Classification and diagnosis of the idiopathic interstitial pneumonias are difficult for surgical pathologists. Part of the problem is that the histologic abnormalities in these diseases are relatively nonspecific, consisting of variable amounts of inflammation and fibrosis. Furthermore, diagnosis requires substantial clinical input with correlation of clinical and pathologic findings. Another problem is that clinicians and pathologists have used different classification systems and terminology. In the clinical literature, the idiopathic interstitial pneumonias were usually lumped together under the category of idiopathic pulmonary fibrosis (IPF) or cryptogenic fibrosing alveolitis, whereas multiple categories based on Liebow's classification (Table 1) were utilized by pathologists. 1,5,7 Controversy has followed the latter classification over the years, especially with regard to the separation of usual interstitial pneumonia (UIP) and desquamative interstitial pneumonia (DIP). With recent refinement of diagnostic features, most individuals accept the notion that these two entities are separate diseases with different pathogenesis and prognosis. It has also become clear that a histologically distinct, rapidly progressive form of idiopathic interstitial pneumonia exists. This disease, known as acute interstitial pneumonia (AIP) or Hamman–Rich disease, was described initially by Hamman and Rich but was later lumped with UIP by Liebow, 7 who considered it the early stage of this disease. Other forms of interstitial pneumonia described by Liebow, 7 including lymphoid interstitial pneumonia (LIP) and giant-cell interstitial pneumonia (GIP), remain well-recognized diseases. LIP is now considered a lymphoproliferative disorder and GIP is related to hard-metal pneumoconiosis, and therefore they are not included in the idiopathic interstitial pneumonia classification. Liebow's 7 final category of bronchiolitis obliterans with usual interstitial pneumonia consisted of a mixture of different entities, including UIP or diffuse alveolar damage with areas of bronchiolitis obliterans as well as cases of bronchiolitis obliterans organizing pneumonia (BOOP), and the term is no longer used.
It has become clear in recent years that cases of interstitial pneumonia exist that do not fit histologically with UIP, DIP, or AIP. They have been designated by some as cellular interstitial pneumonia or unclassified interstitial pneumonia. 5,10 The first large series to focus on these cases utilized the descriptive term nonspecific interstitial pneumonia/fibrosis (NSIP), 4 and this name has gained broad acceptance as noted by Travis et al. 10 in this issue of the American Journal of Surgical Pathology. These four forms of interstitial pneumonia comprise the current classification of the idiopathic interstitial pneumonias (Table 2). UIP is the most common followed by NSIP, whereas DIP and AIP are relatively rare. Respiratory bronchiolitis interstitial lung disease is another form of idiopathic interstitial pneumonia that is thought by some to be the early stage of DIP and thus may be lumped with DIP. The utility of this classification is to separate those diseases with different clinical presentation or course and a generally better prognosis from UIP, which is usually unresponsive to therapy and fatal. Although Travis et al. 10 include BOOP in their classification of idiopathic interstitial pneumonia, this disorder is not an interstitial process. While BOOP is characterized by fibrosis and chronic inflammation like the interstitial pneumonias, it differs in that the reaction affects predominantly the airspaces rather than the interstitium. This difference is seen radiographically as well as pathologically where patchy airspace rather than interstitial opacities are the usual finding.
When first described in 1994, NSIP was considered to be a wastebasket term for all unclassifiable interstitial pneumonias. 4 Patients with known or suspected etiologies, such as inhalational exposures, drug reactions, acute lung injury, or associated connective tissue diseases, for example, were included along with those who had no known etiology. Patients with underlying immune compromise or suspected infections, however, were excluded. Since 1994 there have been seven additional reports of NSIP, including that of Travis et al., 1–3,6,8–10 and the concept of NSIP has evolved from a broad, inclusive term into a specific form of idiopathic interstitial pneumonia. Histologically it is a uniform-appearing, cellular interstitial pneumonia characterized by a lymphoplasmacytic infiltrate within alveolar septa. Varying amounts of fibrosis consisting predominantly of collagen are admixed with the chronic inflammation, and cases can be divided into cellular and fibrotic variants. Alveolar pneumocyte hyperplasia often accompanies the other changes, and patchy intra-alveolar macrophage accumulation may occur. Small foci of intraluminal fibrosis resembling BOOP are common but are always overshadowed by the more extensive interstitial pneumonia. The diagnostic histologic features of NSIP and the other idiopathic interstitial pneumonias are summarized in Table 3. All reported series have confirmed a better prognosis for NSIP than for UIP, with an average mortality rate of 16% for NSIP patients (range, 0%–29%). These figures contrast with mortality rates for UIP which range from 60% to more than 90%. When NSIP cases are divided into those with and without fibrosis, deaths have been reported only in the fibrotic group, including 5 of 26 patients (19%) in the study by Katzenstein and Fiorelli, 4 2 of 15 patients (13%) in the study by Nagai et al., 8 and 9 of 22 patients (41%) in the series by Travis et al. 10 Although the mortality rate noted by Travis et al. 10 is higher than others, most deaths occurred after 5 years. The overall 5-year survival rate for NSIP in that series was 90% and the 10-year survival rate 35% compared with a 43% 5-year and 15% 10-year survival for UIP.
