Mid-Regional Pro-Adrenomedullin (MR-proADM) as a Biomarker for Sepsis and Septic Shock: Narrative Review

Early identification and diagnosis of sepsis and septic shock is vitally important; despite appropriate management, mortality and morbidity rates remain high. For this reason, many biomarkers and screening systems have been investigated in accordance with the precision medicine concept. A narrative review was conducted to assess the role of mid-regional pro-adrenomedullin (MR-proADM) as a biomarker for sepsis and septic shock. Relevant studies were collected via an electronic PubMed, Web of Science, and The Cochrane Library search. The review focused on both diagnosis and prognosis in patients with sepsis and septic shock and specifically in subpopulations of patients with sepsis and septic shock with burns or malignant tumors. No exclusion criteria regarding age, sex, intensive care unit admission, follow-up duration, or co morbidities were used so as to maximize sensitivity and due to lack of randomized controlled trials, opinion paper and reviews were also included in this review. A total of 22 studies, one opinion paper, and one review paper were investigated. MR-proADM levels were found to be useful in assessing patients’ initial evolution and become even more useful during follow-up with increased area under curve values in the mortality prognosis by exceeding values of 0.8 in the data shown in several studies. These results also improve along with other biomarkers or severity scores and especially correlate with the organ failure degree. The results of this study indicate that MR-proADM is a good biomarker for the diagnosis and prognosis of sepsis and septic shock patients as well as for organ failure. Although several publications have discussed its role as a biomarker for pneumonia, its value as a biomarker for sepsis and septic shock should now be assessed in randomized controlled trials and more collaborative prospective studies with larger patient samples.


Introduction
Adrenomedullin (ADM) plays an important role in the inflammation process and in progression from sepsis to septic shock. Secreted from various organs and tissues, it is mainly produced by vascular endothelial cells and plays a part in vasodilator, positive inotropic, diuretic, natriuretic, and bronchodilator functions [1,2]. Because of the rapid clearance of circulating ADM from the circulation and its binding protein (complement factor H), it is difficult to detect plasma ADM levels by standard immunoassay. The mid-regional fragment of pro-adrenomedullin (MR-proADM) is more stable and demonstrates the levels of the active ADM peptide [3].
Sepsis and septic shock remain a major healthcare problem worldwide, with high mortality rates. Early identification is vital for reducing these rates [4]. For this reason, many biomarkers and screening systems have been investigated in accordance with the precision medicine concept [5].
Despite the fact that the sepsis and septic shock definitions were updated and severe sepsis with systemic inflammatory response syndrome criteria have been passed out of use according to the third international consensus (Sepsis-3), sepsis is defined as organ dysfunction caused by dysregulated response of a host to an infection and the sequential organ failure assessment (SOFA) score scale has been approved for use in defining organ dysfunction. However, this scale has some limitations because several parameters need to be assessed and cut off points were established for intensive care unit patients [6]. Although there is no golden standard biomarker for sepsis and septic shock patients in terms of diagnosis and prognosis at present, MR-proADM seems to be a good alternative for SOFA score for the assessment of organ failure as well as for predicting mortality of septic patients.
In this paper, we review the role of MR-proADM as a biomarker for sepsis and septic shock. Our hypothesis was that MR-proADM may be a prognostic marker of severity in septic patients and has an important role to play in the initial evaluation and the clinical management of this condition.

Materials and Methods
We conducted a literature review to investigate the role of MR-proADM in sepsis and septic shock, applying the methodology described below.

Information Sources
An electronic PubMed, Web of Science, and The Cochrane Library search was performed.

Search
The search was performed using the following strategy and key words: mid-regional pro-adrenomedullin [

Study Selection
Although new definitions of sepsis and septic shock are available [6], in the interests of comprehensiveness we also included the clinical criteria or confirmed diagnosis for sepsis, severe sepsis, and septic shock [7]. The three target conditions of interest for this review are defined as sepsis, severe sepsis, or septic shock. No exclusion criteria regarding age, sex, intensive care unit admission, follow-up duration, or co morbidities were imposed.

Data Items
We extracted data onto a study-specific pro forma, which included the following list of clinical questions under the PICO framework (Population-Intervention-Comparison-Outcome) format (Table 1). MR-proADM was identified as a diagnostic and prognostic marker, stratifying the mortality risk in septic patients. The measure of the index test as MR-proADM was performed in blood samples. The thresholds used in this review were those generated and presented within each study included, as there is currently no accepted standard; therefore, it was not possible to specify a threshold value beforehand.

