Acute Hemorrhagic Necrotizing Encephalopathy in a COVID-19 Patient With a Response to Intravenous Immunoglobulin Therapy

Introduction

Acute hemorrhagic necrotizing encephalopathy (AHNE) is an uncommon, rapidly progressive neurological condition, typically associated with viral infections such as influenza, parainfluenza, varicella, human herpesvirus, and enterovirus. It predominantly affects children, but adult cases, while rare, have been documented. The exact pathophysiology remains incompletely understood, but current evidence suggests that AHNE arises from a parainfectious, immune-mediated “cytokine storm” rather than from direct viral invasion. This inflammatory response disrupts the blood–brain barrier, resulting in vasogenic edema, vascular leakage, and hemorrhagic necrosis, particularly affecting deep brain structures such as the bilateral thalami, brainstem, and cerebellum. Importantly, AHNE is not classified as an inflammatory encephalitis, and cerebrospinal fluid (CSF) typically shows elevated protein without pleocytosis, which is a distinguishing diagnostic feature [1–3].

Genetic susceptibility may play a role in the development of AHNE. Mutations in the RAN binding protein 2 (RANBP2) gene, thermolabile polymorphisms in carnitine palmitoyltransferase II (CPT II), and certain Human Leukocyte Antigen (HLA) genotypes have been implicated in increasing individual vulnerability to this condition [1–3].

Clinically, AHNE presents with a rapid neurological decline typically following a febrile systemic illness. Encephalopathy is the hallmark presentation and can be accompanied by fever, seizures, acute neurological deterioration, vomiting, hallucinations, ataxia, or coma [1,3,4].

The prognosis is generally poor. Mortality is estimated at approximately 30%, and fewer than 10% of patients achieve complete recovery. Among survivors, lasting neurological sequelae such as cognitive impairment, spasticity, and motor deficits are common. In some cases, progressive neurological deterioration continues beyond the acute phase. Adults tend to have more extensive thalamic involvement and worse outcomes compared to children [5,6].

We present a case of a 67-year-old Bahraini male who developed AHNE in the context of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection. The patient responded positively to intravenous immunoglobulin (IVIG) therapy and achieved significant recovery despite brainstem involvement, a factor typically linked to a poor prognosis. This case adds to the growing body of literature on SARS-CoV-2-associated AHNE in adults and stresses the importance of considering AHNE in the differential diagnosis of adults presenting with unexplained febrile acute encephalopathy, especially in the setting of recent or concurrent viral infections like SARS-CoV-2. Furthermore, our report shows the potential neuroinvasive properties of SARS-CoV-2 in patients with minimal respiratory symptoms, the therapeutic role of immunotherapy in improving outcomes, and the urgent need to develop adult-specific prognostic predictors.

Case Report

A 67-year-old Bahraini male known to have type 2 diabetes mellitus (on vildagliptin/metformin 50mg/1000 mg and metformin 1 g once daily), hypertension (on valsartan/hydrochlorothiazide 80mg/12.5 mg once daily), and extended left hemi-colectomy in 2019 for high-grade focal dysplastic polyps (no adjuvant therapy required), vaccinated with 2 doses of Sinopharm, was brought to the Accident and Emergency Department by ambulance with progressive impairment of consciousness.

At 6: 00 AM on the day of presentation, he was found by his wife to be acutely confused, disoriented to time, place, and person, unable to recognize her, and speaking incoherently. This abrupt neurological decline was preceded by a 2-week history of intermittent, tension-like headaches and a 1-week history of fever, which became persistent 3 days prior to admission. Confusion and altered level of consciousness developed approximately 24 hours before arrival to the emergency department. There was no history of associated visual disturbances, nausea, vomiting, limb weakness, dysphagia, abnormal movements, gait instability, or urinary or fecal incontinence. There was no reported history of recent weight loss, gastrointestinal symptoms, sick contacts, head trauma, substance misuse, or carbon monoxide exposure. His family history was negative for hereditary neurodegenerative disorders.

