Lung (1990) Suppl:993-997
New York Inc. 1990
Idiopathic Interstitial Fibrosis D. Lambe University of Edinburgh, Edinburgh, United Kingdom
Abstract. The lung has only a limited pattern of response to injury. Idiopathic interstitial fibrosis is a descriptive term, describing the end stage of one pattern of pulmonary fibrosis in which the inflammation lies within the structure of the alveolar walls and around vessels and bronchioli, in the interstitium. The disease is characteristically subpleural in distribution and provides a marked restrictive consequence on pulmonary function which progresses to hypopoxic core pulmonale. This is despite the apparent normality of much of the lung at autopsy. Six to fifteen percent develop lung cancer, the proportion depending on the autopsy rate. A nonspecific nature of the changes limits the diagnostic value of transbronchial biopsy, open lung biopsy may be of value in assessing and grading the disease in relationship to therapy. Key words: Lung--Inflammation--Interstitial fibrosis--Honeycomb--Biopsy The term interstitial fibrosis without qualification is used by pathologists as a purely descriptive term to describe an entirely nonspecific pattern of scarring in the lung. It can also be used, usually by clinicians, to refer to the poorly understood but well recognized clinical entity which passes under a variety of names: idiopathic interstitial pulmonary fibrosis, idiopathic fibrosing alveolitis, cryptogenic fibrosing alveolitis, diffuse interstitial pulmonary fibrosis, the usual interstitial pneumonitis (UIP), etc. The common theme to all these names is the presence of an inflammatory process progressing to an interstitial fibrosis. The lung has only a limited pattern of response to injury. Inflammatory processes show three broad patterns with predominant involvement of airways, especially bronchioli; intraluminal inflammation involving the alveolar space and the lumen of the more distal airways as seen in a simple pneumonia; Offprint requests to: Dr. D. Lambe, Department of Pathology, Medical School of the University of Edinburgh, Teviot Place, GB-Edinburgh, EH8 9AG, UK.
994
D. Lambe
and interstitial inflammation in which the predominant process lies within the structure of the alveolar wails and around vessels and bronchioli. The interstitial space consists of the sheaths of connective tissue around the bronchiovascular bundles and extending along the veins and septa to the pleura. There is a less conspicuous parenchymal interstitial space lying within the alveolar walls. Most interstitial diseases of the lung involve both the perivascular and alveolar wall interstitium. Arguments about the site and nature of the interstitial space are one of the reasons for the plethora of names applied to idiopathic pulmonary fibrosis [6]. Scarfing is the usual and expected sequel to nonresolved inflammatory processes. One would anticipate that an interstitial fibrosis was the end result of an interstitial inflammatory process, and so it may be. In addition, intraluminal exudate may organize and become incorporated into the alveolar wall to become interstitial. The most extreme example of this is the case of those subjects with ARDS who survived long enough to undergo organization of the intraalveolar exudate. Though originally entirely intralnminar after about four weeks the pattern may be almost entirely of an interstitial fibrosis. The histological features of idiopathic interstitial fibrosis are those of an interstitial inflammatory process consisting of variable numbers of lymphocytes, plasma cells with a small but often inconspicuous polymorph component. Depending on the stage of the disease, there is a variable amount of fibrosis, with first a widening of the involved alveolar walls and later a loss of the architectural pattem with a replacement by bands of fibrosis in which variable sized spaces lined by hypertrophied type-2 pneumocytes or by bronchiolar epithelium appear. These larger air spaces are probably derived from residual respiratory or terminal bronchioli. The overall pattern of fibrosis with alternating air spaces which appear to increase in size, perhaps as the scarfing distorts and retracts the adjacent lung tissue, gives rise to a characteristic pattern of scarring, the "honeycomb lung." The appearance of such honeycomb lung may vary, the air spaces being anything from 1-2 mm to 5-10 mm in diameter, characteristically lined by shiny intact epithelium. The relative proportions of air space to fibrosis in such honeycomb lung probably depends