© Copyright 2007 by the Society for Neuro-Oncology
Received April 5, 2006
Accepted September 5, 2006
Preoperative thrombocytosis predicts poor survival in patients with glioblastoma
Marc A. Brockmann 1*,
Alf Giese 2,
Kathrin Mueller 3,
Finjap Janvier Kaba 4,
Frank Lohr 4,
Christel Weiss 5,
Stefan Gottschalk 6,
Ingo Nolte 1,
Jan Leppert 7,
Jochen Tuettenberg 8,
Christoph Groden 1
1 University Hospital Mannheim, Department of Neuroradiology, Mannheim, 68167 Germany
2 Department of Neurosurgery, Georg-August-University of Göttingen, 37075 Göttingen, Germany
3 University Hospital Mannheim, Departments of Neuroradiology, Mannheim, 68167 Germany
4 University Hospital Mannheim, Department of Radiation Therapy, Mannheim, 68167 Germany
5 University Hospital Mannheim, Department of Statistics in Medicine, Mannheim, 68167 Germany
6 University Hospital Schleswig-Holstein, Campus Luebeck, Department of Neuroradiology, 23538 Luebeck, Germany
7 University Hospital Schleswig-Holstein, Campus Luebeck, Department of Neurosurgery, 23538 Luebeck, Germany
8 University Hospital Mannheim, Department of Neurosurgery, Mannheim, 68167 Germany
 |
Abstract |
|---|
Thrombocytosis, which is defined as a platelet count greater than 400 platelets/nl, has been found to be an independent predictor of shorter survival in various tumors. Release of growth factors from tumors has been proposed to increase platelet counts. Preoperative platelet counts and other clinical and hematological parameters were reviewed from the records of 153 patients diagnosed between 1999 and 2004 with histologically confirmed glioblastoma in order to evaluate the prognostic significance of preoperative thrombocytosis in these patients. The relationship between thrombocytosis and survival was initially analyzed in all patients regardless of further therapy. Univariate log-rank tests showed that the median survival time of 29 patients with preoperative thrombocytosis (19%) was significantly shorter (4 months; 95% confidence interval [95% CI], 3-6 months) compared to 124 patients with normal platelet counts (11 months; 95% CI, 8-13 months; p = 0.0006). Multivariate analysis (Cox proportional hazards model) confirmed preoperative platelet count, age, prothrombin time, and activated partial thromboplastin time to be prognostic factors of survival (all p < 0.05). In a subset of patients (only operated patients with radiation therapy with or without additional chemotherapy), survival was likewise significantly shorter when preoperative thrombocytosis was diagnosed (6 months; 95% CI, 4-12 months) compared to patients with normal platelet count (13 months; 95% CI, 11-15 months; p = 0.0359). In multivariate analysis, age, platelet count, preoperative prothrombin time, and degree of tumor resection retained significance as prognostic factors of survival (all p < 0.05). The results of our study demonstrate preoperative thrombocytosis to be a prognostic factor associated with shorter survival time in patients with glioblastoma.
Key Words:
angiogenesis, malignant glioma, platelets, prognostic parameter
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
