| 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 |
|||||||||
* From the Departments of Pathology (Dr. Churg) and Radiology (Dr. Müller), University of British Columbia, Vancouver, BC, Canada.
Correspondence to: Andrew Churg, MD, Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver, BC, Canada V6T 2B5; e-mail: achurg{at}interchange.ubc.ca
Key Words: hypersensitivity pneumonitis • idiopathic interstitial pneumonia • nonspecific interstitial pneumonia • usual interstitial pneumonia
The 2002 American Thoracic Society (ATS)/European Respiratory Society (ERS) classification1 of the idiopathic interstitial pneumonias represents a major advance in our understanding of these entities. The classification defines the features of a number of entities, emphasizes the necessity of correlating pathologic findings on biopsy with radiologic and clinical data to arrive at a diagnosis, and in particular sets out idiopathic pulmonary fibrosis (usual interstitial pneumonia [UIP]) as a disease with a poor prognosis.
Nonetheless, in our experience, clinicians frequently ask an important question: apart from UIP, and apart from the obvious need to rule out a cause, how important is it to separate most of the idiopathic interstitial pneumonias from each other in terms of treatment and prognosis, and what does one do with equivocal cases or cases in which the clinical, radiologic, and pathologic findings do not allow exact classification into one of these conditions? Another way to look at this is the old issue of lumpers vs splitters: how much of what we view as separate entities are really separations of minor degree where the exact classification makes no practical difference?
This is not an unusual or trivial problem. For example, it is now clear that even experienced pulmonary pathologists disagree about the exact classification of the idiopathic interstitial pneumonias in a significant number of cases, and the separation of fibrotic nonspecific interstitial pneumonia (NSIP) from UIP is a particular problem.2 Similarly, separation of NSIP that has a moderate number of airspace macrophages from desquamative interstitial pneumonia (DIP) may not be possible. Moreover, a histologic diagnosis of NSIP often does not constitute a final diagnosis, since it might reflect hypersensitivity pneumonitis, collagen vascular disease, or drug-induced lung disease; and in fact collagen vascular disease and drug reactions can look like UIP, cryptogenic organizing pneumonia (COP), and various unclassifiable interstitial pneumonias as well. Small amounts of organizing pneumonia (OP pattern) may be seen in virtually any of the idiopathic interstitial pneumonias; at what point does that change the diagnosis? Further, although the ATS/ERS classification quite rightly does not include hypersensitivity pneumonitis, there is often considerable overlap between the morphologic and radiologic features of hypersensitivity pneumonitis and several of the idiopathic interstitial pneumonias, since hypersensitivity pneumonitis can have fairly extensive areas that look like OP or can be a perfect mimic of NSIP3
The ATS/ERS classification is clearly the "gold standard." However, in many cases even expert clinicians, pathologists, and radiologists fail to reach a consensus as to the diagnosis. For this situation, we propose an alternative way to look at the idiopathic interstitial pneumonias and pathologically and radiologically similar conditions such as hypersensitivity pneumonitis, interstitial lung disease in collagen vascular disease patients, and drug-related interstitial lung disease, based on separating radiologic or pathologic findings (Table 1 ) into three fundamental groups that reflect the prognosis. For this purpose, we have excluded two entities from the ATS/ERS classification: acute interstitial pneumonia, which is simply idiopathic ARDS and has the generally poor prognosis of ARDS; and lymphocytic interstitial pneumonia, most cases of which are really lymphoproliferative disorders (many related to AIDS, some really lymphomas) that generally require a completely different therapeutic approach. And for the reasons indicated above, we have added hypersensitivity pneumonitis.
