Acute Disseminated Encephalomyelitis

Updated: Nov 08, 2018
  • Author: J Nicholas Brenton, MD; Chief Editor: Tarakad S Ramachandran, MBBS, MBA, MPH, FAAN, FACP, FAHA, FRCP, FRCPC, FRS, LRCP, MRCP, MRCS  more...
  • Print


Acute disseminated encephalomyelitis (ADEM) is an immune-mediated inflammatory demyelinating condition that predominately affects the white matter of the brain and spinal cord. The disorder manifests as an acute-onset encephalopathy associated with polyfocal neurologic deficits and is typically self-limiting. [37, 46, 48] ADEM bears a striking clinical and pathological resemblance to other acute demyelinating syndromes (ADS) of childhood, including multiple sclerosis (MS). ADEM in children is readily distinguishable from alternative diagnoses on the basis of clinical features and findings on neuroimaging and laboratory investigations. However, given that ADEM lacks a specific identified biological marker rendering a reliable laboratory diagnosis, long-term follow-up is important as there are instances where an illness initially diagnosed as ADEM is ultimately replaced with a diagnosis of MS. [1]

The onset of ADEM usually occurs in the wake of a clearly identifiable febrile prodromal illness or immunization and in association with prominent constitutional signs and encephalopathy of varied degrees. ADEM is typically a monophasic disease of pre-pubertal children; whereas, MS is typically a chronic relapsing and remitting disease of young adults. Abnormalities of findings on cerebrospinal fluid (CSF) immunoglobulin studies are less common in ADEM. However, the division between these processes is indistinct, suggesting a clinical continuum. Moreover, other conditions along the suggested continuum include optic neuritis, transverse myelitis, and neuromyelitis optica - clinical entities that may occur as manifestations of either MS or ADEM. [2] Other boundaries of ADEM merge indistinctly with a wide variety of inflammatory encephalitic and vasculitic illnesses as well as monosymptomatic, postinfectious illnesses that should remain distinctfromADEM, such as acute cerebellar ataxia (ACA). A furtherindistinct boundary is shared by ADEM and Guillain-Barré syndrome as manifested in cases of Miller-Fisher syndrome and encephalomyeloradiculoneuropathy (EMRN).

Susceptibility to either ADEM or MS is likely the product of multiple factors, including a complex interrelationship of genetics and exposure to infectious agents and other environmental factors. Of particular interest are the indications that susceptibility to either condition is in part age-related. Most cases of ADEM possibly occur as the result of an inflammatory response provoked by pre-pubertal infection with a virus, vaccine, or other infectious agent. Typically, the manifestations of ADEM occur quickly after this pre-pubertal febrile systemic illness and are monophasic. In a minority of cases, patients with ADEM experience one or two pre-pubertal recurrences followed by remission. MS, on the other hand, typically manifests as a relapsing-remitting illness in ensuing adolescence or young adulthood, a significant and unexplained latency of effect with apparent permanency of immune dysregulation. Bouts of MS occur without a febrile prodrome. Uncommonly, MS develops in pre-pubertal individuals andADEMdevelops in post-pubertal individuals. In very rare instances, individuals manifest pre-pubertal ADEM and, after long latency, MS in adolescence.



Multiple sclerosis (MS) and acute disseminated encephalomyelitis (ADEM) bear a close pathological resemblance, each resembling the pathology of experimental allergic encephalomyelitis (EAE). The prominence of perivenular round cell inflammation in either illness is a feature that is shared with many forms of encephalitis, but patchy demyelination with preservation of axon cylinders and the prominence of microglial cells in the inflammatory exudate are not.

The pathology of various developmental stages of the MS plaque is more fully characterized than the pathology of the lesions of ADEM. This is because most patients with ADEM recover completely and without apparent pathological residua. Few biopsies have been obtained or submitted to postmortem analysis. MS plaques are known to exhibit organization features, especially in the margins of active plaques that are not found in cases of ADEM. On the other hand, the general pathological similarities suggest but do not confirm the possibility that ADEM is a forme fruste of MS that is somehow effectively and permanently controlled after one, or possibly a few, demyelinating bouts.

