Platelet count has been reported to have predictive value in various cancer entities. In the case of breast cancer, evidence about involvement of platelets is still incomplete. Our objective was to assess the influence of pretreatment thrombocytosis on survival and establish its prognostic relevance for breast cancer patients. We performed a retrospective, multivariate analysis of 4,300 patients with early-stage breast cancer. All subjects participated in one of five prospective, randomized, multicenter trials conducted by the Austrian Breast and Colorectal Cancer Study Group. Thrombocytosis was defined as a platelet count exceeding 400 G/L. Median follow-up was 52 months. Univariate and multiple Cox regression models were calculated for overall survival (OS), breast cancer-related survival and disease-free survival (DFS). Pretreatment thrombocytosis was observed in 161 patients (3.7%). Estimated median OS, breast cancer-related survival and DFS for patients with versus those without thrombocytosis was 71.0 versus 99.5, 72.0 versus 100.9, and 80.4 versus 88.4 months, respectively (p = 0.0054, p = 0.0095, p = 0.0199). A multiple Cox regression model including tumor and nodal status, grading, age, hormone receptor status and pretreatment thrombocytosis identified pretreatment thrombocytosis as an independent predictive factor for OS (p = 0.0064) and breast cancer-related survival (p = 0.0162). Multivariate analysis failed to identify pretreatment thrombocytosis as an independent risk factor for DFS (p = 0.1355). In our retrospective study, elevated platelet counts at time of diagnosis were associated with poor prognosis in breast cancer. We hypothesize that platelets may contribute to the pathophysiology of hematogenous metastasis.