October - December 2006: 
Volume 19, Issue 4

Click on the image to download the Issue in PDF format.


Relationship of MMP9 and tissue inhibitor of MMP1 (TIMMP1) to functional and ultrasonographic markers in sarcoidosis and idiopathic pulmonary fibrosis
Metalloproteinases and their tissue inhibitors act on the metabolism of collagen in the interstitial tissue matrix and are involved in the pathogenesis of tissue remodeling and fibrosis. The study objective is to investigate the MMP9 and TIMMP1 status in patients with sarcoidosis and idiopathic pulmonary fibrosis (IPF), and their relationship to functional and ultrasonographic markers. Patient and methods: Serum levels of MMP9 and TIMMP1 in 22 sarcoidosis patients, in 10 IPF patients and 12 control subjects were determined (ELISA, Human Biotrak, Amersam). In sarcoidosis patients, disease stage (Ga167 scintigraphy) and left ventricular diastolic function (pulse and tissue Doppler) were further assessed. In IPF patients, systolic pulmonary artery pressure (SPAP) was measured; all patients underwent comprehensive lung function testing. Results: Serum MMP9 and TIMMP1 levels were similar among patients with either pathology; however, they were significantly higher in patients with sarcoidosis (p<0.001, p<0.01), as well as in patients with IPF (p<0.01, p<0.05) compared to controls. Among sarcoidosis patients, there was a significant relationship between MMP9 levels and both FEF25-75 (p<0.05) and disease activity (p<0.01), whereas among IPF patients MMP9 levels were associated with values for DLCO (p<0.05). 41% of sarcoidosis patients showed left ventricular diastolic dysfunction in terms of impaired relaxation and 40% of IPF patients had SPAP >35 mm Hg. No significant relationship between ultrasonographic findings and MMP9 and TIMMP1 levels, and the molar ratio MMP9/TIMMP1 was found. Conclusion: MMP9 levels reflect lung function impairment in sarcoidosis and IPF patients. Nevertheless, they cannot reliably indicate cardiovascular involvement. However, in sarcoidosis they may represent a useful marker of disease activity. Pneumon 2006; 19(4):357-366.
Full text


Matrix metalloproteinases (MMPs) are a group of endopeptidases that degrade extracellular matrix proteins, such as collagen, gelatin, elastin and fibronectin. Their dynamic balance with tissue inhibitors of their activity regulates the turnover of tissue matrix in the processes of aging and normal tissue regeneration.1 Recent studies lay special emphasis on the disturbance of this balance in tissue remodeling and angiogenesis, which may result in diseases such as bronchial asthma, chronic obstructive pulmonary disease, cancer, pulmonary hypertension, and, most recently, sarcoidosis and idiopathic pulmonary fibrosis.2 Pathological similarities in the pathogenesis of tissue remodeling and fibrosis have been demonstrated between the last two conditions;3 further evidence from bronchoalveolar lavage (BAL) analyses and immunohistochemical studies suggests that MMPs, in particular MMP9 (gelatinase), and their inhibitors are implicated in their progression.

In the present study, in the quest for reliable clinical markers to monitor inflammation and fibrosis, serum levels of MMP9 and the tissue inhibitor of MMP1 (TIMMP1) in patients with sarcoidosis and idiopathic pulmonary fibrosis were measured using the standard immunoenzymic assay ELISA. The study investigates whether the balance between MMP9 and TIMMP1 in peripheral blood relates to disease activity in sarcoidosis, and lung function impairment and cardiovascular involvement in both conditions.


Twenty two non-smoking patients with histologically confirmed sarcoidosis were studied. These subjects were classified by disease stage on the basis of radiographic findings on chest x-ray as follows:5 stage I, eight patients with mediastinal/portal lymphadenopathy, without parenchymal involvement; stage II, eight patients with lymphadenopathy and parenchymal infiltrations, without evidence of pulmonary fibrosis; and stage III, six patients with pulmonary infiltrations or fibrosis, without lymphadenopathy. Disease activity was evaluated using whole-body scintigraphy with 67Ga-citrate (Lamda/Panda sign).6 The disease was active in 14 patients and latent in 8 patients.