on the size of the original acinar units and possibly on the presence or absence of preexisting microscopic emphysema associated with past cigarette smoking. The lumen of air spaces in the involved lung of idiopathic pulmonary fibrosis show characteristically hypertrophied alveolar lining cells with the lumen itself containing varying numbers of macrophages and occasional lymphocytes. In the extreme case, those cases to which the term Desquamative Interstitial Pneumonitis (DIP) is applied, the lumen may be packed with macrophages. Originally this appearance of packed alveolar space was interpreted as being the consequence of desquamated type-2 pneumocytes. It is now a simple matter to apply immunocytochemical techniques and identify the ceils as nonepithelium macrophages. None of the histological features of idiopathic pulmonary fibrosis are characteristic, thus any of the individual components may be seen in other conditions. What is characteristic however, is the predominantly subpleural position of the inflammatory process and of the developing scarfing. The disease does
Idiopathic Interstitial Fibrosis
995
not follow a simple progression from an inflammatory phase through progressively increasing scarring to end stage disease. It appears to be more locally episodic with individual acinar units or lobular units showing active inflammation while adjacent units have progressed through to end stage scarring. There is the impression that the disease progresses in such an episodic manner from the pleura in towards the center of the lung; the honeycomb scarring present immediately under the pleura may show little evidence of the inflammation which may be present at the inner margin of a band of such honeycombing. In the developed areas of scarring it is characteristic to find irregular bands of smooth muscle proliferation which appear to be more prominent in the end stage fibrosis of idiopathic pulmonary fibrosis than of other fibrosing lung conditions. The pulmonary vasculature shows very prominent intimal and medial abnormalities which may be partly a response to the adjacent scarring and also to the local alveolar hypoxia. What can the pathologist do to help the clinician in the diagnosis and management of idiopathic interstitial fibrosis? All the histological features described are individually entirely nonspecific. It is not possible to make a diagnosis of idiopathic pulmonary fibrosis on material derived from transbronchial biopsy. The most that can be expected is the possible exclusion of conditions having more characteristic histological appearances, such as lymphatic spread by tumor or sarcoidosis. It is important that when describing the appearances in transbronchial biopsies the pathologist avoids the term interstitial fibrosis unless he qualifies it as being "nonspecific." It is easy for the children to read into a descriptive term of interstitial fibrosis the specific diagnosis of the entity, idiopathic interstitial fibrosis. The dangers of extrapolation of the appearances in small transbronchial biopsies was emphasized by Wall et al., when they showed that any attempt to extrapolate upon the predominant appearance of interstitial inflammation or scarring had no validity and gave rise to significant misdiagnosis. Open lung biopsy allows all the components of idiopathic interstitial fibrosis to be identified and allows one to exclude other specific diagnoses. However, even with the open lung biopsy the component histological processes have no absolutely characteristic appearances and the diagnosis is still based on the circumstantial evidence of the clinical situation, a range of appearances and the absence of other specific diagnoses. Selection of material by the surgeon for biopsy is of the greatest importance. As described above, the disease does not progress through characteristic stages throughout the areas of involved lung and it is possible to select old disease under the pleura which shows predominant scarfing and no evidence of inflammation, or from the margins of lesions, areas showing predominantly inflammatory changes with little scarfing. This may make little difference to the overall diagnosis but may significantly bias attempts to assess the "activity" of the diseased process upon which some clinicians base their therapeutic regimes. There is a broad correspondence between the amount of fibrosis and the lack of therapeutic response and between the amount of inflammatory infiltrate and of the desquamative component and clinical response to therapy. Those cases showing the features of DIP usually have less fibrosis and, in
