OBJECTIVES: To evaluate the significance of thrombocytosis for determining survival in patients with localized renal cell carcinoma (RCC) who underwent radical nephrectomy with curative intent. METHODS: The study population comprised 151 patients who underwent radical nephrectomy with a diagnosis of localized RCC. The inclusion criteria were at least one preoperative platelet count and a tissue diagnosis of RCC. Thrombocytosis was defined as a platelet count greater than 400,000/mm3. The charts of these patients were retrospectively reviewed and the relationship between the preoperative platelet counts and survival was evaluated according to the histologic variants. RESULTS: The mean follow-up was 37.1 +/- 24.1 months (median 34, range 3 to 91). Of these 151 patients, 21 had a preoperative platelet count greater than 400,000/mm3. The thrombocytosis rate was significantly greater in those with advanced T stage (P = 0.045). Lymph node positivity also correlated with thrombocytosis (P = 0.028). However, no statistically significant correlation was found between grade and histologic subtypes and thrombocytosis (P = 0.053 and P = 0.354, respectively). Ten (47.6%) of 21 patients with thrombocytosis and 21 (16.2%) of the remaining 130 patients died of disease progression (P = 0.002). The Kaplan-Meier cause-specific survival curves showed that patients who had thrombocytosis had a mean survival of 45.2 months compared with 76.6 months in patients without thrombocytosis (P = 0.0002). CONCLUSIONS: Preoperative thrombocytosis is a significant predictor for determining prognosis in patients with localized RCC. Thrombocytosis was found more frequently in patients with advanced stage RCC, and patients with preoperative thrombocytosis had a worse survival compared with those with normal platelet counts.