The study also included 10 non-smoking patients with idiopathic pulmonary fibrosis. Diagnosis was based on ATS/ERS diagnostic criteria with histological confirmation in two patients and standard radiographic findings in high resolution computed tomography (HRCT) in all patients.7

Flows, lung volumes and capacities were measured in all patients using spirometry; CO diffusion was evaluated using the single re-breath method (Ganshorn, GMBH). Partial pressures of arterial blood gases were determined in blood samples from the brachial artery while breathing room air (Radiometer Copenhagen, ABL). In patients with sarcoidosis, serum angiotensin-converting enzyme (SACE) levels were determined using an enzymic method.

Left ventricular diastolic function was evaluated in sarcoidosis patients using pulse and tissue Doppler (General Electric Vivid 3 Ultrasound Machine). By simultaneous guidance of the two-dimensional heart image and placement of the transducer in the apical position for a 4-chamber projection, the stroke volume sample from pulse Doppler was placed in left ventricular outflow, in particular in the apex of the leaflet aperture. The following diastolic function parameters were assessed: fast diastolic filling (E wave; cm/sec); slow diastolic filling (A wave; cm/sec), E/A ratio, flow delay time (DT; msec), diastolic velocity at the mitral annulus (Em; cm/sec). In addition, the thickness of interventricular septum (IVSd; mm) and left ventricular systolic function (LVEF %) were also assessed. Diastolic function was considered impaired if three of the five parameters were abnormal, provided that E/A <1 or/and Em <8 cm/sec.

In patients with idiopathic pulmonary fibrosis, systolic pulmonary artery pressure (SPAP; mm Hg) measurements were made using continuous Doppler and applying the modified Bernoulli formula on the tricuspid regurgitation wave.8 Pulmonary hypertension was diagnosed if SPAP was higher than 35 mm Hg.

MMP9 and TIMMP1 levels in the serum of all patients and 12 control subjects (mean age 50.2±7 years, nonsmokers, with normal lung function) were measured using the quantitative immunoenzymic method ELISA (Human Biotrak, Amersam Biosciences).9 The method identifies the precursor form of MMP9 (pro-MMP9, 92 kDa) and can determine its free form and its TIMMP1-bound form.In addition, it detects total TIMMP1 levels, representing the sum of the free plus the MMPs-bound form.

Serum MMP9 and TIMMP1 levels were determined according to the following protocol:9 in the first step of incubation, MMP9 and TIMMP1 molecules bind on microplates pre-incubated with monoclonal antibodies. In the second step, polyclonal horseradish peroxidaselabeled antibodies are added and fixed complexes are formed. The amount of peroxidase was determined by adding tetramethylbenzidine (TMB). The reaction was terminated by the addition of acid, which produced a colored solution measured by spectrophotometry at 450 nm. The concentrations of MMP9 and TIMMP1 were determined using a standard logistic curve involving 4 parameters (4PL). Enzyme concentrations are expressed in ng/mL and the sensitivity of the analytic assay is 0.6 ng/mL for MMP9 and 1.25 ng/mL for TIMMP1.

Anthropometric characteristics and lung function parameters of study subjects are shown in Table 1.

Statistical analysis

All values are presented as mean ± standard deviation (SD). Mann-Whitney U test was used for the comparison of the values of the same quantitative parameter between two independent groups. Spearman correlation coefficient was used to determine the degree of correlation between two variables. Comparisons between mean values of two or more data groups were made using analysis of variance (ANOVA). Statistical significance was assumed for p values lower than 0.05 (p <0.05). The statistical package SPSS was used for all statistical analyses.


