Abstract: Objective To develop a predictive model for differentiating septic shock from sepsis. Thus, treating potential septic shock patients as soon as possible. Method A retrospective analysis was performed on 82 patients with sepsis hospitalized in Dongguan People's Hospital from 1 January 2018 to 31 December 2020, and they were divided into sepsis group and septic shock group according to whether they developed into septic shock. Single factor analysis was performed on the general clinical data of the patients. The receiver operating characteristic (ROC) curve was used to determine the cut-off value of continuous variables with statistical significance between the two groups. Binary data conversion was performed on the continuous variables according to the cut-off value. Multivariate binary logistic regression analysis was used to further screen the indexes with predictive value for septic shock, and the corresponding score was set up according to the β regression coefficient of each variable to establish the septic shock prediction model. Finally, 64 patients admitted with sepsis from 1 January 2022 to 31 December 2023 will be used to validate the model. Result Univariate analysis showed that age and gender had no statistical significance between the two groups, while the other observation indexes had statistical significance. However, only procalcitonin (PCT) ≥12μg/L,C-reactive protein (CRP) ≥181mg/L and neutrophil to lymphocyte ratio (NLR) ≥17 were included in the multivariate logistic regression model finally. The prediction model equation was as follows: Y=2.471×PCT+1.76×CRP+1.009×NLR, and the cut-off value was 2.62, that is, when Y value ≥2.62, sepsis was highly likely to develop into septic shock. The sensitivity, specificity and accuracy of the model were 89.5%, 63.6% and 85.9%, respectively. The validation results showed a sensitivity of 88.2%, a specificity of 83.3% and an accuracy of 85.9%．Conclusion The scoring model provides a simple and feasible way of facilitating a differential diagnosis of septic shock and sepsis, thus, providing evidence for the timely treatment of high-risk patients with septic shock.

Key words: Sepsis, Septic shock, Clinical score prediction model