Neuron-specific enolase and Tau protein as biomarkers for sepsis-associated delirium: a cross-sectional pilot study

ABSTRACT

Objective  Sepsis-associated delirium is a common cerebral manifestation in patients with sepsis, potentially caused by a combination of neuroinflammation and other neurophysiological disorders. This study investigated the expression of neuron-specific enolase and Tau protein as biomarkers in patients with sepsis-associated delirium. While neuron-specific enolase and Tau protein are known to be associated with brain injury, their diagnostic potential in patients with sepsis-associated delirium is not well understood.

Methods  This cross-sectional pilot study evaluated plasma levels of neuron-specific enolase and Tau protein in patients with delirium and sepsis to explore their potential for identifying sepsis in patients admitted to the emergency department.

Results  A total of 25 patients with delirium were analyzed, 56% of whom had sepsis. Patients with sepsis exhibited significantly higher neuron-specific enolase levels (2.7ng/mL [95%CI= 2.2-3.2] versus 1.3 ng/mL [95%CI= 0.8-2.5], p<0.003) and Tau protein levels (96.1pg/mL [95%CI= 77.0-111.3] versus 43.0pg/mL [95%CI= 31.2-84.5], p<0.003) compared to patients without sepsis. Neuron-specific enolase and Tau protein thresholds of >2.08ng/mL and >59.27pg/mL, respectively, demonstrated 90% specificity for identifying sepsis in patients.

Conclusion  Neuron-specific enolase and Tau protein levels were significantly higher in patients with sepsis than in those without, underscoring their potential ability to identify the infectious etiology of delirium in older patients admitted to emergency departments. Clinical Trials #RBR-233bct.

Biomarkers; Delirium; Emergency medicine; Neuroinflammatory diseases; Sepsis

Highlights

■ Biomarkers of brain injury, such as neuron-specific enolase and Tau proteins, are higher in older patients with sepsis and delirium.

■ Diagnosing sepsis in patients with delirium can be challenging.

■ Early identification of sepsis is key to managing sepsisassociated delirium.

In Brief

In this study, Pinto et al. identified significantly higher levels of neuron-specific enolase and Tau protein in older patients with sepsis-associated delirium in the emergency department, suggesting the potential of these biomarkers as diagnostic tools in this population.

INTRODUCTION

Delirium is defined by the acute onset of fluctuating levels of consciousness, impaired attention, disorganized thinking, and cognitive dysfunction.¹ It is frequently observed in emergency departments (EDs) and is associated with longer hospital stays, higher mortality rates, increased admissions to intensive care units (ICUs), and long-term cognitive impairment.²,³

Various conditions, including metabolic disorders, intoxication, prolonged immobilization, and, most commonly, infections, can trigger delirium in susceptible patients.⁴ However, because there are no definitive blood or imaging tests to diagnose delirium or identify its underlying cause, patients often use multiple invasive procedures to determine the etiology and guide appropriate treatment.⁵

The most widely accepted theory of delirium pathophysiology involves systemic injury and inflammation, leading to the blood-brain barrier (BBB) disruption, glial and neuronal activation, increased inflammation, and subsequent cell death.⁶ Similarly, sepsis can cause endothelial and BBB damage, resulting in the release of neuronal proteins into the bloodstream.⁷ Neuron-specific enolase (NSE) is a recognized biomarker of neuronal death and has prognostic value in patients who are neurocritical. Tau protein, which plays a key role in stabilizing axonal microtubules, has been linked to cognitive impairment. Both NSE and Tau are serum biomarkers indicative of brain injury and have been associated with delirium and, more recently, sepsis.⁸,⁹

OBJECTIVE

This study evaluated the potential correlation between plasma levels of brain cell injury biomarkers-neuron-specific enolase and Tau protein- and delirium in patients. Additionally, it sought to identify sepsis as a potential cause of delirium in patients admitted to emergency departments.

METHODS

Study design, setting, and population

This cross-sectional pilot study analyzed patients aged 65 years or older with acute medical emergencies admitted to the ED of a tertiary university hospital (Hospital das Clínicas) in São Paulo, Brazil, between September 30, 2019, and March 17, 2020. Hospital das Clínicas is a 2,200-bed facility dedicated to caring for high-complexity medical and surgical patients. This study is part of an ongoing research project on Biomarkers and Clinical Scores in Critically Ill Patients, which investigates serum biomarkers in older patients in the ED and their outcomes.¹⁰

We evaluated all older patients admitted with delirium for inclusion in this study. Patients were excluded if they met any of the following criteria: hospitalization for more than 24 h in any healthcare facility, hospitalization within 30 days before admission, anticipated hospital discharge in less than 48 h, or receipt of end-of-life care. Additional exclusions included patients without delirium at admission, those unable to provide consent, delirious patients without a reliable informant, patients who had undergone thrombolysis or had contraindications for venipuncture, those facing significant language barriers, readmissions, previously included patients, those who refused participation, and those lost to follow-up (Figure 1).

