Objective To evaluate the accuracy of trans-xiphoid inferior vena cava diameter variation (v-IVCTX), trans-liver inferior vena cava diameter variation (v-IVCTL), internal jugular vein diameter variation (v-IJV) and femoral vein diameter variation (v-FV) in predicting the fluid responsiveness in the patients with septic shock complicated with myocardial injury. Methods Fifty patients with septic shock complicated with myocardial depression admitted to intensive care unit of our hospital, aged ≥18 yr, were selected.The hemodynamics was monitored by PiCCO.Cardiac output (CO) and stroke volume were measured, and v-IVCTX, v-IVCTL, v-IJV and v-FV were measured by ultrasound and then calculated.Fluid replacement test was performed after meeting the standard of fluid resuscitation, and positive fluid responsiveness was defined as increase in CO after volume expansion>10%.Patients were divided into positive fluid responsiveness group and negative fluid responsiveness group.The receiver operating characteristic curve was drawn to evaluate the accuracy of the respiratory variation of each vein diameter in predicting fluid responsiveness. Results Compared with negative fluid responsiveness group, v-IVCTX, v-IVCTL and v-IJV were significantly increased (P 0.05). The cutoff value of v-IVCTX, v-IVCTL and v-IJV in predicting fluid responsiveness was 16.5%, 14.5% and 12%, respectively, the sensitivity was 80%, 76% and 84%, respectively, and the specificity was 72%, 64% and 44%, respectively.The area under the curve (95% confidence interval) of v-IVCTX, v-IVCTL and v-IJV was 0.777 (0.642-0.911), 0.741 (0.605-0.876), and 0.694 (0.549-0.838), respectively. Conclusion v-IVCTX and v-IVCTL both can predict the fluid responsiveness in the patients with septic shock complicated with myocardial injury. Key words: Shock, septic; Vascular capacitance; Inferior vena cava; Jugular veins; Femoral vein