Biomarkers are valuable tools for the prediction of mortality in patients with sepsis. However, the use of a single biomarker to predict patient outcomes is challenging owing to the complexity and redundancy of the immune response to infections.We aimed to conduct a prospective observational analysis to investigate the prognostic value of pentraxin 3, interleukin 6, procalcitonin, and lactate combined in predicting the 28-day mortality rate in patients with sepsis or septic shock (n = 160; sepsis, 78; sepsis shock, 82). Two methods (the frequency sum of values above the cutoff, and the multivariate logistic regression model) were used to assess the prognostic value of the biomarker combination.In the receiver operating characteristic curve analyses, the combination of the 4 biomarkers was better than the Sequential Organ Failure Assessment (SOFA) score in predicting the 28-day mortality rate, regardless of whether the frequency sum of values above the cutoff or the multivariate logistic model was used for the analysis. The addition of the SOFA score to the biomarker combination did not result in a better performance for the prediction of mortality.The combined biomarker approach showed good performance in predicting 28-day all-cause mortality among patients diagnosed with either sepsis or septic shock according to the Sepsis-3 definitions. Furthermore, it was superior to the SOFA score in predicting mortality.