COVID-19-induced new-onset psychosis: a case report.

Introduction

Novel coronavirus (COVID-19 or SARS-CoV- 2) can cause neurological symptoms including loss of smell and/or taste, confusion, inability to concentrate, and short-term memory loss (Montalvan et al., 2020). Previous studies have also shown patients infected with SARS-CoV-2 can present with new-onset psychiatric illnesses including psychosis (Javed and Shad, 2021). A retrospective cohort study published in April of 2021 examined the incidence of neurological or psychiatric diagnosis in 236,379 patients following their COVID-19 diagnosis and found that 1.4% of them developed a psychotic disorder within 6 months (Maxime et al., 2021). The 12-month prevalence of psychotic disorders in the general population is 4.03 in 1000, which is significantly lower than the prevalence of psychotic disorders after SARS-CoV- 2 infection (Moreno-Küstner et al., 2018). To this date, there are at least 40 case reports including 48 patients who presented with psychotic symptoms associated with SARS-CoV- 2 but many of these cases lacked clinically relevant details and delirium was frequently not excluded (Smith et al., 2021). In this case report, we describe a patient who developed new-onset psychosis days after recovering from COVID pneumonia. This patient required psychiatric hospitalization but responded rapidly to antipsychotic treatment.

Case

Ms. S, a 32-year-old African American female, self-presented to the ED in August 2021 for anxiety and paranoia. The patient is a well-educated individual with no past psychiatric history and completed a master's degree and held a job successfully as a data analyst. Seventeen days before this ED encounter, the patient was hospitalized for acute hypoxic respiratory failure secondary to coronavirus infection confirmed by PCR testing. She was successfully treated with oxygen and intramuscular dexamethasone 16 mg in total, a steroid commonly used in coronavirus infection. The patient was discharged after 7 days of hospitalization with oral dexamethasone 6 mg/day.

Two days after she returned home, she started to have severe anxiety. She finished 6 days of dexamethasone and her anxiety worsened with paranoia believing a spiritual group bribed her mom to poison her with dexamethasone. She could not sleep and felt extremely anxious with a racing heartbeat. On exam, she was anxious and paranoid yet cooperative. She had pressured but coherent speech. She became tearful and angry when talking about her mom and said “I don't understand why she did this to me.” She gave a very detailed theory of how her mom and friend poisoned the water in her apartment. She said her friend “used the bathroom in my apartment and when she left, all the pill bottles are open and the poison got into the shower vapor. ” She had a fair insight about her mental illness by acknowledging she “couldn't think right” and actively sought help.

Her vital signs were stable. Physical exam including neurological exam was unremarkable. She was alert and oriented to person, time, place, and situation. She was cooperative and had good eye contact. Her speech was pressured with anxious affect. Thought process was racing. She denied suicidal or homicidal ideation but voiced paranoid delusions of being poisoned. She denied hallucinations and was not observed to be responding to internal stimuli. Her labs showed an increased white blood cell count of 13.8 × 103/µL (normal < 10.5 × 103/µL) most likely resulting from steroid use. Comprehensive metabolic panel was normal, head CT without contrast was unremarkable, thyroid stimulating hormone (TSH) and lipid panel were normal, and rapid plasma reagin (RPR) and HIV testing in March 2021 were negative. C-reactive protein (CRP) 9.3 (normal < 0.5 mg/dl) was elevated during last hospitalization seventeen days ago. Urine drug screen and blood alcohol level were negative.

The patient received Haldol injection of 2.5 mg twice and Ativan 2 mg in the ED after which the patient became calmer. On the psychiatric floor, the patient received risperidone 2 mg at bedtime and clonazepam 0.5 mg at bedtime. She became less anxious and paranoid rapidly. She realized she was delusional and her mom did not poison her on 3rd day of treatment. Upon discharge on day 8, she started to question her belief that her friend poisoned her and returned to baseline except for the mild loss of smell and taste, and mild difficulty with concentration. She was discharged home with Risperidone 2 mg at bedtime. Before her follow-up appointment three months after the hospitalization, her sense of smell and taste returned and she has returned to baseline completely. Risperidone was discontinued based on her rapid and steady recovery. At the 6-month follow-up visit, she continued to remain free from psychosis or anxiety. She moved afterward and stopped following up.

Discussion

This case, with new onset of psychosis after Covid infection, presented similarly to three cases reported by Ferrando in May 2020 and one other case reported by Chacko in September 2020. In Ferrando's report, three patients with new-onset psychosis tested positive for COVID although they did not have any respiratory manifestations (Ferrando et al., 2020). These three patients all presented with persecutory delusions and paranoia. In Chacko's study, a 52-year-old man with no past psychiatric history presented with mutism and anxiety and later tested positive for COVID antibodies (Chacko et al., 2020). To be noted, this case occurred before vaccinations against COVID-19 became available and antibody test was used to indicate the previous infection. In a review paper on COVID-19-associated psychosis, many of the 48 patients who were included had persecutory delusions, paranoia, and anxiety as the major symptoms (Smith et al., 2021). Like our patient, most of them were alert and oriented during the initial presentation. Most patients responded well to anti-psychotics except for the patient in the Chacko study who required electroconvulsive therapy (ECT) due to severe agitation and mutism. Although fear of COVID-19 has been shown to trigger brief psychotic disorders in some people (Brown et al., 2020), none of Ferrando and Chacko's patients expressed overly concern for COVID and our patient has already recovered from COVID pneumonia when her psychiatric symptoms started to develop.

