RESUMO
PURPOSE: To evaluate the still controversially discussed prognostic role of preoperative platelet level (PPL) and thrombocytosis (TC) in patients who undergo surgery for renal cell carcinoma (RCC) based on the largest patient series reported to date. METHODS: A total of 3,139 patients, who underwent radical or nephron-sparing nephrectomy at four centres, were subdivided based on a threshold for preoperative platelets of 400 × 10(9) cells/L. Univariate and multivariable Cox regression analyses were applied to determine the prognostic influence of PPL and TC on cancer-specific survival (CSS) for patients with localized and metastatic disease at presentation. RESULTS: Group 1 (PPL ≤ 400/nl) and Group 2 (PPL > 400/nl) included 2,862 (91 %) and 277 patients (9 %), respectively. With a median follow-up (FU) of 69.5 months (IQR: 35-105), CSS of all patients after 5 years was 84.6 % in Group 1 versus 53.4 % in Group 2 (p < 0.001). At multivariable analysis, TC (HR:1.337; p = 0.007) and continuous PPL (HR:1.001; p = 0.002) independently predicted a decreased survival. However, integration of these parameters into multivariable models for the entire study group and for patients with localized tumours did only result in marginal improvement of the model quality (0.66 and 1.04 %, respectively). Interestingly, neither TC (p = 0.257) nor PPL (p = 0.132) significantly influenced survival in M1 patients. CONCLUSIONS: Preoperative TC turned out an independent predictor for decreased CSS in patients undergoing surgery for localized RCC. However, significant improvement of multivariable models comprising standard clinical and pathological parameters by the inclusion of TC is not achieved. In metastatic disease, TC did not reveal an independent influence on CSS.