gain have contributed significantly to increasing patient survival [6–12]. Neonatal screening is important for early diagnosis and follow-up care. This study evaluates the evolution of radiologic findings in children with CF during the first 6 years of age by using the Brasfield score model to analyze the differences in the mean score when comparing the species of the bacteria infecting the lung as well as the timing of the diagnosis.
Materials and methods The Research Ethics Committee of the Federal University of Minas Gerais approved this study. We obtained authorization from patients or families to review the radiographs included in the study. In this cross-sectional study chest radiographs of children from a CF referral center were evaluated by Brasfield score. CF was diagnosed by a sweat test in all of the children. This study included children with an early diagnosis by neonatal screening, which was initiated at the CF center in 2004, and those who had been diagnosed prior to 2004 on clinical grounds presenting after the neonatal period. Poor-quality radiographs were excluded. If two radiographs from the same child were obtained fewer than 15 days apart, the second radiograph was excluded. The data collection, including clinical data records and radiographs, was performed between July 2011 and July 2012. The chest radiographs had been obtained according to the out-patient CF clinic routine, i.e. at least one per year and during exacerbations. A complete evaluation of all radiographs was made by two radiologists with 18 and 8 years of experience in pediatric radiology to determine the most appropriate results for each radiograph. The radiologists were blinded to patient history (infection history and early vs. late diagnosis). The Brasfield score is divided into five categories representing the characteristic radiologic aspects of cystic fibrosis: air trapping, linear markings, nodular-cystic lesions, large lesions and general severity. The first three radiologic findings are classified according to severity from 0 to 4. Large lesions can be classified as follows: 0 if there are no large lesions; 3 in the presence of a single, segmental or lobar atelectasis or consolidation; or 5 in the presence of multiple atelectasis or consolidations. General severity is a subjective score item that allows the observer to classify from 0 to 5 considering the overall radiography alterations, reserving the score of 5 for cases presenting with pneumothorax, cardiomegaly or other complications arising from the severity of the pulmonary disease. The five categories are added together and then subtracted from 25 to give the final score (3 is the most severe score) [5].
The results from sample airway cultures were collected from the medical records. The cultures were obtained according to the service routine, i.e. from sputum whenever possible and from tracheal aspirate when the child was unable to expectorate or did not have phlegm. We included data from the date of the first cultures positive for P. aeruginosa and oxacillin-sensitive S. aureus. We included the total number of positive cultures for each one of these bacteria, P. aeruginosa and oxacillin-sensitive S. aureus, up to the date of the chest radiograph and classified them as 0, 1, 2 and 3 or more. As proposed by Canton et al. [11], three or more positive culture categories separated by an interval of greater than 1 month and within 6 months was defined as chronic colonization [11]. The analyzed variables were the Brasfield score and its five categories, the age of the child upon chest radiography, the number of positive cultures up to the time of the chest radiograph, and the timing of the diagnosis (early or delayed). Comparisons of Brasfield score and its categories with the other variables — age group of the children, early or delayed diagnosis, number of P. aeruginosa and oxacillin-sensitive S. aureus cultures — were performed using the Kruskal– Wallis non-parametric test [8]. The results were considered significant at P
ORIGINAL ARTICLE
Pulmonary radiographic findings in young children with cystic fibrosis Jesiana Ferreira Pedrosa & Cássio da Cunha Ibiapina & Cristina Gonçalves Alvim & Paulo Augusto Moreira Camargos & Fabiana Paiva Martins & Elizabet Vilar Guimarães & Renata Marcos Bedran
Received: 7 February 2014 / Revised: 19 May 2014 / Accepted: 29 June 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Pulmonary evaluation is one of the greatest challenges in children with cystic fibrosis who are younger than 6 years. Although chest CT can provide the most information for initial and progressive changes in cystic fibrosis, the radiation exposures can lead to significant cumulative exposure in children as they are followed with serial scanning to monitor early and progressive disease. Therefore the systematized study of chest radiographs using the Brasfield score has been used scientifically with the aim of evaluating the evolution of pulmonary abnormalities in children with cystic fibrosis. Objective This study was performed to assess the radiologic findings in children younger than 6 years with cystic fibrosis. We used the Brasfield score to compare radiographs performed in patients with Pseudomonas aeruginosa vs. oxacillin-sensitive Staphylococcus aureus and to compare radiographs in children with early vs. delayed diagnosis. Materials and methods A total of 254 chest radiographs from 67 children who had undergone material cultures of the airways as part of routine care were evaluated in this crosssectional study. The statistical analysis was performed by the Kruskal–Wallis test, with a significance level of 5%. Results Approximately 35.8% of chest radiographs had a Brasfield score lower than 21 points, which is compatible with potentially irreversible pulmonary disease. Brasfield scores decreased (this score decreases with increasing disease severity) in older children, and both bacterial colonization with Pseudomonas and later diagnosis were associated with lower (greater disease) scores. J. F. Pedrosa (*) : C. da Cunha Ibiapina : C. G. Alvim : P. A. M. Camargos : F. P. Martins : E. V. Guimarães : R. M. Bedran Departamento de Anatomia e Imagem (IMA), Faculdade de Medicina, Universidade Federal de Minas Gerais, Av. Prof. Alfredo Balena, 190, Belo Horizonte/MG, Minas Gerais 30.310-100, Brazil e-mail: [email protected]
