John Libbey Eurotext

Cryptogenic new-onset refractory status epilepticus (NORSE) following blood transfusion in a patient with severe anemia Volume 23, issue 2, April 2021


  • Figure 1
  • Figure 2

Based on the current consensus definition, new-onset refractory status epilepticus (NORSE) is “a clinical presentation, not a specific diagnosis, in a patient without active epilepsy or other preexisting relevant neurological disorder, with new onset of refractory status epilepticus without a clear acute or active structurale, toxic or metabolic cause.” [1]. Approximately 50% of NORSE cases have an eventual identifiable cause -inflammatory and autoimmune, infectious, genetic/congenital, and toxic, metabolic, or drug-related- with the other 50% remaining cryptogenic [2, 3]. To our knowledge, none have previously been attributed to blood transfusion.

Case study

A 46-year-old woman with a remote history of thyroidectomy for papillary thyroid cancer presented with six months of intermittent menorrhagia, followed by one month of continuous menorrhagia and progressive fatigue. No other prodromal or neurological symptoms were evident. Initial laboratory workup revealed hemoglobin at 2.6 g/dL. After rapidly receiving three units of packed red blood cells, she became tachycardic and confused, with subsequent generalized tonic-clonic seizures that responded to lorazepam. She was then started on levetiracetam. Serum calcium level after transfusion was 7.5 mg/dL. Lumbar puncture performed on the first hospital day showed mild neutrophilic pleocytosis (5-15 WBCs) without oligoclonal bands. She was started on empiric vancomycin, ceftriaxone, and acyclovir, but these were later discontinued after a CSF meningoencephalitis panel returned negative. Due to worsening encephalopathy, continuous EEG was performed and revealed frequent right frontal seizures. Subsequently, levetiracetam dose was increased and she was started on lacosamide. Brain MRI showed T2 hyperintense and diffusion-restricting lesions in the bilateral frontal and insular cortices and medial thalami (figure 1A, B). Additionally, the bifrontal lesions had subtle enhancement on post-contrast sequences (figure 1C). Brain MR angiography showed normal vasculature. She had no hypotension or cardiac arrest to suggest anoxic injury. On hospital Day 4, EEG showed 1-2-Hz right frontal polymorphic waves with superimposed fast rhythmic activity consistent with non-convulsive status epilepticus (figure 2). Despite escalation of anticonvulsant therapy to a combination of levetiracetam, lacosamide, phenytoin, clobazam, propofol (70 mcg/kg/min) and midazolam (11 mg/h) infusions, she developed super-refractory status epilepticus, ultimately requiring pentobarbital infusion to achieve 80% burst suppression. Given that comprehensive infectious workup was negative, she received a five-day course of empiric high-dose intravenous methylprednisolone (1 g daily) for possible autoimmune encephalitis starting on hospital Day 5. By Day 11, her EEG improved dramatically and she was weaned off pentobarbital without further seizures. Intravenous immunoglobulin therapy and plasma exchange were considered but ultimately not pursued due to clinical improvement. Infectious workup was negative for a bacterial, viral, fungal, or prion process. Inflammatory and autoimmune workup including autoimmune encephalitis antibody panel was negative. Malignancy workup, including serum malignancy markers (CA-125, CA-19-9), endometrial biopsy, CT scan of chest, abdomen, pelvis, and PET imaging, revealed a uterine polyp without hyperplastic or carcinomatous features, but was otherwise unrevealing. Her anemia was attributed to this uterine polyp. She required a tracheostomy and gastrostomy tube. By Day 24, she started following simple axial commands. By Day 31, she was conversant with improving encephalopathy. She was discharged to an acute inpatient rehabilitation center and after another four weeks, she was nearly back to baseline though with slower processing speed.


Neurological complications are rare following blood transfusion. Seizures after transfusion have been described in the context of citrate-induced hypocalcemia and PRES. While our patient had a history of mild hypocalcemia (7.3 mg/dL; thought to be due to parathyroid injury after prior thyroidectomy), her serum calcium levels remained stable after transfusion. Therefore, her seizures were unlikely to be related to citrate toxicity. There are also reports of posterior reversible encephalopathy syndrome (PRES) and reversible cerebral vasoconstriction syndrome (RCVS) following transfusion for chronic or subacute anemia. Of the published post-transfusion cases of PRES, most patients had anemia due to menorrhagia from uterine fibroids [4-8] with additional cases of aplastic and iron deficiency anemia [9]. Only two patients with menorrhagia developed status epilepticus requiring anesthetic agents and barbiturates [6-8]. In all cases, neurological deficits resolved within two weeks. No patients required tracheostomy, enteral feeding, or long-term inpatient rehabilitation. In one retrospective analysis of seven women who developed RCVS after blood transfusion, three developed non-refractory seizures five to seven days after transfusion [10].

All reported cases of transfusion-related PRES and RCVS occurred in the setting of subacute to chronic, but not acute anemia. The precise mechanisms of these entities remain speculative. One proposed mechanism involves compensatory cerebral vasodilation in response to chronic systemic hypoxia, such as in severe anemia [11]. With rapid transfusion, the rise in hematocrit and blood viscosity may induce sudden reversal of compensatory vasodilation [12]with resultant endothelial damage, vasogenic edema, and parenchymal irritation. Our patient's MRI lesions are suggestive of multifocal edema and do not show the posterior-predominant distribution typical of PRES. While we are not aware of any specific relationship between multifocal edema and NORSE, we suspect the underlying mechanism may be similar to PRES. A final alternate mechanism is paradoxical worsening of oxygen delivery related to blood transfusion secondary to packed RBC storage lesions [13], although the effects of this on the central nervous system have not been well described previously. While this ultimately remains a case of cryptogenic NORSE, there is a compelling argument for rapid blood transfusion as a provoking factor. The severe and unexpected outcome in our patient warrants more careful consideration of transfusion strategies in patients presenting with severe, chronic anemia.

Supplementary data

Summary didactic slides are available on the website.


None of the authors have any conflict of interest to declare.