Altered mental status is a common chief complaint of patients who present to the emergency department and a frequent reason for neurological consultation.¹ The most common discharge etiologies accounting for altered mental status are neurologic (28%) and toxicologic (21%), followed by trauma (14%), psychiatric (14%), infectious (10%), endocrine/metabolic (5%), pulmonary (3%), oncologic (3%), cardiovascular (1%), gastrointestinal (1%), and renal (1%).¹ The extreme form of altered mental status is coma, characterized by the absence of arousal and awareness of self and environment, lasting for more than 1 hour.² In such patients, timely medical and surgical interventions can be life-saving, and the initial evaluation of such patients is critical for establishing a correct diagnosis.² Thus, the Emergency Neurological Life Support Guidelines suggest an algorithm for the initial assessment and treatment of patients with coma based on general physical examinations, airway stabilization, breathing, circulation, cervical spine precautions, neurological assessments, recommended stat labs, and brain imaging.²

According to these guidelines, brain computed tomography (CT) and CT angiography are recommended as the primary imaging modality. Brain imaging helps clinicians identify the structural etiologies of coma including cerebral infarction, intracranial hemorrhage, brain masses, cerebral edema, and acute hydrocephalus causing increased intracranial pressure (ICP). The acquired information regarding vascular patency is useful for administering reperfusion therapies for acute stroke or predicting vasospasm in patients with subarachnoid hemorrhage (SAH). Bedside neurosonology can aid in the evaluation and management of such patients. In certain situations, it may even be preferred over CTbased imaging. For example, in primary care or prehospital settings, neurosonology is much more accessible than is brain CT. In addition, ultrasound is more practical for situations in which information regarding serial temporal changes of intracranial parameters are required.

Point-of-care cardiothoracic ultrasound is another technique often employed for neurocritical patients. Neurocritical patients frequently exhibit other medical problems. Hence, the timely evaluation of cardiac comorbidities^3,4 and airway or pulmonary problems⁵ is essential. Furthermore, evaluations of a patient’s hemodynamic status is helpful in situations where cerebral perfusion pressure optimization is needed.⁶ Point-of-care ultrasound can assists in the diagnosis and monitoring of such patients.

This review describes the various ultrasound methods that are used to evaluate patients with altered mental status or the neurocritically ill, with a focus on the detection of large vessel occlusion (LVO) in ischemic stroke, monitoring of vasospasm after SAH, noninvasive identification of ICP, and practical use of point-of-care cardiothoracic ultrasound for neurocritical patients.

The detection of intracranial LVO by transcranial Doppler ultrasound (TCD) focuses on the Thrombolysis In Brain Ischemia (TIBI) grade,⁷ which consists of six grades (Fig. 1).⁸ Key findings to be considered are a >30% decrease in mean flow velocity, flattening of systolic acceleration, absence of diastolic flow, or no detectable flow. In this grading system, TIBI waveforms of grades 0 (no flow), 1 (minimal), 2 (blunted), and 3 (dampened) are considered representative of occlusions.⁷ The TIBI grade can also be affected by the insonated arterial section and its distance from the arterial occlusion.⁹ Secondary findings, including an increase in flow velocity of the adjacent arterial vasculature, the development of prominent collaterals, or flow reversal distal to the occlusive segment, may support identification.^10,11 Using both primary and secondary findings, a TCD study including 190 patients identified arterial occlusion with an overall sensitivity of 83.0% and specificity of 94.4% compared to digital subtraction angiography or magnetic resonance angiography. The findings were higher for middle cerebral artery (MCA) (sensitivity, 93%; specificity, 98%) and distal internal carotid artery (ICA) (sensitivity, 81%; specificity, 96%) occlusions, and lower for basilar artery occlusions (sensitivity, 60%; specificity, 96%).¹² The method of urgent combined carotid/transcranial ultrasound testing for detection of LVO patients with acute cerebral ischemia has also been developed. Simultaneous evaluation by a stroke neurologist and experienced neurosonographer enables focused TCD and carotid/vertebral duplex scans to be performed within 10 minutes. In a study of 150 consecutive patients, this fast-track neurovascular ultrasound examination detected lesions amenable for interventional treatment, including occlusion, near occlusion, stenosis >50%, or thrombi, in arteries with 100% sensitivity and 100% specificity.¹³ Detailed methods for fast-track neurovascular ultrasound are provided in Table 1. The power motion-mode Doppler (PMD) window can also aid in diagnosing arterial occlusions, in which a high resistance signature with absent diastolic flow suggests arterial occlusion (Fig. 2). Using the previously mentioned TCD criteria and PMD criteria, occlusion was detected with 100% sensitivity and 94.5% specificity in the anterior circulation, while the sensitivity was lower in the posterior circulation (sensitivity, 93%; specificity, 98%).¹⁴

