OBJECTIVE: To determine whether preoperative platelets to lymphocyte ratio (PLR) could predict disease stage, surgical outcome, and survival in patients with epithelial ovarian cancer (EOC). METHODS: Medical records of EOC patients who had surgery between January 2004 and December 2010 were reviewed. Clinicopathological and complete blood count data were collected. The optimal predictive value of PLR to predict advanced stage, suboptimal surgery, and survival was determined and compared with those of thrombocytosis (> or =400,000 cells/mm3) and neutrophil to lymphocyte ratio (NLR) > or =2.6. RESULTS: A total of 166 EOC patients were included in the study. PLR of 200 yielded better predictive values than those of thrombocytosis and NLR > or =2.6. The area under curve (AUC), sensitivity, specificity, positive and negative predictive values, and accuracy of PLR to predict advanced stage were: 0.66, 59.0%, 72.7%, 65.7%, 66.7%, and 66.3%, respectively. The corresponding values to predict suboptimal surgery were: 0.70, 70.0%, 69.8%, 50.0%, 84.4%, and 69.9%. The patients who had PLR> or =200 had significantly shorter progression-free and overall survivals than those with PLR<200. Stage, grade, surgical outcome, thrombocytosis, and PLR were significant prognostic factors for survivals by univariable analyses while only stage remained significant by multivariable analysis. CONCLUSION: PLR had potential clinical value in predicting advanced stage disease or suboptimal surgery. PLR was a better prognostic indicator for survivals of EOC patients compared to thrombocytosis or NLR>2.6.
Area Under Curve ; Blood Cell Count ; Blood Platelets ; Humans ; Lymphocytes ; Medical Records ; Neoplasms, Glandular and Epithelial ; Neutrophils ; Ovarian Neoplasms ; Sensitivity and Specificity ; Thrombocytosis

OBJECTIVE: To determine whether preoperative platelets to lymphocyte ratio (PLR) could predict disease stage, surgical outcome, and survival in patients with epithelial ovarian cancer (EOC). METHODS: Medical records of EOC patients who had surgery between January 2004 and December 2010 were reviewed. Clinicopathological and complete blood count data were collected. The optimal predictive value of PLR to predict advanced stage, suboptimal surgery, and survival was determined and compared with those of thrombocytosis (> or =400,000 cells/mm3) and neutrophil to lymphocyte ratio (NLR) > or =2.6. RESULTS: A total of 166 EOC patients were included in the study. PLR of 200 yielded better predictive values than those of thrombocytosis and NLR > or =2.6. The area under curve (AUC), sensitivity, specificity, positive and negative predictive values, and accuracy of PLR to predict advanced stage were: 0.66, 59.0%, 72.7%, 65.7%, 66.7%, and 66.3%, respectively. The corresponding values to predict suboptimal surgery were: 0.70, 70.0%, 69.8%, 50.0%, 84.4%, and 69.9%. The patients who had PLR> or =200 had significantly shorter progression-free and overall survivals than those with PLR<200. Stage, grade, surgical outcome, thrombocytosis, and PLR were significant prognostic factors for survivals by univariable analyses while only stage remained significant by multivariable analysis. CONCLUSION: PLR had potential clinical value in predicting advanced stage disease or suboptimal surgery. PLR was a better prognostic indicator for survivals of EOC patients compared to thrombocytosis or NLR>2.6.
Area Under Curve ; Blood Cell Count ; Blood Platelets ; Humans ; Lymphocytes ; Medical Records ; Neoplasms, Glandular and Epithelial ; Neutrophils ; Ovarian Neoplasms ; Sensitivity and Specificity ; Thrombocytosis