Dr. Aneesh B. Singhal: This 60-year-old woman with hypertension, diabetes mellitus, and schizophrenia presented to this hospital with acute psychosis and focal neurologic deficits on the left side during the Covid-19 pandemic. One week earlier, she had begun to have headache and cough, and more recently, vomiting and diarrhea had developed; an initial evaluation confirmed the diagnosis of SARS-CoV-2 infection. Although infection with SARS-CoV-2 provides an explanation for her prodromal symptoms, could this patient’s acute psychosis and weakness on the left side also be associated with Covid-19? Because our understanding of the clinical manifestations of SARS-CoV-2 infection is still evolving, it is important to consider other causes that may be contributing to her presentation.
Could this patient’s abrupt onset of screaming from the fear of dying and an inability to move indicate an episode of brief psychosis? Brief psychosis is usually precipitated by extreme life stress and can be a component of schizophrenia.1 In this patient with schizophrenia, factors such as social isolation, insomnia, a language barrier, fear of the development of Covid-19, and housing insecurity may have triggered psychosis.2 However, brief psychosis associated with schizophrenia typically lasts more than 1 day, so it is unlikely to account for her altered mental status.
Hypoglycemia, which can manifest as psychosis,3 should be considered in this case, given the patient’s development of lethargy and subsequent decrease in symptoms after the administration of dextrose. However, her glucose level was 60 mg per deciliter, and symptoms of hypoglycemia usually develop when the level drops below 45 mg per deciliter (2.5 mmol per liter). Other metabolic or toxic encephalopathies can be ruled out on the basis of her prompt cognitive improvement and negative results of toxicologic screening.
Neuropsychiatric Manifestations of Covid-19
Can this patient’s apparent episode of psychosis be a manifestation of encephalitis associated with SARS-CoV-2 infection, resulting either from direct infection or from an immune, inflammatory, cytokine, or neural receptor–mediated mechanism?8 Coronaviruses primarily affect the lungs but have widespread systemic and cerebral effects. Neuropsychiatric symptoms have been associated with other coronavirus infections; hence, an association with Covid-19 would not be surprising.6,9 There is a long association between viral pandemics and psychiatric symptoms such as acute psychosis, delirium, insomnia, mania, depression, and suicidality. The medical literature from the time of the Russian (1889–1890) and Spanish (1918–1920) influenza pandemics is replete with reports of fearful experiences (e.g., “psychoses of influenza” and “influenza nervosa”).10 Similarly, von Economo’s encephalitis (encephalitis lethargica) (1917–1928)11 was associated with several neuropsychiatric features. Encephalitis remains a possible cause of this patient’s brief psychosis, but I will also consider the diagnosis of stroke.
The neuropsychiatric consequences of stroke, such as depression, are common and well recognized. Acute stroke can unmask latent symptoms of schizophrenia and infrequently can trigger psychosis, mania, delirium, hallucinations, delusions, and other neuropsychiatric syndromes such as Anton’s syndrome, the Capgras syndrome, and the Cotard syndrome. Stroke-induced psychosis has been associated with lesions that affect the prefrontal and occipital cortexes and with disruption of neural networks resulting from subcortical lesions in the thalamus and basal ganglia (often on the right side),12 as well as the midbrain, pons, and cerebellum.
Both encephalitis and stroke are possible causes of the patient’s brief psychosis. Can the patient’s headache and focal neurologic deficits help narrow the differential diagnosis to the most likely diagnosis?
Headache and Focal Neurologic Deficits
In this patient with ongoing headache and focal neurologic deficits, we must consider the diagnosis of community-acquired bacterial meningitis.13 The absence of fever and of signs of meningeal irritation makes bacterial meningitis unlikely. Cerebrospinal fluid (CSF) analysis is required for diagnosis; however, lumbar puncture was not performed in this case, presumably because of the ability to promptly perform imaging of the head.13
Headache and cognitive changes are common in patients with viral encephalitis. Coronavirus infections, including SARS-CoV-2 infection, can cause meningitis, encephalitis, myelitis, and acute necrotizing encephalopathy, particularly in patients with moderate-to-severe illness.6,14
Embolic stroke (especially to the middle and posterior cerebral arteries), lacunar stroke, cerebral arteriopathies, venous sinus thrombosis, and brain hemorrhages are all associated with headache. The 1-week duration of the patient’s preceding headache with associated viral symptoms makes it unlikely that stroke was the cause of her headache. However, stroke cannot be ruled out as a cause of her focal neurologic symptoms. Headache disorders — such as migraine with aura, headache and neurologic deficits with CSF lymphocytosis (also known as the HaNDL syndrome), and familial hemiplegic migraine — are unlikely, given the absence of recurrent migraine headaches.15
Correlation of Clinical and Imaging Findings
Does the head imaging performed in the emergency department help us to determine the diagnosis in this case? MRI of the head revealed an acute brain lesion, effectively ruling out conditions that are often invoked in patients who previously had a stroke or psychiatric disease and now have new focal deficits (e.g., poststroke recrudescence [the reemergence of previous stroke-related deficits]16 and functional neurologic disorders17).
