Biomarkers in sepsis: towards precision medicine for the prevention of adverse outcomes

In 2016, an International Consensus has defined sepsis as a “life-threatening organ dysfunction caused by a dysregulated host response to infection” and septic shock as “a subset of sepsis in which profound circulatory, cellular, and metabolic abnormalities are associated with a greater risk of mortality compared to sepsis alone.” Globally, sepsis is an urgent public health issue and one of the leading causes of death in emergency departments, intensive care units (ICUs) and hospitals. The World Health Assembly resolution recommends the promotion of research to develop innovative means to prevent, diagnose, and treat sepsis, by including sepsis among the priorities in the research area for funding institutions4 in order to obtain the optimization of hospital resources and therapies. Current clinical biomarkers detected in urine or blood represent useful diagnostic and prognostic tools.5 The first, can be used to distinguish sepsis from non-infectious critical illness or to determine causative pathogens in order to promptly begin the best possible treatment. Prognostic biomarkers can help to identify patient’s risk profiles, predicting adverse clinical outcomes and allowing clinicians to better focus on preventing complications and death.5, 6 Furthermore, prognostic biomarkers permit the identification of pathophysiologic pathways targeting personalized therapies. Recently new attention is being focused on adrenomedullin and more specifically on its stable mid-regional fragment pro-adrenomedullin (MR-proADM)7 which has been demonstrated to be one of the most reliable biomarkers to predict the onset of pneumonia, other diseases and mortality. Adrenomedullin is a strong vasodilating, antimicrobial, immunomodulant peptide. It is synthetized by endocrine cells and then secreted systemically by various cell types, involved in the inflammatory reaction, in response to sepsis, hypoxia, or oxidative stress. Most of these factors are originated from the acute pathophysiological changes induced by sepsis and are not associated with comorbidities. Pro-adrenomedullin measured upon admission has been demonstrated to be a good predictor of severity and outcome for community-acquired pneumonia, improving prognostic accuracy. In this issue of Minerva Anestesiologica, Valenzuela-Sanchez et al. report the results of a prospective observational single-center study conducted on 120 consecutive ICU septic patients. This paper offers new important data to understand the importance of plasma levels of biomarkers in sepsis along the same line of recently published data. They found that the MRproADM clearance during five days of ICU stay was related to a good prognosis. In details, at hours following admission, the median MR-proADR plasma levels in surviving sepsis patients fell to 1.65 nmol/L but remained higher in the non-survivors. On day 5 the survivors showed greater MR-proADM. In accordance with other studies15 the Cox proportional hazards suggested by the authors showed that the MR-proADR plasma levels at 48 hours and at five day after ICU admission remained higher in patients not surviving sepsis compared to the survivors. Despite these indisputably interesting results, the paper leaves some open questions. Firstly, the reproducibility of the results should be replicated in a larger cohort. Indeed, it is not infrequent that findings from small, exploratory studies are not confirmed in larger trials. The work of Valenzuela-Sanchez et al.13 does not represent an exception. Secondly, it is to be investigated whether the detection of MR pro-ADM in plasma levels is useful to predict the pathology’s outcome. Indeed, its real usefulness to evaluate the organ dysfunction in septic patients is limited especially during the first 24 hours after ICU admission, and the correlation with the SOFA over time was not sufficiently studied, obstructing solid conclusion on organ failure predictivity. Thirdly, the role of MR pro-ADM for the diagnosis of infection in patients with suspected sepsis is weak because high levels were also found in other diseases such as acute heart and respiratory failure, acute kidney injury, cirrhosis and cancer. It is however true that any biomarker, including procalcitonin (PCT), has per se a perfect sensitivity and specificity in everyday clinical practice. On the other hand, also the combined use of severity scores and biomarkers to support the clinical assessment might be useful, but their predictive accuracy is still suboptimal.The study by Valenzuela-Sanchez et al. offers an interesting suggestion showing that the persistence of high values or the reduction in clearance of the biomarker on day 5 indicated the persistence of the infection. Even though this finding must be validated in a larger cohort of patients, in principle it might help clinicians in the future to change the therapeutic strategy. Lastly, the sepsis response in humans varies with time, and also the time range in which a specific biomarker can be specific and useful changes. This is why it might be difficult to ably assess the usefulness of the biomarker also in experimental models. There are many exciting and important questions waiting to be answered, before the role of biomarkers in sepsis are fully elucidated.