Risk of Bias in Individual Studies
Studies were initially selected by the main author (UO), who performed title and abstract screening. We then obtained the full text of each potentially eligible study.

Summary Measures
Mean and median levels, odd ratios, and p values were stated for each included study, but a meta-analysis could not be performed due to the heterogeneity of the results.

Synthesis of Results
Analysis of all outcomes was performed, including the authors, year of publication, country, study design, total number and classification of patients, and the comparison of MR-proADM with other biomarkers or scoring systems. The results are also sorted by number of patients respectively.

Additional Analyses
The reference lists of all relevant studies were also checked.

Study Selection and Characteristics
A total of 22 studies, one opinion paper, and one review paper were investigated ( Table 2).

Risk of Bias within Studies
Two articles were excluded due to the high risk of bias or concerns regarding applicability (using the QUADAS-2 tool) [8]. Additional screening identified one review and one opinion paper with no meta-analyses or randomized controlled trials.
As the study was designed as a narrative literature review, no ethical committee permission was needed. In adults (>18 year old) Patients with sepsis, severe sepsis, or septic shock Determining of pro-adrenomedullin/MR-proADM levels Other biomarkers or score systems Assessing of diagnostic or prognostic value (via mortality rates) of pro-adrenomedullin/MR-proADM levels In children Patients with sepsis, severe sepsis, or septic shock Determining of pro-adrenomedullin/MR-proADM levels Other biomarkers or score systems Assessing of diagnostic or prognostic value (via mortality rates) of pro-adrenomedullin/MR-proADM levels PCT: procalcitonin; CRP: C-reactive protein; MR-proADM: mid-regional pro-adrenomedullin; TBSA: total body surface area; MR-proANP: mid-regional pro-anti natriuretic peptide; CT-proET-1: C terminal pro-endothelin 1; PSI: pneumonia severity index; APACHE-II and SOFA (sequential organ failure assessment): score systems.      The MR-proADM levels are more effective than procalcitonin (PCT) and C-reactive protein (CRP) levels to determine an unfavorable outcome and the risk of mortality in patients with sepsis admitted to the ICU. It has also proved useful in patients diagnosed with organ dysfunction of infectious etiology ICU: intensive care unit; SIRS: systemic inflammatory response syndrome; IL: interleukin; TNF: tumor necrosis factor; MCP: monocyte chemoattractant protein; SAPS: simplified acute physiology score; hsTNT: high sensitivity troponin T; CAP: community acquired pneumonia; PCT: procalcitonin; CRP: C-reactive protein; MR-proADM: mid-regional pro-adrenomedullin; TBSA: total body surface area; PSI: pneumonia severity index; APACHE-II and SOFA: score systems; AUC: area under the receiver operating characteristic curve. * American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference and Surviving Sepsis Campaign Criteria. ** SEPSIS-3 Criteria.
A multinational, prospective, observational cohort study named TRIAGE included a total number of 7132 patients with medical emergencies shows that for the prediction of death, ProADM significantly improved regression models, also improved clinical models for prediction of intensive care unit admission and high initial treatment urgency [17].
Enguix-Armada et al. analyzed 388 patients to determine whether the combination of C-reactive protein (CRP), procalcitonin (PCT), presepsin, or SCD14-ST and MR-proADM measured in the first 24 h after intensive care unit (ICU) admission improved diagnostic and prognostic management in both severe sepsis and septic shock. They found that MR-proADM showed a better prognostic value (p < 0.001) particularly in cases of septic shock (p < 0.01), increasing in line with the APACHE-II score [18].
In a cohort of 128 patients referred to the emergency department with high fever and suspected severe infections such as sepsis, lower respiratory tract infections, urinary tract infections, gastrointestinal infections, soft tissue infections, central nervous system infections, or osteomyelitis, Travaglino et al. found a correlation in the sepsis-septic shock groups between the APACHE II score and MR-proADM [21]. Investigating 120 patients, Valenzuela Sanchez et al. concluded that initial MR-proADM levels helped to identify the infectious origin in patients with systemic inflammatory response syndrome (SIRS) and organ dysfunction [22]. MR-proADM levels and its clearance on day 5 after admission were the most effective biomarkers for determining unfavorable evolution and the risk of mortality in patients with severe sepsis admitted to the ICU [22].
In 2005, Christ-Crain et al.'s observational study of MR-proADM as a prognostic marker in sepsis in 101 patients proposed its use as a new prognostic biomarker in critically ill patients with varying severities of sepsis. The ability of MR-proADM to predict outcomes was similar to that of APACHE II and SAPS II scores [23]. De la Torre-Prados et al. found that MR-proADM showed the best association with 28-day mortality and also had prognostic value (log rank test: p = 0.001). Statistical significance was also seen in the Cox regression analysis (p < 0.001) for 100 patients with a relative risk 1.26 times higher than baseline for each mmol/L of increase in MR-proADM [24].
In 71 patients with severe sepsis and septic shock, Andaluz-Ojeda et al. also concluded that proADM was the only biomarker to show significant differences between survivors and non-survivors for concentration in plasma at the three time points analyzed (day 1, 3, and 7); it was the biomarker with the best diagnostic accuracy for predicting mortality in the ICU in the AUROC analysis [25]. In 48 septic patients (15 of whom had septic shock), Schuetz et al. found mean MR-proADM levels at admission to be below 1 nmol/L in non-septic critical patients, 2.6 nmol/L in patients with sepsis, and 8 nmol/L (p < 0.01) in patients with septic shock [26]. Lundberg et al. also showed that MR-proADM and high sensitivity troponin T (hsTNT) levels were significantly higher among 7-day and 28-day non-survivors [27].

