Objective To investigate the efficacies of heart rate (HR), mean artery pressure (MAP),central venous pressure (CVP), pulmonary artery wedge pressure (PAWP) and stroke volume variation (SVV) in assessing low stroke volume index (SVI) caused by volume deficiency in major non-cardiac surgeries. Methods Twelve patients undergoing sacral or retroperitoneal tumor resection were enrolled in this study. After induction and onset of mechanical ventilation, radial artery was cannulated and connected to FloTrac for BP and SVV monitoring. Swan-Ganz catheter was placed via internal jugular vein into the pulmonary artery for CVP,PAWP, and SVI monitoring. HR, MAP, CVP, PAWP, SVV, and SVI were recorded every 30 min. If phenylephrine was injected within 5 min before the time point for data-record, the record should be postponed to 5 min after the injection. If vasoconstrictor or cardiotonic was injected intravenously and continuously to keep the normal blood pressure, the data-record in this patient should be stopped then, otherwise it should be stopped when the concentration of sevoflurane was turned lower at the end stage of the operation. The receiver operating characteristic (ROC) curves of HR, MAP, CVP, PAWP, and SVV were made according to a low stroke volume which was less than 25 ml/m2 and AUC(area under curve) were counted. Results The AUC of SVV for assessment of a low SVI was 0.837, while AUC of HR, MAP, CVP and PAWP were 0.615, 0.647, 0.623, and 0.658, respectively. The highest Youden’s index was at the point when SVV was 9.5% in its ROC curve, with a sensitivity of 82.4% and specificity of 74.6%. Conclusion SVV is a middle-level effective variable to assess the abnormally low SVI caused by volume deficiency, while HR, MAP, CVP and PAWP have a low efficacy for the same assessment.