Early detection of sepsis

Sepsis is among the most common causes of death in hospitalised patients. Hospital mortality ranges from 28.3–41.1%. Current guidelines recommend that anti-infectious therapy such as antimicrobial therapy and surgical source control should be initiated as soon as possible to optimise outcomes.¹ However, severe sepsis often remains unrecognised. Current data demonstrate that initiation of adequate therapy is often delayed for hours. Results from culture-based pathogen detection are available only after several days and cannot help the physician in the first treatment decision. 

 

Conventional diagnosis relies on clinically suspected infection and the new onset of organ dysfunction but initial symptoms are often unspecific and are not recognised as a complicated infection. In 2003, the PIRO (Predisposition – Infection – Response – Organ dysfunction) concept for improved characterisation and staging of patients with sepsis was developed. Although this concept did not find its way in clinical practice, the authors stated that new biomolecular methods and biomarkers should be future tools to aid the diagnosis of sepsis. In the last decade, many of these methods have been developed and are available for the clinician.² However, the availability of clinical studies regarding the impact of such techniques on the clinical course of the patient is limited.

 

Biomarker

C-reactive protein

C-reactive (CRP) is an acute phase protein and is released from the liver four to six hours after stimulation predominantly of IL-6 and peaks at 36 hours. CRP has been shown to aid in the diagnosis of infection such as pneumonia, acute appendicitis or infectious complications after colorectal surgery. However, CRP has a slow kinetic after onset of infection, is elevated also in minor infections, and is elevated in many non-infectious causes of inflammation such as trauma, surgery or rheumatic disorders. Indeed, CRP was unable to predict infectious complication after gastro-oesophageal cancer surgery and pancreatic surgery. No prospective randomised studies about the impact of CRP guided treatment algorithms on the clinical course of patients with severe sepsis or septic shock is available. Diagnostic accuracy to differentiate bacterial from non-infectious causes of infection was only moderate (sensitivity 0.75; specificity 0.67).³

 

The performance to differentiate patients with severe sepsis from non-infectious causes of systemic inflammation is even lower. In general, diagnostic accuracy is inferior to procalcitonin. These properties make CRP only of limited use for the application in critical care and are due to (i) the slow kinetic of CRP levels after onset of infection, (ii) CRP increases during minor infection and may not reflect severity of infection, and (iii) CRP is elevated after non-infectious causes of inflammation such as trauma, surgery or rheumatic disorders. CRP is a poor predictor of mortality in this patient population and therefore cannot identify a high risk population. CRP levels are decreasing over the first 48 hours when successful antimicrobial therapy is initiated.⁴ Thus, CRP may be a good marker to monitor success of antimicrobial therapy. 

 

Procalcitonin

Procalcitonin (PCT) is the pro-hormone of calcitonin which is normally produced in the C-cells of the thyroid glands. While all PCT is cleaved to calcitonin in healthy humans, presence of sepsis causes a massive release of PCT into the bloodstream within 4–12 hours after onset of infection. PCT levels are associated with outcome; patients with septic shock present the highest PCT serum concentrations (4–45ng/ml) while patients with uncomplicated lower respiratory tract infections have PCT levels between 0.1–0.5ng/ml. A recent meta-analysis including 3244 surgical and medical patients from 30 studies calculated a sensitivity of 0.77 and a specificity of 0.79 to discriminate sepsis from non-infectious causes of sepsis.⁵ Thus, PCT is a helpful marker in the early diagnosis of sepsis. 

 

However, there are many relevant conditions where PCT can be elevated although infection is not present such as cardiac arrest, trauma, severe surgery, and some autoimmune diseases. It has to be noted that patients with medullary thyroid carcinoma present high PCT serum concentrations of about 100ng/ml and higher independent from an infectious disease. Despite its shortcomings, PCT is currently the most investigated biomarker for early sepsis diagnosis and can aid the physician in this respect.

 

PCT serum concentrations decrease with a half-life of about 24 hours when the infection is sufficiently treated. It has been shown that adequately decreasing in opposite to constantly elevated PCT levels are associated with an improved outcome. Several studies have therefore addressed the question whether PCT can be used to identify the appropriate time to finish antimicrobial therapy. Indeed, it was shown that PCT-guided treatment algorithms resulted in a significant reduction in duration of antimicrobial therapy without jeopardising the treatment results.⁶

 

These studies were mainly done in community acquired lower respiratory tract infections. Some smaller studies addressed this question in the critical care setting on patients with severe sepsis and septic shock. Although these studies delivered promising results it is currently unclear whether a PCT algorithm can be safely and efficiently applied in this patient population. Safety and efficacy of PCT-guided therapy in patients with severe sepsis or septic shock is currently tested in a prospective randomised multicentre study (SISPCT-study; ClinicalTrials-ID: NCT00832039). 

