Diagnostics of Central and Autonomic Nervous System Dysfunction in Patients with Sepsis-Associated Encephalopathy

*Irina Alexandrovna Savvina, Daria V. Ryzhkova, Kristina M. Bykova, Konstantin E. Lebedev, Anna Olegovna Petrova, Natalya V. Dryagina, Elena G. Potemkina and Eleonora T. Nazaralieva*

## **Abstract**

This chapter is devoted to monitoring of central and autonomic nervous system (ANS) in patients with verified sepsis to recognize the specific functional and anatomic changes in the brain and its important autonomic centers which is named sepsis-associated encephalopathy (SAE). Fluctuation of conscience level from agitation to delirium and coma, muscle tone, and severity of pain syndrome is evaluated with different scales (SOFA, SAPS II, RASS, CAM-ICU, FOUR, PBSS, BPS, MRC, MAS, CNS). Multimodal neuromonitoring includes EEG, EPs, ENMG, cerebral oxymetry, saturation in the bulb of the jugular vein, TCD, and neuroimaging (MRI, PET). Dysfunction of autonomic brainstem structures is detected with variational cardiointervalometry, pupillometry, thermometry (peripheral and central), photoplethysmography assessment of perfusion index, quantitative assessment of muscle strength on the MRC scale and MAS, and diagnostics of the severity of the PSH syndrome. Monitoring data help clinicians to make decisions on SAE patient management tactics.

**Keywords:** sepsis-associated encephalopathy, sepsis, neurophysiological monitoring, neuroimaging, multimodal assessment of neurological status, autonomic nervous system dysfunction

### **1. Introduction**

The dysfunction of the central nervous system due to sepsis is realized in the form of sepsis-associated encephalopathy (SAE) with different levels of neurological deficit. SAE is an early manifestation of an infection of the body and may appear before other systemic signs of sepsis become apparent. SAE occurs in 70% of septic

patients in critical condition and is associated with a higher mortality rate, longer hospital stay, poorer quality of life, and lower cognitive function scores than in other patients with sepsis without septic encephalopathy [1–3]. SAE can act as an independent predictor of mortality in severe sepsis and septic shock. The degree of its severity corresponds to the severity of the septic process [4].

The use of SOFA and SAPS II evaluation scales allows for determining the severity of the patient's condition for further solving therapeutic, prognostic, and legal issues. These scales are a clinical tool for assessing the patients' condition to predict the likelihood of developing a particular outcome, including death, disability, or long-term neurocognitive consequences. SAPS II was found to be more accurate than APACHE II, a similar intensive care classification score, and is commonly used in studies to compare morbidity and outcomes among patients.

### **2. Diagnostics of central nervous system dysfunction in patients with SAE**

Patients with verified sepsis according to the international consensus definitions of sepsis and septic shock sepsis-3 (Third International Consensus Definitions for Sepsis and Septic Shock, 2016), undergo clinical, laboratorial, radiological, and functional examination in order to identify nonspecific and specific functional and structural changes in the central nervous system inherent in acute cerebral injury, caused by a systemic infection. According to a number of authors, the absence of specific markers for the diagnosis of SAE complicates the diagnosis [4–6].

#### **2.1 Neurological assessment**

Impaired consciousness is revealed in a significant proportion of patients with sepsis, and it may even precede the manifestations of the underlying disease. At the same time, the presence of signs of a disorder of consciousness sharply worsens the prognosis. There are various clinical variants of the course of SAE from acute minimal changes in mental status and behavioral disorders (inattention, disorientation, and excitement) to delirium and coma [7]. Delirium is defined as a single syndrome in which a number of different etiological causes cause a relatively consistent pattern of acute generalized cognitive disorders [8].

A patient with sepsis and SAE is given a comprehensive diagnosis of disorders of the level of consciousness, in the case of coma—brainstem reflexes to assess the depth of coma, in the case of fluctuations in the level of consciousness—agitation, excitement, aggression, and confusion to a hypo-active state, for objective dynamic monitoring of a life-threatening condition, timely decision-making for therapeutic tactics for delirium relief, transfer of the patient to a state of medically induced coma (neurovegetative stabilization), and respiratory support. Dynamic behavioral assessment of the severity of pain syndrome in a patient with SAE gives clinicians a tool for choosing the regimen and doses of analgesic drugs from NSAIDs and paracetamol to opioids in the structure of neurovegetative blockade.

