Brain Injury in Critical Illness and a Note on the Virus

Published:December 11, 2020DOI:
      Robertson SW, Ely EW, Wilson JE. Manifestations of critical illness brain injury. In: Vincent JL, ed. Annual Update in Intensive Care and Emergency Medicine 2020. London, UK: Springer Nature; 2020:457-467.
      The Annual Update in Intensive Care and Emergency Medicine for 2020 contains an overview of important patterns of neurologic dysfunction observed with critical illness. Topics range from coma with features initially described in the 1600s to other recent and important work describing the impact of delirium and catatonia in the intensive care unit (ICU) coming from the group at Vanderbilt, which has taken the lead in describing neurologic implications of critical care practice and many of the unforeseen and unfortunate side effects. This column is developed from the Vanderbilt review to build on our understanding of the atypical neurologic examination that may be found in the critically ill patient receiving care from a transport team.


      Wijdicks EFM. The Comatose Patient. 2nd ed. New York, NY: Oxford University Press; 2014.
      McClenathan BM, Thakor NV, Hoesch RE. Pathophysiology of acute coma and disorders of consciousness: considerations for diagnosis and management. Semin Neurol. 2013;33:91-109.
      Iyer VN, Mandrekar JN, Danielson RD, et al. Validity of the FOUR score coma scale in the medical intensive care unit. Mayo Clin Proc. 2009;84:694-701.
      A coma is characterized by closed eyes and unresponsiveness with absence of an appropriate response to even the most vigorous stimulation. Patients in a coma do not localize stimuli or make any defensive movements in response to pain. Extremity movement in a coma is generally limited to posturing with flexion or extension of extremities with painful stimuli. A sudden coma is a neurologic emergency; it indicates acute onset of one or more fatal events by potentially reversible conditions resulting in severe impairment of brain activity.
      A coma is a disorder of profoundly impaired arousal and, possibly as a by-product, awareness. The presence of a coma indicates dysfunction of one or more components of neural networks mediating the intrinsic ability to maintain and regulate the state of alertness. These networks include several brainstem nuclei, most notably the reticular activating system of the pons, the ventral medulla, and their projections to the midbrain, thalamus, hypothalamus, and forebrain. These brainstem structures interact with each other and with the cerebral cortex in a complex interplay of activation and inhibition mediated by neurotransmitters including acetylcholine, gamma-aminobutyric acid, dopamine, and serotonin.
      One of the oldest measures of the level of consciousness in common clinical use today is the Glasgow Coma Scale (GCS). The GCS grades the quality of eye opening (4 points), verbal responses (5 points), and motor responses (6 points) to auditory and tactile stimuli. It was initially validated for prognostication after traumatic brain injury (TBI). For example, a GCS of 8 is a threshold to intubate a patient for airway protection and defines a coma in brain injury. An important limitation of the GCS comes with intubation, which eliminates evaluation of a verbal response. The GCS in an intubated patient is typically followed by a “T,” and the highest possible score decreases from 15 to 11 T, where 1 is the lowest score on any component of the GCS. The Full Outline of UnResponsiveness (FOUR) score is another well-validated scale that is based on eye responses, motor response, brainstem reflexes (including the response of pupils to light, sensitivity of the cornea to touch, and production of cough with airway suctioning), and respiratory patterns. Compared with the GCS, the FOUR score has the advantage of avoiding dependence on verbal responses; thus, it is not confounded by the inability to speak (eg, due to intubation). Another tool with value in the evaluation of altered mentation including a coma is the Richmond Agitation-Sedation Scale (RASS). The RASS is a 10-point scale ranging from −5 (coma) to +4 (severely agitated), with 0 being normal. The RASS differs from the GCS and FOUR score in two respects: first, the RASS focuses on arousal and does not require intact comprehension or command following to obtain a normal score and, second, the RASS includes values indicating hyperarousal or agitation. This scale is validated as a monitoring tool to guide sedative titration and to evaluate agitation in the adult ICU.
      Electroencephalography (EEG) is a useful technology to assess causes of coma. EEG can assess potential etiologies of a coma including subclinical status epilepticus and rule out conditions that mimic a coma such as locked-in syndrome, a condition characterized by global motor impairment with preserved sensory awareness. Electroencephalographic patterns most commonly described in a coma include burst suppression, very low voltage or discontinuous activity, and generalized periodic discharges. In the appropriate clinical setting, the presence or absence of these patterns can aid in the determination of a prognosis in patients with a coma.
      The initial management of the comatose patient involves stabilization. Airway protection, assurance of hemodynamic stability, and respiratory support are potentially lifesaving measures in the minutes to hours after the onset of a coma. Historic data provide important clues to the source of a coma. Physical examination, ideally with little or no sedation, should screen for signs of trauma and focal neurologic deficits. Early laboratory studies should include blood glucose measurement, an electrocardiogram, a complete blood count, a comprehensive metabolic panel, and a brain computed tomography scan. Additional evaluation, such as toxicology screens, cerebrospinal fluid studies, EEG, and/or brain magnetic resonance imaging, should be guided by the clinical history and examination. Treatment in the acute phase of a coma largely consists of addressing the underlying etiology and minimizing the risks of adverse neurologic sequelae (eg, resuscitation and targeted temperature management in the case of cardiac arrest). Subsequent care is guided by the differential diagnosis and risk of complications. In general, it is important to avoid medications that may confound coma assessment such as sedatives or other central nervous system depressant agents.
      Clinical observations driving the understanding and diagnosis of a coma date to the 1600s. Although we commonly associate a coma with hypoxia or an injury, infections and metabolic, oncologic, rheumatologic, and peripartum events are found in a list of 100 classic vignettes associated with comas as listed by Wijdicks.
      In general, coma portends a poor prognosis in the critically ill population. The depth and duration of a TBI-induced coma are predictors of functional status at 3 months postinjury. A lower GCS score at presentation is associated with increased in-hospital mortality in nontrauma patients as well. Clinical signs of a coma after cardiac arrest strongly predict death or a poor neurologic outcome. Mechanically ventilated patients with burst suppression (an EEG-based measure consistent with a coma) are twice as likely to die by 6 months postdischarge as those without burst suppression. The trajectory of level of arousal and interaction with the environment also inform prognosis. Serial assessment is important, as is an exhaustive search for and remediation of reversible contributing factors.