All patients were middle-aged; there were no significant differences in age between patients in the two disease groups and controls. MMP9 and TIMMP1 levels and the MMP9/TIMMP1 ratio were significantly higher in sera from sarcoidosis patients (p<0.001; p<0.01; p<0.001) and IPF patients (p<0.01; p<0.05; p<0.05) compared to controls (Fig. 1). Although MMP9 levels in sarcoidosis patients were higher than in IPF patients, this difference was not of statistical significance. There was a significant correlation between MMP9 and TIMMP1 levels in both diseases (sarcoidosis: r 0.49, p<0.05; idiopathic pulmonary fibrosis: r 0.60, p<0.05).

As regards lung function and SACE levels, no difference between patients with active and latent disease was noted (Table 1). Diffusion capacity was significantly impaired in patients with pulmonary fibrosis compared to patients with sarcoidosis (p<0.001).

MMP9 levels in sarcoidosis patients showed a significant correlation with FEF25-75 (p –0.43, p<0.05; Fig. 2) both in the total of patients and separately in the two sub-groups of active and latent disease. A statistically significant correlation of MMP9 levels with active sarcoidosis based on 67Ga-citrate scintigraphy (r 0.50, p<0.05) was also found; values were significantly higher in the 14 patients with active sarcoidosis compared to the 8 patients with latent disease (p<0.05; Fig. 3). Nevertheless, MMP9 levels showed no association with SACE levels and disease stage. With respect to SACE levels, there was no significant difference in any of the three sarcoidosis stages, and no correlation with active disease or FEF25-75.

TIMMP1 levels and MMP9/TIMMP1 ratio were not associated with other disease parameters in patients with sarcoidosis. On the other hand, analysis of variance (ANOVA) did not show statistically significant differences in MMP9 and TIMMP1 levels, and MMP9/TIMMP1 ratio between the three stages of the disease.

In patients with idiopathic pulmonary fibrosis, MMP9 levels correlated significantly with DLCO (r 0.60, p<0.05). However, no other correlation with any other lung function parameter was detected; further, TIMMP1 levels and MMP9/TIMMP1 ratio correlated with none of the assessed lung function parameters.

Pulmonary hypertension (SPAP >35 mm Hg) was identified in 4/10 (40%) pulmonary fibrosis patients; however, the mean systolic pulmonary artery pressure (29±7 mm Hg) was not associated with MMP9 and TIMMP1 levels and MMP9/TIMMP1 ratio.

Left ventricular diastolic dysfunction, primarily affecting relaxation phase, was detected in 9/22 (41%) patients with sarcoidosis. Left ventricular diastolic dysfunction presented in 8/14 (57%) patients with active disease and 1/8 (12.5%) patient with latent disease (p<0.001). All patients had normal left ventricular systolic function (EF >45%) and normally thick interventricular septum (IVSD), showing no signs of cardiomyopathy; none of the patients presented pericarditis.

The assessed diastolic parameters showed no statistical correlation with MMP9 and TIMMP1 levels, and MMP9/ TIMMP1 ratio. Furthermore, there was no significant difference in MMP9 and TIMMP1 levels and MMP9/ TIMMP1 ratio between patients with diastolic dysfunction and patients with normal heart function (540.4±174 ng/ mL, 180.5±50 ng/mL, 2.9±1.4 versus 485.3±130 ng/L, 156.6±65 ng/mL, 3.32±1.5, respectively).


According to the findings of the present study, MMP9 levels, tissue inhibitor of MMP1 levels and MMP/ TIMMP1 ratio are increased in patients with sarcoidosis. Furthermore, MMP9 levels are associated with active disease, as assessed by 67Ga-citrate scintigraphy and the degree of small airway obstruction (FEF25-75). In patients with idiopathic pulmonary fibrosis, MMP9 and TIMMP1 levels and MMP9/TIMMP1 ratio in serum are increased; MMP9 levels, in particular, are associated with CO diffusion capacity, indicating an impairment in lung function.