Assessment of muscle tone is extremely important due to the rapid development of critical condition polyneuromyopathy in patients with sepsis, which undoubtedly affects the rehabilitation potential of a septic patient when the condition is stabilized. We consider it appropriate and recommend using the following scales for patients with sepsis and SAE to assess the dysfunction of the central and autonomous parts of the nervous system:

*DOI: http://dx.doi.org/10.5772/intechopen.108392 Diagnostics of Central and Autonomic Nervous System Dysfunction in Patients…*


Surviving patients are often left with long-term cognitive impairments that can lead to disability. There is increasing evidence that SAE can cause long-term cognitive impairment, including changes in the speed of mental processing, decreased memory, attention, and visual–spatial abilities [9]. They can persist for several years after the diagnosis of SAE, affecting functional abilities, quality of life, and the ability to return to work. Depending on the degree of disorders (e.g., mild, moderate, or severe), the presence of this disease can be a great burden for both family members and caregivers. Many patients diagnosed with SAE have chronic neurocognitive effects for 6 years after discharge from the intensive care unit [10]. A small series of cases showed that several months after discharge from the intensive care unit, all patients had severe memory and attention disorders [11]. These restrictions arose when performing basic everyday tasks, such as eating, bathing, and various household activities. Other important neuropsychiatric diseases that are present in surviving patients are depression and anxiety. The prevalence and severity of these mental disorders in patients vary from 10–58% [12].

#### **2.2 Laboratory diagnostic markers**

With the development of SAE from the standpoint of pathophysiology, the endothelium and blood–brain barrier (BBB) are damaged, resulting in the destruction of proteins with their subsequent release into the bloodstream. These peptides can be considered as markers of BBB integrity. In the study by Barichello et al., plasma levels

of occludin, claudin-5, zonula occludens-1 (ZO-1), procalcitonin, and lactate were evaluated in patients with sepsis [13]. Occludin and ZO-1 were elevated and positively correlated with the severity of the disease. The prognostic value of the ZO-1 level for in-hospital mortality was comparable to the lactate level [14].

In another study, Wang et al., the diameter of the optic nerve sheath (ONSD) was measured using ultrasound in rabbits with SAE and the levels of neurospecific enolase (NSE), astroglial protein S100B, tumor necrosis factor-α (TNF-α) were determined by enzyme immunoassay. According to the results of the study, changes in ONSD (increased over time) correlated with NSE and S100B. ONSD monitoring has shown a high prognostic value in SAE [14].

It is known that S100B and NSE are biomarkers of brain damage in traumatic brain injury, cerebral stroke, and hypoxic–ischemic encephalopathy. A retrospective analysis of a sepsis cohort reported that an NSE concentration > 12.5 μg/L was independently associated with 23% and 29% increased risk of 30-day mortality and delirium, respectively [15]. In the studies of Yao et al., the clinical significance of S100B and NSE in blood serum for the diagnosis of SAE and evaluation of its prognosis was studied. Studies have shown that serum concentrations of S100B in patients with SAE were significantly higher than in patients without SAE. The efficacy and sensitivity of serum S100B in the diagnosis of SAE were high, but they had low specificity. Compared with NSE, serum S100B was better suited both for the diagnosis of SAE and predicting the outcome of sepsis [16]. Also, in a study by Piazza et al., it was shown that the level of S100B protein in patients with SAE was elevated but did not correlate with the Glasgow Coma scale (GCS) at admission, the electroencephalogram pattern (EEG) or the SOFA scale score. It follows from the above that in sepsis, an increase in S100B does not allow physicians to predict neurological recovery [17].

The prognostic value of light (NfL) and heavy (NfH) neurofilament chain levels as biomarkers of neuroaxonal injury in patients with sepsis-associated encephalopathy were measured by Ehler J. et al. in the prospective, pilot observational study (2019) [18]. According to the results of researchers, the plasma and CSF NfL levels were higher in patients with SAE and correlated with the severity of SAE and a poor functional outcome. As the result, there was found a correlation between the higher value of CSF NfL levels and mortality.