Conclusion The evaluation of radiographs using the Brasfield score demonstrated the most important pulmonary findings in cystic fibrosis and identified the age group when these alterations began to appear more pronounced. Keywords Cystic fibrosis . Chest radiography . Brasfield score . Bacterial infection . Children
Introduction Cystic fibrosis (CF) lung disease begins in early childhood. Evidence of the presence of potentially irreversible pulmonary findings is detected in the chest radiographs of up to 50% of children by the age of 2 years [1–3]. The pulmonary evaluation of children younger than 6 years is limited, particularly in low–middle income settings, because of the unavailability of spirometers and chest CT, the difficulty in gaining the cooperation of young children when performing spirometry and the reluctance to expose young children to the higher dose of ionizing radiation attributed to chest CT. Chest radiography has been the most commonly used tool for evaluating pulmonary disease in this age group, together with the microbiological monitoring of the airway [1–3]. The systematized study of chest radiographs by radiographic scoring has allowed the objective analyses of pulmonary lesions and the comparison of disease severity among patients. Authors have recently shown that quantitative chest radiology is the best procedure for frequent assessment of bronchopulmonary disease in CF [4]. The Brasfield score was created in 1979 and has been extensively used scientifically with the aim of evaluating the evolution of pulmonary abnormalities in children with CF. The Brasfield score is simple, has a high degree of intra- and inter-observer reproducibility, and correlates with the child’s clinical status and with pulmonary function tests [5]. Greater
Pediatr Radiol
knowledge of the disease, early diagnosis, follow-up of patients with the aim of controlling pulmonary infections, and appropriate weight gain have contributed significantly to increasing patient survival [6–12]. Neonatal screening is important for early diagnosis and follow-up care. This study evaluates the evolution of radiologic findings in children with CF during the first 6 years of age by using the Brasfield score model to analyze the differences in the mean score when comparing the species of the bacteria infecting the lung as well as the timing of the diagnosis.
Materials and methods The Research Ethics Committee of the Federal University of Minas Gerais approved this study. We obtained authorization from patients or families to review the radiographs included in the study. In this cross-sectional study chest radiographs of children from a CF referral center were evaluated by Brasfield score. CF was diagnosed by a sweat test in all of the children. This study included children with an early diagnosis by neonatal screening, which was initiated at the CF center in 2004, and those who had been diagnosed prior to 2004 on clinical grounds presenting after the neonatal period. Poor-quality radiographs were excluded. If two radiographs from the same child were obtained fewer than 15 days apart, the second radiograph was excluded. The data collection, including clinical data records and radiographs, was performed between July 2011 and July 2012. The chest radiographs had been obtained according to the out-patient CF clinic routine, i.e. at least one per year and during exacerbations. A complete evaluation of all radiographs was made by two radiologists with 18 and 8 years of experience in pediatric radiology to determine the most appropriate results for each radiograph. The radiologists were blinded to patient history (infection history and early vs. late diagnosis). The Brasfield score is divided into five categories representing the characteristic radiologic aspects of cystic fibrosis: air trapping, linear markings, nodular-cystic lesions, large lesions and general severity. The first three radiologic findings are classified according to severity from 0 to 4. Large lesions can be classified as follows: 0 if there are no large lesions; 3 in the presence of a single, segmental or lobar atelectasis or consolidation; or 5 in the presence of multiple atelectasis or consolidations. General severity is a subjective score item that allows the observer to classify from 0 to 5 considering the overall radiography alterations, reserving the score of 5 for cases presenting with pneumothorax, cardiomegaly or other complications arising from the severity of the pulmonary disease. The five categories are added together and then subtracted from 25 to give the final score (3 is the most severe score) [5].
The results from sample airway cultures were collected from the medical records. The cultures were obtained according to the service routine, i.e. from sputum whenever possible and from tracheal aspirate when the child was unable to expectorate or did not have phlegm. We included data from the date of the first cultures positive for P. aeruginosa and oxacillin-sensitive S. aureus. We included the total number of positive cultures for each one of these bacteria, P. aeruginosa and oxacillin-sensitive S. aureus, up to the date of the chest radiograph and classified them as 0, 1, 2 and 3 or more. As proposed by Canton et al. [11], three or more positive culture categories separated by an interval of greater than 1 month and within 6 months was defined as chronic colonization [11]. The analyzed variables were the Brasfield score and its five categories, the age of the child upon chest radiography, the number of positive cultures up to the time of the chest radiograph, and the timing of the diagnosis (early or delayed). Comparisons of Brasfield score and its categories with the other variables — age group of the children, early or delayed diagnosis, number of P. aeruginosa and oxacillin-sensitive S. aureus cultures — were performed using the Kruskal– Wallis non-parametric test [8]. The results were considered significant at P
