Objective To analyze the relationship between early plasma crystal osmotic pressure and clinical characteristics in patients undergoing extracorporeal membrane oxygenation (ECMO), and to explore its predictive value for prognosis.

Methods Patients receiving ECMO support therapy were selected as the study subjects and divided into survival group and death group based on their 28-day survival status. Clinical data of the two groups were compared. Multivariate logistic regression analysis was employed to identify factors influencing patient prognosis and to analyze the independent correlations of early plasma crystal osmotic pressure, as a continuous, binary, and quintile variable, with patient prognosis. The locally weighted scatterplot smoothing (LOWESS) method was used to analyze the association between early plasma crystal osmotic pressure and patient mortality. The predictive value of early plasma crystal osmotic pressure for patient mortality risk was assessed using the receiver operating characteristic (ROC) curve, and patients were subsequently divided into two groups based on the optimal cutoff value obtained. The Kaplan-Meier method was utilized to plot survival curves and compare survivalrates between the two groups.

Results A total of 142 patients were included, with 70 in the survival group and 72 in the death group. The death group exhibited higher levels of continuous renal replacement therapy, serum sodium, serum potassium, early plasma crystal osmotic pressure, alanineaminotransferase (ALT), aspartate aminotransferase (AST), procalcitonin (PCT) and D-dimer compared to the survival group, while the duration of ECMO support was shorter in the death group, these differences were statistically significant (P < 0.05). Multivariate logistic regression analysis revealed that elevated serum potassium, early plasma crystal osmotic pressure and ALT were risk factors for mortality in ECMO patients (P < 0.05). Early plasma crystal osmotic pressure demonstrated an independent correlation with patient prognosis (P_{for trend} < 0.001) and exhibited a certain nonlinear relationship withpatient mortality. The area under the ROC curve for early plasma crystal osmotic pressure in predicting mortality in ECMO patients was 0.680 (0.592 to 0.768, P < 0.001), with a sensitivity of 0.542, a specificity of 0.800, and an optimal cutoff value of 313.46 mmol/L. The survival rate of the group ≤313.46 mmol/L was higher than that of the group >313.46 mmol/L (P < 0.001).

Conclusion Early plasma crystal osmotic pressure is influenced by serum sodium, serum potassium, and ALT in ECMO patients and exhibits a relatively high discriminative ability in predicting patient prognosis.