996
D. Lambe
comparison to the classical histological pattern, have a longer survival and a much higher response to steroids [1]. While in the classical idiopathic pulmonary fibrosis the prognosis and response to therapy seemed to be negatively correlated to the amount of fibrosis and positively correlated to the amount of inflammation in a biopsy [3, 11]. The use of immunofluorescent techniques on frozen sections of open lung biopsies to study the possible pathogenetic mechanisms and potentially the disease activity does not seem to be of diagnostic value [2]. Bronchoalveolar lavage (BAL) is a simpler and safer technique than lung biopsy. Attempts to use differential accounts from BAL to guide management have had varied success. Overall an increase in lymphocytes is more likely to be found in responders and an increase in polymorphs and eosinophils seen in association with poor response to therapy [3, 5]. Such correlations are identified between groups of cases and the information is less specific in the case of individual patients. There is still doubt as to whether BAL data truly reflect the interstitial inflammatory state of the lung [3]. Despite recent advances in immunology and the application of immunological techniques to histological material and the use of BAL, there has been no real advance in our understanding of the nature of idiopathic interstitial fibrosis, nor in the help that the pathologist can give the clinician since the first proper histological description by Liewbow and Carrington [4]. With the exception of a small number of patients having a predominantly DIP histological pattern [1] all cases of idiopathic interstitial fibrosis progressed to develop hypoxic cor pulmonale. The honeycomb type of end stage pulmonary fibrosis has a characteristic subpleural distribution in these cases, localized to the posterior and inferior surfaces of the lower lobe with relativesparing of the anterior aspect of the lower lobe and the posterior aspect of the upper adjacent to the fissure. Despite much apparently normal lung remaining the subpleural fibrosis seems to impair alveolar ventilation to produce alveolar hypoxia and possibly affects the perfusion of the lower lobe (McCarthy and Cherniack, 1973) with the development of cor pulmonale. Intercurrent infection is a common final incident but there is also an increased number of cases who develop malignant lung tumors. The proportion of cases developing lung cancer varies from 6 to over 15% in different series [8-10] partly dependent on the autopsy rate as many as these carcinomas are clinically occult and diagnosed at post mortem examination.
References 1. Carrington CB, Gaensler EA, Coutu RE, Fitzgerald MX, Gupta RG (1978) Natural history and treated course of usual and desquamative interstitial pneumonia. N Engl J Med 298:802-809 2. Coffin B, Dewar A, Rodriguez-Roisin R, Turner-Warwick M (1985) Fine structural changes in cryptogenic fibrosing alveolitis and asbestos. J Pathol 147:107-119 3. Haslam PL, Turton C, Heard B, et al (1980) Bronchoalveolar tavage in pulmonary fibrosis: comparison of cells obtained with lung biopsy and clinical features. Thorax 35:9-18 4. Liebow AA, Carrington CB (1969) The interstitial pneumonias. In: Simon M, Potchen EJ, Lemay M, eds. Frontiers of pulmonary radiology. New York: Grune & Stratton, 102-141
Idiopathic Interstitial Fibrosis
997
5. Rudd RM, Haslam PL, Turner-Warwick M (1981) Cryptogenic fibrosing alveolitis: relationships of pulmonary physiology and bronchoalveolar lavage to response to treatment and prognosis. Am Rev Respir Dis I24:1-8 6. Scadding JC (1974) Diffuse pulmonary alveolar fibrosis. Thorax 29:271-281 7. Wall CP, Gaensler EA, Carrington CB, Hayes JA (1981) Comparison of transbronchial and open lung biopsies in chronic infiltrative lung diseases. Am Rev Resp Dis 123:280-285 8. Stack BHR, Choo-Kang YFJ, Herd BE (1972) Prognosis of cryptogenic fibrosing alveolitis. Thorax 27:535-542 9. Haddad R, Massaro D (1968) Idiopathic diffuse interstitial pulmonary fibrosis (fibrosing alveolitis)--atypical epithelial proliferation and lung cancer. Am J Med 45:211-218 10. Turner-Warwick M, Lebowitz M, Burrows B, et al (1980) Cryptogenic fibrosing alveolitis and lung cancer. Thorax 35:496-499 11. Wright PH, Herd BE, Steel SJ, Turner-Warwick M (1981) Cryptogenic fibrosing alveolitis: assessment by graded trephine lung biopsy histology compared with clinical, radiographic and physiological features. British Journal of Disease of the Chest 75:61-70
