Diffusion MRI abnormalities after prolonged febrile seizures with encephalopathy
Citation Manager Formats
Make Comment
See Comments

Abstract
Background: Patients with encephalopathy heralded by a prolonged seizure as the initial symptom often have abnormal subcortical white matter on diffusion-weighted MRI (DWI).
Objective: To determine if these patients share other common features.
Methods: Patients with encephalopathy heralded by a prolonged seizure and followed by the identification of abnormal subcortical white matter on MRI were collected retrospectively. Their clinical, laboratory, and radiologic data were reviewed.
Results: Seventeen patients were identified, ages 10 months to 4 years. All had a prolonged febrile seizure (longer than 1 hour in 12 patients) as their initial symptom. Subsequent seizures, most often in clusters of complex partial seizures, were seen 4 to 6 days after the initial seizure in 16 patients. Outcome ranged from almost normal to severe mental retardation. MRI performed within 2 days of presentation showed no abnormality. Subcortical white matter lesions were observed on DWI between 3 and 9 days in all 17 patients. T2-weighted images showed linear high intensity of subcortical U fibers in 13 patients. The lesions were predominantly frontal or frontoparietal in location with sparing of the perirolandic region. The diffusion abnormality disappeared between days 9 and 25, and cerebral atrophy was detected later than 2 weeks. Three patients having only frontal lesions had relatively good clinical outcome.
Conclusions: Although the pathophysiologic mechanism remains unknown, these patients seem to have a distinctive encephalopathy syndrome. MRI is helpful in establishing the diagnosis of this encephalopathy.
Encephalitis can exhibit a wide range of CNS manifestations, with varying severity. When neurologic manifestations suggest encephalitis, but inflammatory cells are not found in the brain or CSF, the condition is identified by the less specific term “encephalopathy”; examples of this are Reye syndrome and influenza-associated encephalitis/encephalopathy (IAEE).1–3
MRI is accepted as a more sensitive technique than CT for the diagnosis of encephalitis/encephalopathy; diffusion-weighted imaging (DWI) is particularly useful for detecting early changes in the brain. Recently, some encephalitis/encephalopathy syndromes such as acute necrotizing encephalopathy (ANE),4 hemorrhagic shock and encephalopathy syndrome (HSES),5 and clinically mild encephalitis/encephalopathy with a reversible splenial lesion (MERS)6 have been established based on clinicoradiologic features.
One report described a patient with encephalopathy who presented with febrile prolonged seizures as the initial symptoms and developed abnormalities of the subcortical white matter that were most conspicuous on DWI.7 Similar patients have been seen in pediatric clinics in Japan, but the exact nature of their clinical and radiologic features remains uncertain. To determine whether such patients are affected by a distinct clinicoradiologic entity, the clinical, radiologic, and laboratory findings of 17 Asian patients with similar disorders were retrospectively reviewed.
Methods.
A questionnaire was sent to the members of the Annual Zao Conference on Pediatric Neurology and to some members of Japanese Society of Pediatric Neurology and Japanese Society of Neuroradiology after institutional review board approval from Kameda Medical Center to see if they had encountered patients with encephalitis/encephalopathy who presented with prolonged seizures as the initial symptoms and developed subcortical white matter lesions on MRI that were most conspicuous on DWI. We reviewed MRI scans and clinical charts of these patients after obtaining informed consent from their parents. We additionally accumulated information about symptoms, clinical diagnosis, medications, treatments, prognosis, and results of CSF analysis and EEG. The diagnosis of encephalitis was established if the patients had acute onset of brain dysfunction, such as seizures and disorders of consciousness, with inflammatory changes, such as pleocytosis of the CSF. When there was no evidence of inflammatory changes in the CSF, we used the term “encephalopathy.”
MRI scans of three patients with a prolonged febrile seizure alone (at ages 8 months to 2 years) at Kameda Medical Center were also reviewed as disease control.
Results.
We identified 17 children who met our entry criteria among the 20 patients whose clinical records and MRI examinations were referred for this study. Two patients were excluded because of the poor image quality of their MRI. Another patient was excluded after the episode for which she was referred was diagnosed as hypoxic–ischemic encephalopathy due to drowning. The clinical records and radiologic examinations of the remaining 17 patients were reviewed by the authors and are the basis of this study.
The ages of the 17 patients (9 boys and 8 girls) ranged from 10 months to 4 years (table). Three patients (Patients 2, 3, and 5) had developmental delay, and one patient (Patient 1) had a history of febrile seizures. All 17 patients presented with a seizure longer than 30 minutes (longer than 1 hour in12 patients) as their initial neurologic symptoms within a day of the onset of fever. Twelve patients had continuous disturbance of consciousness level or hemiparesis (Patient 1) after the initial prolonged seizure, including three (Patients 3, 11, and 17) who needed mechanical ventilation. But the other five patients (Patients 5, 7, 10, 13, and 15) had normal, clear consciousness with no neurologic symptoms on the next day. Second seizures, most often clusters of complex partial seizures associated with deterioration of consciousness level, were seen 4 to 6 days after the initial prolonged seizure in 16 of the 17 patients. Patient 5 became drowsy and manifested involuntary hand-washing movements on day 4 but had no second seizure. Their treatments varied, including various anticonvulsive drugs for all patients and methylprednisolone (mPSL) pulse therapy for nine patients. Outcome of the 17 patients ranged from almost normal in one child (Patient 10), to mild mental retardation in three (Patients 5, 13, and 14), to severe mental retardation, paralysis, and epilepsy (table). Three of the five patients with rapid recovery of their consciousness level after the initial prolonged seizure had relatively good outcomes (Patients 5, 10, and 13); however, the other two (Patients 7 and 15) had moderate or severe mental retardation.
Table Clinical data for patients with encephalopathy
Infectious agents were identified in 10 patients (table). The pathogens included influenza A and B (four patients), human herpes virus (HHV) 6 and 7 (causes of exanthema subitum, four patients), varicella zoster virus, and adenovirus. Analyses of CSF revealed absence of pleocytosis and normal protein levels in the 17 patients, leading to the diagnosis of encephalopathy. Serum aminotransferase was elevated in 13 patients (more than threefold in Patients 1, 2, 16, and 17), but ammonia was elevated in only 1 of 16 examined patients (Patient 4). Hyperglycemia (glucose level over 200 mg/dL at the initial seizure) was seen in four patients (Patients 4, 6, 12, and 14). But no patient had hypoglycemia, which was a characteristic of classic Reye syndrome. No patient had disseminated intravascular coagulation, which is characteristic of HSES. EEG showed slow basic activity or epileptic discharges in 15 of the 16 patients examined in the acute stage.
In 5 of the 17 patients (Patients 1, 2, 11, 12, and 14), the initial MRI study (figure 1, A) was performed within 2 days of the initial prolonged seizure and before the second seizure; these studies showed no acute parenchymal lesion (Patient 2 having chronic white matter and thalamic injuries).
Figure 1. T2-weighted imaging (A, B, E, H), diffusion-weighted imaging (DWI) (C, F), and apparent diffusion coefficient (ADC) maps (D, G) for Patient 11. MRI at day 2 showed no parenchymal lesion on fast spin echo T2-weighted imaging (A, repetition time [TR]/echo time [TE] = 4,000/98 milliseconds) and DWI. T2-weighted imaging at day 8 (B) revealed T2 prolongation in the subcortical white matter, especially along the U-fibers. DWI (C; spin echo echo planar imaging, TR/TE = 10,000/125, b factor = 1,000) showed high signal lesions in the subcortical white matter with reduced diffusion. Reduced diffusion is verified on the ADC image (D; ADC in the frontal white matter = 0.311 × 10−3 mm3/s). MRI at day 15 (E through G) showed T2 prolongation with increased diffusion in the white matter (ADC in the frontal white matter = 1.120 × 10−3 mm3/second). Cerebral atrophy was recognized on day 22 (H).
Subcortical white matter lesions were observed in all 17 patients scanned between 3 and 9 days after the initial prolonged seizure and were most conspicuous on DWI (figure 1, C). Most of these studies were performed a few days after the second seizure (15 patients) or after clinical deterioration (Patient 5). In Patient 14, the second MRI (on day 3) was performed before the second seizure (on day 4). The calculated ADC map demonstrated decreased ADC values of the subcortical lesions in all 12 patients so examined (figure 1, D). T2-weighted images or fluid-attenuated inversion recovery (FLAIR) images showed linear high intensity along the U-fibers in 13 patients (figure 1, B); cortical hyperintensity was less prominent than U-fiber abnormality in 8 patients. ADC values of the cortex could not be confidently measured because the thin nature and undulating configuration of the cortex did not allow confident measurement (without possible contamination by CSF). The lesions were predominantly frontal or frontoparietal in location with sparing of the perirolandic region; they were symmetric in 16 patients but asymmetric in Patient 1, in whom the left side was more affected (this child had a clinical diagnosis of hemiconvulsion hemiplegia epilepsy syndrome).
The subcortical high signal intensity on DWI disappeared in all 17 patients on the follow-up studies performed between days 9 and 25 (figure 1, F). High signal intensity in the cortex overlying affected subcortical white matter was seen on DWI during this time period in 11 patients. The ADC maps showed increased ADC within the subcortical white matter in three of seven examined patients (figure 1, G) compared with age-matched controls at this time (normal data not shown). T2-weighted or FLAIR images showed high signal intensity in the subcortical white matter in 13 patients (figure 1, E). Cerebral atrophy was observed in all 16 patients examined later than 2 weeks (figure 1, H). Clinical and radiologic information for the 17 patients is shown in tables E-1 and E-2 on the Neurology Web site (go to www.neurology.org).
MRI of three patients with a prolonged febrile seizure alone (no subsequent encephalopathy) scanned on day 3 or 4 after the seizure showed no parenchymal abnormalities on any sequence. No follow-up study was performed for these patients.
The serial images obtained in Patient 14 (figure 2) are of considerable interest because the subcortical lesions were recognized before the second seizure, and MR spectroscopy revealed increased glutamine/glutamate complex (Glx). The initial MRI on day 1 showed no parenchymal lesions. Because of persisting lack of consciousness and frontal dominant high-voltage slow waves on EEG, a second MRI was performed on day 3, revealing reduced diffusion in the frontal subcortical white matter. MR spectroscopy in the left frontal white matter, with the same methods previously reported,8 showed increased concentration of Glx (12.4 mM) compared with the values obtained from nine age-matched control subjects (7.6 ± 1.0 mM). In addition, N-acetyl aspartate (NAA) (4.6 mM vs control values of 6.8 ± 0.4 mM) was decreased. mPSL pulse therapy was started on day 3; a mild seizure was observed once on day 4. MRI on day 9 showed high signal in the frontal cortex on DWI and in the frontal cortex and white matter on T2-weighted imaging. MRI on day 16 showed mild cerebral atrophy and T2 prolongation in the frontal white matter. MR spectroscopy in the affected left frontal white matter showed markedly decreased NAA of 2.5 mM and normal Glx of 5.9 mM at this time. The patient's motor abilities recovered almost fully after 2 months, but moderate mental retardation, especially speech delay, has persisted.
Figure 2. MRI and MR spectroscopy for Patient 14. (A through C) Performed on day 3. Diffusion-weighted imaging (A) showed reduced diffusion in the frontal subcortical white matter without abnormality on the T2-weighted image (B). MR spectroscopy (C; repetition time/echo time = 5,000/30 milliseconds) in the left frontal white matter revealed increased concentration of glutamine/glutamate complex of 12.4 mM (2.05 to 2.50 ppm) with decreased N-acetyl aspartate (2.02 ppm) of 4.6 mM. (D through F) Performed on day 16. Diffusion-weighted imaging (D) and T2-weighted imaging (E) showed mild cerebral atrophy and T2 prolongation in the frontal white matter. MR spectroscopy (F) showed markedly decreased N-acetyl aspartate of 2.5 mM and normal glutamine/glutamate complex of 5.9 mM.
Discussion.
The most important finding of this study is that all 17 children presented with similar clinical and radiologic features. All were younger than 4 years (most under 2 years) and had a prolonged seizure within a day of the onset of fever as an initial neurologic symptom, followed by secondary seizures and deterioration of consciousness level at days 4 to 6, finally resulting in variable levels of neurologic sequelae. Radiologically, MRI showed no acute abnormality during the first 2 days but revealed injury with reduced diffusion in the frontal or frontoparietal subcortical white matter, with sparing of perirolandic region, during days 3 to 9. The diffusion abnormality of the subcortical white matter disappeared between days 9 and 25, finally resulting in cerebral atrophy.
Reye syndrome, an acute encephalopathy with fatty degeneration of the liver that is closely associated with administration of aspirin,1 was one of the differential diagnoses of the 17 patients in this study. Reye syndrome presents with a stereotypic, biphasic course, that is, a prodromal febrile illness, usually influenza B or chickenpox, followed by an abrupt onset of vomiting and seizures, with coma or death usually ensuing within an interval of 5 to 7 days. The neurologic symptoms usually develop while the patient is recovering from a viral infection. Thus, the prolonged febrile seizure on day 1 observed in the current 17 children is unusual in Reye syndrome. No patient in this study received aspirin as an antipyretic or had hypoglycemia, both of which are characteristic of young patients with Reye syndrome. Therefore, the clinical features in this study are distinct from those of Reye syndrome. Other encephalopathy syndromes, such as ANE (symmetric involvement of thalamus on MRI),4 HSES (an acute onset of encephalopathy with fever, shock, watery diarrhea, renal and hepatic dysfunction, and severe disseminated intravascular coagulation),5 and MERS (a reversible splenial lesion on MRI)6 are also ruled out by clinical and radiologic features.
Influenza was isolated from the throats of four patients in this study, establishing the clinical diagnosis of IAEE, the most common encephalopathy in Japan. The neuroradiologic findings in patients with IAEE have been divided into six categories:3,6 Category 1, no abnormal lesions; Category 2, diffuse involvement of cerebral cortex that is identifiable on MRI approximately 4 days after the onset of CNS manifestations; Category 3, diffuse brain edema; Category 4, symmetric involvement of thalamus (ANE); Category 5, postinfectious focal encephalitis; and Category 6, a reversible splenial lesion (MERS). Linear subcortical T2 prolongation, previously described to be a deep cortical lesion and illustrated in a figure of a patient with category 2,3 appears identical to the subcortical T2 abnormality seen between 3 and 9 days in the children in our study. From the clinical perspective, IAEE is divided into at least four types: ANE, HSES, the acute brain swelling type, and the status epilepticus type.9 Patients with the status epilepticus type of IAEE present with a prolonged febrile seizure as the initial neurologic symptom; this is followed by recurrent seizures and lobar edema after 4 days and results in cerebral atrophy after months. These features seem to be identical to the encephalopathy in the current series. Therefore, it is likely that the category 2 and status epileptic types of IAEE are identical to the encephalopathy reported in this study, even though DWI and detailed follow-up MRI studies were not available in the previous reports.
Similar clinical symptoms and radiologic findings were reported in a 16-month-old Japanese patient with HHV-6 encephalopathy.7 MRI demonstrated linear subcortical T2 prolongation and reduced diffusion on day 7, which disappeared on day 16, and cerebral atrophy on day 42. As HHV-6 or 7 was actually isolated in four patients in this study, it seems likely that influenza and HHV-6 and 7 are likely to be common pathogens associated with this type of encephalopathy.
As far as we know, there has been no report of a patient other than East Asian with this type of encephalopathy. One possible explanation for this observation is that Asian people might have genetic backgrounds that facilitate the development of this type of encephalopathy, as well as IAEE,2 although the exact mechanism by which these syndromes develop is still uncertain. Another possibility is that, because of the relatively high incidence of IAEE in Japan, we commonly perform brain MRI in patients with a prolonged seizure or suspected encephalopathy, leading to detection of the dynamic MRI changes in this type of encephalopathy.
Five of 17 patients in this study showed rapid recovery of their consciousness levels after the initial prolonged seizure; thus, a tentative initial diagnosis of febrile seizure was made. Because four of the five patients developed subsequent mental retardation, it is important for clinicians to be able to differentiate these two conditions (febrile seizure alone and encephalopathy). However, no laboratory or neuroradiologic finding to suggest this type of encephalopathy was found on day 1 or 2. Patient 14 was treated with mPSL pulse therapy on day 3 (before second seizure) because the MRI showed subcortical lesions. Despite a mild second seizure on day 4, outcome was relatively good. Therefore, it seems reasonable to propose that MRI should be performed on day 3 in patients with a febrile prolonged seizure to rule out this type of encephalopathy. Studies are under way to identify methods for earlier diagnosis and treatment.
The cause of the MRI findings in these children is unknown. Does the initial prolonged seizure cause reduced diffusion in the subcortical white matter on days 3 to 9? Previous reports of MRI in patients with status epilepticus have shown high signal on T2-weighted imaging with reduced diffusion in the affected cortex, ipsilateral thalamus, and contralateral cerebellum during status epilepticus or within a few days of status epilepticus.10,11 Pathologic studies showed selective neuronal necrosis within these areas.11,12 The subcortical white matter was unremarkable in patients with status epilepticus.10 The cortical signal returned to normal during the following weeks or months, but focal brain atrophy developed and T2 prolongation remained. These MRI findings differ from those of patients in this study, who had no abnormality on days 1 or 2. Another MRI study of 35 children with status epilepticus (21 with a prolonged febrile seizure) showed no subcortical abnormalities within 5 days of the event,13 similar to the three patients with an isolated prolonged febrile seizure in this study. Therefore, the subcortical T2 and diffusion abnormalities seen in this study seem unlikely to have resulted from the initial prolonged seizure.
One potential explanation for the timing of the radiologic findings is that the brain injury is a result of excitotoxic damage. The majority of excitatory neurons in the human cerebral cortex release an excitatory neurotransmitter, glutamate, which is taken up from the synaptic cleft by surrounding astrocytes and metabolized into a relatively harmless compound, glutamine. Glutamine is then returned to the surrounding glutamatergic neurons; this process is referred to as glutamate–glutamine cycling.14 If glutamate is released in quantities that cannot be processed by astrocytes or if the astrocytes are not functioning properly, the high levels of glutamate that accumulate in the interstitial spaces can cause cellular damage; this condition is referred to as excitotoxicity. In patients with hepatic encephalopathy or urea cycle disorders, MR spectroscopy reveals an increased concentration of Glx, reflecting ammonia-induced glutamine accumulation.8 CSF glutamine concentration is actually extraordinarily high in patients with urea cycle disorders.8 Ammonia-induced glutamine accumulation in astrocytes creates an osmotic gradient that causes a shift of water into the astrocytes, resulting in astrocytic swelling or edema.
Under excitotoxic conditions such as seizures or ischemia, astrocytes are thought to be neuroprotective due to their ability to clear extracellular glutamate (glial glutamate detoxification).15 MR spectroscopy in Patient 14 of this study revealed increased Glx concentration in the affected white matter along with imaging findings of reduced subcortical diffusion on day 3. This finding seems unlikely to be a result of hepatic dysfunction because this patient, like most other patients in this study, had a normal level of ammonia. The most likely cause, therefore, is glutamate accumulation due to hyperactivity of glutamatergic neurons. Glutamate can induce an increase in astrocytic cell volume with a resulting decrease of the extracellular space in primary culture from cerebral cortex of newborn rat.16,17 Glutamate induced astrocytic swelling and edema may play a part in the reduced diffusion seen in the subcortical white matter. Normalization of the Glx concentration on MR spectroscopy performed on day 16 may explain the transience of the reduced diffusion, if glutamate-induced astrocytic swelling and edema are reversible. Progressively decreased NAA on MR spectroscopy on days 3 and 9 suggested persistent neuronal damage, which may have resulted in the cerebral atrophy.
