Background  Sepsis is one of the leading causes of death in patients admitted to the Intensive Care Unit (ICU), which is resulted from the dysregulated host response to infection and in turn causing multiple organ dysfunction. Identifying potential sepsis in early stage and conducting adequate and timely interventions proactively are the key to reduce the incidence and improve the prognosis.

  Objective  To conduct nomogram based on the combined use of inflammatory markers for predicting the occurrence of sepsis in ICU patients, so as to help clinicians identify potential septic patients in the early stage.

  Methods  Patients admitted to the ICU of the First Medical Center, Chinese PLA General Hospital from August 2017 to December 2020 were enrolled in this study. Patients meeting the study criteria were randomly assigned into a training cohort (70%) and a validation cohort (30%). Based on the results of the logistic regression, the training cohort was used to construct a nomogram by the “rms” package of R software. The receiver operating characteristic curves (ROC) were established and the area under the curve (AUC) was calculated to evaluate the discrimination performance of the nomogram. The sensitivity and specificity were also calculated. The calibration curve was drawn by 1000 bootstrap resampling. The validation cohort was used to assess the generality and stability of the nomogram. The probability of sepsis was calculated for each patient in the validation cohort according to the established nomogram. Finally, the predictive performance of the nomogram was compared with the single inflammatory markers and sequential organ failure assessment (SOFA) score.

  Results  A total of 2 074 patients were enrolled, with 1 451 patients randomly assigned to the training cohort and 623 patients randomly assigned to the validation cohort. Five inflammatory markers (WBC, CRP, IL-6, NLR and PCT) were used to construct nomogram according to the results of the logistic regression. And the nomogram indicated a good predictive performance for sepsis, with an AUC of 0.854 (95% CI: 0.835-0.872), sensitivity of 0.820 and specificity of 0.737, respectively, which was superior to the single inflammatory markers and SOFA score (all P<0.05).

  Conclusion  The nomogram based on the combined inflammatory markers can be used for predicting sepsis in the ICU, further helping clinicians to identify potential septic patients in early stage.