Objectives: The pleural invasion (PL) score is a useful prognostic indicator in lung cancer. However, in many cases, the cancer may exfoliate itself into the pleural cavity and may progress to a malignant pleural effusion without invading the parietal pleura. This stage is not currently evaluated, but it is detectable by means of the pleural lavage cytology (PLC). However, PLC's contribution to TNM staging has not yet been clarified. The purpose of this investigation was to demonstrate the usefulness of PLC in the precise staging of patients with such an occult pleural dissemination.
Methods: A total of 3231 patients who were included in a multi-institutional database were studied retrospectively. PLC was performed by washing the thoracic cavity with a small amount of physiological saline immediately after opening the thoracic cavity during lung cancer surgery.
Results: The incidence of positive PLC findings was 4.58%. In comparison with the negative group, the survival curves were significantly worse (P < 0.001) and the incidence of recurrence with pleuritis carcinomatosa was significantly higher (P < 0.001). According to the subset analysis, the survival difference was prominent in earlier stage groups and lower PL score groups. The positive findings were confirmed to be a significantly poor prognostic indicator (P = 0.016) by multivariate analysis using the Cox proportional hazard model (Cox analysis). However, integration of the positive findings with the PL score was attempted for the convenience of TNM staging. To find the accurate PL score for positive PLC findings, the Cox analysis was re-estimated using the PL score upgraded stepwise. The most reliable model with the highest score for the likelihood ratio χ(2) statistic was obtained by scoring positive findings as PL3. So, it was considered to be the most reliable conversion.
Conclusions: Examining PLC in clinical practice is useful for detecting occult pleural dissemination before the appearance of a malignant pleural effusion. Evidence of positive findings should be treated as supplemental information to the precise diagnosis of TNM staging. Scoring positive PLC findings as PL3 (=T3) was appropriate.