Matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 in diagnosis of pleural effusion of malignant origin

OBJECTIVES The aim of the present study was to evaluate the diagnostic accuracy of matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 in differentiating benign from malignant exudative pleural effusions. METHODS This is a unicentre observational study including 97 consecutive patients with exudative pleural effusions. Metalloproteinase-9, tissue inhibitor of metalloproteinase-1, lactate dehydrogenase, ferritin, carcinoembryonic antigen and carbohydrate antigen 15-3 were measured in pleural effusion and serum by enzyme-linked immunosorbent assay. The activity of metalloproteinase-9 was also evaluated by substrate zymography. The data were correlated with final diagnosis of pleural effusions to evaluate the diagnostic accuracy. RESULTS Of the 97 eligible patients, 6 were excluded. Of the 91 patients included in the study, 70 had malignant pleural effusions and 21 had benign pleural effusions. Both in sera and pleural effusions, matrix metalloproteinase-9 (P < 0.0001), tissue inhibitor of metalloproteinase-1 (P < 0.0001) and carcinoembryonic antigen (P < 0.0001) levels were higher in neoplastic patients than in benign group. Zymography analysis showed a most prominent band at a molecular weight of 92 kDa (metalloproteinase-9) whereas a less intense band was observed at 72 kDa (metalloproteinase-2). A significant correlation was found between metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 levels in pleural effusion (P < 0.0001; r = 0.8) and serum (P < 0.03; r = 0.2). Pleural effusion metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 levels showed higher value of sensitivity (97 and 91%, respectively) and specificity (90 and 95%, respectively) compared with other standard markers. Serum metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 levels showed similar results. Among 70 neoplastic patients, 29 had negative pleural cytology. Of these, 25 presented elevated levels of metalloproteinase-9 and tissue inhibitor of metalloproteinase-1, whereas 4 patients had elevated levels of one of the two markers. CONCLUSIONS Our results showed that metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 might be valuable markers in differentiating benign from malignant pleural effusions. Their levels are neither influenced by the histology and tumour origin nor by the presence of tumour cells in pleural effusions. Thus, their use in clinical practice could help in the selection of patients needing more invasive procedures, such as thoracoscopic biopsy.