New York Inc. 1990
Idiopathic Interstitial Fibrosis D. Lambe University of Edinburgh, Edinburgh, United Kingdom
Abstract. The lung has only a limited pattern of response to injury. Idiopathic interstitial fibrosis is a descriptive term, describing the end stage of one pattern of pulmonary fibrosis in which the inflammation lies within the structure of the alveolar walls and around vessels and bronchioli, in the interstitium. The disease is characteristically subpleural in distribution and provides a marked restrictive consequence on pulmonary function which progresses to hypopoxic core pulmonale. This is despite the apparent normality of much of the lung at autopsy. Six to fifteen percent develop lung cancer, the proportion depending on the autopsy rate. A nonspecific nature of the changes limits the diagnostic value of transbronchial biopsy, open lung biopsy may be of value in assessing and grading the disease in relationship to therapy. Key words: Lung--Inflammation--Interstitial fibrosis--Honeycomb--Biopsy The term interstitial fibrosis without qualification is used by pathologists as a purely descriptive term to describe an entirely nonspecific pattern of scarring in the lung. It can also be used, usually by clinicians, to refer to the poorly understood but well recognized clinical entity which passes under a variety of names: idiopathic interstitial pulmonary fibrosis, idiopathic fibrosing alveolitis, cryptogenic fibrosing alveolitis, diffuse interstitial pulmonary fibrosis, the usual interstitial pneumonitis (UIP), etc. The common theme to all these names is the presence of an inflammatory process progressing to an interstitial fibrosis. The lung has only a limited pattern of response to injury. Inflammatory processes show three broad patterns with predominant involvement of airways, especially bronchioli; intraluminal inflammation involving the alveolar space and the lumen of the more distal airways as seen in a simple pneumonia; Offprint requests to: Dr. D. Lambe, Department of Pathology, Medical School of the University of Edinburgh, Teviot Place, GB-Edinburgh, EH8 9AG, UK.
994
D. Lambe
and interstitial inflammation in which the predominant process lies within the structure of the alveolar wails and around vessels and bronchioli. The interstitial space consists of the sheaths of connective tissue around the bronchiovascular bundles and extending along the veins and septa to the pleura. There is a less conspicuous parenchymal interstitial space lying within the alveolar walls. Most interstitial diseases of the lung involve both the perivascular and alveolar wall interstitium. Arguments about the site and nature of the interstitial space are one of the reasons for the plethora of names applied to idiopathic pulmonary fibrosis [6]. Scarfing is the usual and expected sequel to nonresolved inflammatory processes. One would anticipate that an interstitial fibrosis was the end result of an interstitial inflammatory process, and so it may be. In addition, intraluminal exudate may organize and become incorporated into the alveolar wall to become interstitial. The most extreme example of this is the case of those subjects with ARDS who survived long enough to undergo organization of the intraalveolar exudate. Though originally entirely intralnminar after about four weeks the pattern may be almost entirely of an interstitial fibrosis. The histological features of idiopathic interstitial fibrosis are those of an interstitial inflammatory process consisting of variable numbers of lymphocytes, plasma cells with a small but often inconspicuous polymorph component. Depending on the stage of the disease, there is a variable amount of fibrosis, with first a widening of the involved alveolar walls and later a loss of the architectural pattem with a replacement by bands of fibrosis in which variable sized spaces lined by hypertrophied type-2 pneumocytes or by bronchiolar epithelium appear. These larger air spaces are probably derived from residual respiratory or terminal bronchioli. The overall pattern of fibrosis with alternating air spaces which appear to increase in size, perhaps as the scarfing distorts and retracts the adjacent lung tissue, gives rise to a characteristic pattern of scarring, the "honeycomb lung." The appearance of such honeycomb lung may vary, the air spaces being anything from 1-2 mm to 5-10 mm in diameter, characteristically lined by shiny intact epithelium. The relative proportions of air space to fibrosis in such honeycomb lung probably depends on the size