|
|
Linear fibrosis without architectural distortion is associated with a distinctly worse prognosis than purely cellular lesions. This phenomenon has been well documented for patients with NSIP.4 Patients with hypersensitivity pneumonitis and fibrosis (ie, chronic hypersensitivity pneumonitis) are known to do worse than those without fibrosis,56 and some of these patients have a fibrotic NSIP-like fibrosis on biopsy.37 However, the prognosis for patients with NSIP and linear fibrosis appears to be considerably better than that of patients with a pathologic pattern of UIP. Patients with chronic hypersensitivity pneumonitis and a UIP picture probably also have a poor prognosis,7 but this remains to be proven. We propose that drug reactions, collagen vascular disease cases, and unclassifiable interstitial pneumonias with linear or UIP-like fibrosis will behave as fibrotic NSIP or UIP, respectively, and there do in fact exist published data to support the worse prognosis of fibrotic compared to cellular NSIP pattern in patients with collagen vascular disease.8
In terms of high-resolution CT (HRCT) findings, similar separations can be made: chronic interstitial diseases characterized by airspace opacification (ground-glass opacities or consolidation) without reticulation (ie, without underlying fibrosis) include COP, cellular NSIP, most cases of DIP, and subacute hypersensitivity pneumonitis (Fig 2 , top left, A). These patients usually respond to therapy and have a good prognosis.91011 Diseases that may manifest with extensive ground-glass opacities and relatively mild reticulation (reticulation representing < 25% of the parenchymal abnormalities) include mixed fibrotic and cellular NSIP, some cases of DIP, and chronic hypersensitivity pneumonitis. Patients with this combination tend to have a worse prognosis than patients with purely ground-glass opacities but a better prognosis than patients with predominant reticulation, a finding that may be seen in patients with fibrotic NSIP (Fig 2, top right, B), chronic hypersensitivity pneumonitis, and UIP.41213 In patients with UIP, the extent of fibrosis on HRCT has been shown to be one of the most reliable indexes of poor prognosis.1415 The patients with the worst prognosis are those with honeycombing (Fig 2, bottom, C), the vast majority of whom have UIP,1617 although some have chronic hypersensitivity pneumonitis.
|
We put this proposal forward in part as observation and in part as a hypothesis that needs to be tested. In particular, there is a lack of detailed information on whether the amount of fibrosis vs amount of cellularity is important (or does any amount of fibrosis immediately produce a worse outcome?). It is also unclear as to what degree the pattern of fibrosis in hypersensitivity pneumonitis (UIP-like vs NSIP-like) affects prognosis, and the same is true of the roughly one third of DIP cases with a poor prognosis, and which may resemble fibrotic NSIP on HRCT.18 A further problem that needs to be examined are cases of pathologic UIP that radiologically do not have honeycombing, and that appear to have a slower clinical course than UIP, which does have radiologic honeycombing19; this observation implies that radiologic honeycombing is probably the single most important predicator of a poor prognosis, and suggests that a combined pathologic/radiologic impression may produce finer discriminations that diagnoses based on either modality alone.
Readers may object that we are reverting to an older view of the world and losing the important advances incorporated into the ATS/ERS classification. As well, specific diagnoses in theory allow correlation of particular disease entities and outcome, although problems in assigning a name to a morphologic/radiologic appearance confound this process. Ignoring specific diagnoses is not our intent, and making a specific diagnosis whenever possible must remain the ultimate goal, as it may have other important treatment implications: patients with DIP/RBILD need to be counseled to quit smoking; in patients with hypersensitivity pneumonitis, one wants to attempt to find the offending antigen (although it is far from clear whether this always makes a difference); and for drug reactions, the drug needs to be identified and discontinued. In addition, within some diseases there are other factors that also impact prognosis, for example, the number of fibroblast foci or diffusing capacity at presentation in UIP, or the exact antigen in hypersensitivity pneumonitis.
We have not included the clinician’s impression of the disease process as part of our criteria, but this is done quite purposefully, because our real question is as follows: given a patient with evidence of an interstitial lung disease that fits into the general categories we have been discussing, what are the important pathologic and radiologic features that determine likely response to treatment and prognosis? The scheme we propose provides the clinician with a useful and simple categorization that indicates what information is crucial to obtain from the pathologist and radiologist (Is the process purely cellular? Is there evidence of fibrosis and, if so, how much and in what pattern?) and appears to provide a reasonably good guide to prognosis—and prognosis is what patients really want to know.
Footnotes
Abbreviations: ATS = American Thoracic Society; COP = cryptogenic organizing pneumonia; DIP = desquamative interstitial pneumonia; ERS = European Respiratory Society; HRCT = high-resolution CT; NSIP = nonspecific interstitial pneumonia; OP = organizing pneumonia; RBILD = respiratory bronchiolitis with interstitial lung disease; UIP = usual interstitial pneumonia
The authors have no conflicts of interest to disclose.
Received for publication March 28, 2006. Accepted for publication June 15, 2006.
References
This article has been cited by other articles:
|
|
 |
|
  A. Baydur Mechanical Ventilation in Interstitial Lung Disease: Which Patients Are Likely to Benefit? Chest, May 1, 2008; 133(5): 1062 - 1063. [Full Text] [PDF]
|
|
|
|
|
|
A. N. Rubinowitz, M. Moon, and R. Homer A 34-Year-Old Man Presenting With Gradual Onset of Shortness of Breath and Interstitial Lung Disease Chest, April 1, 2008; 133(4): 1041 - 1047. [Full Text] [PDF]
|
|
|
|
|
|
K. K. Brown and M. I. Schwarz Classifying interstitial lung disease: remembrance of things past. Chest, November 1, 2006; 130(5): 1289 - 1291. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
