Respiratory Research (Jan 2024)

Diagnostic accuracy of thoracic CT to differentiate transudative from exudative pleural effusion prior to thoracentesis

  • Yan Zhang,
  • Yang Zhang,
  • Wei Wang,
  • Xiaoyu Feng,
  • Jiahuan Guo,
  • Bo Chen,
  • Fuyun Zhang,
  • Huanhuan Wang,
  • Mengnan Fan,
  • Yingwei Zhu,
  • Yuxia Sun,
  • Tongsheng Wang,
  • Yimin Mao,
  • Pengfei Gao

DOI
https://doi.org/10.1186/s12931-024-02681-w
Journal volume & issue
Vol. 25, no. 1
pp. 1 – 9

Abstract

Read online

Abstract Background Computed tomography (CT) scan is commonly performed for pleural effusion diagnostis in the clinic. However, there are limited data assessing the accuracy of thoracic CT for the separation of transudative from exudative effusions. The study aimed to determine the diagnostic value of thoracic CT in distinguishing transudates from exudates in patients with pleural effusion. Methods This is a two-center retrospective analysis of patients with pleural effusion, a total of 209 patients were included from The First Affiliated Hospital of Henan University of Science and Technology as the derivation cohort (Luoyang cohort), and 195 patients from the First Affiliated Hospital of Zhengzhou University as the validation cohort (Zhengzhou cohort). Patients who underwent thoracic CT scan followed by diagnostic thoracentesis were enrolled. The optimal cut-points of CT value in pleural fluid (PF) and PF to blood CT value ratio for predicting a transudative vs. exudative pleural effusions were determined in the derivation cohort and further verified in the validation cohort. Results In the Derivation (Luoyang) cohort, patients with exudates had significantly higher CT value [13.01 (10.01–16.11) vs. 4.89 (2.31–9.83) HU] and PF to blood CT value ratio [0.37 (0.27–0.53) vs. 0.16 (0.07–0.26)] than those with transudates. With a cut-off value of 10.81 HU, the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of CT value were 0.85, 88.89%, 68.90%, 43.96%, and 95.76%, respectively. The optimum cut-value for PF to blood CT value ratio was 0.27 with AUC of 0.86, yielding a sensitivity of 61.11%, specificity of 86.36%, PPV of 78.57%, and NPV of 73.08%. These were further verified in the Validation (Zhengzhou) cohort. Conclusions CT value and PF to blood CT value ratio showed good differential abilities in predicting transudates from exudates, which may help to avoid unnecessary thoracentesis.

Keywords