Sarcoidosis

Sarcoidosis has been an interest of mine since about the second year in medical school. I was fascinated by this disease that could produce granulomas all over the body and then spontaneously disappear without a trace. On the other hand some cases progressed to devastating disease. What were the determinants, the controlling factors that caused such a switch? Even more basic perhaps what kind of inciting agent could produce such a variable response? Or was this aspect related to host characteristics?

Prevalence of References

I used to think at that time that I knew a great deal about that disease. As I have progressed in my reading and education, I feel less sure about my understanding and appreciation for the multitude of presentations and ramifications of that disease. In the past 11 years over 7000 references on sarcoidosis have been published. This in itself is a statement of the magnitude of the problem or at least that which is perceived by investigators.

Overview of Talk

I do not want to inundate you with the multitude of ways in which sarcoidosis can present since every organ system in the body can be affected by this process. Nor will I discuss the numerous treatment regimens attempted in this disease. I would like to give you some idea about the newer insights into the disease mechanisms and proposed markers of disease activity and whether or not these have added to our management of the disease.

Definition

We will start by defining sarcoidosis. The chapter in Harrison's textbook of medicine defines sarcoidosis as "a chronic multisystem disorder of unknown etiology characterized in affected organs by an accumulation of lymphocytes and mononuclear phagocytes, noncaseating epithelioid granulomas, and derangements of the normal tissue architecture." A more accurate definition most likely will not appear until the etiology is more clearly defined. Equally important in making the diagnosis of sarcoidosis is ruling out mycobacterial, fungal and neoplastic diseases and hypersensitivity pneumonitis.

Historical Perspective

This disease has fascinated a number of prominent physicians for over 100 years. During that time there appears to have been several phases in our understanding of sarcoidosis. Initially, the disease was thought to be only cutaneous.

Summary of Historical Trends

1869-1899 Cutaneous
Hutchinsons' Papillary Psoriasis - represented skin plaques skin plaques in various locations.
Boeck's Multiple Sarkoids - characteristic nodules that were once called sarcoids.
Besnier's Lupus Pernio - violaceous skin lesions often occurring on the face and extremities, a lesion conferring a poor prognosis.

1902-1909 Bone lesions associated with skin lesions in chronic forms of the disease.

1909-1940's Multisystem disease, first heralded by Jorgen Schaumann in 1914 who also noted the potential occurrence of organ involvement in the absence of skin lesions.

1946-1953 Lung
Lofgren's Bilateral Hilar Lymphoma Syndrome
Association with erythema nodosum, fever and arthralgias conferred favorable prognosis.

Until this time our knowledge about sarcoidosis remained descriptive without much insight into mechanisms of disease.

Diagnosis

Pathology

The characteristic lesion of sarcoidosis is the granuloma characteristically present in a repetitive monotonous pattern all granulomas existing in the same stage of development.

Each granuloma consists of a collection of epithelioid cells (former macrophages) and multinucleated giant cells of the Langhans or foreign body types surrounded by lymphocytes and lesser numbers of macrophages and plasma cells in a compact granulomatous structure. This is surrounded by a rim of fibroblasts. In a given tissue the pattern is of regular repetitive granulomas without necrosis, although some hyalinization can be seen especially in the later fibrotic stages.

Within the giant cells are occasionally seen asteroid bodies or

Schaumann bodies or

crystalline inclusions. Inclusion bodies are seen in a number of other granulomatous diseases and are therefore not specific for sarcoidosis. The surrounding lung tissue may be entirely normal without much interstitial pneumonia.

Granulomatous vasculitis is frequently found in sarcoidosis and consists of noncaseating granulomas within the intima and media of vessels. These may compress the lumen but do not cause necrosis of vessel walls. In sarcoidosis the lesions are more likely perilymphatic, especially in the interlobular septa whereas peribronchial lesions suggest hypersensitivity pneumonitis.

Because other diseases may include noncaseous granulomas (berylliosis, infection, hypersensitivity pneumonias) it is important to have at least two tissue sources of noncaseous granulomas or one tissue source with minimal symptoms. Many of the diseases on this list can be eliminated on the basis of clinical findings and historical information. Especially important is a detailed occupational and environmental history from the patient.