 

Interleukin-6

Interleukin (IL)-6 is directly induced by the primary cytokines of sepsis tumour necrosis factor (TNF) and IL-1. IL-6 is the fastest biomarker as it reaches peak levels within two hours after the infectious stimulus. Since this biomarker persists much longer in the bloodstream than TNF and IL-1 and it is closely linked to the systemic inflammatory response during sepsis IL-6 seems to be predestined for the diagnosis of sepsis. Serum levels of IL-6 are closely related to the severity and outcome of sepsis in patients and decrease in patients where the infection is controlled.

 

However, large prospective studies to investigate the performance of this biomarker are missing. Available studies are showing conflicting results about the ability to accurately identify patients with sepsis. As true for other biomarkers of sepsis, major surgery and major trauma can induce the release of IL-6 without presence of infection. The role of IL-6 as sepsis biomarker remains to be established. Recently, a study on a larger cohort of neonates and infants showed a good diagnostic accuracy for diagnosing sepsis in a paediatric population.⁷

 

Other biomarkers

The soluble triggering receptor expressed on myeloid cells-1 (sTREM-1) is a member of the immunoglobulin superfamily, which is up-regulated on phagocytes after exposure to bacteria and fungi and then released into body fluids. High sTREM-1 levels are associated with a poor outcome. Clinical data for this biomarker are still limited but available studies report a diagnostic accuracy for differentiating sepsis from non-infectious causes of inflammation which is similar to PCT (sensitivity 0.79, specificity 0.8).⁸ However, other inflammatory diseases as well as pancreatitis without infection may affect sTREM-1-levels. Therefore, sTREM-1 seems to be an interesting biomarker for the diagnosis of sepsis but larger studies are warranted.

 

Lipopolysaccharide (LPS)-binding protein (LBP) is involved in the signal transduction of infection as it forms a complex with LPS and then initiates the release of cytokines and pro-inflammatory mediators by binding to CD14 and the TOLL-like receptor ⁴. While serum LBP concentrations average from 5–10 µg/ml in healthy humans, sepsis causes an increase to 30–40µg/ml within 24 hours. However, differentiation of infectious from non-infectious causes of systemic inflammation is poor.⁹ Furthermore, LPS concentrations are not associated with outcome and do not allow the monitoring of an adequate treatment response. Currently, LBP does not play a role in the diagnosis of sepsis.

 

Soluble urokinase plasminogen activator receptor (suPAR) appears in the blood and other body fluid after it is cleaved from the membrane of immune cells. High suPAR levels of greater than 12ng/ml are associated with a poor outcome. However, current data do not support the application of suPAR as a biomarker for the diagnosis of sepsis as this protein seems to be a general marker of inflammation rather than specific for infection.

 

Biomarkers panels are another approach to increase the diagnostic accuracy. Measurement of a single biomarker may not be adequate to reflect the complex pathophysiology of sepsis. Therefore, some studies combined several biomarkers to a panel which resulted in a better separation of sepsis and non-infectious systemic inflammation than one biomarker alone,¹⁰ However, it is currently unclear which combination of biomarkers is most appropriate. 

 

Pathogen detection

Another option to diagnose sepsis is to directly identify the underlying pathogen. The results of microbiological samples currently do not play a role in the treatment decisions of patients with suspected sepsis since results of microbiological samples may only be available up to 72 hours after sampling. Furthermore, blood culture is only positive in 30% of the patients with sepsis. This shortcoming of culture-based pathogen detection causes a diagnostic dilemma since guidelines require that an adequate antimicrobial therapy should be initiated as soon as possible.

 

Pathogen detection based on multiplex polymerase chain reaction (PCR), which detects specific sequences of bacterial and fungal ribosomal RNA in the blood, might offer an solution to this problem11 since results of a PCR may theoretically be available within one working day. Several systems are commercially available which have been investigated regarding their accuracy to predict positive blood cultures. In general, multiplex PCR produces twice as many positive results than a single set of blood cultures, which still leaves more than half of the septic patients with a negative PCR. A meta analysis of 34 studies calculated a pooled sensitivity for combined bacteremia and fungemia of 0.75 and a combined specificity of 0.92.¹² Better results have been reported for the detection of fungemia alone (sensitivity 0.95, specificity 0.92).¹³ Indeed, antifungal therapy based on PCR-based detection of fungi improved outcome of patients after bone-marrow transplantation. However, such studies in the critical care setting are missing.

 

The method is limited by the fact that only pathogens of the assay’s PCR target list can be discovered and identification of antibiotic resistance is very limited (that is, methicillin-resistant staphylococci, vancomycin-resistant enterococci). Thus, PCR-based pathogen detection cannot replace culture-based diagnosis and would therefore significantly increase costs. Some studies to cost effectiveness exist but results are not yet conclusive. Data about the application of the PCR into clinical practice revealed an average time to result of 24 hours which significantly exceeds the expected eight hours.¹⁴ Faster availability of the results would need 24 hours a day and seven days a week coverage of technicians and equipment.