Figure 1
Flowchart for the selection of patients

The study protocol was approved by the Ethics Committee for the Analysis of Research Project (CAPPesp) of the Hospital das Clinicas, Faculdade de Medicina, Universidade de São Paulo (CAAE: 77169716.2.0000.0068; # 2.327.363). Written informed consent was obtained from all participants. We report our results in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.(¹¹

Delirium assessment

We used the Confusion Assessment Method (CAM) algorithm for admission assessment.¹² Our standardized interview protocol included a brief neuropsychiatric history, cognitive screening (10-CS), attention testing (reciting the days of the week backward and the vigilance A test), level of consciousness assessment (Richmond Agitation and Sedation Scale [RASS]), and reviews of the electronic medical records.¹³-¹⁶

Raters underwent training sessions before the study began, which involved simulations and bedside evaluations. We achieved high inter-rater reliability for diagnosing CAM-based delirium (>95%). In cases where raters were unsure about the presence of delirium, two experienced physicians re-evaluated the assessments to confirm the final diagnosis.

Brain injury biomarker measurements

Sampling procedures were performed after patient inclusion. Three registered nurses collected 30ml of blood via venipuncture. Blood samples for brain injury biomarker analyses were immediately centrifuged for 10 min. Plasma was preserved at -20°C for up to 48 h before being transferred to a -80°C freezer for long-term storage and further processing.

We used the Milliplex^® magnetic bead immunoassay and the MAGPIX^® System (Merck Millipore, USA) to measure cytokine plasma levels.

Statistical analyses

Our primary outcome was the diagnosis of sepsis at admission, based on the Sepsis-3 criteria, which included the presence of a suspected or confirmed infection and a change of two or more points in the Sequential Organ Failure Assessment (SOFA) score from baseline. Sepsis was confirmed through a complete review of medical records, including a detailed evaluation of clinical symptoms, medical history, physical examination findings, and laboratory results such as blood cultures and other relevant tests. Although radiological data were not directly used for sepsis diagnosis, they were reviewed for additional context regarding infection-related findings, and microbiological data were considered if available. The primary focus was on clinical and laboratory evidence consistent with the Sepsis-3 criteria.¹⁷ All enrolled patients were included in the analysis on an intention-to-treat basis.

The number of participants was determined based on feasibility, available resources, research staff capacity, and the number of eligible patients, which aligned with current recommendations.¹⁸ Despite our ED providing medical care to 800 older patients monthly, with 30% being eligible for hospitalization, most had been transferred from less complex hospitals and hospitalized for more than 24 h at the time of recruitment. Recruitment was completed one month ahead of schedule due to the onset of the COVID-19 pandemic.

We used the Shapiro-Wilk test to assess the normality of variable distributions. Sepsis occurrence was analyzed using unpaired t-tests or Mann-Whitney U tests, depending on the normality of variable distributions. Categorical variables were analyzed using Pearson’s χ² test. All analyses were performed using Stata software (StataCorp, 2021, Stata Statistical Software: Release 17, College Station, TX: StataCorp LLC).

RESULTS

Patient demographic and clinical characteristics

We analyzed 1,201 patients, of whom 25 (2.1%) were older patients diagnosed with delirium at admission. The cohort was predominantly female (n=14, 56%), with a mean age of 78.2 (± 8.0) years. Patients were hospitalized for a median of 8 (4-15) days, and 9 (36%) died (Figure 1, Table 1).