On initial assessment, he was hemodynamically stable (blood pressure 145/74, heart rate 98 beats per minute, oxygen saturation 96% on room air), febrile (38.9°C), and had a random blood glucose level of 9.8 mmol/L. Later on the same day, he required 5 liters of oxygen via face mask (fraction of inspired oxygen [FiO2] 40%). On examination, he was conscious but disoriented, producing incomprehensible speech spontaneously, opening eyes to painful stimuli, localizing to pain, and unable to follow simple commands (Glasgow coma scale [GCS] 9). The result of a cranial nerve examination was normal, with no restriction in eye movements, and optic disc assessment showed no papilledema. There were no signs of meningism. Motor examination revealed 4-limb weakness grade 2–3 on the Muscle Power Scale (MRC), generalized hyperreflexia, and upgoing planter responses. Sensory, cerebellar, and gait examinations could not be reliably assessed due to his condition.

He was admitted to the intensive care unit (ICU) on Feb 18, 2023 and was followed by the Neurology service, with the initial impression of encephalitis versus stroke. Basic routine laboratory test results were unremarkable for leukocytosis, electrolyte disturbance, and raised ammonia, excluding metabolic causes, but ferritin (191 ng/mL, reference range 30–400 ng/mL) and interleukin-6 (IL-6) (45.8 pg/mL, reference range 0–7 pg/mL) were elevated, suggesting an inflammatory or infectious etiology. Results were negative for autoimmune encephalitis panel (including contactin-associated protein-like 2 (CASPR2), leucine-rich glioma inactivated 1 (LGI1), dipeptidyl-peptidase-like protein-6 (DPPX), G-protein coupled receptors for gamma-aminobutyric acid (GABA B-receptor), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA 1/2 receptor), N-methyl-D-aspartate (anti-NMDA), glycine receptor, and myelin oligodendrocyte glycoprotein (MOG antibodies) and vasculitis screening (antinuclear antibody [ANA]), and anti-double-stranded DNA antibodies (anti-dsDNA). A toxicology screen was not indicated as per hospital policy, as there was no clinical suspicion of drug misuse.

A non-contrast computed tomography (CT) scan of the brain was ordered to rule out intra- or extra-axial hemorrhage, midline shift, obvious space-occupying lesions, or acute cerebrovascular insult. Electrocardiography (ECG) showed normal sinus rhythm, and no consolidation was noted on chest X-ray.

The patient was started empirically on acyclovir 750 mg intravenous (IV) and ceftriaxone 2 g IV for 3 days. Lumbar puncture (LP) was done, which showed normal opening pressure, clear CSF, and notably high CSF protein, with no pleocytosis. Meningitis/encephalitis polymerase chain reaction (PCR panel) (including Herpes Simplex Virus (HSV), Varicella Zoster Virus (VZV), Cytomegalomirus (CMV), Human Herpes Virus 6 (HHV-6), and bacterial pathogens), and CSF culture were negative, which excluded typical viral or bacterial causes of meningoencephalitis. Tuberculosis PCR was also negative in the CSF sample.

Given the patient’s febrile presentation, a viral nasopharyngeal respiratory panel was obtained and returned positive for SARS-CoV-2. In response, the Coronavirus Disease-19 (COVID-19) treatment protocol was initiated, including a 5-day course of remdesivir, starting with a 200 mg IV loading dose followed by 100 mg IV daily.

Brain magnetic resonance imaging (MRI) was done to rule out ischemic infarction, which showed multiple lesions of low T1, high T2, and fluid-attenuated inversion recovery (FLAIR) seen in the bilateral thalami and pons (Figures 1–3). The lesions showed restricted diffusion. Multiple signal-void foci were also seen with these lesions on susceptibility-weighted imaging (SWI) (Figure 4), suggesting hemorrhage. On magnetic resonance angiography (MRA), basilar and right vertebral arteries were attenuated. Acute disseminated encephalomyelitis (ADEM) was considered unlikely given the absence of CSF pleocytosis and the symmetrical distribution of lesions in our patient. The typical asymmetric, patchy white-matter lesions and relapsing course seen in ADEM were not observed.

COVID-19-related hemorrhagic necrotizing encephalitis was suspected to be the most likely culprit in our case, given the temporal relationship with SARS-CoV-2 infection and elevated ferritin and IL-6 levels. Furthermore, neuroradiological findings (symmetric brain lesions involving bilateral thalami, periventricular white matter, and brainstem) were in keeping with the impression of possible AHNE. The patient scored 4 on the Acute Necrotizing Encephalopathy severity score (ANE-SS), indicating a moderate risk typically associated with severe sequelae or death; however, this tool was developed for pediatric populations and may not accurately predict outcomes in adults [7].

