Intraoperative pleural lavage cytology is an independent prognostic indicator for staging non–small cell lung cancer

Abstract

ObjectivesFor patients undergoing lung resection for cancer, macroscopic evidence of metastasis is clearly associated with adverse prognosis. However, less is known about the significance of tumor cells detected by using tests such as pleural lavage cytology. To ascertain the frequency and quantify the effect of this finding on survival, we performed a prospective study of intraoperative pleural lavage cytology. MethodsPleural lavage cytology consisted of cytologic analysis of 100 mL of saline irrigated over the lung surface immediately after thoracotomy. Patients were excluded if they had an existing effusion, extreme adhesions, or lateral chest wall invasion or if resection was not performed. Survival was calculated by means of Kaplan-Meier analysis and compared by using log-rank tests. Cox regression was used to ascertain independent predictors of prognosis. ResultsFrom 1995 through 2003, we performed pleural lavage cytology on 292 patients undergoing thoracotomy for lung cancer. The mean age was 64 (SD, 10) years, and 196 (67%) patients were men. Of 292 samples, 13 (4.5%) showed evidence of malignant cells. The median time to follow-up was 15 months (interquartile range, 1-40 months), with a median survival of 49 months for patients with negative pleural lavage cytology results and 13 months for patients with positive pleural lavage cytology results (P = .002). Univariate prognostic predictors were positive pleural lavage cytology status (P = .03), stage (P = .03), adenocarcinoma (P = .06), and parietal pleural involvement (P = .01). In the final multivariate model only positive pleural lavage cytology status (P = .006) and stage (P = .03) remained significant. ConclusionsIntraoperative pleural lavage cytology is a simple addition to intrathoracic staging and an independent predictor of prognosis. Positive results potentially affect survival by upstaging patients to stage IIIB or greater.