of the original acinar units and possibly on the presence or absence of preexisting microscopic emphysema associated with past cigarette smoking. The lumen of air spaces in the involved lung of idiopathic pulmonary fibrosis show characteristically hypertrophied alveolar lining cells with the lumen itself containing varying numbers of macrophages and occasional lymphocytes. In the extreme case, those cases to which the term Desquamative Interstitial Pneumonitis (DIP) is applied, the lumen may be packed with macrophages. Originally this appearance of packed alveolar space was interpreted as being the consequence of desquamated type-2 pneumocytes. It is now a simple matter to apply immunocytochemical techniques and identify the ceils as nonepithelium macrophages. None of the histological features of idiopathic pulmonary fibrosis are characteristic, thus any of the individual components may be seen in other conditions. What is characteristic however, is the predominantly subpleural position of the inflammatory process and of the developing scarfing. The disease does
Idiopathic Interstitial Fibrosis
995
not follow a simple progression from an inflammatory phase through progressively increasing scarring to end stage disease. It appears to be more locally episodic with individual acinar units or lobular units showing active inflammation while adjacent units have progressed through to end stage scarring. There is the impression that the disease progresses in such an episodic manner from the pleura in towards the center of the lung; the honeycomb scarring present immediately under the pleura may show little evidence of the inflammation which may be present at the inner margin of a band of such honeycombing. In the developed areas of scarring it is characteristic to find irregular bands of smooth muscle proliferation which appear to be more prominent in the end stage fibrosis of idiopathic pulmonary fibrosis than of other fibrosing lung conditions. The pulmonary vasculature shows very prominent intimal and medial abnormalities which may be partly a response to the adjacent scarring and also to the local alveolar hypoxia. What can the pathologist do to help the clinician in the diagnosis and management of idiopathic interstitial fibrosis? All the histological features described are individually entirely nonspecific. It is not possible to make a diagnosis of idiopathic pulmonary fibrosis on material derived from transbronchial biopsy. The most that can be expected is the possible exclusion of conditions having more characteristic histological appearances, such as lymphatic spread by tumor or sarcoidosis. It is important that when describing the appearances in transbronchial biopsies the pathologist avoids the term interstitial fibrosis unless he qualifies it as being "nonspecific." It is easy for the children to read into a descriptive term of interstitial fibrosis the specific diagnosis of the entity, idiopathic interstitial fibrosis. The dangers of extrapolation of the appearances in small transbronchial biopsies was emphasized by Wall et al., when they showed that any attempt to extrapolate upon the predominant appearance of interstitial inflammation or scarring had no validity and gave rise to significant misdiagnosis. Open lung biopsy allows all the components of idiopathic interstitial fibrosis to be identified and allows one to exclude other specific diagnoses. However, even with the open lung biopsy the component histological processes have no absolutely characteristic appearances and the diagnosis is still based on the circumstantial evidence of the clinical situation, a range of appearances and the absence of other specific diagnoses. Selection of material by the surgeon for biopsy is of the greatest importance. As described above, the disease does not progress through characteristic stages throughout the areas of involved lung and it is possible to select old disease under the pleura which shows predominant scarfing and no evidence of inflammation, or from the margins of lesions, areas showing predominantly inflammatory changes with little scarfing. This may make little difference to the overall diagnosis but may significantly bias attempts to assess the "activity" of the diseased process upon which some clinicians base their therapeutic regimes. There is a broad correspondence between the amount of fibrosis and the lack of therapeutic response and between the amount of inflammatory infiltrate and of the desquamative component and clinical response to therapy. Those cases showing the features of DIP usually have less fibrosis and, in
