Abstract
Sepsis is a pathological process in which, uncontrolled as a result of dysregulation, the body’s immune response to infection leads to organ failure that threatens the patient’s life. The development of septic shock aggravates the course of sepsis and significantly affects the outcome of the pathological process. Therefore, the development of prognostic tools (markers, models, algorithms) is an important component in the treatment of patients with sepsis. Objectives. To develop models capable of predicting the likelihood of septic shock and death in patients based on a combination of identified prognostic factors. Materials. To identify factors predisposing to the development of septic shock, 2 groups of patients were formed: the 1st group (109 people) included patients with sepsis without the development of shock, and the 2nd group (72 people) - patients in whom sepsis progressed to septic shock. Similarly, to determine predictors of death, groups of patients with a favorable (129 people) and unfavorable (fatal, 49 people) outcome were formed. Results. The analysis identified factors associated with the development of septic shock: urea, OR 2.3 (1.1-4.7), p = 0.026; APTT, OR 4.3 (1.4-16), p = 0.017; lactate, OR 2.4 (1.2-5), p = 0.02; protein, OR 1.3 (0.6-3), p = 0.475. Factors associated with death were established: CRP, AUC 0.69 (0.60-0.78); lactate, AUC 0.72 (0.63-0.82); fibrinogen A, AUC 0.69 (0.59-0.78); Pelod-2 scores, AUC 0.77 (0.69 -0.85); pSofa, AUC 0,72 (0,63-0,81); Phoenix, AUC 0,80 (0,71-0,88); Prism, AUC 0.70 (0.61-0.79) and sum of organ deficiencies, AUC 0.82 (0.74-0.88). Conclusions. The presented models for predicting the development of septic shock and adverse outcomes in sepsis are accessible and easy to use.