Skip to main content

Reliability of biomarkers of sepsis during extracorporeal therapies: the clinician needs to know what is eliminated and what is not


The evolution of renal replacement therapy (RRT) techniques, and the increasing number of critically ill patients receiving extracorporeal therapies, has presented clinicians with a significant problem: if biomarkers are removed by RRT, can they still be considered reliable in their role of guiding diagnosis and treatment?

The most commonly used RRT techniques in intensive care units (ICUs) can be classified into three categories: continuous renal replacement therapy (CRRT), intermittent hemodialysis (IHD), and hybrid techniques such as those performed with sorbent devices and plasma exchange (PE). These techniques remove substances from the plasma via convection, adsorption, or a combination of the two. Various factors determine the degree of removal, including molecular weight (MW) and charge, and the type of membrane and RRT technique used. IHD has a cut-off of 5 kDa in most cases and the risk of eliminating biomarkers is small. For CRRT, the cut-off value of the membranes is about 35 kDa, and as a result, filtration of a significant number of biomarkers may occur. New highly adsorptive membranes (HAMs), such as the acrylonitrile 69-surface treated (AN69-ST), are being used more frequently in ICUs [1]. This means that biomarkers with a MW above 35 kDa, while not removed by convection, may potentially be removed in a significant quantity by adsorption. With hybrid devices like CytoSorb, removal of hydrophobic substances with a MW up to 55 kDa occurs via selective binding [2]. PE has a cut-off of 1000 kDa and removes not only biomarkers but also a range of other substances including clotting factors and immunoglobulins. Clearance of a substance cannot always be predicted from MW and RRT membrane characteristics alone, highlighting the need for further studies to determine biomarker levels pre- and post-device for different CRRT techniques. For example, the relatively small MW (25 kDa) of high mobility group protein B1 (HMGB1), a damage-associated molecular pattern (DAMP) and marker of outcome, in theory does not prohibit its removal by convection. However, HMGB-1 is not eliminated by convection and is only effectively cleared through adsorption by HAMs like AN69-ST [3]. This occurs because it has a flat shape, and this prevents its passage through a CRRT membrane, despite its small MW. The degree of biomarkers removal by RRT, with the consequent effect on their serum levels, is essential information for clinicians (Fig. 1).

Fig. 1
figure 1

Biomarker molecular weight and removal by CRRT membranes

Biomarkers eliminated by CRRT and sorbents

C-reactive protein (CRP) is the most commonly used biomarker of inflammation. While often thought of as a pentamer with a MW of 125 kDa, CRP is predominantly present as a monomer (mCRP, MW 22–25 kDa) in the blood of septic patients and as such is removed by all forms of CRRT. Substantial amounts can also be eliminated via adsorption, by both conventional CRRT membranes and the CytoSorb device [4].

Procalcitonin (PCT), a biomarker used to detect (and exclude) the presence of infection and to monitor response to treatment, has a MW of 13.5 kDa and has been detected in the ultrafiltrate of patients undergoing CRRT [5]. Most of the PCT is eliminated by convection, but adsorption also contributes to elimination during the first hours of treatment [5].

B-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP), biomarkers of cardiac dysfunction and outcome in sepsis, are also highly likely to be easily cleared by CRRT given their low MWs (3.5 kDa for BNP and 8.5 kDa for NT-proBNP) [6].

Mid-regional pro-adrenomedullin (MR-pro-ADM), a biomarker of sepsis severity and response to treatment, has a MW between 4 and 5.5 kDa, and its plasma concentration has been shown to decrease by 45–65% if a high-flux membrane is used [7].

Recently, presepsin has also been identified as a diagnostic biomarker of sepsis [8]. It has a MW of 13 kDa, which theoretically means that it could be subject to significant convective elimination.

Given that RRT artificially decreases creatinine levels, a patient under RRT should be considered as having the full acute kidney injury in any score.

Biomarkers not eliminated by CRRT or sorbents but needing further investigation

Endocan is a diagnostic and prognostic biomarker for sepsis and acute respiratory distress syndrome [9]. CRRT with a membrane cut-off of 35 kDa is unlikely to remove endocan (MW 40 kDa), but removal may possibly occur by adsorption when HAMs are used [10].

Pentraxin 3 (PTX3), a marker of sepsis severity and a diagnostic marker for ventilator-associated pneumonia [11], has a MW of 35 kDa and thus, in theory, can be removed by CRRT. However, a recent study demonstrated little or no clearance or absorption by the filter during CVVH [12].

Heparin binding protein (HBP), a predictor of sepsis-induced organ dysfunction [13], has a MW of 37 kDa and as such should not be removed by convection. HBP has been detected in the effluent of patients undergoing CRRT, without a consistent decrease in plasma levels [14]. Studies are needed to investigate whether adsorption is possible.

Osteopontin (OPN) is a predictor of outcome in critically ill patients [15]. A highly negatively charged protein with a MW of 32 kDa, osteopontin can theoretically be removed by CRRT, but at this time evidence is lacking.

Table 1 summarizes all biomarkers described in this review with their MW, ability to be removed by convection and/or adsorption, and whether a study focusing on removal via RRT has been performed.