Another advantage of TCD is that it can be hand-carried to be performed at bedside. The success of endovascular treatment trials has increased the demand for identifying LVO patients.^15-19 Earlier identification in prehospital settings will aid in the triage of patients needing comprehensive stroke centers and increase the number of patients that would benefit from endovascular therapy.²⁰ The Regensburg stroke mobile project focuses on such issues, and the authors hypothesized that a neurologist equipped with a portable ultrasound device would be able to achieve diagnostic accuracy “in the field” that is similar to that achieved with in-hospital advanced neuroimaging (CT angiography and magnetic resonance angiography). Transcranial color-coded sonography (TCCS) assessments have been performed in prehospital “911” emergency stroke situations, wherein portable duplex ultrasound devices were used to visualize the bilateral MCAs. In 113 eligible patients, TCCS had a sensitivity of 90% and specificity of 98% for detecting M1 occlusions.²¹ However, 10 occlusions were identified from among the 102 performed TCCS procedures, which is a low yield considering the clinical effort. Furthermore, up to 20% of patients have poor temporal windows, which is another limitation.²² Utilizing prehospital stroke scales to predict LVO, such as the Field Assessment Stroke Triage for Emergency Destination (FAST-ED).²³ or Stroke vision, aphasia, neglect (VAN).²⁴ assessment, and preferentially performing TCD for patients in whom LVO is highly suspected may be able to overcome such limitations.

Neurosonologic examinations may also play a role in monitoring patients after recanalization procedures. Indeed, TCD can be used as both a surrogate marker for reperfusion and a method for detecting embolism, thrombosis, hypoperfusion, and hyperperfusion.²⁵ Moreover, carotid duplex tests can aid in evaluating the status of the carotid artery²⁶ or extracranial vertebral artery²⁷ after stenting. Due to alterations of blood flow and vessel compliance, modified criteria for the systolic velocities or ICA/common carotid artery ratio for instent restenosis are used for the ICA.²⁶

Cerebral vasospasm is a leading cause of morbidity and mortality after aneurysmal SAH. Up to 70% of patients experience angiographic vasospasm, which typically occurs between 5 and 14 days after SAH onset. Symptomatic vasospasm occurs in ~30% of patients. Delayed ischemic neurologic deficits occur in ~50% of patients with angiographic vasospasm.²⁸ Current American Society of Neuroimaging guidelines highlight the utility of TCD for diagnosing cerebral vasospasm in both the anterior and posterior circulation following SAH.²⁵ TCD can be employed to detect vasospasm development before it becomes clinically apparent (days 2–5) and identify progression to the severe spasm phase, when the development of delayed ischemic deficits is the greatest (days 5–12). Thus, TCD findings may assist in the planning of appropriate therapeutic interventions.²⁹ However, there is some debate over the clinical utility of TCD for vasospasm monitoring. A recent meta-analysis by Mastantuono et al.³⁰ revealed conservative results, demonstrating a pooled sensitivity of 66.7% (95% confidence interval [CI], 55.9–75.9) and specificity of 89.5% (95% CI, 80.3–94.7) for TCD in the detection of vasospasm in the MCA. TCCS was not convincingly shown to be more accurate than TCD. The authors concluded that both TCD and TCCS are capable of detecting vasospasm in the MCA, but neither is useful for excluding the possibility of vasospasm. Further, for arteries other than MCA, the usefulness of TCD remains unclear.³⁰

As the intracranial compartment is covered by thick skull, gold standard methods of measuring ICP are inherently invasive in nature. Invasive ICP monitoring is especially important after severe traumatic brain injury.³¹ However, due to complications including infection or hemorrhage,^32,33 methods of reliably estimating the ICP noninvasively are being developed. Among them, TCD-based detection of cerebral blood flow (CBF) changes and ultrasound-based detection of optic nerve sheath diameter (ONSD) enlargements are readily available, safe, and reliable methods that can be performed at the bedside.³⁴