Table 1
Characteristics of the patients

	Septic patients (n=14)	Non-septic patients (n=11)	p value
Age, median (IQR), years	76 (12.3)	79 (18)	0.3*
Female, n (%)	6 (42.8)	8 (72.7)	0.1‡
Education level, median (IQR), years	8 (5.2)	3 (4)	0.007*
Charlson Comorbidity Index, median (IQR)	2 (2.2)	3 (2)	0.9*
Polypharmacy, n (%)	8 (57.1)	4 (36.3)	0.3‡
Katz Index, median (IQR)	4.5 (6)	4 (11)	0.6*
Frail, n (%)	6 (42.8)	5 (45.4)	0.9‡
Dementia, n (%)	2 (14.2)	1 (9.0)	0.7‡
ICU admission, n (%)	4 (28.5)	4 (36.3)	0.7‡
In-hospital mortality, n (%)	4 (28.5)	5 (45.4)	0.4‡

• Variables are expressed as absolute numbers (%), except for age, education level, and Charlson Comorbidity Index, which are expressed as medians (interquartile range [IQR]).
• *p values were calculated using the nonparametric Mann-Whitney test for continuous quantitative variables; ^‡ p values were calculated using the χ² test for categorical variables.
• ICU: intensive care unit.

Among the patients with delirium, 14 (56%) were diagnosed with sepsis. The most common site of infection in these patients with sepsis was pneumonia (n=6, 42.8%), followed by skin and soft tissue infections (n=3, 21.4%), urinary tract infections (n=2, 14.2%), and abdominal infections (n=2, 14.2%). Patients without sepsis had conditions including stroke (n=4, 36.3%), acute heart failure (n=2, 18.1%), and acute upper gastrointestinal bleeding (n=2, 18.1%).

Main results

Patients with sepsis had significantly higher NSE and Tau protein levels than those without sepsis (Table 2, 3 and Figure 2).

Table 2
Characteristics of biomarkers

	Septic patients (n=14)	Non-septic patients (n=11)	p value*
NSE, median (IQR), [95% CI]	2.7 (0.9), [2.2–3.2]	1.3 (1.0), [0.8–2.5]	0.003
Tau, median (IQR), [95% CI]	96.1 (30.1), [77.0–111.3]	43.0 (41.6), [31.2–84.5]	0.003

• * p values were calculated using the nonparametric Mann-Whitney U test.

Table 3
Characteristics of the neuron-specific enolase and Tau protein

	Value	Sensitivity %	Specificity %	LR +	LR -	AUC
NSE	2.08	85.7	81.2	4.7	0.17	0.84
Tau	59.27	92.9	72.7	3.4	0.09	0.84

• NSE: neuron-specific enolase; LR: likelihood ratio.

Figure 2
(A) Neuron-specific enolase in patients with and without sepsis and (B) Tau protein in patients with and without sepsis

DISCUSSION

This study analyzed older patients with delirium and found that plasma levels of NSE and Tau protein were higher in those with sepsis. Sepsis is a potentially fatal condition caused by a dysregulated systemic response to infection,¹⁷ with the brain being particularly vulnerable to its effects. Symptoms range from mild confusion to deep coma.¹⁹ A common neurological complication is sepsis-associated delirium (SAD), which is thought to result from a combination of neuroinflammation, disturbances in cerebral perfusion, BBB dysfunction, and altered neurotransmission.²⁰

Effective management of SAD depends on the early detection and treatment of underlying sepsis, as controlling SAD after it develops is challenging.²⁰ Delirium may be the only clinical manifestation of sepsis in older patients, and diagnosing it can be delayed due to the need to investigate its etiology.²¹ However, the utility of biomarkers, neuroimaging, and electroencephalograms for diagnosing SAD remains controversial.²²

Plasma levels of NSE and Tau protein have been associated with brain injury and delirium.²³,²⁴ Neuron-specific enolase is a promising peripheral blood marker of brain injury as it is a cytosolic enzyme primarily expressed in neurons and neuroendocrine cells, with a high concentration in the brain.²⁵ Neuron-specific enolase has demonstrated value in assessing the severity of brain injury and early prognosis following traumatic brain injury,²⁶ cardiac arrest,²⁷ and sepsis.²⁸

Tau is a microtubule-associated protein highly expressed in neurons, which regulates axonal microtubule stability. Its levels increase in response to various neurodegenerative conditions and following stroke.²⁹,³⁰ Elevated Tau protein levels can also result from injury, synaptic activity, or synaptic pruning, making their interpretation complex.³¹ Plasma Tau protein levels have been linked to the incidence and severity of postoperative delirium.²⁴ To our knowledge, this study is the first to demonstrate an association between Tau protein levels and SAD.

This study has some limitations. First, while NSE and Tau protein levels were measured in peripheral blood, these values may not fully correspond to those in the brain. However, this was an ED-based study designed to be feasible and reproducible.

