Differential Diagnosis By Analysis of Pleural Effusion

Ko, Won Ki; Lee, Jun Gu; Jung, Jae Ho; Park, Mu Suk; Jeong, Nak Yeong; Kim, Young Sam; Yang, Dong Gyoo; Yoo, Nae Choon; Ahn, Chul Min; Kim, Sung Kyu

Abstract

BACKGROUND
Pleural effusion is one of most common clinical mainifestations associated with a variety of pulmonary disease such as malignancy, tuberculosis, and pneumonia. However, there are no useful laboratory tests to determine the specific cause of pleural effusion. Therefore, an attempt was made to analyze the various types of pleural effusion and search for useful laboratory tests for pleural effusion in order to differentiate between the disease, especially between a malignant pleural effusion and a non-malignant pleural effusion.
METHODS
93 patients with a pleural effusion, who visited the Severance hospital from January 1998 to August 1999, were enrolled in this study. Ultrasound-guided thoracentesis was done and a confirmational diagnosis was made by a gram stain, bacterial culture, Ziehl-Neelsen stain, a mycobacterial culture, a pleural biopsy and cytology.
RESULTS
The male to female ratio was 56:37 and the average age was 47.1±21.8 years. There were 16 cases with a malignant effusion, 12 cases with a para-malignant effusion, 36 cases with tuberculosis, 22 cases with a para-pneumonic effusion, and 7 cases with transudate. The LDH2 fraction was significantly higher in the para-malignant effusion group compared to the para-pneumonic effusion group [30.6±64.% and 20.2±7.5%, respectively (p<0.05)] and both the LDH and LDH2 fraction was significantly in the para-malignant effusion group compared to those with tuberculosis [16.4±7.2% vs. 7.6±4.7%, and 30.6±6.4% vs. 17.6±6.3% respectively (p<0.05)]. The pleural effusion/serum LDH4 fraction ratio was significantly lower in the malignant effusion group compared to those with tuberculosis [1.5±0.8 vs. 2.1±0.6, respectively (p<0.05)]. The LDH4 fraction and the pleural effusion/serum LDH4 fraction ratio was significantly lower in the para-malignant effusion group compared to those with tuberculosis [17.0±5.8% vs. 23.5±4.6% and 1.3±0.4 vs. 2.1±0.6, respectively(p<0.05)].
CONCLUSION
These results suggest that the LDH isoenzyme was the only useful biochemical test for a differential diagnosis of the various disease. In particular, the most useful test was the pleural effusion/serum LDH4 fraction ratio to distinguish between a para-malignant effusion and a tuberculous effusion.