Point-of-care ultrasonography at the bedside may involve the use of a series of focused ultrasonographic examinations to efficiently diagnose or exclude certain conditions in patients presenting with particular symptoms, including hypotension, chest pain, or dyspnea.⁶⁴ Alternatively, it may aid the physician in diagnosing and monitoring critically ill patients. Point-of-care ultrasonography is especially valuable in patients requiring neurocritical care, due to the occurrence of frequent systemic comorbidities in these patients. Ultrasound can be used at the bedside in a variety of situations including for the 1) evaluation of cardiothoracic comorbidities such as atrial septal defect and patent foramen ovale, global/segmental ventricular dysfunction, pericardial effusion/cardiac tamponade, and sources of emboli upon admission to the neurocritical care unit;⁶⁵ 2) differential diagnosis of acutely developed symptoms including hypotension, chest pain, and dyspnea;^66,67 3) serial follow-up of pulmonary or cardiologic dysfunctions such as improvement or aggravation of pulmonary edema or pneumonia; and 4) evaluation and management of volume status in situations where cerebral perfusion pressure optimization is needed, including hypertensive therapy for ischemic stroke⁶⁸ or preventing delayed cerebral ischemia after SAH.⁶⁹

For these patients, lung ultrasound, echocardiography, and vascular ultrasound (inferior vena cava [IVC], aorta, and deep veins of the lower extremity) can be sequentially performed.^67,70 While there is no universally accepted sequence, scanning of the lung and IVC with the abdominal aorta, followed by heart and abdominal evaluations when required, is practical [⁷¹]. Sonography of the lower-extremity deep veins is performed in clinical situations when needed. Throughout this process, evidence of the “8 Es,” a mnemonic for the eight potential goals to identify, is sought.⁵⁰ The eight Es are empty thorax (pneumothorax), edematous lung, extended focused assessment with sonography for trauma (E-FAST), effusion (pericardial), equality (right ventricular [RV] dysfunction), ejection fraction (left ventricular [LV] dysfunction), exit (aorta), entrance (IVC), and endocardial movements.

Point-of-care ultrasonography can be performed practically using two probes (Fig. 5). The first step uses a curvilinear probe and starts with lung ultrasound. For rapid assessment, lung ultrasound ⁷² can be performed at two anterior lung points and one posterolateral alveolar and/or pleural syndrome (PLAPS) point for each lung. Five sonographic lung patterns can be distinguished: normal lung pattern, pneumothorax, interstitial syndrome, alveolar consolidation, and pleural effusion.⁷¹ A normal lung pattern (Fig. 6A) is defined as A-lines (an artifact formed by repetition of the pleural line) with lung sliding (dynamic movement of the pleural line, which moves synchronously with respiration) in the anterior lung, without alveolar consolidation or pleural effusion at the PLAPS point. A normal lung pattern may be truly normal or observed in airway disease.⁷³ or pulmonary embolism⁷⁴ in cases of dyspnea. Diffuse interstitial syndrome pattern is defined as multiple bilateral B-lines (Fig. 6D) with at least two positive scans on each side of the thorax.⁷⁵ Abnormal B-lines are classified as ≥3 lines per intercostal space, while ≤2 lines is considered normal.⁷¹ Diffuse interstitial syndrome may indicate pulmonary edema, pulmonary fibrosis, or interstitial pneumonia.⁷⁶ Combined echocardiographic evidence of LV dysfunction with bilateral B-lines may be indicative of cardiogenic pulmonary edema. A sonographic diagnosis of pneumothorax comprises three steps: 1) absence of lung sliding (presence of lung sliding is observed as a seashore sign [a stratified pattern above the pleural line, and sandy appearance below the pleural line in M-mode], while the absence of lung sliding is observed as a stratosphere sign [homogeneous stratified pattern in M-mode]; Fig. 6B;⁷⁷⁾ 2) absence of B-lines and presence of A-lines; and 3) confirmation of lung points⁷⁸ (alternating seashore and stratosphere signs). Pleural effusion or alveolar interstitial patterns can be identified at the PLAPS point (Fig. 6C). Consolidated lung regions are visualized as tissue-like patterns.

