| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Denver, CO
Dr. Brown is Acting Head, Pulmonary/Critical Care Medicine Division, and Director, Interstitial Lung Disease Program, National Jewish Medical and Research Center. Dr. Schwarz is James C. Campbell Professor of Pulmonary Medicine, Division of Pulmonary Sciences and Critical Care Medicine, University of Colorado Health Sciences Center.
Correspondence to: Kevin K. Brown, MD, Acting Head, Pulmonary/Critical Care Medicine Division, Director, Interstitial Lung Disease Program, National Jewish Medical and Research Center, 1400 Jackson St, Denver, CO 80206; e-mail: brownk{at}njc.org
Physicians confronted with a patient with an interstitial lung disease (ILD) face a number of major problems. First, its prognosis can range from death within weeks to complete recovery, and clarifying these outcomes for any individual patient is crucial. We currently stratify an individual’s prognosis by defining clinico-radiographic-pathologic syndromes. Making a definitive diagnosis of one of these syndromes generally requires information from each of the clinical, radiographic, and pathologic pieces, with the data obtained by surgical lung biopsy providing the most authoritative information. However advances in high-resolution CT (HRCT) technology have forced a second problem: when is a surgical lung biopsy, with its attendant risk of complication, necessary for diagnosis? Thirdly, even when obtained, surgically obtained lung tissue may not be diagnostic and may even be difficult to classify according to current criteria. What is to be done in these circumstances? Lastly, the treatment options (irrespective of the underlying etiology or histology) are often associated with significant side effects, are of limited efficacy, or are both. The answer to these problems has eluded clinicians for decades.
In this issue of CHEST (see page 1566), Churg (a pathologist) and Müller (a radiologist)1 propose a scheme for classification of ILD to overcome some of these problems. Based on the histologic and radiologic features of disease rather than a specific diagnosis, their scheme hypothesizes that particular histologic and HRCT radiographic patterns or features are correlated with outcome, ie, both survival and response to therapy, and therefore improve on our ability to manage patients. Their proposal bears close scrutiny, as it has a strong historical precedent and new information provides support for this hypothesis.
The diffuse lung diseases are currently classified by underlying cause: infections, genetic disorders, exposures to a variety of inhaled (eg, avian antigen) or systemic substances (eg, medications), or underlying systemic disorders (eg, rheumatoid arthritis). After potentially identifiable causes are specifically sought out and excluded through a comprehensive evaluation, a diagnosis of an idiopathic interstitial pneumonia (IIP) can be made. This group includes idiopathic pulmonary fibrosis, nonspecific interstitial pneumonia (NSIP), cryptogenic organizing pneumonia, respiratory bronchiolitis ILD and desquamative interstitial pneumonia, acute interstitial pneumonia, and lymphocytic interstitial pneumonia. These disorders are definitively distinguished by the underlying histopathologic pattern seen on surgical lung biopsy, and this distinction is important as each pathologically defined disorder differs, often dramatically, in terms of outcome, and response (or lack thereof) to current therapy.2
The difficulties inherent in classifying ILD using the current criteria are not to be underestimated. While the importance of the history and examination cannot be overestimated, they are not alone diagnostic. The nonspecificity of laboratory values is well known, and while pulmonary physiology provides information regarding the pattern and severity of abnormality, it generally has little to say about the underlying cause of disease. Surgical lung biopsy has been recognized as providing the most definitive diagnostic and prognostic information. However, it cannot be interpreted in isolation,3 not all patterns are diagnostic and, more importantly, interpretation is subject to significant both intraobserver and interobserver variability.4
The rapid advance of HRCT technology and the ability to interpret the results have provided us with a powerful tool and forced reevaluation of the requirement for a surgical lung biopsy in some cases. New data suggest that certain typical HRCT patterns are highly correlated with particular histopathologic patterns,5 although the corollary, that the absence of typical HRCT findings excludes the pathologic pattern, is not true.6 It has also been shown that in specific clinical settings, prognosis is more closely associated with HRCT features than with other clinical variables.78 Unfortunately, HRCT scanning shares the limitations associated with lung biopsies: intraobserver and interobserver variability,9 the common presence of nonspecific patterns, and the inability to interpret them without clinical input.
Prior to the more recent classification of ILD, it was intuitive that pathology would provide a definitive diagnosis and that cellular disease should be responsive to treatment and associated with improved survival when compared to those disorders characterized by fibrosis. Subsequent investigation confirmed this intuition, and we now know that ILD defined by cellular histologic reactions, such as cryptogenic organizing pneumonia, eosinophilic pneumonia, and cellular NSIP, either respond well to antiinflammatory or immunomodulatory therapy or have a prolonged survival when compared to fibrosing disorders.210 In fact, we now suspect that the cases responsive to immunomodulatory therapy reported in historically defined "idiopathic pulmonary fibrosis" cohorts represented the more cellular disorders (eg, NSIP) rather than the pathologic usual interstitial pneumonia pattern that currently defines idiopathic pulmonary fibrosis.
With this background, the current proposal of Churg and Müller1 forces us to refocus on this distinction between cellularity and fibrosis and consider whether our current classification scheme should be updated or backdated. They utilize the American Thoracic Society/European Respiratory Society IIP classification scheme (adding the features seen in hypersensitivity pneumonitis) to suggest that pathologic and radiographic features of cellularity and fibrosis offer clearer information about outcome. However, we know that both pathology and underlying cause are important in determining outcome. For example, when the diagnosis of drug-induced ILD or hypersensitivity pneumonitis is made, the most complete recovery requires cessation of exposure regardless of the requirement for additional medication therapy. As a variety of radiologic or pathologic patterns can be seen in these disorders, both are primarily clinical diagnoses based on the timing of appropriate exposures. We also know that even with similar pathologic or radiographic findings outcome may differ between IIP and ILD associated with connective tissue disease.1112
Whether the radiologic and pathologic patterns and features highlighted by Churg and Müller1 will be less prone to intraobserver and interobserver variability, provide information that is independent of the underlying clinical diagnosis, and be more directly correlated with outcome remains to be proved. A number of prospective studies of specific individual pathologic and radiographic features in a diverse range of diffuse lung diseases diagnosed using current criteria will be needed to determine if this proposed schema provides added value. So for now, when confronted with a new patient with ILD, honor both the present and the past by noting the significance of cellularity and fibrosis, but do not forget the importance of the specific diagnosis.
Footnotes
The authors have no conflicts of interest to disclose on this topic.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