Importantly, in both conditions, the levels of pro- MMP9, both free and bound forms, determined by ELISA, were associated with the levels of free and MMPs-bound TIMMP1, which enhances the reliability of the analytic method. Another interesting finding is that increased levels of MMP9 are associated with a compensatory increase in the respective tissue inhibitor, thus suggesting a homeostasis process in matrix turnover and tissue damage repair. It should be noted, though, that the increase in TIMMP1 had a lower level of statistical significance compared to increased levels of MMP9 in both disease groups.

This is the first study to evaluate serum MMP9 and TIMMP1 levels in patients with sarcoidosis and idiopathic pulmonary fibrosis using ELISA. This analytic method was first used for the evaluation of MMP9 and TIMMP1 status in peripheral blood by Bosse et al in patients with bronchial asthma; MMP9/TIMMP1 ratio was strongly associated with response to steroids.10 The method has already been used reliably in BAL fluid and sputum from patients with chronic obstructive pulmonary disease (COPD) and bronchial asthma, where increased MMP9 levels were found that correlated with the degree of lung function impairment.11-14 In patients with cryptogenic organizing pneumonia (COP), ELISA studies have demonstrated increased MMP9 levels and abnormal MMP9/TIMMP1 ratio in BAL fluid, which were associated with macrophage, neutrophil and lymphocyte counts.15

In patients with idiopathic pulmonary fibrosis, the presence of metalloproteinases in peripheral blood is consistent with relevant findings of immunohistochemical studies in lung biopsies and zymography studies in BALF. In particular, increased MMP9 levels have been found in BALF from patients with idiopathic pulmonary fibrosis by Fukada et al, and the relationship between MMP9 levels and the effect of treatment on the course of the disease has been examined.16,17 Increased MMP9 levels in BALF from these patients apparently indicate inflammation and suggest MMP9 production by neutrophils, epithelial cells and fibroblasts.16 Suga et al have shown higher MMP9 levels in patients with rapidly progressing disease, that were associated with higher neutrophil counts in BALF.18 In addition, immunohistochemical staining has demonstrated the role of metaplastic epithelial cells and their apoptosis in the imbalance between metalloproteinases and their inhibitors.19 It is worthy of note that gene technology has recently been used in studies aiming to interpret these results.20,21

The association of MMP9 levels with lung diffusion capacity found in our study reflects the increased accumulation of extracellular interstitial tissue at the alveolar level and its infiltration by fibroblasts with increased metalloproteinase production, as described in the literature.22 The clinical implication of this finding is first demonstrated here; the important role of MMP9 in tissue remodeling and the pathogenesis of fibrosis is essentially pointed out, although this relationship was not at present confirmed in BAL studies on our study subjects.

Among sarcoidosis patients, MMP9 levels were higher in patients with active disease compared to patients with latent disease, and were positively associated with 67Gacitrate scintigraphy results. This association has been suggested by findings in BAL fluid of patients with active sarcoidosis who had increased MMP2 production by alveolar macrophages.23 To our knowledge, the association between peripheral blood metalloproteinases and active sarcoidosis is described for the first time in the relevant literature. It should be noted that MMP1 polymorphism studies have recently indicated genetic predisposition to the clinical features of active disease and involvement of more than three organs.24

Conventionally, active sarcoidosis is defined by clinical criteria, biochemical markers such as SACE, and imaging studies.25 Whole-body scintigraphy using 67Ga-citrate has the higher sensitivity (94%). Interestingly, ELISA results were statistically associated with the reference method, whereas SACE levels were not associated with disease activity. Our results showed that MMP9 levels had no association with SACE levels and disease stage. However, a greater number of patients are required to investigate the sensitivity, specificity, and positive and negative prognostic value of the method with reference to 67Ga-citrate scintigraphy and SACE.