Next, vascular ultrasound of the IVC and aorta is performed. The IVC comprises the entrance to the heart, and sonographic IVC assessments can be used to predict preload or volume responsiveness. Respiratory variation can be evaluated by the caval index ([(IVC expiratory diameter – inspiratory diameter) / IVC expiratory diameter] × 100) (Fig. 7A). The predictive power of the caval index is consistent in intubated patients. In septic shock patients, respiratory variation in the IVC diameter of >12% predicted increases in cardiac output after volume loading with positive and negative predictive values of 93% and 92%, respectively.⁷⁹ In another study, a caval index threshold of 18% discriminated fluid responders and non-responders with 90% sensitivity and 90% specificity.⁸⁰ However, its predictive ability is less powerful for patients with spontaneous breathing. Initial reports showed that a caval index value >50% suggested a right atrial pressure of <10 mmHg.⁸¹ In contrast, a more recent study revealed that bedside sonographic measurement of the caval index during spontaneous breathing did not predict fluid responsiveness in a heterogeneous emergency department population.⁸² Currently, the optimal approach for measuring the caval index in patients with spontaneous breathing is to instruct them to “sniff,” resulting in brief, rapid expiration.⁸³ Practically, the combined use of the maximal IVC diameter and collapsibility is recommended by the European Association of Echocardiography.⁸⁴ An IVC diameter ≤2.1 cm that collapses >50% with a sniff suggests a normal right atrial pressure of 3 mmHg (range, 0–5 mmHg), whereas an IVC diameter >2.1 cm that collapses <50% with a sniff suggests a high right atrial pressure of 15 mmHg (range, 10–20 mmHg). For indeterminate cases in which the IVC diameter and collapse do not fit this paradigm, an intermediate value of 8 mmHg (range, 5–10 mmHg) may be used. Preferably, secondary indices of elevated right atrial pressure should be integrated (Table 2).⁸⁴ Insonation of the abdominal aorta can be performed to evaluate the exits of the heart and to reveal evidence of aortic dissection or aneurysm.

Using the phased array probe, echocardiography can be sequentially performed. Potential targets to search for in neurocritical care patients are pericardial effusion with or without tamponade physiology (diastolic right ventricle collapse), pulmonary embolism causing RV dysfunction (RV enlargement and paradoxical interventricular septal movement), LV systolic dysfunction or acute myocardial infarction (LV and RV regional wall motion abnormalities) and mechanical complications (papillary muscular rupture and LV wall rupture), and signs of hypovolemic or distributive shock (hypokinetic small-sized left ventricle or hyperkinetic normal-sized left ventricle).⁶⁷ The parasternal long axis view can be initially insonated. The global ejection fraction is easily evaluated by eyeballing⁸⁵ in this view, along with regional wall motion. RV dilatation, paradoxical septal motion, pericardial effusion, and the ascending aorta can also be evaluated. The probe is then turned 90º to the patient’s left shoulder to insonate the parasternal short axis view. In this view, regional wall motion can be closely evaluated. Clues for problems in equality such as RV dilatation or paradoxical septal motion can also be observed in this view. Diseases causing increases in RV pressure, including pulmonary embolism and pulmonary hypertension, would produce RV enlargement and LV compression, resulting in a D-shaped LV.⁸⁶ In contrast, cardiac tamponade would yield paradoxical diastolic RV collapse (Fig. 7B).⁸⁷ The apical four-chamber view can subsequently be performed. Overall LV function can be evaluated in this view, along with comparisons of LV to RV size and function or wall motion abnormalities. For patients in whom apical views cannot be obtained due to intubation or postural difficulties, a subcostal four-chamber view can be used. A suprasternal notch view can be employed for evaluating the cardiac exit, especially in patients with chest pain, but assessments in this view do not need to be routinely performed in neurocritical care patients.

Ultrasound is likewise used to diagnose lower-extremity deep vein thrombosis in evaluation of patients with dyspnea who have a normal lung pattern,⁷³ or as a screening procedure in the critical care unit. A protocol utilizing a simplified compression technique at the common femoral vein and popliteal vein has been developed, because in previous studies, almost all cases of deep vein thrombosis involved either the popliteal vein or the common femoral vein.⁸⁸ A linear probe is generally used, and the femoral vein is insonated from just distal to the femoral triangle to the point at which the common femoral artery bifurcates. The popliteal vein can be visualized in the popliteal fossa, wherein the vein is usually located more superficially to the popliteal artery. If the vein collapses to a thin line upon external pressure, it is presumed to be patent, whereas if the vein does not collapse, it is presumed that the vessel lumen contains a clot (Fig. 8).

Point-of-care ultrasound, including TCD, carotid duplex studies, and orbital and cardiothoracic ultrasound, is an effective tool that can help neurologists diagnose and manage patients presenting with altered mental status, as well as patients in the neurocritical care setting.