Second, many patients were not enrolled during the study period. Recruitment occurred daily in the morning, but some patients were transferred to wards or ICUs within hours of ED admission. Additional factors, such as difficulties obtaining consent, loss of follow-up, and logistical challenges, further contributed to patient losses. To address these issues, we implemented measures to mitigate patient attrition. These included enhanced follow-up efforts through multiple phone and mail attempts to track and retain participants. We also reviewed and simplified the consent procedures to better accommodate patient needs, along with providing additional support when required.

Finally, although our findings suggest that plasma NSE and Tau protein changes are associated with sepsis in patients with delirium, the underlying mechanisms remain unclear. These changes could result from endothelial dysfunction, neuronal injury, synaptic alterations, or a combination of these processes.

CONCLUSION

Neuron-specific enolase and Tau protein levels are established biomarkers of brain injury. In this study, among the older patients with delirium, these biomarker levels were higher in those with sepsis patients compared to those without sepsis. We propose that plasma levels of neuron-specific enolase and Tau protein be further investigated as potential tools for identifying the infectious etiology of delirium in older patients in emergency departments.

ACKNOWLEDGMENT

This study was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; grant numbers # 2016/14.566-4 and # 2023/04.997-1).

REFERENCES

• ¹ Garcez FB, Apolinario D, Campora F, Curiati JA, Jacob-Filho W, Avelino-Silva TJ. Delirium and post-discharge dementia: results from a cohort of older adults without baseline cognitive impairment. Age Ageing. 2019;48(6):845-51.
• ² Kennedy M, Enander RA, Tadiri SP, Wolfe RE, Shapiro NI, Marcantonio ER. Delirium risk prediction, healthcare use and mortality of elderly adults in the emergency department. J Am Geriatr Soc. 2014;62(3):462-9.
• ³ Alencar JC, Carvalho FB, Veiga AD, Stanzani AS, Maia IW, Silva LO, et al. Delirium during Covid-19 pandemic: were we over-reporting an overlooked syndrome? JBMEDE. 2022;2(3):e22015.
• ⁴ Oliveira J E Silva L, Berning MJ, Stanich JA, Gerberi DJ, Murad MH, Han JH, et al. Risk Factors for Delirium in Older Adults in the Emergency Department: A Systematic Review and Meta-Analysis. Ann Emerg Med. 2021;78(4):549-65.
• ⁵ Marcantonio ER. Delirium in Hospitalized Older Adults. N Engl J Med. 2017;377(15):1456-66.
• ⁶ Maldonado JR. Neuropathogenesis of delirium: review of current etiologic theories and common pathways. Am J Geriatr Psychiatry. 2013;21(12):1190-222. Review.
• ⁷ Barichello T, Generoso JS, Collodel A, Petronilho F, Dal-Pizzol F. The blood-brain barrier dysfunction in sepsis. Tissue Barriers. 2021;9(1):1840912.
• ⁸ Mietani K, Hasegawa-Moriyama M, Inoue R, Ogata T, Shimojo N, Kurano M, et al. Elevated neuron-specific enolase level is associated with postoperative delirium and detection of phosphorylated neurofilament heavy subunit: A prospective observational study. PLoS One. 2021;16(11):e0259217.
• ⁹ McKay TB, Qu J, Liang F, Mueller A, Wiener-Kronish J, Xie Z, et al. Tau as a serum biomarker of delirium after major cardiac surgery: a single centre case-control study. Br J Anaesth. 2022;129(1):e13-6.
• ¹⁰ de Alencar JC, Garcez FB, Pinto AA, Silva LO, Soler LM, Fernandez SS, et al. Brain injury biomarkers do not predict delirium in acutely ill older patients: a prospective cohort study. Sci Rep. 2023;13(1):4964.
• ¹¹ Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M; STROBE Initiative. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007;4(10):e297. Review.
• ¹² Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Ann Intern Med. 1990;113(12):941-8.
• ¹³ Apolinario D, Lichtenthaler DG, Magaldi RM, Soares AT, Busse AL, Amaral JR, et al. Using temporal orientation, category fluency, and word recall for detecting cognitive impairment: the 10-point cognitive screener (10-CS). Int J Geriatr Psychiatry. 2016;31(1):4-12.