      Hughes CG, Pandharipande PP, Ely EW. Delirium: Acute Brain Dysfunction in the Critically Ill. London, UK: Springer Nature; 2020.
      Devlin JW, Skrobik Y, Gélinas C, et al. Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit Care Med. 2018;46:e825-e873.
      Lindroth H, Khan BA, Carpenter JS, et al. Delirium severity trajectories and outcomes in ICU patients: defining a dynamic symptom phenotype. Ann Am Thorac Soc. 2020;17:1094-1103.
      Delirium is the most common form of acute brain dysfunction in the ICU and is manifested by inattention, disorientation, altered arousal, changes in response to standard stimuli, and at times abnormalities in thought and perception. Approximately 70% of patients on mechanical ventilation will experience delirium during their critical illness, and almost a third of the days in the ICU are days spent with delirium. Delirium is commonly associated with motor abnormalities, including hypoactive (decreased movement and arousal), hyperactive (increased movement and agitation), and mixed (exhibiting features of both motor subtypes that may fluctuate) presentations. The hypoactive form of delirium predominates (perhaps because of prompt treatment of hyperactive patients with sedation); however, it is underdiagnosed and is associated with worse outcomes.
      In the critical care setting, the most widely used instrument to screen for delirium is the Confusion Assessment Method for the ICU (CAM-ICU). CAM-ICU can be conducted in 1 or 2 minutes and is 98% accurate in detecting delirium when compared with formal evaluation by a psychiatrist. First, patients are evaluated for level of consciousness. If the patient responds to verbal commands, evaluation for delirium may proceed. The assessment includes identification of fluctuating mental status or changes from baseline, detection of inattention, and either disorganized thinking or altered level of consciousness. A diagnosis of delirium includes the first and second of the characteristics above with either the third or fourth.
      An important recent study further examined trajectories and outcomes of ICU patients experiencing delirium. Data from over 500 patients included in a randomized controlled trial were re-examined, and patterns of delirium presentation were categorized based on severity and duration of symptoms. Seven days of data from a pharmacologic management study for delirium were evaluated for each patient. Five patterns of delirium were identified using the CAM-ICU scoring system. Mild-brief trajectory patients had mild brief symptoms, whereas severe-rapid recovery patients experienced more dramatic symptoms but improved rapidly, typically within 60 hours. Mild-accelerating delirium begins with mild to moderate symptoms that quickly worsen, demonstrating progress to severe delirium. Severe-slow delirium demonstrates severe symptoms with only a gradual decline over the 7-day observation interval. Severe-nonrecovers delirium included severe symptoms throughout the 7-day observation period. The distribution of different delirium phenotypes varied in this study population. Patients with mild-accelerating delirium made up 7% of the population, whereas the largest group of patients was in the severe-nonrecovers group, representing 40% of the patients evaluated. Thirty-day mortality also varied among the groups of patients with delirium. Mild-accelerating patients experienced 31% mortality over 30 days, whereas patients with mild-brief delirium experienced a 3% mortality. Clearly, this study warrants validation, but the conclusions are supported by careful data collection and innovative statistical analysis to demonstrate different severity trajectories of delirium.
      Distinct etiologic mechanisms leading to the development of delirium remain unknown; however, preliminary evidence suggests regional brainstem dysfunction and compromise of regulatory response pathways. Neuroinflammation leading to upper brainstem dysfunction may specifically contribute to the development of delirium; however, given the variety of presentation patterns, there may be several mechanisms contributing to the development of delirium instead of one distinct causal pathway.
      Antipsychotic medications have historically been the treatment of choice for delirium; however, recent findings suggest that haloperidol (a typical antipsychotic) and ziprasidone (an atypical antipsychotic) have no effect on the duration of delirium in the ICU. Despite these disappointing findings, antipsychotic medications may still have an indication for the treatment of psychotic symptoms (delusions, hallucinations, and agitation) secondary to critical illness delirium, although further research is needed. Delirium is characterized by alterations in the sleep/wake cycle; therefore, melatonin (a hormone released by the pineal gland with a key role in circadian rhythm regulation) or ramelteon (a melatonin receptor stimulant) may have a role in the treatment or prevention of delirium. Nonpharmacologic interventions such as adherence to the ABCDEF (Assess, prevent, and manage pain; Both spontaneous awakening and breathing trials; Choice of analgesia and sedation; Delirium assess, prevent, and manage; Early mobility and exercise; and Family engagement/empowerment) bundle have shown the greatest effectiveness in reducing delirium occurrence in the ICU.
      Delirium is known to be independently predictive of increased in-hospital and postdischarge mortality, new-onset dementia akin to an Alzheimer-type dementia, depression, posttraumatic stress disorder, longer length of stay in the hospital, new institutionalization at discharge, inability to return to work, and increased cost of care, among others.