Differential Diagnosis of Granulomatous Disease

Hypersensitivity Pneumonitis
Berylliosis
Zirconium Disease
Silicosis
Starch Exposure
Crohn's Disease
Fungal Disease

Mycobacteria:
Tuberculosis
Atypical Mycobacteriosis
Leprosy

Spirochetes:
Syphilis

Other:
Hodgkins Disease

The liver and scalene lymph nodes, although frequently positive in sarcoidosis are so nonspecific as to be meaningless adjuncts to the diagnosis.

In summary, the major Classical Pathologic Findings include:

1. Monotonous pattern of epithelioid granuloma
2. Giant Cells
3. Inclusion Bodies in Giant Cells
4. Multiple Organ Involvement

Alveolitis

"Alveolar interstitial infiltration of macrophages and lymphocytes is present in the early phases of sarcoidosis and may precede granuloma formation. It appears to be inversely related to the frequency of granulomas." Rosen, 1978

In 1978 Rosen and colleagues studied the pathology of early phases of pulmonary sarcoidosis. They found a high percentage of interstitial pneumonitis with macrophages and lymphocytes (macrophage predominance) which was thought to precede granuloma formation but associated with normal lung and with granulomas elsewhere in the lung. Cells are NOT in the alveolar spaces and there are not hyaline membranes. This description of an interstitial pneumonitis appears to contradict statements made in established textbooks of pulmonary disease. The classic description given in Katzenstein's textbook of lung pathology includes a statement that if interstitial pneumonia is a prominent component then a vigorous search for agents potentially causing hypersensitivity pneumonitis should be investigated. The importance of this description will become apparent in the next few minutes.

Pathogenesis/Etiology

What have we learned in the last 10 years concerning the pathogenesis sarcoidosis? Today the current thrust of research is based on immunologic investigation into granuloma formation in general and comparison with other granulomatous diseases which have specific or at least better defined antigenic causes. In general the disease is associated with depressed peripheral delayed hypersensitivity, peripheral lymphopenia, increased total and T lymphocytes and OKT4:T8 subsets locally, hyperactive B cells and increased circulating immune-complexes. I will summarize some of the findings of interest although I hasten to say that almost every phase of the investigation is controversial and little is universally accepted. The basic philosophy is that an antigen or group of antigens as of yet undefined, stimulate a response in macrophages and lymphocytes that we call an alveolitis and which consists of increased accumulation of these cell types in the alveolar interstitium. This leads to granuloma formation, the hallmark of sarcoidosis. The granuloma is a functional entity capable of self perpetuation and the production of agents capable of injuring the lung parenchyma and stimulating lung fibroblasts to proliferate and lay down collagen in the form of pulmonary fibrosis. This rather simplistic theory has been examined more closely, mostly with the aid of bronchoalveolar lavage in order to sample the lung compartments directly. This technique consists of inserting a fiberoptic bronchoscope into the conducting airways of the lung, wedging the bronchoscope in a peripheral airway (usually only 4th or 5th order branch) and instilling a solution, usually saline, into the subsegment. The instilled solution is aspirated back and the fluid studied. This fluid is referred to by some as alveolar lining fluid although larger airways are also bathed in the liquid prior to aspiration. The fluid can be studied for cellular and biochemical composition, and the cells can be isolated and tested functionally in vitro. The fluid can also be tested for its ability to affect cellular function in vitro. Using this technique it has been determined that cells from sarcoid patients contain high proportions and absolute amounts/ml of fluid of lymphocytes. Normally macrophages comprise 90% and lymphocytes 10% or less of BAL cells.In sarcoidosis the total number of cells is several times the number found in controls. Furthermore, the percentage of lymphocytes is increased to about 33% of the total but may be as high as 55-60% of total cells.