 

Instead of directly detecting pathogens in the blood sample other techniques focus on a faster pathogen detection in the blood culture. Several methods have been introduced into clinical practice and some of them have the potential to reach the time to result of the PCR.¹⁵ These methods do however need a positive blood culture as a prerequisite. In this context, MALDI-TOF (matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry) is a very promising method. Introduction of MALDI-TOF together with an antibiotic stewardship programme resulted in an a shorter time to appropriate antimicrobial therapy.¹⁶ More of such studies are needed to evaluate the impact of diagnostic methods on the care of patients with sepsis.

 

Conclusion

There is currently no biomarker or biomolecular technique available which alone allows a rapid and reliable discrimination between sepsis and other causes of systemic inflammation. Thus, diagnosis and initiation of therapy remains a clinical decision by assessing the patient’s history, possible symptoms of infection, and development of acute organ dysfunction. However, biomarkers can aid and shorten this decision process when taking into account the shortcomings of biomarkers. PCT is currently the most investigated biomarker for this purpose and the only biomarker, which has been integrated into treatment algorithms.

 

CRP and IL-6 are inferior to PCT for the diagnosis of sepsis in most of the studies but also less well investigated. Likewise, PCR based pathogen detection may shorten the time to prescription of an appropriate antimicrobial therapy but cannot out-rule the presence of infection when negative. Currently, the improvement of time to pathogen detection by bimolecular techniques is a promising way to aid the physician in the fast prescription of adequate antimicrobial therapy. There is a lack of clinical studies, which investigate the incorporation of new diagnostic approaches into clinical algorithms.

 

1. Dellinger RP et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med 2013;39(2):165–228. 
2. Bloos F, Reinhart K. Rapid diagnosis of sepsis. Virulence 2014;5(1):154–160. 
3. Simon L et al. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a systematic review and meta-analysis. Clin Infect Dis 2004;39(2):206–17. 
4. Póvoa P et al. C-reactive protein, an early marker of community-acquired sepsis resolution: a multi-center prospective observational study. Crit Care 2011;15(4):R169
5. Wacker C et al. Procalcitonin as a diagnostic marker for sepsis: a systematic review and meta-analysis. Lancet Infect Dis 2013;13(5):426–35. 
6. Schuetz P, Briel M, Mueller B. Clinical outcomes associated with procalcitonin algorithms to guide antibiotic therapy in respiratory tract infections. JAMA 2013;309(7):717–8. 
7. Neunhoeffer F et al. Serum Concentrations of Interleukin-6, Procalcitonin, and C-Reactive Protein: Discrimination of Septical Complications and Systemic Inflammatory Response Syndrome after Pediatric Surgery. Eur J Pediatr Surg 2016;26(2):180–5.
8. Wu Y et al. Accuracy of plasma sTREM-1 for sepsis diagnosis in systemic inflammatory patients: a systematic review and meta-analysis. Critical Care 2012;16(6):R229. 
9. Sakr Y et al. Lipopolysaccharide binding protein in a surgical intensive care unit: a marker of sepsis? Crit Care Med 2008;36(7):2014–22. 
10. Langley RJ et al. Integrative “omic” analysis of experimental bacteremia identifies a metabolic signature that distinguishes human sepsis from systemic inflammatory response syndromes. Am J Respir Crit Care Med 2014;190(4):445–55. 
11. Pletz MW, Wellinghausen N, Welte T. Will polymerase chain reaction (PCR)-based diagnostics improve outcome in septic patients? A clinical view. Intensive Care Med 2011;37(7):1069–76. 
12. Chang S-S et al. Multiplex PCR system for rapid detection of pathogens in patients with presumed sepsis – a systemic review and meta-analysis. PLoS ONE 2013;8(5):e62323. 
13. Avni T, Leibovici L, Paul M. PCR diagnosis of invasive candidiasis: systematic review and meta-analysis. J Clin Microbiol 2011;49(2):665–70. 
14. Bloos F et al. Evaluation of a polymerase chain reaction assay for pathogen detection in septic patients under routine condition: an observational study. PLoS ONE 2012;7(9):e46003. 
15. Liesenfeld O et al. Molecular diagnosis of sepsis: New aspects and recent developments. Eur J Microbiol Immunol (Bp) 2014;4(1):1–25. 
16. Huang AM et al. Impact of rapid organism identification via matrix-assisted laser desorption/ionization time-of-flight combined with antimicrobial stewardship team intervention in adult patients with bacteremia and candidemia. Clin Infect Dis 2013;57(9):1237–45.