Objective: To evaluate the role of volume management guided by extravascular lung water index(EVLWI) in improving the clinical outcomes and cardiac function for patients with cardiogenic shock. Methods: This study was a single-center, prospective cohort study. Patients with cardiogenic shock admitted to the Department of Cardiovascular Medicine, Shanghai East Hospital from July 2022 to December 2023 were enrolled. Patients were matched 1∶1 by propensity score and divided into EVLWI group and control group. In the control group, the volume management strategy was determined by the attending physician based mainly on conventional factors, including clinical features, biochemical assessments, and certain blood pressure measurements. In EVLWI group, the volume management plan was formulated by integrating conventional factors with EVLWI derived from pulse index continuous cardiac output (PiCCO) monitoring. Baseline clinical data, in-hospital treatment, and hemodynamic data were collected. Major adverse cardiovascular events and cardiac function related parameters were compared at 30 d after treatment between the two groups. Baseline EVLWI levels were compared between the non-survivors and the survivors in the EVLWI group. The receiver operating characteristic curve was plotted to assess the accuracy of baseline EVLWI and central venous pressure in predicting all-cause mortality at 30 d after treatment in patients with cardiogenic shock, and subgroup analysis was performed according to ischemic/non-ischemic etiology and with/without use of inotropic drugs. Kaplan-Meier curve was used for survival analysis, with log-rank tests comparing all-cause mortality, cardiac death, and readmission rate for heart failure at 30 d after treatment. Results: A total of 200 patients with cardiogenic shock were enrolled, aged (71.35±12.82) years, 144(72%) males, EVLWI group and control group 100 patients each. Compared with the control group, EVLWI group had lower all-cause mortality (16%(16/100) vs. 42%(42/100), log-rank P<0.01), cardiac death (14%(14/100) vs. 34%(34/100), log-rank P<0.01), and readmission rate for heart failure (4%(4/100) vs. 12%(12/100), log-rank P=0.03) at 30 d after treatment. Subgroup analysis showed that EVLWI-guided volume management was associated with lower all-cause mortality at 30 d after treatment in patients with cardiogenic shock of ischemic or non-ischemic etiology and with or without inotropic drugs (all P<0.05). In EVLWI group, baseline EVLWI levels were higher in non-survivors than those in survivors [(15.99±6.47) ml/kg vs.(9.75±2.55) ml/kg, P<0.01]. The baseline EVLWI could predicting all-cause mortality at 30 d after treatment in patients with cardiogenic shock, with an area under the receiver operating characteristic curve of 0.84 (95%CI: 0.75-0.94, P<0.01), while the baseline central venous pressure had no predicting value (AUC=0.54, 95%CI: 0.40-0.69, P=0.60). The optimal cutoff value of EVLWI in pridicting all-cause mortality at 30 d after treatment in patients with cardiogenic shock was >10.3 ml/kg. With the optimization of hemodynamic parameters, left ventricular ejection fraction was improved in EVLWI group, and serum levels of N-terminal pro-brain natriuretic peptide, creatinine, alanine aminotransferase and lactic acid were decreased (all P<0.05). Conclusion: EVLWI-guided volume management exerts a beneficial effect on therapeutic decision-making and improves clinical outcomes and cardiac function in patients with cardiogenic shock.