Another important finding, reported for the first time, is that MMP9 levels in patients with sarcoidosis were significantly associated with mid-expiratory flow in the total of patients, and separately in patients with active or latent disease. Involvement of the small airways is a common finding in sarcoidosis and reflects pathological changes such as endobronchial granulomas, peribronchial fibrosis, depositions in interstitial connective tissue, or bronchial hyperreactivity.25 The observed association may reflect the progression of tissue remodeling in small airways and increased production of MMP9 by sarcoidal granulomas. In addition, immunohistochemical studies have shown increased MMP9 and MMP1 levels and low TIMMP1 levels in multinucleated gigantocytes in granulomatous tissue taken by lung and heart biopsy from patients with sarcoidosis.27

Involvement of cardiac myocytes has been implicated in the development of myocardial sarcoidosis; infiltrative cardiomyopathy is pathologically documented in 27% of the patients at autopsy.28 Indicative of myocardial disease are ultrasonographic findings consistent with left ventricular diastolic dysfunction which have been identified in 50% of the patients with normal systolic function.29

In our patients, diastolic relaxation dysfunction was identified in 41% of the patients without ultrasonographic evidence of pericarditis or infiltrative cardiomyopathy. It is noteworthy that serum MMP9 and TIMMP1 levels and MMP9/TIMMP1 ratio were not associated with these findings, suggesting that these markers do not reflect myocardial involvement in sarcoidosis. However, none of the sarcoidosis patients met the clinical criteria for myocardial involvement; thus, it can be assumed that myocardial remodeling may not be reflected in peripheral blood in early stages of the disease.

Similarly, diagnosis of pulmonary hypertension in 40% of the patients with idiopathic pulmonary fibrosis was not associated with serum MMP9 and TIMMP1 levels, and MMP9/TIMMP1 ratio, despite the established role of metalloproteinases in the pathogenesis of pulmonary hypertension.31 MMP9 levels were also recently evaluated as markers for monitoring the therapeutic effect of an inhibitor of endothelin receptors (bosentan) in patients with pulmonary hypertension and scleroderma.32 However, in our study, the number of patients with idiopathic pulmonary fibrosis is small and reliable conclusions require a larger patient population.

The selection of 67Ga scintigraphy as a method for evaluating sarcoidosis activity, although highly sensitive, is not universally accepted and poses significant limitations, according to international literature.33 Its use in the present study provides evidence of active disease not considered pathognomonic; hence, further studies are required to define the role of the method in monitoring of the disease activity. Recent reports point out the potential role of magnetic resonance imaging in early diagnosis of myocardial involvement, which may present with increased signal of the interventricular septum.

The limitations of the study include inability to investigate the effect of treatment on study parameters because of the small number of patients. Although all patients were clinically stable, three patients with idiopathic pulmonary fibrosis and respiratory failure received cortidosteroids at fixed doses and immunosuppressive treatment. On corticosteroid treatment were also eight patients with sarcoidosis, whereas the remaining six patients were included in the study at disease diagnosis, before treatment was initiated. However, our conclusions are supported by the fact that all study subjects were non-smokers, since smoking has been shown to increase MMP9 levels and affect the MMP9/TIMMP1 ratio.35

In conclusion, despite the above-described limitations, our results provide strong evidence about the important role of matrix metalloproteinase 9 (MMP9) and its balance with its tissue inhibitor TIMMP1 in the peripheral blood of patients with sarcoidosis and idiopathic pulmonary fibrosis. In sarcoidosis, in particular, clinical correlations with impaired lung function and active disease may be interpreted as a reflection of tissue remodeling and inflammation in the lungs.

Determinations of metalloproteinase levels in peripheral blood may become a new reliable marker for the monitoring of disease progression and, perhaps, response to treatment in the future, using simple and accessible diagnostic methods. On the other hand, they will be useful in the evaluation of the effect of pharmaceutical inhibitors of metalloproteinases, which are already examined in clinical studies, with a view to develop new perspectives in therapeutic management.