Brain lesions with restricted diffusion typically indicate stroke, but given this patient’s presentation, I will consider a broader differential diagnosis.18 Encephalitic and demyelinating lesions can have restricted diffusion, and demyelination has been reported in association with Covid-19–related encephalitis.6,14,19 However, demyelinating lesions are usually patchy and associated with leptomeningeal enhancement and moderate-to-severe respiratory illness, characteristics that are not consistent with those seen in this patient. Hypoglycemia can result in the development of lesions of diverse shapes, sizes, and locations that have restricted diffusion and may simulate lacunar infarction. However, lesions associated with hypoglycemia usually result in cerebral swelling, are reversible, and are most often associated with severe, prolonged hypoglycemia, which was not present in this patient.20 Lesions with restricted diffusion can occur with migraine disorders21; however, this patient’s headache shows no features of migraine,15 and lesions associated with migraine typically affect the posterior circulation and cross arterial territories. Seizure-induced lesions with restricted diffusion are unlikely in this patient, given that they usually develop after prolonged seizures or status epilepticus and typically appear in the cortical, hippocampal, or pulvinar region.22 Overall, the size and location of this patient’s lesion with restricted diffusion, combined with the abrupt onset of hemiparesis, strongly support a diagnosis of acute ischemic stroke.
This patient had pure motor hemiparesis, which is a classic lacunar stroke syndrome.23 The punctate size of the lesion and its location in the contralateral posterior limb of the internal capsule in this patient with chronic hypertension suggest small-vessel disease (lipohyalinosis) as the mechanism of the stroke.
Other mechanisms are also possible. This patient was taking risperidone, which has been associated with stroke.24 Primary angiitis of the central nervous system (CNS) and a reversible cerebral vasoconstriction syndrome are often invoked in patients with headaches and stroke; however, the clinical and imaging features of these entities were not observed in this patient.25 Finally, cardiac and artery-to-artery embolism cannot be firmly ruled out without appropriate testing, although both seem unlikely on the basis of the clinical and imaging characteristics favoring the presence of lipohyalinotic small-vessel disease.
Stroke and Covid-19
Coagulopathy and inflammation (black box) are central to thrombosis within cerebral blood vessels and cardiac embolism, the two major mechanisms of ischemic stroke in SARS-CoV-2 infection. Other possible mechanisms of stroke and sources of cardiac embolism are listed in the lighter boxes. Patients with SARS-CoV-2 infection may have a higher risk of stroke because of coexisting factors such as advanced age, cardiovascular disease, nonadherence to medications, and cerebral microvascular dysfunction. ACE2 denotes angiotensin-converting enzyme 2, and RAAS renin–angiotensin–aldosterone system.
Is there a relationship between this patient’s stroke and Covid-19? Upper respiratory infections are known to trigger stroke; a case–crossover study showed an odds ratio of 2.88 (95% confidence interval, 1.86 to 4.47) for stroke occurring within 15 days after a hospital visit for an influenza-like illness, as compared with the same calendar period 1 or 2 years before the stroke.26 Emerging data suggest a much higher risk of stroke with Covid-19 than with influenza.27 Multiple studies have reported on the frequency and characteristics of stroke in patients with Covid-19.5,6,28 Ischemic stroke appears to affect men and older adults with cardiovascular risk factors most frequently. Potential mechanisms of stroke associated with Covid-19 include coagulopathy, cardiac embolism due to myocardial injury, and nonadherence to medication, all of which may be common during this pandemic (Figure 2). However, most patients with Covid-19 who had a stroke also had concurrent advanced systemic illness. This patient had mild viral illness at the time of stroke, and the d-dimer level at the time of admission was not markedly abnormal. It is conceivable that the small-vessel stroke resulted from viral endotheliitis or systemic inflammation, since her subacute headache and transient psychosis could be consistent with viral encephalitis. SARS-CoV-2 may invade the endothelium29; in addition, it is associated with vasculitis,30,31 and pathological evidence indicates that SARS-CoV-2 RNA may be present within brain tissue.32,33
In conclusion, this patient’s brief psychosis was probably related to direct and indirect effects of SARS-CoV-2 infection, including the socioeconomic effect of Covid-19 in combination with underlying schizophrenia, and to the new infarct in the right hemisphere. The small-vessel stroke resulting from lipohyalinosis was probably triggered by Covid-19, although direct viral mechanisms cannot be ruled out.