• ¹⁴ Tieges Z, Brown LJ, MacLullich AM. Objective assessment of attention in delirium: a narrative review. Int J Geriatr Psychiatry. 2014;29(12):1185-97. Review.
• ¹⁵ Sessler CN, Gosnell MS, Grap MJ, Brophy GM, O'Neal PV, Keane KA, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166(10):1338-44.
• ¹⁶ Krewulak KD, Hiploylee C, Ely EW, Stelfox HT, Inouye SK, Fiest KM. Adaptation and Validation of a Chart-Based Delirium Detection Tool for the ICU (CHART-DEL-ICU). J Am Geriatr Soc. 2021;69(4):1027-34.
• ¹⁷ Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-10.
• ¹⁸ Bacchetti P, McCulloch CE, Segal MR. Simple, defensible sample sizes based on cost efficiency. Biometrics. 2008;64(2):577-85.
• ¹⁹ Widmann CN, Heneka MT. Long-term cerebral consequences of sepsis. Lancet Neurol. 2014;13(6):630-6.
• ²⁰ Atterton B, Paulino MC, Povoa P, Martin-Loeches I. Sepsis Associated Delirium. Medicina (Kaunas). 2020;56(5):240.
• ²¹ Kennedy M, Helfand BK, Gou RY, Gartaganis SL, Webb M, Moccia JM, et al. Delirium in Older Patients With COVID-19 Presenting to the Emergency Department. JAMA Netw Open. 2020;3(11):e2029540.
• ²² Toft K, Tontsch J, Abdelhamid S, Steiner L, Siegemund M, Hollinger A. Serum biomarkers of delirium in the elderly: a narrative review. Ann Intensive Care. 2019;9(1):76. Review.
• ²³ Anderson BJ, Reilly JP, Shashaty MG, Palakshappa JA, Wysoczanski A, Dunn TG, et al. Admission plasma levels of the neuronal injury marker neuron-specific enolase are associated with mortality and delirium in sepsis. J Crit Care. 2016;36:18-23.
• ²⁴ Ballweg T, White M, Parker M, Casey C, Bo A, Farahbakhsh Z, et al. Association between plasma tau and postoperative delirium incidence and severity: a prospective observational study. Br J Anaesth. 2021;126(2):458-66.
• ²⁵ Taylor CB, Royds JA, Parsons MA, Timperley WR. Diagnostic aspects of enolase isozymes. Isozymes Curr Top Biol Med Res. 1983;11:95-119.
• ²⁶ Vos PE, Lamers KJ, Hendriks JC, van Haaren M, Beems T, Zimmerman C, et al. Glial and neuronal proteins in serum predict outcome after severe traumatic brain injury. Neurology. 2004;62(8):1303-10.
• ²⁷ Meynaar IA, Oudemans-van Straaten HM, van der Wetering J, Verlooy P, Slaats EH, Bosman RJ, et al. Serum neuron-specific enolase predicts outcome in post-anoxic coma: a prospective cohort study. Intensive Care Med. 2003;29(2):189-95.
• ²⁸ Nguyen DN, Spapen H, Su F, Schiettecatte J, Shi L, Hachimi-Idrissi S, et al. Elevated serum levels of S-100beta protein and neuron-specific enolase are associated with brain injury in patients with severe sepsis and septic shock. Crit Care Med. 2006;34(7):1967-74.
• ²⁹ Bateman RJ, Blennow K, Doody R, Hendrix S, Lovestone S, Salloway S, et al. Plasma Biomarkers of AD Emerging as Essential Tools for Drug Development: An EU/US CTAD Task Force Report. J Prev Alzheimers Dis. 2019;6(3):169-73.
• ³⁰ Pase MP, Himali JJ, Aparicio HJ, Romero JR, Satizabal CL, Maillard P, et al. Plasma total-tau as a biomarker of stroke risk in the community. Ann Neurol. 2019;86(3):463-7.
• ³¹ Sato C, Barthélemy NR, Mawuenyega KG, Patterson BW, Gordon BA, Jockel-Balsarotti J, et al. Tau Kinetics in Neurons and the Human Central Nervous System. Neuron. 2018;97(6):1284-98.e7. Erratum in: Neuron. 2018;98(4):861-4.

• AVAILABILITY OF DATA AND MATERIAL
  Data supporting the findings of this study are available from the corresponding author upon request.
• This manuscript has not been published previously and is not under consideration for publication elsewhere, either in whole or in part, in any language.

Edited by

• Associate Editor:
  Antonio Lucio Teixeira Biggs Institute, University of Texas Health Science Center, San Antonio, TX, USA ORCID: https://orcid.org/0000-0002-9621-5422

Data availability

Data supporting the findings of this study are available from the corresponding author upon request.

Publication Dates

• Publication in this collection
  07 Apr 2025
• Date of issue
  2025

History

• Received
  21 June 2024
• Accepted
  20 Oct 2024