In 1983 Gary Hunninghake showed that specifically T-helper lymphocytes were elevated during active phases of the disease. This may pre-date clinical and radiographic abnormalities. The T-cells have a higher T4/T8 ratio due to enhanced T4 levels. This is in contrast to peripheral blood lymphocytes which may be lowered overall and may have normal or lower T4/T8 ratios. Thus the peripheral blood immune effector cell response is felt not to adequately reflect the lung activity in sarcoidosis. The T cells are activated as determined by their ability to proliferate spontaneously. Furthermore, there appeared to be a bimodal distribution of lymphocyte numbers among various patients such that one group had 28% lymphocytes and the other had <= 28% lymphocytes. The hypothesis was that those in the higher BAL lymphocytosis group had a high intensity alveolitis that might be clinically relevant. Although the percentage is decreased, the absolute numbers of alveolar macrophages in BAL are increased although not to the same degree as are the lymphocytes. The results of bronchoalveolar lavage cell counting was believed to correlate with the degree of interstitial inflammatory cell infiltration histologically. It is of interest to point out the macrophage predominance in the original histologic description of this alveolitis by Rosen in 1978 in contrast to the lymphocyte predominance of BAL studies by Hunninghake and others subsequently. Confusion may come from the fact that the total percentage of macrophages is higher than the percentage of lymphocytes in all cases but the relative increase in lymphocytes is greater than that for macrophages thus justifying the term lymphocytic alveolitis. Alveolitis thusly defined has been described in other extrathoracic granulomatous diseases with no clinical or radiographic evidence of lung involvement i.e. subclinical alveolitis. However, most reports lack histologic confirmation, and, therefore the diagnosis of alveolitis in most cases rests on finding increased lymphocytes in BAL. We will get back to this point later.

Keogh in 1983 demonstrated that high intensity alveolitis often reverted to low intensity alveolitis and to a lesser extent vice versa. Although they claim that high intensity alveolitis more often progressed to deterioration, reversion from high to low intensity alveolitis makes this an unreliable clinical predictor. Furthermore, other reports fail to detect functional deterioration of subclinical alveolitis. As an example, farmers with subclinical alveolitis did not develop hypersensitivity pneumonitis in a four year follow-up study. In fact increased T cells were not found even though T4/T8 ratios were somewhat elevated. Both cell types, i.e. macrophages and lymphocytes, are present in activated states. Macrophage activation is characterized by increased ability to release reactive oxygen species and enzymes, spontaneous release of interleukin-1 and increased ability to interact with the coagulation-fibrinolytic system. Macrophages also express Ia-like antigens on their surface.

In 1984 Gary Hunninghake demonstrated the spontaneous release of interleukin-1 by alveolar macrophages from patients with sarcoidosis compared with normals and those with idiopathic pulmonary fibrosis. The assay used was a functional assay based on ability to induce proliferation of thymocytes. T lymphocyte activation is manifested by active proliferation, formation of rosettes with sheep erythrocytes, and spontaneous release of Migration Inhibition Factor, Macrophage Chemotactic Factor, IL-2, gamma IF and factors that stimulate B cells to differentiate into immunoglobulin-secreting cells. Also in 1984 Gary Hunninghake demonstrated the ability of lymphocytes to release Il-2 based on an assay of cellular proliferation. Proposed mechanisms: IL-1 and IL-2 are chemotactic for blood T-cells and this increased accumulation in the lung may account for the relative peripheral blood T-lymphopenia often observed. Activated lung T-cells release monocyte chemotactic factor which may attract blood monocytes to the lung and specifically to sites of disease. To what extent local alveolar macrophages and T-lymphocytes undergo cell division to enhance their numbers is unclear, however, both IL-1 and IL-2 can function as T-lymphocyte growth factors. Now the activated lymphocytes produce macrophage migration inhibitory factor which fixes the attracted macrophages in place at the site of injury. This along with immune interferon (gamma interferon) induces macrophage differentiation and stimulates the formation of granulomas. This pattern represents a mechanism common among a number of granulomatous disorders affecting the lung and other tissues and probably indicates that tissues are capable of reacting in only a limited number of ways to offensive agents.

Activated T-lymphocytes also produce B-lymphocyte growth factor and B-lymphocyte differentiation factor which stimulate B-lymphocytes to produce antigen- stimulated antibodies as well as polyclonal antibodies. Granuloma may then be responsible for its own perpetuation and for progression to fibrosis although some modulation must occur since fibrosis does not always accompany the granulomas of sarcoidosis.

Mosely, in 1986, in Hunninghake's lab demonstrated that BAL cells from sarcoidosis patients released interferon and growth factor activity for human lung fibroblasts. Patients whose cells released interferon had elevated fibroblast growth compared with cells that do not spontaneously release interferon. This may represent a synergistic effect between gamma interferon and interleukin- 1 on fibroblast growth.