MR-proADM Levels and Sepsis in Burn Patients
Burns are also associated with increased levels of MR-proADM, but the dynamics regarding sepsis onset after burn injury need to be studied in greater depth. Assessing PCT and MR-proADM as biomarkers of sepsis in 42 burn patients with >15% total body surface area, Gille et al. found that in the patients who developed sepsis (n = 27, 64.3%) MR-proADM and PCT levels were significantly higher (p < 0.001) on days categorized as septic, with an increase of 31% and at least 0.015 nmol/L (area under curve 0.76) or of >39% and at least 0.15 µg/L (area under the curve 0.83), respectively, for the detection of sepsis. Interestingly, the increase in MR-proADM was recorded one day earlier than in PCT, so the authors concluded that it may be more suitable for use in early recognition of sepsis [30].

MR-proADM Levels in Sepsis with Underlying Malignancies
In malignancies, the origin of fever should be distinguished from infectious and non-infectious causes, such as tumor burden, blood transfusions, or medications. Biomarkers have also been investigated for the diagnosis of infection [31]. Al Shuaibi et al. studied 340 febrile patients (103 with definite sepsis, and 159 with systemic inflammatory response syndrome, SIRS) with hematologic malignancies for the presence of sepsis, SIRS, documented infections, and response to antimicrobial therapy. Only initial MR-proADM levels were found to be significantly higher in patients with localized bacterial infections than in those with no documented infection (p = 0.019) and significantly higher in patients with definite sepsis than in those with SIRS (p = 0.023) [32].