996
D. Lambe
comparison to the classical histological pattern, have a longer survival and a much higher response to steroids [1]. While in the classical idiopathic pulmonary fibrosis the prognosis and response to therapy seemed to be negatively correlated to the amount of fibrosis and positively correlated to the amount of inflammation in a biopsy [3, 11]. The use of immunofluorescent techniques on frozen sections of open lung biopsies to study the possible pathogenetic mechanisms and potentially the disease activity does not seem to be of diagnostic value [2]. Bronchoalveolar lavage (BAL) is a simpler and safer technique than lung biopsy. Attempts to use differential accounts from BAL to guide management have had varied success. Overall an increase in lymphocytes is more likely to be found in responders and an increase in polymorphs and eosinophils seen in association with poor response to therapy [3, 5]. Such correlations are identified between groups of cases and the information is less specific in the case of individual patients. There is still doubt as to whether BAL data truly reflect the interstitial inflammatory state of the lung [3]. Despite recent advances in immunology and the application of immunological techniques to histological material and the use of BAL, there has been no real advance in our understanding of the nature of idiopathic interstitial fibrosis, nor in the help that the pathologist can give the clinician since the first proper histological description by Liewbow and Carrington [4]. With the exception of a small number of patients having a predominantly DIP histological pattern [1] all cases of idiopathic interstitial fibrosis progressed to develop hypoxic cor pulmonale. The honeycomb type of end stage pulmonary fibrosis has a characteristic subpleural distribution in these cases, localized to the posterior and inferior surfaces of the lower lobe with relativesparing of the anterior aspect of the lower lobe and the posterior aspect of the upper adjacent to the fissure. Despite much apparently normal lung remaining the subpleural fibrosis seems to impair alveolar ventilation to produce alveolar hypoxia and possibly affects the perfusion of the lower lobe (McCarthy and Cherniack, 1973) with the development of cor pulmonale. Intercurrent infection is a common final incident but there is also an increased number of cases who develop malignant lung tumors. The proportion of cases developing lung cancer varies from 6 to over 15% in different series [8-10] partly dependent on the autopsy rate as many as these carcinomas are clinically occult and diagnosed at post mortem examination.
References 1. Carrington CB, Gaensler EA, Coutu RE, Fitzgerald MX, Gupta RG (1978) Natural history and treated course of usual and desquamative interstitial pneumonia. N Engl J Med 298:802-809 2. Coffin B, Dewar A, Rodriguez-Roisin R, Turner-Warwick M (1985) Fine structural changes in cryptogenic fibrosing alveolitis and asbestos. J Pathol 147:107-119 3. Haslam PL, Turton C, Heard B, et al (1980) Bronchoalveolar tavage in pulmonary fibrosis: comparison of cells obtained with lung biopsy and clinical features. Thorax 35:9-18 4. Liebow AA, Carrington CB (1969) The interstitial pneumonias. In: Simon M, Potchen EJ, Lemay M, eds. Frontiers of pulmonary radiology. New York: Grune & Stratton, 102-141
Idiopathic Interstitial Fibrosis
997
5. Rudd RM, Haslam PL, Turner-Warwick M (1981) Cryptogenic fibrosing alveolitis: relationships of pulmonary physiology and bronchoalveolar lavage to response to treatment and prognosis. Am Rev Respir Dis I24:1-8 6. Scadding JC (1974) Diffuse pulmonary alveolar fibrosis. Thorax 29:271-281 7. Wall CP, Gaensler EA, Carrington CB, Hayes JA (1981) Comparison of transbronchial and open lung biopsies in chronic infiltrative lung diseases. Am Rev Resp Dis 123:280-285 8. Stack BHR, Choo-Kang YFJ, Herd BE (1972) Prognosis of cryptogenic fibrosing alveolitis. Thorax 27:535-542 9. Haddad R, Massaro D (1968) Idiopathic diffuse interstitial pulmonary fibrosis (fibrosing alveolitis)--atypical epithelial proliferation and lung cancer. Am J Med 45:211-218 10. Turner-Warwick M, Lebowitz M, Burrows B, et al (1980) Cryptogenic fibrosing alveolitis and lung cancer. Thorax 35:496-499 11. Wright PH, Herd BE, Steel SJ, Turner-Warwick M (1981) Cryptogenic fibrosing alveolitis: assessment by graded trephine lung biopsy histology compared with clinical, radiographic and physiological features. British Journal of Disease of the Chest 75:61-70