Fibronectin and alveolar macrophage derived growth factor are also released by alveolar macrophages from patients with sarcoidosis although their role is less well defined in this disease. Fibronectin aids in recruiting macrophages, promotes adherence of macrophages to extracellular matrix, and promotes fibroblast proliferation. Macrophage derived growth factor promotes fibroblast proliferation. Fibroblasts are the major extracellular matrix cell capable of producing collagen which is laid down excessively in fibrotic disease states. A potential controlling factor, PGE2 is released by alveolar macrophages and inhibits granuloma formation in animal models of granulomatous diseases. PGE2 also inhibits fibroblast growth in vitro presumably by inhibiting IL-1 induced release of IL-2 by T cells. Furthermore, Wolter in Peter Ward's group in 1983 demonstrated that the capacity of alveolar macrophages to release PGE2 is decreased in patients with sarcoidosis. Thus diminished protection from granuloma formation may be another factor in the pathogenesis of the sarcoid granulomatous lesion.

Activity

The utility of explaining the factors involved in the pathogenesis will be realized only if they contribute to our ability to predict and modify disease progression. In that context the initial step in therapy would be to develop markers capable of determining whether or not the disease was active and whether or not it was going to progress. Traditionally, radiographic stages were described in order to assess disease activity.

Radiographic Stages

Stage 0 Normal chest x-ray, non-caseating granulomas, can have (3%) low DLCO

Stage I Bilateral hilar lymphadenopathy, often with right (61%) paratracheal nodes, normal lung fields, NCG on path in over 90% of cases. Don't need tissue if patient is asymptomatic.

Stage II BHL plus pulmonary infiltrates. (25.3%)

Stage III Pulmonary infiltrates without BHL. (9.7%)

Stage IV Pulmonary fibrosis, honeycombing. (1%)

Currently most pulmonologists and chest radiologist lump Stages III and IV together for practical purposes. You can't distinguish fibrosis from infiltrates with certainty without honeycombing. It was initially believed that these were true stages, i.e. Stage 0 progressed to stage I and stage I progressed to stage II, etc. This may occur but it does not necessarily occur. Furthermore, even stage III can spontaneously remit as was the situation for the case presented at the beginning of this talk. Despite the lack of sequential progression the stages do predict disease course to some extent.

In 1967 in a world wide survey of 1609 patients with sarcoidosis, Siltzbach demonstrated that

% of Stage Regressing

Stage I 60 [hilar adenopathy alone]
Stage II 46 [hilar adenopathy plus parenchymal infiltrates]
Stage III 12 [parenchymal infiltrates alone]

73% of all sarcoid patients have a normal chest radiograph within two years of presentation.

So to reiterate, even though presentation with a given stage is associated with a particular prognosis, the stages do not necessarily represent phases of progression. Furthermore, the stages do not correlate with pulmonary function. One of the major problems we are dealing with is that we don't have a good handle on what activity means. We assume this to indicate active granuloma formation and that this in turn reflects destruction of normal lung architecture which may ultimately result in deranged function. A number of parameters have been reported to correlate with activity. A number of these are based on products made by granulomas or by various cells of granulomas either spontaneously or upon stimulation in vitro.

SACE

Serum angiotensin converting enzyme (SACE) is the most widely used test in the follow-up of sarcoidosis patients. It is a membrane bound glycoprotein found in activated alveolar macrophages and at the surface of epithelioid cells and consequently may reflect development and extent of granulomas. It may be normal in early and acute disease reflecting the small number of granulomas present at this stage. It is poorly correlated with pulmonary function and variably related with radiographic stage. It correlates with the degree of extrapulmonary lesions. Steroids decrease SACE with a delay after onset of treatment. A minimum dose of prednisone is required to normalized SACE. After discontinuing steroids, a transient increase in SACE may occur before final normalization. In spontaneous recovery of sarcoidosis without steroids, even previously persistently elevated SACE levels may normalize. It is positive in 60% of sarcoidosis patients. Some investigators believe it to be a convenient and practical marker of activity.

ACE in various diseases

Sarcoidosis

Conditions with Elevated Serum Angiotensin Converting Enzyme

60 % of Sarcoidosis Patients Have Elevated SACE

Serum angiotensin converting enzyme levels however are nonspecific and are not helpful in diagnosing sarcoidosis, especially since levels are increased in a few diseases which are often difficult to distinguish from sarcoidosis.

