© 2010 THE AUTHORS BJU INTERNATIONAL © 2 0 1 0 B J U I N T E R N A T I O N A L | 1 0 8 , 7 3 – 7 9 | doi:10.1111/j.1464-410X.2010.09889.x 73 2010 THE AUTHORS; BJU INTERNATIONAL 2010 BJU INTERNATIONAL Urological Oncology PLATELET LEVELS AND SURGERY FOR RENAL MALIGNANCY WOSNITZER ET AL. Role of preoperative platelet level in clinical and pathological outcomes after surgery for renal cortical malignancies Matthew Wosnitzer*, Allison Polland*, Qiong Hai † , Gregory Hruby* and James McKiernan* *Department of Urology, and † Mailman School of Public Health, Columbia University, New York, NY, USA Accepted for publication 21 July 2010 • Groups differed significantly inmedian tumour size (5.06 vs. 7.28 cm) (P ≤ 0.001), pathologic T stage (P = 0.002), and metastases (P ≤ 0.0001). No significant difference existed regarding histologic findings at surgery. • With median follow-up of 24 months, PLT > 400 × 10 9 cells/liter was associated with decreased overall (OS) and disease- specific survival (DSS) using log rank test (P ≤ 0.0001). • On multivariate analysis, controlling for TNM stage, histology, and tumour diameter, PLT > 400 × 10 9 cells/liter independently predicted decreased OS (HR 1.67, P = 0.007) and DSS (HR 2.39, P = 0.001). As a continuous variable, PLT predicted OS (HR 1.002, P = 0.005) and DSS (HR 1.003, P = 0.004). • With metastatic patients excluded, PLT was significantly associated with OS and DSS, but was not an independent predictor. CONCLUSION • PLT is a clinically significant independent predictor of OS and DSS in continuous and categorical analyses in patients undergoing renal cortical malignancy surgery. PLT may be clinically useful for risk stratifying patients undergoing surgery for renal cancer, especially for prognosis assessment of patients with renal cortical malignancy and micrometastatic disease at surgery. KEYWORDS renal cell carcinoma, preoperative platelet count, thrombocytosis, partial nephrectomy, radical nephrectomy, recurrence What's known on the subject? and What does the study add? Preoperative platelet count (PLT), which reflects the systemic inflammatory response, has been previously identified in several small studies as useful prognostic factor in a number of cancers including renal cell carcinoma. This study is the largest study to our knowledge to investigate the relationship between platelets and surgical outcomes in patients with renal cell carcinoma. Thrombocytosis (PLT > 400) was associated with and was an independent predictor of decreased overall and disease-specific survival in patients undergoing extirpative renal surgery, and was associated with larger tumour size, and higher TNM stage at surgery. Study Type – Therapy (case series) Level of Evidence 4 OBJECTIVE • To investigate preoperative platelet level (PLT) as a prognostic factor for pathologic and clinical outcomes following surgery for renal cortical malignancy. PATIENTS AND METHODS • 1422 patients underwent radical or partial nephrectomy at our institution from 1988– 2009 for renal cortical lesions. • The cohort with available PLT values was divided into group 1 (PLT ≤ 400 × 10 9 ) and (PLT > 400 × 10 9 ) based on institutional laboratory upper threshold (400 × 10 9 cells/ liter). RESULTS • 961 patients were divided into groups 1 (n = 870) and 2 (n = 91), with mean age at surgery of 61 and 60 years, 70.6% and 50.6% males (P ≤ 0.0001), 56% undergoing radical nephrectomy in each group, 39.1% vs. 22% undergoing partial nephrectomy (P = 0.001) respectively. BJUI BJU INTERNATIONAL INTRODUCTION The National Cancer Institute estimated that 57 760 individuals (35 430 men and 22 330 women) would be diagnosed with renal cancer and that 12 980 individuals would die of cancer of the kidney or renal pelvis in 2009 [1]. Renal cell carcinoma (RCC) is one of the most lethal urological cancers with varying rates of survival after kidney resection. Approximately one-third of patients with RCC have metastatic disease at initial diagnosis and up to 50% will develop distant metastases after initial diagnosis. This shows the need to identify those patients most likely to die from RCC who may benefit from adjuvant treatment. Currently, the 2010 American Joint Cancer Committee/Union Internationale Contre le Cancer TNM staging system is the most widely used predictor, followed by a number of other scoring systems including the University of California Los Angeles Integrated Staging System [2], and the ‘Stage Size Grade Necrosis’ [3] and Kattan [4] prognostic algorithms. There are a number of limitations in the predictive value