Based on the clinical impression of AHNE, he was started on IV methylprednisolone at a dose of 1 g daily for 5 days on day 2 of admission. However, his neurological status worsened, with progression to quadriparesis and ophthalmoplegia. The ICU team proceeded with endotracheal intubation due to the GCS score decreasing to 8. Given the lack of response to corticosteroids, a trial of IVIG was initiated on day 7, with 0.4 g/kg administered daily for 4 consecutive days. During the following weeks, a tracheostomy was performed by the ICU team due to poor weaning from mechanical ventilation.

By day 24 of admission, he demonstrated improvement in movement of extra-ocular muscles, obeying simple axial commands such as eye closure and sticking the tongue out, with delayed recovery of limb strength. The patient was admitted for 4 months, during which time he received extensive rehabilitation in the form of physical therapy, where improvement was noted on active and passive movements. He was able to walk around the room with assistance and gradually was able to perform some minimal physical activity independently. He was reviewed by speech and swallow therapists and was successfully weaned off naso-gastric feeding and transitioned to oral feeding. His tracheostomy was later closed and removed after consultation and review by a respiratory therapist and ear, nose, and throat (ENT) specialist.

On discharge, his neurological examination showed normal cranial nerve examination, spastic dysarthria, bilateral upper- and lower-limb spasticity more on the right than the left, pyramidal pattern of weakness more proximally, hyper-reflexia, positive clonus, and extensor planters. He could walk with minimal assistance and eat independently, but still required assistance with some activities of daily living, such as dressing and going to the bathroom.A 6-month MRI brain follow-up as an outpatient showed a marked reduction of bilateral basal ganglia hyperintensities (Figure 5).

Discussion

This case report shows the therapeutic potential of IVIG when corticosteroids are ineffective. AHNE is a rare and rapidly progressive neurological condition, usually associated with viral infections. It is thought to result from a parainfectious, immune-mediated cytokine storm that compromises the blood–brain barrier, leading to vasogenic edema, vascular leakage, and hemorrhagic necrosis, most commonly affecting the bilateral thalami, brainstem, and cerebellum [1]. Although SARS-CoV-2 primarily affects the respiratory system, involvement of the central and peripheral nervous systems is relatively common. Reported central nervous system complications include meningitis, encephalitis, seizures, both ischemic and hemorrhagic strokes, and, as in our patient, AHNE [7–11].

The diagnosis of AHNE in our patient was established based on a triad of clinical, biochemical, and radiological features, as well as ruling out plausible differentials: (1) Acute febrile encephalopathy, (2) CSF examination showing normal cell counts and increased protein concentration, and (3) Characteristic neuroimaging findings of symmetrical lesions involving bilateral thalami, brainstem and periventricular regions. These findings are consistent with the Mizuguchi criteria. Differential diagnoses such as ADEM, acute viral encephalitis, cerebral venous sinus thrombosis, and metabolic or toxic encephalopathies were systematically excluded [7–11].

Current management of AHNE is based on limited evidence. Case reports and series advocate the early immunomodulatory therapy (high-dose corticosteroids, IVIG, and plasmapheresis) based on the presumed cytokine-mediated pathogenesis [12,13]. Corticosteroids are commonly used as first-line treatment, but reports on its efficacy are inconsistent. Our patient showed no response to high-dose methylprednisolone, prompting IVIG initiation [14]. Current evidence suggests that the therapeutic effects of IVIG operate through multiple, potentially overlapping mechanisms that modulate various aspects of the inflammatory response; therefore, it was deemed an appropriate treatment option in this case [15]. Plasmapheresis was not used in this case, but it may be a reasonable option in similar refractory cases.

Several case reports have demonstrated the potential efficacy and safety of IVIG in treating COVID-19-associated encephalopathy, which helped support our decision to use a trial of IVIG. Delamarre et al described a 51-year-old man with COVID-related AHNE who recovered fully 8 days after receiving IVIG [16]. Muccioli et al reported the cases of 5 elderly patients with COVID-related encephalopathy who also recovered after IVIG without adverse effects [17]. Abdel-Mannan et al and Delorme et al similarly reported favorable outcomes in younger and older patients, respectively, following IVIG. Collectively, these cases show IVIG is a promising and well-tolerated therapeutic option across age groups for COVID-19-associated encephalopathy, particularly in cases marked by rapid neurological deterioration [18–20].

