LAUSR

Background: Thoracentesis with cytological examination of pleural fluid is the initial test of choice for evaluation of pleural effusions in patients with suspected malignant pleural effusion (MPE). There is limited data on the sensitivity of thoracentesis stratified by tumor type. A better understanding of stratified sensitivities is of clinical interest, and may guide early and appropriate referral for pleural biopsy. Objective: The primary objective was sensitivity of thoracentesis with pleural fluid cytology stratified by tumor type. Methods: This is a retrospective cohort study of consecutive patients with a solid tumor malignancy with proven or strong suspicion for metastatic disease with new pleural effusions that underwent an initial thoracentesis. Only patients with metastatic disease were included. Results: Of the 725 patients examined, 63% had pleural fluid cytology positive for malignancy. Sensitivity of thoracentesis varied from a low of 0.38 (95% CI 0.13–0.68) in head and neck malignancy, 0.38 (95% CI 0.15–0.65) in sarcoma, and 0.53 (95% CI 0.34–0.72) in renal cancer to a high of 93 (95% CI 88–97) in breast cancer, and 100 (95% CI 0.82–100) in pancreatic cancer. Factors associated with an increased risk of MPE included larger amount of fluid drained (p = 0.014) and higher pleural fluid protein (p = 0.002). The only factor associated with decreased risk of MPE if first cytology was negative for malignancy was the presence of contralateral effusion (p = 0.005). Conclusions: Sensitivity of thoracentesis for solid tumors varies significantly depending on the type of tumor and is lowest in those with sarcomas, head and neck malignancies, and renal cell cancers.