1. Woessner JF. MMPs and TIMMP1s: an historical perspective. Methods Mol Biol 2001; 151:1-23
2. O’Connor CM, Fitz Gerald MX. Matrix metalloproteases and lung disease.Thorax 1994; 49:602-609.
3. Peters CA, Freeman MR, Fernadez CH, Stephan J, Wiederschain DG, Moses MH. Dysregulated proteolytic balance as the basis of excess extracellular matrix in fibrotic disease. Am J Physiol 1997; 272:1960-1965.
4. Henry MT, McMahom K, Mackarel AJ, Prikk K, Sorsa T, Maisi P, Sepper R, Fitz Gerald MX, O’Connor C. M Matrix metalloproteinases and tissue inhibitor of metalloproteinase-1 in sarcoidosis and IPF. Eur Respir J 2002; 20:1220-1227.
5. du Bois R, Baughman P. Diffuse lung disease, Arnold 2004, pp113.
6. Alberts Chr, van der Schoot JB, Groen AS. 67Ga scintigraphy as an index of disease activity in pulmonary sarcoidosis. Eur J Nucl Med 1981; 6:205-12.
7. ATS/ERS International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Car Med 2002; 15:165(2):277-304.
8. Laaban JP, Diebold B, Zelinski R, Laffay M, Raffoul H. Noninvasive estimation of systolic pulmonary artery pressure using Doppler echocardiography in patients with COPD. Chest 1989; 96:1258-62.
9. Bergmann U, Michaelis J, Oberhoff R, Knauper V, Beckmann R, Tschesche H. Enzyme linked immunosorbent assays (ELISA) for the quantitative determonation of human leucocytes collagenase and gelatinase. J Clin Chem Clin Biochem 1989; 27(6):351-9.
10. Bosse M, Chakir J, Rouabhia M, Boulet LP, Audette M, Laviolette A. Serum matrix metalloproteinase-1: tissue inhibitor of metalloproteinase-1 ratio correlates with steroid responsiveness in moderate to severe asthma. Am J Respir Crit Care Med 1999; 159:596-602.
11. Cataldo D, Munaut C, Noel A, et al. MMP-2 and MMP-9 linked gelatinolytic activity in the sputum from patients with asthma and chronic obstructive pulmonary disease.Int Arch Allergy Immunol 2000; 123:259-267.
12. Mautino G, Henriguet C, Gougat C, et al. Increased expression of tissue inhibitor of metalloproteinase-1 and loss of correlation with matrix metalloproteinase-9 by macrophages in asthma. Lab Invest 1999; 79:39-47.
13. Cataldo D, Munaut C, Noel A, Frankenne F, Bartsch P, Louis R. Matrix metalloproteinases and TIMMP1 production by peripheral blood granulocytes from COPD patients and asthmatics. Allergy 2001; 56(2):145-51.
14. Vignola AM, Ricobono L, Mirabella A, et al. Sputum metalloproteinase-9/tissue inhibitor of metalloproteinase ratio correlates with airflow obstruction in asthma and chronic bronchitis. Am J Respir Crit Care Med 1998; 158:1945-1950.
15. Choi KH, Lee HB, Jeong MY, et al. The role of matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase- 1 in cryptogenic organising pneumonia. Chest 2002; 121:1478-1485.
16. Fukada Y, Ishizaki M, Kudoh S, Kitaichi M, Yamanaka N. Localization of matrix metalloproteinases 1,2 and 9 and tissue inhibitor of metalloproteinase-2 in interstitial lung disease. Lab Invest 1998; 78:687-698.
17. Lemjabbar H, Gosset P, Lechapt-Zalcman E, et al. Overexpression of alveolar macrophage gelatinase B (MMP-9) in patients with idiopathic pulmonary fibrosis. Effects of steroid and immunosuppressive treatment. Am J Respir Cell Mol Biol 1999; 20:903-913.
18. Suga M, Iyonaga K, Okamoto T, et al. Characteristic elevation of matrix metalloproteinase activity in idiopathic interstitial pneumonias. Am J Respir Crit Care Med 2000; 162:1949-1956.