## **2.3 Neuroimaging**

#### *2.3.1 Magnetic resonance imaging*

Neuroimaging in patients with SAE gives different results. In some patients, the results of magnetic resonance imaging (MRI) of the brain reflect the normal state of the cortical–subcortical structures and the brainstem, in other patients, acute damage is observed. In the case of acute injury, MRI reveals, in particular, multiple ischemic strokes or white matter lesions in the semi-oval center (mainly at the level of Virchow-Robin spaces). MRI signs of septic acute cerebral injury are presented in **Figure 1.**

According to Sharshar et al., the severity of damage in the central nervous system is associated with the severity of sepsis and inversely correlates with the scores on the GCS [19]. A lesion in the white matter of the brain corresponds to vasogenic edema (e.g., an increased signal on a T2-weighted image and a reduced signal on a diffusionweighted image). Data published by Piazza et al., indicate that the size and number of foci of brain damage vary significantly among patients with SAE [20].

*DOI: http://dx.doi.org/10.5772/intechopen.108392 Diagnostics of Central and Autonomic Nervous System Dysfunction in Patients…*

#### **Figure 1.**

*A, B. MRI of the brain of patient P., 75 y.o., on the 5th day of diagnostics of SAE. T2- WI (A), flair pulse sequence (B). In the subcortical, periventricular parts of the white matter of the brain, multiple different-sized foci of hyperintense signal are visualized, ranging in size from 3 to 14 mm, with a tendency to merge, in places of a draining nature, probably of a dyscirculatory genesis. Pronounced zones of leukoaraiosis are noted periventrically to both horns of the lateral ventricles. Rating on the scale RASS +2 points; CAM-ICU—Delirium is present.*

Posterior reversible encephalopathy syndrome (PRES) is a clinical radiological syndrome characterized by headache, confusion, seizures, and visual impairment [21], which is associated with edema of the white matter, primarily affecting the posterior regions—the occipital and parietal lobes to varying degrees is often diagnosed with sepsis. According to Fugate, who published the results of a Mayo Clinic study, PRES was associated with sepsis in 7–24% of patients [22].

#### *2.3.2 Positron emission tomography*

Metabolic changes detected in brain structures according to positron emission tomography with 18F-fluorodeoxyglucose (18F-FDG PET) in patients with SAE at different stages of the course of sepsis, hypometabolism of various parts of the cerebral cortex are concerned.

**Figures 2–4** show the PET data of the patients with SAE on the 5th, 6th, and 28th days from the diagnosis of sepsis, SAE. The 18F-FDG PET images revealed signs of diffuse hypometabolism in the cerebral cortex, which may reflect not only structural lesions but also functional ones.

However, it should be taken into account that patients with acute cerebral damage of different severity due to sepsis are often transported for imaging and receive hypnotic agents in the structure of intensive therapy, in particular, propofol, which causes nonspecific changes in metabolism in the cerebral cortex to a lesser extent than other hypnotics and anesthetics (e.g., thiopental sodium).

Thus, neuroimaging in patients with SAE gives different results—from the absence of pathological changes to acute cerebral damage, including vasogenic edema of white matter in the periventricular parts of both hemispheres, a pattern of hypoperfusion of the cerebral cortex, and hypometabolism in various parts of the cerebral cortex.

#### **2.4 Transcranial doppler**

Cerebral perfusion pressure (CPP) under condition of violated cerebrovascular autoregulation can lead to global cerebral ischemia (low CPP), and hemorrhage and

#### **Figure 2.**

*18F-FDG PET of the brain on the 5th day after SAE diagnostics in patient P., 75 years old. PET was performed under sedation with propofol 2 mg/kg/h. Glucose hypometabolism in the cerebral cortex in different areas (marked in green). A decrease in glucose metabolism in the parietal lobes. Score according to RASS is a + 2 points, and in CAM-ICU—Delirium is present.*

brain edema (high CPP). Both events may cause neuronal damage and sepsis-associated delirium (SAD). Pfister et al. introduced that in 16 patients impaired autoregulation 48 hours after the onset of sepsis was found, and SAD was diagnosed [23].

The transcranial doppler (TCD)—derived index of autoregulation Mx is used for the assessment of the status of autoregulation; this index has been already calculated for many different clinical conditions: in traumatic brain injury, carotid artery diseases, vasospasm, volunteers, and in sepsis [23, 24].

In the study, Schramm et al. show that autoregulation is impaired in the majority of patients with severe sepsis and septic shock during the first 2 days [25]. Daily measurement of index Mx in patients with verified sepsis during the first 4 days found increased index Mx in the first 2 days, and, thus, impairment of autoregulation which was associated with the occurrence of SAD, suggesting that autoregulation is one of the factors contributing to SAD.