Gallium Scan Use

67Gallium Scans are also used frequently in following the activity of sarcoidosis. It is neither specific nor useful for diagnosis of sarcoidosis. Gallium is fixed by sarcoid granulomas and is believed to assess activity of T lymphocytes.

It is believed to reflect:
a. the extent of disease at different sites
b. degree of granuloma activity at those sites.

Repeated gallium scans are more sensitive than the chest radiograph and symptoms but no studies have yet compared gallium scans with lung CT scans in sarcoidosis.

Its value is in non-invasively
a. assessing the extent and [to a lesser degree] the degree of inflammatory process
b. choosing preferential sites for BAL or lung biopsy
c. following therapeutic response.

Keogh in 1982 stated that the gallium scan is the best way to assess the intensity of activated alvelar macrophages and that BAL was the best way to assess lymphocytes and neutropils. This may be true but it is unclear what relationship these have to the management of patients. The radiation dose of multiple scans may preclude its widespread use as well as its cost, therefore it is of limited value in management of sarcoidosis.

BAL USE

Cell counts in bronchoalveolar lavage fluid have been used to assess activity. Specifically a high intensity alveolitis has been described as having lymphocytes counts 28% of total cells in the lavage fluid. This high intensity alveolitis is said to reflect disease activity and tendency toward progression of the disease. Alveolar lymphocytosis is believed to be a direct reflection of alveolitis. We will say more about this later. Sequential BAL lymphocytes may be more significant especially after 12 months at which time persistence of lymphocytosis may be a predictor of disease progression. However, all the schemes based only on BAL lymphocyte counts are not reliable enough for disease management. In general, none of the newer modalities have proved useful clinically.

Lack of Clinical Utility of Newer Markers

As stated in a critical review by Chretein in 1988, neither the significance nor the reliability in terms of predictive value and therapeutic guidelines, have so far been clearly demonstrated in any such parameters or biological features despite their theoretical interest. (Chretien, 1988)

This may be partially due to
1. spontaneous regression in many situations - i.e. sarcoidosis can be highly active over limited periods of time and yet clear completely. Furthermore, those uses destined to progress to more fibrosis may have a more insidious onset and evidence of activity may be much less florid.
2. the fact that we have no good definition of disease activity or gold standard with which to compare various measures.

In fact, Margaret Turner-Warwick in a 1986 study and in several other studies has determined that SACE, gallium scans and BAL lymphocyte counts (including T subset determinations) fail to predict radiographic response to steroid therapy accurately enough to be clinically useful in individual patients. As she states in a 1988 article, activity of various individual cellular components of a granuloma is not necessarily synonymous with progression towards irreversible destruction of the lung architecture.

Speculation On Etiology

Previously Considered Etiologic Factors

Since its initial presentation as a cutaneous disease sarcoidosis has aroused a multitude of responses from clinicians as well as basic scientists concerning the etiology of this disease. Early in its history due to its cutaneous predominance it was likened to syphilis and later because of lymph node involvement to lymphoma of the Hodgkins variety. The most obvious correlation because of its histologic hallmark the granuloma was tuberculosis which it was believed to represent in some aberrant fashion a theory which is still not impossible although less likely. A variety of infectious agents have been proposed based on anecdotal reports of culture positivity, histologic visualization or blood and tissue immunoreactivity with infectious antigens. None of these has been born out and none has satisfied Koch's postulates. Basically a fairly extensive search for an antigen has been fruitless.

Routes of Exposure to Suspected Antigen(s)

In thinking about the etiology of sarcoidosis several observations are important to consider. The initial presentation is respiratory, cutaneous or ocular in most of the cases. However, almost all cases have lung involvement. Respiratory: exposed to environment via aerosolization allowing exposure to respiratory surfaces (we'll get back to this later). Lung - The lung presents a problem because larger airways are not often affected functionally (although granulomas are observed when looked for). Alveolitis is felt by some to be the initial manifestation. However, the term is misleading in that it represents an interstitial inflammatory process. There are several things conceptually wrong about this supposition.

a. The interstitium is not that compartment of the lung exposed directly to the environment but the alveolar surface i.e. epithelium is. So we have to figure out how the antigen reaches that compartment.

Requirements of Aerosolized Antigen

In order to be aerosolized and reach distal parts of the lung several features are required. The aerosolized antigens must be of the correct solubility so as not to be diluted but yet be transported to distal lung regions. If the supposed antigen is particulate, the aerosolized antigen(s) need to be of the correct mean aerodynamic diameter and mass to prevent their sedimentation in larger airways but to allow bypassing impaction at branching points and distribution to the distal lower respiratory tract.