The expected timeframe for neurological improvement following IVIG administration in AHNE is not well established; however, recovery is generally observed over weeks rather than days. For example, in 1 reported case, initial improvement was documented on day 14, with further recovery noted by day 35 [9]. Across neurological disorders, clinical improvement after IVIG is typically evaluated within 14 days of treatment initiation. Given the high mortality and morbidity associated with ANE, it is unlikely that such improvements result from the natural course of recovery alone [21].

However, a 2023 systematic review by Karami et al found that among 30 adult COVID-related AHNE cases, outcomes varied widely, and only 2 recovered patients received IVIG [12], suggesting a highly variable treatment response, with no clear superiority. Thus, while IVIG may be beneficial in some cases, its use should be viewed as investigational rather than established.

A comparison with previously reported adult AHNE cases offers valuable context for understanding the significance of our patient’s clinical recovery. Despite treatment with dexamethasone, remdesivir, and convalescent plasma for severe COVID-19, a 77-year-old woman with radiological findings suggestive of AHNE related to severe COVID-19 infection progressed to multi-organ failure and died. Finally, a 68-year-old woman also failed to respond to similar COVID protocol therapy and died from refractory shock and cardiac arrest [13]. Of note, none of the patients in the reviewed cases received IVIG. Although the absence of IVIG administration in these cases does not establish a causal relationship, it raises the question of whether early IVIG intervention could offer therapeutic benefits in selected patients. These cases emphasize the importance of early recognition of AHNE features to guide timely consideration of immunomodulatory therapy. Although these patients had more severe respiratory compromise and worse biochemical profiles at presentation, this does not undermine the potential value of early treatment.

Interestingly, AHNE is typically associated with severe systemic illness; however, neurological symptoms precede respiratory symptoms in some patients with COVID-19. For example, in an early case of SARS-CoV-2-associated meningitis/encephalitis, viral RNA was detected in the CSF despite negative respiratory swab results, suggesting possible isolated neuroinvasion in the absence of respiratory invasion. We hypothesize that the patient’s respiratory compromise, coupled with fulminant neurological deterioration, may have been due to a predominant parainfectious process [22].

Favorable outcome in neurological disorders is defined as a Modified Rankin Score of 0 to 2, which indicates survival of patients with minimal or no neurological disability. Our patient had a score of 3 hence; while he had unfavorable outcomes due to sequelae of disease, he survived despite high mortality rates (reaching 53.3%) reported in a recent systematic review [12,23]. Several factors may have contributed to our patient’s outcome. He was fully vaccinated, which possibly contributed to reducing the severity of systemic inflammation and preventing mortality. He remained hemodynamically stable. His aspartate aminotransferase (AST) levels remained normal, which in pediatric studies has been linked to a better prognosis. Elevated AST reflects cytokine storm and multi-organ dysfunction, which were both absent in this case. Furthermore, he had no biochemical evidence of multi-organ failure, despite radiologic signs (brainstem involvement) typically associated with poor outcomes [24,25].

Conclusions

This case highlights the importance of including AHNE in the differential diagnosis of adults presenting with unexplained acute encephalopathy, particularly in the context of recent or concurrent viral infections such as SARS-CoV-2. Early recognition is critical, as characteristic neuroimaging findings such as symmetrical thalamic lesions with hemorrhagic features alongside elevated inflammatory markers (ferritin and IL-6) and absence of CSF pleocytosis can support timely diagnosis.

Our patient’s favorable outcome following IVIG despite poor prognostic imaging highlights its potential therapeutic value when corticosteroids are ineffective, especially in patients without multi-organ involvement or significant hepatic involvement. Additionally, his recovery demonstrates the limitations of pediatric prognostic tools, such as ANE-SS, in adult cases and suggests the need for development of adult-specific outcome predictors.

While this case supports a possible role for IVIG in adult AHNE, further research is needed to determine its efficacy, optimal timing, and appropriate patient selection. Additionally, a deeper understanding of AHNE pathophysiology in adults is essential to inform treatment decisions. Establishing standardized diagnostic criteria and evidence-based therapeutic protocols will be critical to improving clinical outcomes in this rare but life-threatening condition.