Collateral from the patient's mother confirmed there is no past psychiatric history and no family history of psychiatric illness. New-onset schizophrenia was considered a differential diagnosis. However, the patient's rapid onset of symptoms after COVID diagnosis and rapid recovery with treatment, lack of family history of psychiatric illness, and lack of disorganized behavior or speech, all make psychosis related to COVID a more likely diagnosis than schizophrenia. Treatment with dexamethasone may also have contributed to psychosis in our patient as systemic corticosteroid use is associated with insomnia, mood swings, depression, and psychosis (Cheng et al., 2004). The risk of developing psychiatric adverse effects is dependent on the dosage: The Boston Collaborative Drug Surveillance Program reported ‘psychiatric reactions’ in 1.3% of 463 patients treated with 40 mg/day or less of prednisone, 4.6% of 175 patients dosed with 41–80 mg/day, and 18.4% of 38 patients receiving doses higher than 80 mg/day (Kenna et al., 2011). Symptoms usually present early in the treatment cycle and typically resolve with dosage reduction or discontinuation. Hypomania and mania are the most common presentations of steroid-induced psychiatric illnesses (Lewis and Smith, 1983) while our patient presented with paranoid delusions. Our patient received steroid injections of 10 mg on the first day, and 6 mg on the second day of her hospital stay with no adverse effects. She was discharged with only 6 mg dexamethasone and developed psychosis two days after discharge. She took 28 mg dexamethasone in total before her symptoms developed. The low dosage of dexamethasone makes steroid use less likely to be the cause of the patient's psychiatric illness, but steroid use could have an additive effect on the patient's symptoms.

Though rare, post-infectious psychosis and mania cases have been found in 1918 flu and SARS and MERS (Ferrando et al., 2020). The exact mechanism of neuropsychiatric illness of these viruses including COVID is unknown but several mechanisms have been proposed. Direct viral central nervous system (CNS) invasion is a possible mechanism. Coronavirus can transport from the olfactory bulb to the brain via retrograde axonal transport (Desforges et al., 2019). Another mechanism proposed involves an inflammatory response to the virus. Cytokine storm, a profound and diffuse inflammatory response to COVID-19 infection, is responsible for the severe pulmonary and cardiac complications of the disease and can also be the cause of neuropsychiatric manifestations (Troyer et al., 2020). Activation of many pro-inflammatory cytokines including interleukin (IL) 1, IL-6, TNF-alpha, IL-8, IL-10, and IL-2R in COVID infection can cause neuroinflammatory effects in the brain and subsequent neuropsychiatric symptoms (Ferrando et al., 2020). Four patients in the Ferrando and Chacko studies, along with our patient, have elevated CRP levels which indicates a current inflammatory response. One patient reported by Panarielloa also had increased IL-6 (39 pg/mL) level (Panariello et al., 2020) but data on interleukin level is limited since it is not part of the routine lab work. Measurement of pro-inflammatory markers including CRP and acute phase interleukins can become part of the diagnostics but they are nonspecific markers that can be elevated in many inflammatory diseases. Therefore, COVID-induced psychosis remains a diagnosis made by clinical judgment and exclusion of other diagnoses.

It is crucial to rule out diagnoses including delirium which is common in symptomatic COVID infection requiring hospitalization and autoimmune encephalopathy which is rare but deadly without proper treatment. Delirium was excluded in our case based on the patient's intact orientation and attention, coherent thought process, normal sleep/wake cycle, and progressive and constant nature of her symptoms (while delirium is typically “waxing and weaning”). In a case reported by McAlpine in April 2021, a 30-year-old patient without past medical or psychiatric history developed delusions, paranoia, and hypersomnia followed by insomnia 20 days after a positive nasal swab for COVID-19 (McAlpine et al., 2021). He was treated with haloperidol 5 mg by mouth twice daily with significant improvement of his agitation and delusions. But after discharge, his symptoms recurred and lumbar puncture showed an elevated IgG of 4.8 mg/dL (reference 1.0–3.0 mg/dL). He had a robust response to IVIg of a total 2 g/kg over 3 days and returned to work immediately after discharge. This case presents similarly to COVID-induced psychosis cases in many aspects, including the clinical presentation, the temporal association with COVID, and the initial quick response to antipsychotics. Resistance to antipsychotics and recurrence of symptoms while taking anti-psychotics should prompt clinicians to consider autoimmune encephalopathy as a diagnosis and the threshold of obtaining a lumbar puncture in such cases should be low.

Conclusion

SARS-CoV-2 can affect multiple systems including the neuropsychiatric system. The similarity of the above cases in terms of symptoms and progression warrants further investigation of the neuropsychiatric effect of coronavirus. Diagnosis is based on clinical judgment and early recognition of symptoms is vital. Treatment with anti-psychotics shows rapid improvement of symptoms and delay of treatment could result in high mortality.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.