Abstract
Background Sepsis-3, the recent sepsis definitions, was modified based on a scoring system focused on organ failure; however, it would remain a time-consuming process to detect septic patients using these definitions. Adrenomedullin (AM) is a biomarker for diagnosing sepsis and septic shock, monitoring treatment efficacy, and prognosis. We conducted a study to assess the accuracy of AM for diagnosing and prognosing sepsis and septic shock based on the Sepsis-3 definitions.Methods This is a prospective observational single-center study. Patients admitted to the intensive care unit (ICU) were retrospectively categorized as non-sepsis, sepsis, or septic shock by Sepsis-3 definitions. Total AM (tAM) and mature AM (mAM) were measured upon ICU admission. Receiver operating characteristics (ROC) analyses were performed by calculating the area under the curve (AUC) for diagnosis and prognosis of sepsis and septic shock.ResultsA total of 98 patients were enrolled in the final analysis. Among these, 42, 22, and 34 patients were assigned to non-sepsis, sepsis, and septic shock, respectively. tAM and mAM levels significantly increased according to the severity of sepsis. The AUCs of tAM/mAM for diagnosing sepsis and septic shock were 0.879/0.848 and 0.858/0.830, respectively, whereas those of procalcitonin (PCT)/presepsin (PSEP) were 0.822/0.682 and 0.811/0.661, respectively. The AUCs of tAM/mAM on Day 1 and 3 for predicting 28-day mortality of septic patients were 0.669/0.5741 and 0.931/0.892, respectively, whereas those of sequential organ failure assessment (SOFA) score/lactate were 0.669/0.824 and 0.922/0.794, respectively.Conclusions Both tAM and mAM are reliable, early biomarkers to diagnose sepsis and septic shock according to the Sepsis-3 definitions, and are comparable to PCT. Furthermore, AM level on Day 3 is a reliable biomarker to predict 28-day mortality due to sepsis, which is comparable to that of the SOFA score and lactate level.