Idiopathic Pulmonary Fibrosis
To Biopsy or Not to Biopsy

Anna-Luise A. Katzenstein, M.D.

Crouse Hospital and Upstate Medical University, Syracuse, New York

Jeffrey L. Myers, M.D.

Mayo Clinic, Rochester, Minnesota

	ARTICLE

TOP ARTICLE REFERENCES

The term "idiopathic pulmonary fibrosis" (IPF) historically has encompassed four separate forms of interstitial pneumonia as well as a number of other conditions, each having a different clinical course and prognosis (1). The general consensus currently is that IPF be restricted to usual interstitial pneumonia (UIP) which is the most common idiopathic interstitial pneumonia and the one with the worst prognosis (2, 3). Exciting new data suggesting response of UIP to interferon gamma therapy (4) have spurred the need for accurate diagnosis, both to ensure that individual patients are appropriately treated and also to evaluate efficacy of therapy in prospective clinical trials.

Surgical lung biopsy traditionally has been the "gold standard" for diagnosing UIP and other interstitial lung diseases. In this issue of the AJRCCM (pp. 193-196), Hunninghake and colleagues (5) address the question of whether biopsy is required in all cases. These investigators analyzed accuracy of clinical and radiologic diagnosis in 91 patients with suspected IPF at eight separate institutions. The cases were evaluated clinically and radiographically by pulmonologists at the referring institutions, and a presumptive diagnosis was made. A surgical lung biopsy was performed in all. The clinical, radiologic, and pathologic findings were then reviewed by a clinical core of three pulmonologists, a radiology core of four radiologists, and a pathology core of three pathologists. Pathologic diagnosis required agreement between two of the three pathologists, and the diagnoses of the referring pulmonologist, the clinical core, and the radiologic core were compared. Fifty-four of 91 patients (59%) had UIP on biopsy specimens. The overall accuracy of the clinical diagnosis varied from 68% for the referring pulmonologists to 77% for the clinical core and 75% for the radiology core. However, if only confident diagnoses of IPF or non-IPF were considered, the accuracy for the referring pulmonologists increased to 78%, for the clinical core to 86%, and for the radiology core to 90%. Still, even in the highly accurate radiology core, a confident diagnosis missed four of 30 cases of IPF for a sensitivity of 87%, whereas one of 22 non-IPF patients was misdiagnosed as IPF for a specificity of 95%. Moreover, a confident diagnosis of IPF or non-IPF could be made in only 58% of patients. Thus, overall, a confident clinical diagnosis identified only about half of patients subsequently shown to have IPF. The investigators conclude that surgical lung biopsy is indicated both in patients lacking a confident clinical diagnosis and in those in whom the diagnosis of IPF is considered unlikely. Biopsy may not be needed in patients for whom there is a confident clinical diagnosis of IPF.

The findings of Hunninghake and colleagues complement those published previously by Raghu and associates (6). The latter investigators analyzed the diagnosis of IPF in 59 patients with new-onset interstitial lung disease from a single institution, 29 of whom (49%) were found to have IPF on surgical lung biopsy specimens. The accuracy of a clinical diagnosis of IPF was 62% and of a radiologic diagnosis 76%. Although the sensitivity was relatively low (62% clinically and 78.5% radiologically), the specificity was high (97 and 90%, respectively). The researchers conclude that surgical lung biopsy is indicated mainly for patients in whom the diagnosis is not certain after careful clinical evaluation, while it is not needed in patients with typical clinical and radiologic features.