Table 1 Biomarker molecular weight, potential removal by CRRT or sorbents, and summary of the available studies and the studies that need to be realized of levels in the context of RRT


It is likely that many sepsis biomarkers may be removed by convection, and therefore, their reliability as markers in patients undergoing CRRT is under question. Furthermore, the increasing use of HAMs makes the removal of many biomarkers even more likely. It is possible that some biomarkers may still have utility in the role of guiding diagnosis and treatment of critically ill patients on CRRT; however, further studies exploring biomarker elimination by CRRT are needed to confirm this. The development of new reference ranges for biomarkers in the setting of RRT would also be an interesting avenue of study. Beyond their utility as biomarkers, there are still many other questions to answer, such as whether removal of these, and other, substances by CRRT may result in benefit or harm.

Availability of data and materials

Not applicable.



Renal replacement therapy


Intensive care unit


Continuous renal replacement therapy


Intermittent hemodialysis


Plasma exchange


Molecular weight


Highly adsorptive membrane


Acrylonitrile 69-surface treated




High mobility group protein B1


Damage-associated molecular pattern


C-reactive protein


Monomeric form of CRP




B-type natriuretic peptide


N-terminal prohormone of brain natriuretic peptide


Mid-regional pro-adrenomedullin




Heparin binding protein




  1. Clark WR, Gao D, Lorenzin A, Ronco C. Membranes and sorbents. Contrib Nephrol. 2018;194:70–9.

    Article  PubMed  Google Scholar 

  2. Ankawi G, Xie Y, Yang B, Xie Y, Xie P, Ronco C. What have we learned about the use of Cytosorb adsorption columns? Blood Purif. 2019;48(3):196–202.

    Article  PubMed  CAS  Google Scholar 

  3. Yumoto M, Nishida O, Moriyama K, et al. In vitro evaluation of high mobility group box 1 protein removal with various membranes for continuous hemofiltration. Ther Apher Dial. 2011;15:385–93.

    Article  PubMed  CAS  Google Scholar 

  4. Gao N, Yan C, Zhan G. Changes of serum procalcitonin (PCT), C-reactive protein (CRP), interleukin-17 (IL-17), interleukin-6 (IL-6), high mobility group protein-B1 (HMGB1) and D-dimer in patients with severe acute pancreatitis treated with continuous renal replacement therapy (CRRT) and its clinical significance. Med Sci Monit. 2018;24:5881–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  5. Level C, Chauveau P, Guisset O, et al. Mass transfer, clearance and plasma concentration of procalcitonin during continuous veno-venous hemofiltration in patients with septic shock and acute oliguric renal failure. Crit Care. 2003;6:R160–6.

    Article  Google Scholar 

  6. Wahl HG, Graf S, Renz H, Fassbinder W. Elimination of the cardiac natriuretic peptides B-type natriuretic peptide (BNP) and N-terminal proBNP by hemodialysis. Clin Chem. 2004;50:1071–4.

    Article  PubMed  CAS  Google Scholar 

  7. Mueller T, Gegenhuber A, Kronabethleitner G, Leitner I, Haltmayer M, Dieplinger B. Plasma concentrations of novel cardiac biomarkers before and after hemodialysis session. Clin Biochem. 2015;48:1163–6.

    Article  PubMed  CAS  Google Scholar 

  8. Fujii E, Fujino K, Eguchi Y. An evaluation of clinical inflammatory and coagulation markers in patients with sepsis: a pilot study. Acute Med Surg. 2019;6(2):158–64.

    Article  PubMed  PubMed Central  Google Scholar 

  9. De Freitas CN, Gaudet A, Portier L, et al. Endocan, sepsis, pneumonia, and acute respiratory distress syndrome. Crit Care. 2018;22:280.

    Article  Google Scholar 

  10. Hureau M, Gaudet A, De Freitas CN, et al. Endocan is a reliable biomarker during continuous renal replacement therapy. Crit Care. 2019;23(1):296.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Ketter P, Yu J-J, Cap AP, et al. Pentraxin 3: an immune modulator of infection and useful marker for disease severity assessment in sepsis. Exp Rev Clin Immun. 2016;12(5):501–7.

    Article  CAS  Google Scholar 

  12. Schilder L, Nurmohamed SA, ter Wee PM, et al. Putative novel mediators of acute kidney injury in critically ill patients: handling by continuous venovenous hemofiltration and effect of anticoagulation modalities. BMC Nephrol. 2015;16:178.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  13. Fisher J, Linder A. Heparin-binding protein: a lay player in the pathophysiology of organ dysfunction in sepsis. J Intern Med. 2017;281:562–74.

    Article  PubMed  CAS  Google Scholar 

  14. Samuelsson L, Tydén J, Herwald H, et al. Plasma levels of heparin binding protein (HBP) after the start of CRRT. PLoS One. 2019 Aug 29;14(8):e0221813.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  15. Roderburg C, Benz F, Vargas Cardenas D, et al. Persistently elevated osteopontin serum levels predict mortality in critically ill patients. Crit Care. 2015;19:271.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


We would like to thank Dr. Melissa Jackson for critical review of the manuscript.



Author information

Authors and Affiliations



PMH, SR, and DDB designed the paper. All authors participated in drafting and reviewing. All authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Patrick M. Honore.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare to have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Honore, P.M., Redant, S. & De Bels, D. Reliability of biomarkers of sepsis during extracorporeal therapies: the clinician needs to know what is eliminated and what is not. Crit Care 24, 553 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: