 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Section of Neurology, Departments of Pediatrics
Radiology, St Christopher's Hospital for Children, Drexel University College of Medicine, Philadelphia, Pennsylvania
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods. A retrospective record review was conducted of 13 children who were admitted as inpatients with the diagnosis of ADEM during the period 1998–2003.
Results. Diagnosis was established by clinical signs and symptoms, cerebrospinal fluid changes and multifocal involvement of deep gray and white matter based on MRI. Initial therapy was high-dose methylprednisolone and intravenous immunoglobulin in 12 patients. One child improved spontaneously. Six of 12 children did not improve with corticosteroid treatment. All 6 had an acute progressive course neurologically, and 5 of them also showed a delay in the onset of neuroimaging changes, eventually developing lesions in the deep gray matter and brainstem. This latter group received 5 sessions of plasmapheresis and recovered over the course of several months with varying degrees of residual neurologic deficits.
Conclusions. Presentation of ADEM with delayed development of MRI lesions in deep gray matter and brainstem may herald a prolonged clinical course and lack of response to glucocorticoid therapy. Plasmapheresis might be an effective therapeutic intervention in these patients. The role of plasmapheresis versus corticosteroids and intravenous immunoglobulin as a primary treatment of ADEM needs to be investigated further.
Key Words: MRI • demyelination • white matter • immunoglobulin • immunomodulation
Abbreviations: ADEM, acute disseminated encephalomyelitis • IVIG, intravenous immunoglobulin • CSF, cerebrospinal fluid • FLAIR, fluid attenuated inversion recovery • EDSS, Expanded Disability Status Scale • MS, multiple sclerosis • BDEM, biphasic disseminated encephalomyelitis
Acute disseminated encephalomyelitis (ADEM) is usually a monophasic inflammatory demyelinating disease process that affects the brain and spinal cord and typically occurs after a febrile (often presumed to be viral) prodrome or vaccination. The typical presentation is that of multifocal neurologic disturbance accompanied by change in mental status.1–3
MRI is regarded as the diagnostic imaging modality of choice and typically demonstrates involvement of deep cerebral hemispheric and subcortical white matter4 as well as lesions in the basal ganglia, gray-white junction, diencephalon, brainstem, cerebellum, and spinal cord.5,6 To our knowledge, delayed MRI findings with a lag between clinical symptoms and MRI changes have not been previously described in children.
Treatment with corticosteroids is considered to hasten recovery7,8 and is accepted as the mainstay of therapy. Intravenous immunoglobulin (IVIG) has been reserved for patients who do not respond to corticosteroids.9,10 Use of plasmapheresis to treat this condition is limited to case reports.11 The objective of this study was to describe our experience with ADEM, focusing on (1) the relationship between clinical course and MRI findings and (2) the response to plasmapheresis in a subgroup of patients.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients were included when they had an acute onset of neurologic disturbance and MRI changes involving white matter in a distribution previously described with ADEM.1–3 Patients were excluded when they had infectious meningitis or encephalitis identified by isolation of a bacterial or a viral pathogen. Three patients who had a clinical picture consistent with transverse myelitis had additional lesions in brainstem and thalami indicative of disseminated demyelinating disease. Whenever available, the results of serologic testing, cerebrospinal fluid (CSF) examination, electroencephalogram, and other ancillary testing were included. All patients had serologic testing for Mycoplasma, Epstein-Barr virus, herpes simplex virus, and Lyme disease as well as nasopharyngeal and rectal viral cultures. Patients in whom a viral or bacterial pathogen was identified on culture, serology, or polymerase chain reaction studies were excluded from this series. All patients had a work-up to exclude vasculitis, which typically included a sedimentation rate, antinuclear antibodies, and rheumatoid factor. Magnetic resonance angiography was performed in selected patients in whom it was believed to be indicated.
A 1.5-T Siemens machine was used for the brain MRI studies. T1, T2, fluid attenuated inversion recovery (FLAIR), and diffusion-weighted images were obtained. A pediatric neuroradiologist (E.N.F.) reviewed MRI scans and made special note of imaging changes with respect to the patient's clinical symptoms temporally.
Methylprednisolone was administered intravenously at a dose of 20 to 30 mg/kg per day for 5 days followed by a 2-week tapering dose. Immunoglobulin was administered intravenously at a dose of 0.4 g/kg per day for 5 days. Plasmapheresis when performed was instituted using single-volume exchange by continuous flow centrifugation. Volume that was removed was replaced by 5% albumin. Five plasma exchanges were administered on an every-other-day schedule. All patients who underwent plasmapheresis had previous placement of a central venous catheter and were monitored in the ICU. The functional degree of neurologic disability for all ambulatory patients was assessed using the Kurtzke functional systems12 and Expanded Disability Status Scale (EDSS), as has been done by other investigators.1,13
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Patient 4 was in the fourth week of a hepatitis A infection when she suddenly developed neurologic deficits.14 Two children had inconclusive results to testing for Lyme disease and received a 3-week course of Ceftriaxone. No infectious cause could be identified in the remaining 10 children; however, 8 of 13 children had a preceding prodrome of upper respiratory tract illness or a nonspecific febrile illness 2 to 4 weeks before presentation.
MRI Changes
All of the children in our series eventually developed changes in subcortical white matter and/or deep gray matter, brainstem, cerebellum, and spinal cord. MRI lesions were classified according to the system described by Tenembaum et al.1 One had small white-matter lesions <5 mm (group A); 6 had large white-matter lesions (group B), and 6 had additional bithalamic involvement (group C). No patients had hemorrhagic encephalomyelitis (group D). MRI changes were best seen on FLAIR and T2-weighted sequences.
Six children had MRI changes consistent with ADEM on the first study performed after presentation, within 2 to 3 days of onset of neurologic symptoms. However, 7 had initial MRIs that were normal even on retrospective review. The time lag before lesions could be identified on MRI ranged from 2 days to 25 days after onset of neurologic symptoms. Five of 7 patients with initially normal MRI findings had a rapidly progressive course. Their MRI scans showed gradual evolution of lesions over the next 1 to 4 weeks (Fig 1). Lesions continued to evolve even after clinical improvement. Patient 12, who presented with right hemiparesis and confusion, was treated empirically with corticosteroids and showed rapid resolution of symptoms. A follow-up MRI obtained 1 month later showed a large white-matter lesion in the left parietal cortex at a time when the patient was asymptomatic.
|
Other Studies
Table 1 summarizes the results of CSF studies, which showed pleocytosis and/or elevated protein in 7 of 13 patients at onset. CSF oligoclonal bands were studied in 12 patients and were negative in all cases. Eight children had electroencephalograms available for review: two were normal; the remaining 6 showed slowing of background activity and focal sharp wave activity.
Treatment
As shown in Table 2, 11 of 13 children were treated with high-dose methylprednisolone (20–30 mg/kg per day for 5 days followed by a 2-week taper). One child improved spontaneously with no specific therapy. One child (patient 1), whose presentation was ptosis, ophthalmoplegia, and facial weakness and whose initial MRI was normal, received initial therapy with IVIG (0.4 mg/kg per day for 5 days) for presumed Miller-Fisher syndrome. He also underwent electromyogram and nerve conduction studies because of his striking cranial nerve involvement; these were unremarkable. He subsequently received methylprednisolone without significant improvement. Seven children improved while receiving corticosteroid therapy.
|
Hospital Course
Four children deteriorated rapidly during the first week of hospitalization. Nine had maximal deficits at the time of hospitalization. Six children required mechanical ventilation for respiratory failure, and 4 needed tracheotomies for prolonged ventilatory requirements. These 6 children were the same ones who received plasmapheresis. Duration of hospital stay ranged from 3 to 75 days (median: 30 days).
Outcome
Seven of 13 children had complete recovery with an EDSS score of 0 to 1. All those who received plasmapheresis had a prolonged ICU stay and were discharged to inpatient rehabilitation facilities. At the time of discharge, all had improved. Duration of follow-up ranged from 1.5 to 5 years (median: 2 years). Five of the 6 children who received plasmapheresis had substantial recovery with decannulation of tracheotomies. At the time of last follow-up, 1 had excellent recovery with no deficits; 3 children had a residual spastic paraparesis (EDSS 5–6.5), and 1 child had persistent ophthalmoplegia and cranial nerve dysfunction (EDSS 3.5). One child was lost to follow-up as her family moved out of state.
One child (patient 4) developed new symptoms and new lesions on MRI within 6 months of the initial presentation. She initially presented with flaccid diplegia, with rapid progression to coma, quadriplegia, and bladder dysfunction. Her admission MRI showed normal spinal cord and brain; follow-up studies showed progressive areas of enhancing high signal in the periaqueductal area, basal ganglia, centrum semiovale, and internal capsule. She was given intravenous methylprednisolone and IVIG, followed by 5 courses of plasmapheresis with gradual improvement. She was discharged after a 6-week hospitalization to a rehabilitation facility. Three months after the initial episode of ADEM, she developed alteration in mental status and retrobulbar optic neuritis. MRI showed multiple gadolinium-enhancing lesions in frontal, temporal, and parietal white matter. She received another course of methylprednisolone with clinical improvement and resolution of MRI abnormalities. She has not developed any additional episodes of demyelination in the subsequent 5 years.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Early studies of ADEM reported mortality rates up to 20% with a high incidence of neurologic sequelae in survivors. Recent pediatric reports1,4,13,15 have suggested a more favorable prognosis attributed to treatment with corticosteroids. In our present series, 6 of 13 patients had a prolonged ICU course and multiple deficits at the time of discharge and follow-up. Therefore, there is a subset of children who have a more aggressive course. In our experience, children who had extensive lesions in the brainstem or cervical cord had more deficits compared with those whose lesions were mainly in the cerebellum or subcortical white matter.
MRI is regarded as the imaging modality of choice in diagnosing ADEM.5,6,16 However, a significant proportion (7 of 13) of our patients had normal MRI scans at a time of maximal neurologic deficits. Some did not exhibit significant imaging changes until a few weeks into the illness, leading to potential delay in the diagnosis and treatment, and 1 child (patient 13) showed progression of lesions after clinical improvement. Therefore, a significant lag between clinical course and imaging changes occurs. Honkaniemi et al17 described delayed MRI imaging changes in ADEM in 4 adult patients as did a case report by Murray et al,18 but this has not been reported in children. In our series, 5 of 7 patients with a delay in diagnostic imaging changes had rapid progression of symptoms. Because these patients eventually had a high number of lesions in the deep gray nuclei and brainstem, it is possible that lesions in these areas take a longer time to evolve and become visible with currently used, standard resolution MRI. It is noteworthy that all 4 patients in the adult series17 also had rapid deterioration followed by gradual improvement. It is possible that in patients with a more rapidly progressive course, lesions are not apparent at the time of presentation or shortly thereafter, as compared with patients with a more subacute presentation. This again points to the heterogeneous nature of ADEM and its varied clinical presentation. One could also speculate that the pathogenesis of these cases with delayed MRI changes may be different from those of "typical" ADEM. Our data suggest that the absence of white-matter changes on MRI up to 3 weeks into the illness does not exclude the diagnosis of ADEM; treatment should not be delayed when clinical suspicion is high.
The pathogenesis of ADEM is thought to be disseminated multifocal inflammation and patchy demyelination associated with autoimmune mechanisms in the CNS. ADEM occurs after infection by a virus or may be associated with immunization against rabies, diphtheria-pertussis-tetanus, or other pathogens. Antibodies to these pathogens may exhibit cross-immunoreactivity with myelin basic protein in the white matter. The pathologic findings in ADEM of demyelination and perivenous invasion of lymphocytes and monocytes are similar to those in experimental autoimmune encephalomyelitis, an accepted experimental model for acute demyelination.19 Although ADEM is typically described as a monophasic illness that lasts from 2 to 4 weeks, relapses have been reported,1–3,20,21 raising the question of whether these cases represent instances of multiple sclerosis (MS). One of our patients (patient 5) had relapsing disease. Other investigators1,20 have suggested the term biphasic disseminated encephalomyelitis (BDEM) or multiphasic disseminated encephalomyelitis (MDEM) for these patients. The diagnosis of BDEM or MDEM, in contrast to MS, is supported by the following: florid polysymptomatic presentation, lack of oligoclonal bands in the CSF, predominance of MRI lesions in the subcortical region with relative sparing of the periventricular areas but with possible involvement of the deep gray matter, complete or partial resolution of MRI lesions during convalescence, and recurrence within 6 months of a previous episode.1,20 Our patient with BDEM (patient 5) fulfilled these criteria. According to current diagnostic criteria, presence of new lesions on MRI repeated 3 months after the attack is predictive of MS.22,23 Most children with MS show new lesions on MRI repeated within 2 years even when they remain asymptomatic. None of our children did.
Because of the suggested immune-mediated pathogenesis, immune-modulating agents, especially corticosteroids, have been used in the treatment of ADEM. Despite the lack of controlled studies to prove their efficacy, there is strong anecdotal evidence of their benefit. Recent pediatric case series have reported wide usage of corticosteroids; 74% of 31 children in a recent Australian series4 and 87% of 35 children in a British series20 had good response to therapy. A large pediatric series of 84 patients from Argentina1 also reported good recovery and resolution of MRI lesions with the use of high-dose corticosteroids in 80 of 84 children. The use of IVIG is less widespread but has been described in a few case reports and small series3,9,10,24,25 in children, usually for patients who did not respond to corticosteroids.
Plasmapheresis has a definite role in the treatment of neurologic conditions that are presumed to be immunologically mediated, such as Guillain-Barré syndrome, chronic inflammatory demyelinating polyradiculoneuropathy, polyneuropathies associated with monoclonal gammopathies, and myasthenia gravis.26 It has been suggested that plasmapheresis also may have a role in treating patients with acute demyelination, including MS and ADEM.26,27 However, there have been few reports of its use in ADEM. Two adult case series reported 2 and 4 adults, respectively,10,28 who had rapidly progressive ADEM and improved after plasmapheresis. A recent study from the Mayo Clinic29 reviewed 59 consecutive patients who were treated with plasma exchange for acute, severe attacks of CNS demyelination. Ten of these had ADEM; 4 of 10 had a moderate to marked improvement after plasmapheresis.
To our knowledge, only 3 children who underwent plasmapheresis to treat ADEM after failure to respond to corticosteroids and/or IVIG were reported previously.27,30,31 All 3 reports documented improvement after plasmapheresis.
A drawback of our series is its retrospective nature and heterogeneous characteristics of neurologic deficits. In our series, methylprednisolone was used as initial therapy in almost all instances (11 of 13), followed by IVIG when there was progression or failure of significant neurologic deficits to improve. Six of our patients then received plasmapheresis when they failed to show improvement. At the time of completion of plasmapheresis, all had begun to show improvement. Whether improvement in clinical status can be attributed to plasmapheresis, the previous use of immune-modulating therapies, or the natural course of the disease itself cannot be determined. There were no complications from the use of plasmapheresis in our patients. Potential complications reported by others29,32 include central line sepsis, hypotension, anemia, electrolyte abnormalities, and heparin-induced thrombocytopenia.
Our treatment has evolved to include initial treatment with corticosteroids for 5 days followed by observation for 3 to 5 days, then IVIG for 5 days followed by observation for 3 to 5 days. If there has been no significant improvement or there is worsening of neurologic function, then plasmapheresis is performed for 5 courses every other day.
Plasmapheresis should be considered as a treatment option for patients with ADEM, especially when the course is aggressive or severe disease has not responded to corticosteroids and IVIG. It is unclear whether the use of plasmapheresis early in the course of the disease would alter the prognosis. A multicenter prospective study to address the outcome of the use of different therapeutic modalities in the treatment of ADEM is warranted.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Reprint requests to (D.S.K.) Section of Neurology, St Christopher's Hospital for Children, Erie Avenue at Front Street, Philadelphia, PA 19134. E-mail: divya.khurana{at}drexel.edu
No conflict of interest declared.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2005 American Academy of Pediatrics. All rights reserved. |
||
|
||