19. Ruiz V, Ordonez RM, Berumen J, Ramirez R, Uhal B, Becceril C, Pardo AM, Selman A. Unbalanced collagenases/ TIMP-1 expression and epithelial apoptosis in experimental lung fibrosis. Am J Physiol Lung Cell Mol Physiol 2003; 285(5):1026-1036.
20. Zuo F, Kaminski F, Eugui E, et al. Gene expression analysis reveales matrilysin as a key regulator of pulmonary fibrosis in humans. Proc Nat Acad Sci 2002; 99:6292-7.
21. Edwards ST, Anthony CC, Donnelly S, Dazin PF, Schulman ES, Jones KD, Wolters PJ, et al. c-Kit immunophenotyping and metalloproteinase expression profiles of mast cells in interstitial lung diseases. J Pathol 2005; 206: 279–290.
22. Jeffrey J, Atkinson Α, Robert M. Matrix metalloproteinase- 9 in lung remodeling. Am J Respir Cell Mol Biol 2003; 28:12-24.
23. John M, Oltmanns U, Fietze I, Witt C, Jung K. Increased production of matrix metalloproteinase-2 in alveolar macrophages and regulation by interleukin-10 in patients with acute pulmonary sarcoidosis. Exp Lung Res 2002; 28(1):55-68.
24. Ninomiya S, Niimi T, Shimizu S, Sato S, Achiwa H, Akita K, Maeda H, Ueda R. Matrix metalloproteinase- 1 polymorphism of promoter region in sarcoidosis and tuberculosis patients. Sarcoidosis Vasc Diffuse Lung Dis 2004; 21(1):19-24.
25. Lynch JP, White ES. Pulmonary sarcoidosis in: Sarcoidosis ERS Monograph 2002; 32(10):8:105-128.
26. Klech H, Kohn H, Kummer F, Mostbeck A. Assessment of activity in sarcoidosis and specificity of 67 Gallium scintigraphy, serum ACE levels, chest roentgenography and blood lymphocyte subpopulations. Chest 1982; 82:732-738.
27. Gonzalez AA, Segura AM, Horiba K, Qian S, Yu ZX, Stetler-Stevenson W, Willerson JT, McAllister HA Jr, Ferrans VJ. Matrix metalloproteinases and their tissue inhibitors in the lesions of cardiac and pulmonary sarcoidosis: an immunohistochemical study. Hum Pathol 2002; 33(12):1158-64.
28. Silverman KJ, Hutchins GM, Bulkey BH. Cardiac sarcoidosis. A clinicopathologic study of 84 unselected patients with systemic sarcoidosis. Circulation 1978; 58:1204-11.
29. Angomachalelis N, Hourzamanis A, Vamvalis C, Gavrielidis A. Doppler echocardiographic evaluation of left ventricular diastolic function in patients with systemic sarcoidosis. Postgrad Med J 1992; 68(1):52-56.
30. Shimada T, Shimada K, Sakane T. Diagnosis of cardiac sarcoidosis and the evaluation of the effect of steroid therapy. Am J Med 2001; 110:520-527.
31. Lepetit H, Addahibi S, Fadel E, Frisdal E, Munaut C, et al. Smooth muscle cell matrix metalloproteinases in idiopathic pulmonary arterial hypertension. Eur Respir J 2005; 25(5):834-42.
32. Gianelli G, Iannone F, Marinosci F, Lapadula G, Antonaci S. The effect of bosentan on MMP9 levels in patients with systemic sclerosis-induced pulmonary hypertension. Curr Med Res Opin 2005; 21(3):327-32
33. King T, Schwarz K. Interstitial lung disease, Decker 2003, pp358.
34. Vignaux O, Dhote R, Duboc D, et al. Clinical significance of myocardial magnetic resonance abnormalities in patients with sarcoidosis: 1 year follow up study. Chest 2002; 122:1895-1901.
35. Lim S, Roche N, Oliver BG. Balance of matrix metalloprotease- 9 and tissue inhibitor of metalloprotease-1 from alveolar macrophages in cigarette smokers. Am J Respir Crit Care Med 2000; 162:1355-1360.