      Wilson JE, Carlson R, Duggan MC, et al. Delirium and catatonia in critically ill patients: the delirium and catatonia prospective cohort investigation. Crit Care Med. 2017;45:1837-1844.
      Walther S, Stegmayer K, Wilson JE, Heckers S. Structure and neural mechanisms of catatonia. Lancet Psychiatry. 2019;6:610-619.
      Catatonia is a syndrome of psychomotor abnormalities (decreased, increased, and abnormal motor behaviors such as abnormal flexibility or echo phenomenon) and impaired volition (or will). Catatonia has historically been linked with schizophrenia; however, more recently, it has been described in patients with severe mood and general medical disorders.
      Up to 10% of the general medical population may exhibit catatonic signs during acute hospitalization. Studies into the prevalence of catatonia on a psychiatric liaison service showed an incidence of 1.6% to 8.9% across all medical services, varying across age groups and associated medical conditions. Recent literature has suggested atypical stimulation and neuroinflammatory hypotheses for the development of catatonia, such as increased neural activity and downstream autoimmune effects with specific actions on cellular receptors (e.g., as seen in encephalitis) as potential etiologic theories for catatonia.
      Catatonia has increasingly been recognized as a manifestation of acute brain dysfunction in the ICU, although universally it remains underreported, and in critical illness it is generally not a part of routine screening programs. Classic descriptions of catatonia include depictions of mental status abnormalities consistent with delirium, and classic descriptions of delirium include depictions of catatonia; however, modern diagnostic criteria do not recognize the co-occurrence of delirium and catatonia.
      The most widely used assessment tool for catatonia is the Bush-Francis Catatonia Rating Scale, a 23-item rating scale, which relies on observation, physical examination, and interview of the patient (if possible). The Bush-Francis Catatonia Rating Scale score is calculated from totaling individual items (rated 0-3, with higher scores indicating increased severity of that sign). In recent years, some researchers have begun to explore the association between delirium and catatonia, with estimates that up to one third of critically ill patients will screen positive for behavior consistent with catatonia in the ICU. Despite our increased appreciation of signs for catatonia in critical illness, the clinical relevance of the occurrence of catatonia remains unclear.
      Fortunately, for most patients, prompt recognition of catatonia and treatment with anticonvulsant medications (benzodiazepines or barbiturates) or electroconvulsive therapy (ECT) can be lifesaving regardless of the underling etiology; however, there remains general ambiguity as to whether sedating agents can be safely used in a delirious or comatose individual with catatonia because these medications are associated with the development of delirium. Studies in this area are urgently needed. Catatonia is a potentially lethal condition if left untreated. Its presence puts patients at risk for a variety of medical complications including aspiration, dehydration, malnutrition, contractures, pulmonary embolism, and rhabdomyolysis leading to renal failure. Despite effective treatments, patients with catatonia have increased morbidity compared with their counterparts in the ICU.
      Unfortunately, there are little prospective data regarding the effectiveness of focused treatments for catatonia. Benzodiazepines and ECT are safe and effective, particularly in the acute setting. Lorazepam is the preferred benzodiazepine and may be given sublingually, intravenously, or intramuscularly. Lorazepam is most effective in acute catatonia, which responds well to benzodiazepines in at least two thirds of cases. Unfortunately, a significant number of patients show limited symptom change with benzodiazepine therapy. A poor response to benzodiazepines is seen with increasing patient age or duration of illness and in patients with associated psychosis. The treatment of patients who present with signs of catatonia and delirium is a particular challenge. Antipsychotic drugs are often used to treat delirium but can worsen catatonia, and benzodiazepines, which are often used to treat catatonia, can worsen delirium. When patients do not respond to benzodiazepines in the setting of catatonia, they should be referred for ECT.

