© 2005 American Thoracic Society doi: 10.1164/rccm.2506004
Role of High-Resolution Computed Tomography in Idiopathic Pulmonary FibrosisThe Final Word?Massachusetts General Hospital, Boston, Massachusetts The information available from high-resolution computed tomography (HRCT) scans of the lung has changed our approach to the assessment of chronic infiltrative lung disease. Improved scanner resolution has provided lung parenchymal imaging at the level of the secondary pulmonary lobular level. Such images permit ready identification of the distribution and the types of opacities and other abnormalities identified in chronic interstitial disease, all of which play an important role in diagnosis. Since the advent of HRCT in the late 1980s, many studies have confirmed improved diagnostic accuracy as compared with standard radiographs. The majority of patients suspected of having a chronic infiltrative process in the lungs now undergo an HRCT during their initial investigation. It has been suggested that, in some cases, the imaging data combined with clinical information is adequate for diagnosis, and biopsy may not be required. Such "diagnostic" HRCT studies may occur in such entities as idiopathic pulmonary fibrosis (IPF), lymphangitic spread of carcinoma, and sarcoidosis.
Regarding IPF, identification of usual interstitial pneumonia (UIP) on surgical lung biopsy has been considered the gold standard for diagnosis (1). However, when typical clinical and high-resolution HRCT features are present and assessed by expert clinicians and radiologists, the features may allow a confident diagnosis in more than 50% of cases and eliminate the need for surgical lung biopsy (2). The article in this issue (pp. 488–493) by Lynch and coworkers from the IPF Study Group (3) addresses additional aspects of the important role of HRCT in the diagnosis and follow-up of patients with IPF. The authors specifically focus on the HRCT features of patients with mild to moderate physiologic impairment, the reliability of HRCT interpretation among radiologists both in academic and community settings, and the usefulness of HRCT findings as predictors of mortality (4, 5). This study is unique because of the size of the cohort (i.e., 315 patients with IPF who were enrolled in a randomized, controlled study evaluating the efficacy of IFN- One of the important findings of this study was the concordance between the interpretations of the study-site radiologists, which included radiologists from both academic and community centers, and a core group of expert thoracic radiologists. Concordance regarding the diagnosis of IPF occurred in 86% of the cases. Over the past decade there has been widespread dissemination of knowledge concerning the HRCT features of the infiltrative lung diseases. It is common practice for thinly collimated HRCT images to be included in every thoracic CT study. The advent of multidetector helical CT now permits generation of thin-section images retrospectively from an acquired volumetric set of data. Competent interpretation of HRCT is no longer limited to experts. Identification of the common features of UIP or IPF, such as honeycombing, reticulation, and traction bronchiectasis, is now in the standard skill set of most radiologists. One of the surprising findings of this study was the poor agreement among core radiologists regarding the presence or absence of honeycombing, a very critical feature in the diagnosis of IPF. We agree with the authors that the development of a set of standardized CT images illustrating various HRCT features in interstitial disease, similar to the standard radiographs for International Labour Organization (ILO) classification of the pneumoconioses, is sorely needed. Such standardized images may help reduce discrepancies among readers. Advances in computerized automated methods of disease characterization and quantification may be helpful, as the authors have suggested (6, 7). A consensus diagnosis of IPF by the core radiologists was achieved in 88% of patients who had a histologic diagnosis of UIP before study entry. The typical UIP patterns seen on HRCT thus predict the underlying UIP pathology. However, one of the weaknesses in the design of this study was the lack of centralized and standardized pathologic readings of surgical lung biopsies by consensus of an expert panel of pathologists. The reliability of the histologic diagnosis in this cohort has therefore not been validated. This study also confirms the work of others that the diffusing capacity of carbon monoxide and the HRCT fibrosis score (a combination of the severity of reticulation and honeycombing) are two of the most accurate predictors of mortality (8). The findings in this study certainly support the premise that HRCT is an important part of the evaluation of patients with suspected IPF. The data would support the conclusion that competent interpretation of HRCT findings in infiltrative lung disease can be accomplished by radiologists at regional medical centers. The reliability of the radiologic diagnosis, however, would be greatly enhanced by the development of standardized definitions and standardized images of HRCT findings. There also appears to be convincing evidence that surgical biopsy is not required in patients with classic features of reticulation and honeycombing, which are diagnostic of IPF. HRCT can be considered another important and independent predictor of survival. FOOTNOTES Conflict of Interest Statement: T.C.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1b treatment).