These two studies provide helpful guidelines for determining whether to perform surgical lung biopsy in patients with suspected IPF. One potential shortcoming of the study by Hunninghake and colleagues is that no clinical or radiologic criteria were provided for diagnosing IPF. Instead, they emphasize "experience" of the core pulmonologists, and they question whether their results could be duplicated in other clinical settings lacking such experience. The implication is that experience overrides the use of clinical criteria for diagnosis. Although one could argue that specific criteria for diagnosing IPF have not been validated in prospective studies, detailed criteria have been published by a multidisciplinary group that included several of the current authors (3). Failure to use basic diagnostic criteria detracts somewhat from the usefulness of Hunninghake's study. One problem is that future studies may have difficulty comparing results. Another problem relates to the conclusion that the experienced clinical core physicians are more accurate in diagnosis than referring (and presumably less experienced) pulmonologists. It is likely that this difference in accuracy could have been eliminated or at least minimized if the clinical core and the referring pulmonologists had used the same diagnostic criteria. Although the benefits of experience in clinical practice are undeniable, reliance on experience over objective guidelines does not provide the objectivity that is required for sound scientific investigation.

How often a specific pathologic diagnosis changes patient treatment and outcome is another important issue. This question is especially relevant to patients with advanced interstitial fibrosis who would be expected to have little chance for recovery regardless of diagnosis or therapy. The study by Raghu and associates (6) specifically excluded such patients, but their number in Hunninghake's study is not provided. Clearly, however, the potential impact of biopsy findings on patient care and outcome needs to be considered before undertaking a surgical lung biopsy.

Accuracy of the pathologic diagnosis itself also needs to be addressed. Lung biopsy is the "gold standard" for diagnosis, yet are pathologists always correct? The probability of agreement between only two of the three pathologists in the study by Hunninghake and colleagues with regard to the presence or absence IPF was 0.85, whereas it decreased to 0.72 for a specific diagnosis. The use of standard diagnostic criteria is equally important for pathologists, even experienced ones, as it is for clinicians, and histologic guidelines should be carefully outlined and strictly followed. Even with standard criteria, however, pathologic diagnosis is not a precise science, and there always will be an occasional borderline, difficult-to-diagnose case in which diagnostic agreement will not be attainable even among experts. In practice, diagnostically challenging cases are often resolved after discussion with clinicians and radiologists. This type of interaction, however, was not available in Hunninghake's study because pathologists reviewed slides without the benefit of clinical information. The importance of clinical correlation in lung biopsy interpretation, especially in diagnostically difficult cases, should not be overlooked.

The study by Hunninghake and colleagues, along with that of Raghu and associates, provides evidence that the diagnosis of IPF can be accurately established in some patients without a surgical lung biopsy. This is an important finding for pulmonologists treating patients with this progressive and almost uniformly fatal disease. Whether accuracy of clinical diagnosis is sufficiently high, however, to replace lung biopsy in investigative studies of IPF remains unanswered.

	References

TOP ARTICLE REFERENCES

1. Katzenstein A-LA, Myers JL. Idiopathic pulmonary fibrosis: clinical relevance of pathologic classification. State of the art. Am J Respir Crit Care Med 1998; 157: 1301-1315 [Free Full Text].

2. Bjoraker JA, Ryu JH, Edwin MK, Myers JL, Tazelaar HD, Schroeder DR, Offord KP. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998; 157: 199-203 .

3. King TE Jr,, Costabel U, Cordier J-F, DoPico GA, duBois RM, Lynch D, Lynch JP III,, Myers J, Panos R, Raghu G, Schwartz D, Smith CM. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. Am J Respir Crit Care Med 2000; 161: 646-664 [Free Full Text].

4. Ziesche R, Hofbauer E, Wittmann K, Petkov V, Block L-H. A preliminary study of long-term treatment with interferon gamma-1b and low-dose prednisolone in patients with idiopathic pulmonary fibrosis. N Engl J Med 1999; 341: 1264-1269 [Abstract/Free Full Text].

5. Hunninghake GW, Zimmerman MB, Schwartz DA, King TE Jr,, Lynch J, Hegele R, Waldron J, Colby T, Müller N, Lynch D, et al . . Utility of a lung biopsy for the diagnosis of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2001; 164: 193-196 [Abstract/Free Full Text].

6. Raghu G, Mageto YN, Lockhart D, Schmidt RA, Wood DE, Godwin JD. The accuracy of the clinical diagnosis of new-onset idiopathic pulmonary fibrosis and other interstitial lung disease: a prospective study. Chest 1999; 116: 1168-1174 [Abstract/Free Full Text].