      Closing Comment and the Virus

      Fontana P, Casini A, Robert-Ebadi Helia, et al. Venous thromboembolism in COVID-19: systematic review of reported risks and current guidelines. Swiss Med Wkly. 2020;150:W20301.
      Spano PJ, Shaikh S, Boneva D, et al. Anticoagulant chemoprophylaxis in patients with traumatic brain injuries: a systematic review. J Trauma Acute Care Surg. 2020;88:454-460.
      The reported risk of venous thromboembolism (VTE) in hospitalized coronavirus disease 2019 (COVID-19) patients is 4% to 8% overall and much higher in ICU patients, exceeding 50%. These high VTE rates were seen despite universal thromboprophylaxis and without systematic screening for VTE in many studies. A number of suggestions for this dramatic finding exist. First, ICU patients have VTE risk factors, including immobility, inflammatory states, hypoxia, and central venous catheters. Second, COVID-19 patients have a particular coagulopathy with elevated fibrinolytic biomarkers but without thrombocytopenia or decreased fibrinogen levels. Third, endothelial changes, which may enhance the production of a clot, may be involved. Clearly, additional data are needed to better appreciate strategies to address this problem. Universal and frequently high-dose thromboprophylaxis may be necessary.
      One additional article does not specifically describe COVID patients but is relevant to VTE risk. In a recent review, Spano et al note that the presence of TBI should not preclude anticoagulant thromboprophylaxis. Data from these writers point out that early anticoagulation is associated with reduced VTE incidence without a corresponding increase or exacerbation of intracranial hemorrhage in patients with TBI having a stable repeat head computed tomographic scan.
      Summary Points
      • Delirium, the most common form of brain dysfunction in the ICU, presents with fluctuating levels of arousal and the inability to maintain attention, whereas coma is characterized by profound impairment of arousal and an absent intentional response to external stimuli. Catatonia, a recently recognized manifestation of critical illness brain injury, has a variable presentation including diverse psychomotor, behavioral, and emotional disturbance that may overlap delirium or coma.
      • The key distinctions to help separate these three conditions are abnormal attention and cognition (delirium), depressed consciousness (coma), and abnormal motor behavior seen on physical examination (catatonia). Pathophysiologic mechanisms likely include shared components on a spectrum of acute brain dysfunction in the ICU.
      • Management approaches differ depending on the clinical presentation of brain dysfunction in the ICU (A-F bundle for delirium, risk factor control for coma, and lorazepam challenge or ECT for catatonia with avoidance of antipsychotics). Thus, awareness and recognition of these syndromes are vital.
      • VTE risk is higher in COVID-19 victims than most other patient groups. Aggressive prophylaxis should be considered, even in the setting of TBI.


      Dr. Haitham Hussein from the Department of Neurology at Regions Hospital provided invaluable comments during the preparation of this column. The authors gratefully acknowledge the assistance of Ms. Sherry Willett in preparation of this series for Air Medical Journal.


      David J. Dries, MSE, MD, is with the department of surgery at HealthPartners Medical Group and a professor of surgery and an adjunct clinical professor of emergency medicine at the University of Minnesota in St Paul, MN, and can be reached at [email protected] .
      James A. Dries, BA, is a student at the University of Minnesota.