The association between severe sepsis and septic shock and increased cerebral vascular resistance as assessed by TCD estimated pulsatility index (PI) and resistance index (RI) was previously reported (PI = systolic velocity—(diastolic velocity/mean velocity); RI = systolic velocity—(diastolic velocity/systolic velocity)) [26]. The study of Algebaly et al., showed that pediatric septic patients with SAE had significantly higher PI and RI than their counterparts without SAE [27].

Another study on septic patients by Pierrakos et al., identified TCD-derived PI in the first post-admission day as a significant predictor of subsequent SAE with good sensitivity and specificity [28]. Moreover, the study of Crippa et al., which recruited 100 patients with sepsis, reported a significant association between sepsis-associated brain dysfunction and altered cerebral autoregulation detected by TCD [29].

*DOI: http://dx.doi.org/10.5772/intechopen.108392 Diagnostics of Central and Autonomic Nervous System Dysfunction in Patients…*

#### **Figure 3.**

*18F-FDG PET of the brain in 6 days after SAE diagnostics in patient S., 56 years old. Glucose hypometabolism in various parts of the cerebral cortex (marked in green). The study was performed without sedation. The score according to FOUR scale is 6 points; to PBSS is 10 points; and to CNS is 3,5 points.*

#### **Figure 4.**

*18F-FDG PET of the brain on the 28th day of sepsis and SAE in patient S., 56 years old. Glucose hypometabolism persists in various parts of the cerebral cortex (marked in green). The study was performed on the background of sedation with propofol. The score according to GCS is 13 points; to FOUR scale is 10 points; to PBSS is 11 points; and to CNS is 4,5 points before sedation starts.*

Sanz et al., performed brain imaging of 49 pediatric septic shock patients, and the most frequent acute brain lesion patterns, found on neuroimaging, were ischemia and cerebritis (i.e., cerebral edema/damage in the clinical context of infection) [30].

#### **2.5 Cerebral oxymetry**

Determination of cerebral oxygenation is necessary to measure cerebral perfusion. It is carried out with following methods: invasive (direct)—jugular venous bulb oxygen saturation, brain tissue oxygen tension, and noninvasive (indirect)—near-infrared spectroscopy (NIRS), and computed tomography (CT) perfusion [31–33]. Cerebral perfusion may be altered in sepsis patients [24]. Recently, cerebral oxygenation has also been shown to be significantly lower in septic shock patients with delirium [34], suggesting that poor cerebral perfusion may contribute to delirium development.

In NIRS, a sensor and light source are placed on the forehead, where they emit and receive varying wavelengths of near-infrared light (700–1000 nm) [35], representing changes in oxygenated- and deoxygenated hemoglobin [36]. NIRS noninvasively measures peripheral tissue oxygen saturation (StO2). NIRS measurements for the StO2 initial, StO2 occlusion, and StO2 recovery slope were abnormal in patients with septic shock compared to sepsis patients. The recovery slope was most strongly associated with organ dysfunction and mortality [37].

#### **2.6 Neurophysiological monitoring**

#### *2.6.1 Electroencephalography*

Electroencephalography (EEG) allows rapid electrophysiological measurements of the bioelectric activity of the brain and its dysfunction right at the patient's bedside and complements the clinical and neuroimaging assessment of patients with encephalopathy. Both the progressive slowing of background EEG activity with an increase in cerebral insufficiency and a decrease in EEG reactivity to external stimuli provide important diagnostic and prognostic information [38]. Convulsive seizures and periodic discharges (a type of EEG anomaly) are common in patients in critical condition, require EEG monitoring for detection, and are associated with an unfavorable outcome [39]. Sepsis is a risk factor for both convulsive seizures and periodic discharges, including convulsive epileptic status [40]; however, the prevalence of all these disorders in patients with sepsis has not yet been determined, while EEG changes to external stimulation have become a good prognostic factor in patients with acute traumatic brain injury and encephalopathy [38].

In a prospective analysis of patients with sepsis, and with altered mental status, seizures without convulsions (11%), and periodic discharges (25%) were often detected. It was also proved that the absence of EEG reactivity in the first days of sepsis was associated with mortality both at discharge and after 1 year. It has been shown that EEG reactivity is a predictor of outcomes in critically ill patients with a non-neurological primary diagnosis.