Discussion
In septic patients, early diagnosis and stratification or triage is vitally important for the initiation of appropriate treatment. Our review provides new information on the diagnostic and prognostic role of MR-proADM in sepsis and septic shock.
Statistical analysis could not be performed due to the heterogeneity of the results. In order to define the exact role of MR-proADM as a biomarker for sepsis and septic shock, randomized controlled trials and more collaborative prospective studies should be performed with large numbers of patients.
Due to the complex structure of the septic response, a combination of several tests is likely to be more effective for diagnosis and management than a single ideal biomarker. Angeletti et al. investigated the combined measurement of biomarkers in 104 patients with sepsis and found a better post-test probability with the combination of PCT with MR-proADM; the combination of PCT, MR-proADM, and TNF-α showed the best result in the early diagnosis of sepsis [34]. In both sepsis and localized infections, Angeletti et al. found that the combined use of PCT and MR-proADM increased the post-test probability of diagnosing bacterial infections compared to PCT alone [35].
Although PCT levels are affected by the pathogen (for example, high levels for Gram-negative sepsis), levels of MR-proADM seem to be related to the degree of organ failure and severity of infection rather than the causative agent such as the bacterium or yeast [36]. Angeletti et al. investigated 200 patients with sepsis according to source (Gram-positive and Gram-negative, yeast, and polymicrobial); the AUC for MR-proADM for each were 0.96 (p < 0.01), 0.94 (p < 0.001), 0.99 (p < 0.001), and 0.98 (p < 0.001), respectively [37].
MR-proADM levels for viral infections as the source of sepsis and septic shock have not been previously studied. However, Valenzuela-Sanchez et al. reported that MR-proADM plasma levels measured in influenza A pneumonia without bacterial co-infection were more useful than PCT or CRP for predicting the risk of mortality and need for mechanical ventilation [38,39]. In addition, Guignant et al. found significantly elevated plasma MR-proADM concentrations in patients who developed nosocomial infections after a septic shock episode (p = 0.043) [40]. Thus, the follow-up of MR-proADM levels may represent a major breakthrough in the monitoring of infectious risk in the ICU patient.
MR-proADM levels are also useful in assessing patients' evolution. PCT levels may normalize even though the severity persists, and MR-proADM becomes even more useful during evolution, with increased AUROC values in the mortality prognosis by exceeding values of 0.8 in the data shown in several studies. The results also improve along with other biomarkers or severity scores [19,20,22,27,32,33,40].
Our review has some limitations. First of all, the studies present data from high income or developed countries. In low-and middle-income countries, due to lack of resources, diagnosis and triage of sepsis should be performed prior to hospital admission; therefore, physical examination plays an important role [41]. In a study of African countries, Baelani et al. found that 37% did not have the capability to measure lactate [42]. Secondly, the source of sepsis should be considered for the diagnosis and management. The studies published differentiate between the bacterial sources of sepsis and septic shock, but bacteria are not the only causative pathogens: MR-proADM should also be investigated in fungal, viral, and parasitic infections. Although some randomized controlled trials and meta-analyses of the role of MR-proADM in predicting severity and outcome in community-acquired pneumonia patients have been published in order to define the exact role of MR-proADM as a biomarker for sepsis and septic shock, we recommend the performance of randomized controlled trials and more collaborative prospective studies with large numbers of patients specifically in this group of patients [11,43]. Overall, regarding the role of MR-proADM in sepsis and septic shock patients, this review corroborates a prior report which stressed its importance in combination with other biomarkers or organ failure score systems in accordance with the concept of precision medicine [44,45]. A recent systematic review and meta-analysis that assessed both ADM and MR-proADM in patients with sepsis found that the AUC for sROC (summary receiver operating characteristic) was 0.80 (95% CI, 0.77-0.84), which was greater than the results of published meta-analyses of procalcitonin, troponin, and lactate clearance [46]. Finally, the systematic literature review ended in July 2017 and this is an active area of research.
In spite of these limitations, this article has research implications for the future. Perhaps the most interesting value of MR-proADM in sepsis does not lie in its isolated prognostic value (according to the different studies, the molecule presents a modest AUROC that ranges between 0.7 and 0.8 for mortality) or in its diagnostic role (currently poorly demonstrated) but in the combination with other severity scales (such as the risk of early admission to intensive care unit (REA-ICU) score, PSI, CURB-65, SOFA, or APACHE II scores). In this way, many authors have shown how using a combination of the biomarker plus several scales has significantly improved the prognostic value of the first in the field of pneumonia and sepsis. Recent work by Angeletti et al., Andaluz Ojeda et al., and Spoto S et al. demonstrate the interesting potential of the molecule in this aspect [19,34,47]. Finally, identifying drug-response phenotypes is a priority. As emphasized by a recent ESCMID position paper towards precision medicine in sepsis, when searching for biomarkers it is essential to assess their clinical utility [48]. Investigators need to identify biomarkers that are more or less likely to respond to specific interventions (predictive markers).

Conclusions
MR-proADM has been identified as a prognostic marker that is able to stratify the mortality risk in sepsis patients with different degrees of organ failure. It may be helpful in the early identification and individual risk assessment of sepsis and may also facilitate the subsequent clinical management of sepsis and septic shock. Due to the complex structure of the septic response, it is likely that a combination of several tests may be more effective for diagnosis and management than a single ideal biomarker. MR-proADM levels may also be useful for diagnosis or prognosis of sepsis and septic shock in patients with malignancy or in burn patients.