Alveolitis

The concept of alveolitis appears to be essential for our current theories yet there is something wrong with the dependence upon the presence of alveolitis.

b. The diagnosis of alveolitis is made by BAL i.e. predominance of lymphocytes especially T helper subset, not by histologic evaluation. In fact, there are not good histologic comparison studies. Only one report 10 years ago and one brief mention of alveolar wall thickening by Carrington in 1976.

c. Classical histologic descriptions mention NCG without interstitial reaction, i.e. inflammatory cells and that NCG are the earliest findings. Is the alveolitis an artifact of BAL technique? One study by Semenzato in 1985 demonstrated reasonable correlation between BAL and histologic lymphocyte counts and low intensity but not high intensity alveolitis making the entire procedure too variable for clinical use.
In Hunninghake's 1981 paper the authors describe a good correlation between open lung biopsy cell extracts and BAL cells in normals as well as in patients with idiopathic pulmonary fibrosis and sarcoidosis. Although the mean values appear correlated, the variation was large and the number of patients studied small (9 IPF, 6 sarcoid). Thus, in an individual patient it is not possible to rely on the correlation. Furthermore, since the biopsy material was subjected to extraction rather than using morphometric measures of cell number, the cell counts may be biased by variations in their ability to extract cells. In justifying the use of alveolitis, reports mention that the chest radiograph and lung function testing bear little relationship to the status of the alveolitis. This is not the issue and represents circular reasoning!! The question is: what relationship does the alveolitis (as defined by BAL) have to the activity of the disease and to the progression of the disease? In actual fact, chest radiographic stage is a better predictor of outcome than any parameter we have at the current time. I have major reservations in general on a theoretical basis about the utility of BAL in monitoring disease activity in any interstitial disease. The entire procedure is fraught with sampling error, not because the sample is too small but because the sample may not adequately reflect the compartment of major interest. The sample is of the alveolar space. The disease is of the interstitium. To what extent does BAL reflect events occurring across the alveolar capillary membrane in the interstitium whether it be cell counts or biochemical measurements? Also, using this technique one can never have an appropriate control. In disease when the integrity of the alveolar capillary membrane may be violated BAL may more accurately reflect events in the interstitium. However, in controls with normal alveolar capillary membrane, BAL is unlikely to reflect interstitial events. Therefore, there will never be an adequate control group in which one can compare sampling of the same lung compartment as in the experimental group.

Conclusions

Inhaled antigen is most likely cause since lung is almost always involved. Genetic predisposition is also likely in that some host feature must be present to allow granuloma of the sarcoid variety to form. Multiple antigens are likely to be involved in cooperative provocation of the sarcoid granuloma. Major advances in our understanding of sarcoidosis will have to wait until we have a clear understanding of the etiologic agent or agents. All other studies by there very nature will be indirect. Furthermore, because of the difficulties in diagnosing the disease it is imperative to have tissue documentation along with appropriate clinical supporting evidence that you are dealing with sarcoidosis and not with another granulomatous disease.

This is not intended to be discouraging. Quite the contrary, what we have learned is that
1. We have a clearer understanding of the pathogenesis of granuloma formation.
2. The disease regresses in most situations
3. We have re-confirmed the utility of our time honored measures and predictors of disease progression, namely the chest radiographic patterns and pulmonary function tests.

I would like to end with a quote from Sir Arthur Conon Doyle which summarizes our understanding of sarcoidosis. As Sherlock Holmes put it, "when you have eliminated all which is impossible then whatever remains, however improbable, must be the truth."

SARCOIDOSIS
References
Norman T. Soskel, M. D.
January 11, 1989

Carrington CB, Gaensler EA, Mikus JP, Schachter, AW, Burke BW, Goff AM. Structure and function in sarcoidosis. Ann New York Acad Sci 278:265-283, 1976.

Chretien J. Assessment of sarcoid activity: state of art. In: Grassi C, Rizzato G, Pozzi E, eds. Sarcoidosis and other granulomatous disorders. Proceedings of the XI World Congress on Sarcoidosis and other Granulomatous Disorders, Milan 6-11 September, 1987. Excerpta Medica International Congress series 756, Elsevier Science Publishers B.V., 525-539, 1988.