Although the cellular form of NSIP is easy to diagnose and differentiate from the other forms of idiopathic interstitial pneumonia, difficulty can be encountered in distinguishing the fibrotic form from UIP. In our opinion, NSIP with fibrosis should be diagnosed only in cases characterized by a diffuse, uniform interstitial process in which chronic inflammation is admixed with fibrosis. The presence of honeycomb change and/or fibroblast foci, unless very focal, and a patchy, especially subpleural, distribution of the interstitial changes are features that should suggest UIP. In the series by Katzenstein and Fiorelli, 4 small foci of honeycomb change were present in 12% and rare fibroblast foci were present in 24% of NSIP with fibrosis cases (groups II and III). Higher percentages of honeycomb change were noted in fibrotic NSIP cases both in studies by Nagai et al. 8 (60%) and Travis et al. 10 (91%), and fibroblast foci were found in 55% of cases by Travis et al. 10 These findings suggest that some cases of UIP were included among the fibrotic NSIP cases in those series, and this possibility might explain the higher mortality rates noted for the NSIP cases.
One potentially confusing issue in the diagnosis of the idiopathic interstitial pneumonias is the presence or absence of fibroblast foci. These small collections of fibroblasts and myofibroblasts within myxoid-appearing stroma occur in the interstitium and are covered by epithelium on their luminal surface. They represent foci of organizing acute lung injury and are sites of active collagen synthesis. As such, they are relatively nonspecific and can be found, if searched for, in a wide variety of lung diseases. They are characteristically present in most cases of UIP, and the combination of patchy, nonuniform interstitial scarring and inflammation, honeycomb change, and fibroblast foci is diagnostic of this disease. In fact, fibroblast foci are thought to represent the earliest abnormality of UIP, and thus they are central to understanding the pathogenesis of this disease. They are not present in every case of UIP, however, possibly due to sampling problems, and their absence does not necessarily negate the diagnosis if the pathologic features are otherwise typical and the clinical and radiographic findings fit.
The currently accepted classification of the idiopathic interstitial pneumonias appears to separate these diseases into clinically meaningful and prognostically relevant groups. Although the clinical term IPF was used in the past to encompass all four entities, it is now restricted to cases of UIP. Like any classification scheme, however, it is not perfect, and there will always be a small number of difficult cases that do not fit precisely into any one category. The main area of difficulty is the separation of fibrotic NSIP from UIP. Features that support UIP over fibrotic NSIP include patchy or subpleural distribution of changes, honeycomb areas, and fibroblast foci. In difficult cases the results of high-resolution computed tomography can help, because they often show characteristic peripheral reticular opacities and honeycombing in UIP. Considering, however, that fibrosis itself, especially when extensive, is associated with a worse prognosis in NSIP, the separation of these borderline cases into fibrotic NSIP or UIP may not be all that meaningful clinically. It should be remembered that there will always be controversial cases in which even the experts do not agree, and it may be most prudent in such cases to provide a descriptive diagnosis, such as “interstitial fibrosis with honeycomb change,” for example, rather than attempt to pigeon-hole the changes into a single disease category.
1. Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathological subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998; 157:199–203.
2. Cottin V, Donsbeck A-V, Revel D, Loire R, Cordier J-F. Nonspecific interstitial pneumonia. Individualization of a clinicopathologic entity in a series of 12 patients. Am J Respir Crit Care Med 1998; 158:1286–93.
3. Daniil ZD, Gilchrist FC, Nicholson AG, et al. A histologic pattern of nonspecific interstitial pneumonia is associated with a better prognosis than usual interstitial pneumonia in patients with cryptogenic fibrosing alveolitis. Am J Respir Crit Care Med 1999; 160:899–905.
4. Katzenstein A-LA, Fiorelli RF. Non-specific interstitial pneumonia/fibrosis. Histologic patterns and clinical significance. Am J Surg Pathol 1994; 18:136–47.
5. Katzenstein A-LA, Myers JL. Idiopathic pulmonary fibrosis. Clinical relevance of pathologic classification. State of the art. Am J Respir Crit Care Med 1998; 157:1301–15.
6. Kim TS, Lee KS, Chung MP, et al. Nonspecific interstitial pneumonia with fibrosis: high-resolution CT and pathologic findings. Am J Roentgenol 1998; 171:1645–50.
7. Liebow AA. Definition and classification of interstitial pneumonias in human pathology. Prog Respir Res 1975; 8:1–31.
8. Nagai S, Kitaichi M, Itoh H, Nishimura K, Izumi T, Colby TV. Idiopathic nonspecific interstitial pneumonia/fibrosis: comparison with idiopathic pulmonary fibrosis and BOOP. Eur Respir J 1998; 12:1010–9.
9. Park JS, Lee KS, Kim JS, et al. Nonspecific interstitial pneumonia with fibrosis: radiographic and CT findings in seven patients. Radiology 1995; 195:645–8.
10. Travis WD, Matsui K, Moss J, Ferrans VJ. Idiopathic nonspecific interstitial pneumonia: prognostic significance of cellular and fibrosing patterns: survival comparison with usual interstitial pneumonia and desquamative interstitial pneumonia. Am J Surg Pathol 2000; 24:19–33.