A wide range of alterations in the EEG pattern have been registered in patients with sepsis and SAE: progressive slowing of the normal alpha rhythm, appearance of theta to delta activity, triphasic waves, and malignant burst suppression in severe malignant encephalopathy [5, 41, 42]. According to Young et al., EEG patterns have been classified into four categories of increasing severity of SAE [43]. Examples of EEG changes according to the stage of SAE are demonstrated in **Figures 5** and **6**.

*DOI: http://dx.doi.org/10.5772/intechopen.108392 Diagnostics of Central and Autonomic Nervous System Dysfunction in Patients…*

#### **Figure 5.**

*EEG of patient S., 56 years old, in 6 and 16 days after SAE diagnostics, coma 1. EEG— Electroencephalography. Slow-wave activity dominates. The record is represented by the polymorphic activity of the predominantly theta range. There is an interhemispheric asymmetry: in the left hemisphere, the amplitude of biopotentials is reduced (20 mv); in the right hemisphere, mainly in the frontal-temporal leads, groups of delta waves of increased amplitude (70 mv) in combination with acute waves are recorded. EEG data indicate gross widespread changes with a focus on pathological activity in the frontotemporal region of the right hemisphere.*

#### **Figure 6.**

*EEG of patient S., 56 years old, 28 days after SAE diagnostics. EEG— Electroencephalography. Registration in a state of relaxed wakefulness for an hour, mainly with closed eyes. In the background recording, polymorphic activity with an amplitude of up to 30 μV is visualized. No interhemispheric asymmetry or paroxysmal activity was recorded. The obtained EEG data indicate gross widespread changes of a diffuse nature. The score according to FOUR scale is 10 points; PBSS is 11 points; and CNS is 4,5 points.*

According to Sutter et al., the absence of EEG reactivity is a predictor of mortality but does not predict a functional outcome [38]. According to the concept of Moruzzi G., the reactivity of the EEG to external simulation depends on the preservation of brain stem structures, such as the ascending reticular formation, but the reactivity of the EEG does not necessarily reflect higher brain activity [44]. The results of fundamental research by Moruzzi et al., emphasize the importance of secondary brain dysfunction and, perhaps, precisely brain stem dysfunction in critical conditions. The researchers also found that the sedative effect was associated with a decrease in nonconvulsive seizures or periodic discharges and a lack of EEG reactivity. It is possible that the sedative effect interferes with the reactivity test during the EEG. Nevertheless, as noted in a study by Shehabi et al., continuous sedation is associated with increased mortality [45].

EEG monitoring gives us an excellent opportunity to improve the ability to predict and prevent seizures, predict and potentially prevent neuronal disorders [39].

The diagnosis of SAE may precede neurological changes in patients who underwent EEG monitoring [46]. In two-thirds of the cases, violations on the EEG did not give any clinical manifestations. Nevertheless, the progressive slowing of background EEG activity with an increase in cerebral insufficiency, and a decrease in EEG reactivity to external stimuli, provide important diagnostic and prognostic information.

### *2.6.2 Evoked potentials*

The functional state of the sensory and motor pathways of the brainstem in patients with SAE can be assessed by registering multimodal evoked potentials (EPs): motor-evoked potentials (MEPs), somatosensory-evoked potentials (SSEPs), visualavoked potentials (VAPs), and brainstem auditory-evoked potentials (BAEPs).

EPs measure brain responses to sensory stimulation [47], including responses generated by subcortical structures (BAEPs, SSERs), thalamocortical input to the primary sensory cortices, and intrinsic cortical activity [48]. Abnormal EPs are present in the majority of septic patients [41]. The clinical examination may be less sensitive than electrophysiological methods in the diagnosis of SAE. Rinaldi et al., studied the changes in A-line autoregression index (AAI) induced by postsurgical sepsis [49]. It was shown that the occurrence of sepsis significantly reduces AAI. Thus, the measurement of AAI has the potential to be a reliable diagnostic test to identify subclinical SAE [49].

Alteration in the latency or amplitude of the somatosensory-evoked potentials has been frequently reported in this patient population [50], and it is not influenced by sedative medications [42]. Subcortical (i.e., N20–N23 interlatency) and cortical (N20–N70 interlatency) pathways of SSEPs are impaired early in 34% and 84% of 68 septic patients [51]. EPs are useful for assessing SAE in sedated critically ill patients as in contrast to EEG, their latencies are slightly altered by sedatives [42].