In patients with IAEE, CSF studies have shown increased glutamine and nitrite/nitrate levels together with decreased glutamate levels. This imbalance between glutamate and glutamine suggests increased activity of the glutamate uptake transporter and glutamine synthetase in astrocytes.18,19 Some authors have speculated that the astrocytes may be activated by high nitrite/nitrate and some (unknown) viral factors and cytokines in CSF.18 It is possible that a similar mechanism may play a role in the encephalopathy described in this study, although the pathophysiologic mechanism of this type of encephalopathy remains unknown. Further studies, including analyses of histology, cytokines, nitrite/nitrate, CSF glutamine and glutamate, and MR spectroscopy (preferably with a higher field strength magnet) will be necessary to make this determination.
Acknowledgment
The authors thank the patients and families for their participation. They also thank M. Mizuguchi, MD (Department of Pediatrics, Graduate School of Medicine, University of Tokyo) and H. Yoshikawa, MD (Department of Pediatrics, Miyagi Children's Hospital), for advice and C. Ishii, MD, and H. Uchida, MD (Department of Pediatrics, Showa General Hospital), K. Kobayashi, MD (Department of Pediatrics, Chiba Children's Hospital), T. Maemoto, MD (Department of Pediatrics, Asahi General Hospital), and M. Kubota, MD (Department of Pediatrics, Tokyo Metropolitan Hachioji Children's Hospital) for referring patients and comments.
Footnotes
-
Additional material related to this article can be found on the Neurology Web site. Go to www.neurology.org and scroll down the Table of Contents for the May 9 issue to find the title link for this article.
Commentary, see page 1291
Supported in part by the Research Grant (17A-11) for Nervous and Mental Disorders from the Ministry of Health, Labor, and Welfare of Japan.
Disclosure: The authors report no conflicts of interest.
Received August 29, 2005. Accepted in final form January 19, 2006.
References
- 1.↵
Rudoiph JA, Balistreri WF. Reye syndrome and the mitochondrial hepatopathies. In: Beheman RE, Kliegman RM, Jenson HB, eds. Nelson textbook of pediatrics.7th ed. Philadelphia: Saunders, 2004:1335–1337.
- 2.↵
Morishima T, Togashi T, Yokota S, et al. Encephalitis and encephalopathy associated with an influenza epidemic in Japan. Clin Infect Dis 2002;35:512–517.
- 3.↵
- 4.↵
Mizuguchi M. Acute necrotizing encephalopathy of childhood: a novel form of acute encephalopathy prevalent in Japan and Taiwan. Brain Dev 1997;19:81–92.
- 5.↵
- 6.↵
Tada H, Takanashi J, Barkovich AJ, et al. Clinically mild encephalitis/encephalopathy with a reversible splenial lesion. Neurology 2004;63:1854–1858.
- 7.↵
- 8.↵
Takanashi J, Kurihara A, Tomita M, et al. Distinctly abnormal brain metabolism in late-onset ornithine transcarbamylase deficiency. Neurology 2002;59:210–214.
- 9.↵
Shiomi M. Clinical spectrum of influenza encephalopathy. Jpn J Pediatr Med 2003;35:1676–1681.
- 10.↵
Lansberg MG, O'Brien MW, Norbash AM, Moseley ME, Morrell M, Albers W. MRI abnormalities associated with partial status epilepticus. Neurology 1999;52:1021–1027.
- 11.↵
Men Suleyman, Lee DH, Barron JR, Muñoz DG. Selective neuronal necrosis associated with status epilepticus: MR findings. AJNR Am J Neuroradiol 2000;21:1837–1840.
- 12.
- 13.↵
Scott RC, Gadian DG, King MD, et al. Magnetic resonance imaging findings within 5 days of status epilepticus in childhood. Brain 2003;125:1951–1959.
- 14.↵
- 15.↵
Trendelenburg G, Dirnagl U. Neuroprotective role of astrocytes in cerebral ischemia: focus on ischemic preconditioning. Glia 2005;51:307–320.
- 16.↵
Hansson E. Metabotropic glutamate receptor activation induces astroglial swelling. J Biol Chem 1994;269:21955–21961.
- 17.
Hansson E, Rönnbäck L. Glial neuronal signaling in the central nervous system. FASEB J 2003;17:341–348.
- 18.↵
- 19.
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Hemiplegic Migraine Associated With PRRT2 Variations A Clinical and Genetic Study
Dr. Robert Shapiro and Dr. Amynah Pradhan