The article by Porcel et al9 is important because it demonstrates that BNP measurements are a poor substitute for NT-pro-BNP measurements. Moreover, the pleural fluid NT-pro-BNP levels identify 80% of patients with congestive heart failure that meet the criteria for exudative effusions of Light et al.1 There are still some questions that remain regarding NT-pro-BNP levels in pleural fluid. First, why are the levels in the pleural fluid and serum virtually identical? With most cytokines, the levels are lower in the pleural fluid unless the cytokines are produced in the pleura, in which case the levels are higher in the pleural fluid. If a patient has congestive heart failure and another disease, are the levels still elevated in the pleural fluid? With treatment, do the levels of NT-pro-BNP fall at the same rate in the pleural fluid and in the serum? In making the diagnosis of pleural effusion due to congestive heart failure, is there any reason to do a thoracentesis and measure the pleural fluid NT-pro-BNP level since the levels in the serum and pleural fluid are virtually identical?