Crystal RG. Sarcoidosis. In:Harrison's textbook of medicine, 11th edition, 1987, pp 1445- 1450.

Crystal RG, Roberts WC, Hunninghake GW, Gadek JE, Fulmer JD, Line BR. NIH Conference. Pulmonary sarcoidosis: a disease characterized and perpetuated by activated lung T- lymphocytes. Ann Int Med 94:73-94, 1981.

Hunninghake GW, Bedell GN, Zavala DC, Monick M, Brady M. Role of interleukin-2 release by lung T-cells in active pulmonary sarcoidosis. Am Rev Respir Dis 128:634-638, 1983.

Hunninghake GW. Release of interleukin-1 by alveolar macrophages of patients with active pulmonary sarcoidosis. Am Rev Respir Dis 129:569-572, 1984.

Katzenstein A-L A, Askin FB. Surgical pathology of non-neoplastic lung disease. W. B. Saunders Co., 1982, pp. 152-158.

Kent DC, Houk VN, Elliott RC, Sokolowski JW, Jr., Baker JH, Sorensen K. The definitive evaluation of sarcoidosis. Am Rev Respir Dis 101:721-727, 1970.

Keogh BA, Crystal RG. Alveolitis: the key to the interstitial lung disorders. Thorax 37:1-10, 1982.

Keogh BA, Hunninghake GW, Line BR, Crystal RG. The alveolitis of pulmonary sarcoidosis. Evaluation of natural history and alveolitis-dependent changes in lung function. Am Rev Respir Dis 128:256-265, 1983.

Klatskin G. Hepatic granulomata: problems in interpretation. Ann New York Acad Sci 278:427- 432, 1976.

Reynolds HY. Other granulomatous diseases - how can they help with understanding sarcoidosis? Outlook. In: Grassi C, Rizzato G, Pozzi E, eds. Sarcoidosis and other granulomatous disorders. Proceedings of the XI World Congress on Sarcoidosis and other Granulomatous Disorders, Milan 6-11 September, 1987. Excerpta Medica International Congress series 756, Elsevier Science Publishers B.V., 719-735, 1988.

Rosen Y. Athanassiades TJ, Moon S, Lyons HA. Nongranulomatous interstitial pneumonitis in sarcoidosis. Relationship to development of epithelioid granulomas. Chest 74:122-125, 1978.

Scadding JG and Mitchell DN. Sarcoidosis, 2nd edition. Chapman and Hall, London, 1975.

Schwartz MI. Diagnosis of sarcoidosis. Clin Derm 4:165-167, 1986.

Semenzato G, Chilosi M, Ossi E, Trentin L, Pizzolo G, Cipriani A, Agostini C, Zambello R, Marcer G, Gasparotto G. Bronchoalveolar lavage and lung histology. Comparative analysis of inflammatory and immunocompetent cells in patients with sarcoidosis and hypersensitivity pneumonitis. Am Rev Respir Dis 132:400-404, 1985.

Siltzbach LE. Sarcoidosis: Clinical features and management. Med Clin North Am 51:483-502, 1967.

Siltzbach LE, James DG, Neville E, Turiaf J, Battesti JP, Sharma OP, Hosoda Y, Mikami R, Odaka M. Course and prognosis of sarcoidosis around the world. Am J Med 57:847-852, 1974.

Thomas PD, Hunninghake GW. Current concepts of the pathogenesis of sarcoidosis. Am Rev Respir Dis 135:747-760, 1987.

Turner-Warwick M, McAllister W, Lawrence R, Britten A, Haslam PL. Corticosteroid treatment in pulmonary sarcoidosis: do serial lavage lymphocyte counts, serum angiotensin converting enzyme measurements, and gallium-67 scans help management? Thorax 41:903-913, 1986.

Turner-Warwick M. Treatment of pulmonary sarcoidosis. State of the art. In: Grassi C, Rizzato G, Pozzi E, eds. Sarcoidosis and other granulomatous disorders. Proceedings of the XI World Congress on Sarcoidosis and other Granulomatous Disorders, Milan 6-11 September, 1987. Excerpta Medica International Congress series 756, Elsevier Science Publishers B.V., 621-629, 1988.

Click here to go to table of contents

Click here to send me e-mail