Patients with large tumor-like demyelinating lesions may exhibit a combination of pathological features consistent with both MS and ADEM. The possible relationship between these illnesses is further supported by the similarity of clinical manifestations in either illness and the development of MS during adolescence in a small minority of patients who have had typical ADEM bouts in the first decade of life.

The pathophysiological similarities of these illnesses suggest that the immunologic constitution of susceptible individuals is in some fashion permissive of ADEM, MS, or both and that the degree of susceptibility may describe a gradient with regard to severity and risk for recurrence. The threshold for an initial bout of demyelinating illness may be determined by the combination of this immunologic constitution and the nature of a given antigenic stimulus; the likelihood of recurrence may be determined by the fertility of that constitution for persistence of immuno-dysregulation. Immuno-dysregulation in MS or ADEM may consist of responses that are inadequate, too exuberant, or the combination of both.

If a pathophysiological continuum between MS and ADEM exists, achieving better understanding of the manner in which susceptible individuals with ADEM are able to bring a monophasic or temporarily recurrent immuno-dysregulative response under permanent control is of obvious importance. Cases with characteristics that fall in the indeterminate area of this continuum, such as those that might be labeled multiphasic ADEM, represent an important challenge for accurate classification. In some of these cases, appropriately crediting the immune system with tardy but permanent compensation may be important, thus avoiding inappropriate diagnosis of MS, fraught as that is with psychosocial consequences.

The mechanisms of these demyelinating illnesses remain incompletely understood despite the extraordinary richness and complexity of immunologic abnormalities that have been identified after more than a century of clinical, pathological, and laboratory studies. Experimental observations have depended greatly on EAE, a research model that may be more pertinent to ADEM than MS.

However, the possibility of provoking spontaneously recurrent demyelination with this model further supports the concept that ADEM and MS represent a continuum. Basic studies have shown that, in the earliest stages of inflammation, both MS and ADEM are likely to be mediated by stimulated clones of T-helper cells sensitized to auto-antigens such as myelin proteins. Some studies have even identified serum autoantibodies to various myelin proteins that help to differentiate ADEM from MS. In particular, ADEM appears to be characterized by class-switched IgG autoantibodies, supporting the hypothesis of an antigen-driven immune response in ADEM cases; whereas, MS cases are characterized by serum IgM autoantibodies. [49] The complex ensuing inflammatory cascade entails the local action of cytokines and chemokines as well as lymphokine-induced chemotaxis of other cellular mediators of inflammation (eg, other T cell lines, B cells, microglia, phagocytes).

Pathogenic differences of MS and ADEM are likely to arise in part because of differences in details concerning pro-inflammatory and anti-inflammatory cytokines and chemokines. Interleukin (IL)–1beta, Il-2, IL-4, IL-5, IL-6, IL-8, IL-10, interferon (IFN)–gamma, tumor necrosis factor-alpha, and macrophage inflammatory protein-1-beta are significantly elevated in CSF compared with the CSF of controls. Granulocyte colony-stimulating factor shows a particularly striking elevation at as much as 38-fold greater concentration than is found in the CSF from control subjects. Elevations of IFN-gamma, IL-6, and IL-8 have been significantly correlated with CSF cell counts and protein concentration in individuals with ADEM. The pattern of cytokine elevation suggests that ADEM involves activation of macrophages, microglial cells, and various Th (T helper)–1 and Th2 cells. [3]

Additionally, in 2006, Franciotta et al demonstrated that adults with ADEM have higher CSF concentrations of chemokines that recruit or activate neutrophils (CXL1 and CXL7), monocytes (CCL3 and CCL5), Th1 cells (CXCL10), and Th2 cells (CCL1, CCL17, and CCL22) than healthy normal controls. [4] Moreover, ADEM-associated concentrations of certain of these neutrophils (CXL7 neutrophil activator and the CL1, CCL17, and CCL22 Th2 activators) are higher in the CSF from individuals with ADEM than those with MS. On the other hand, CSF concentrations of the chemokine CCL11 is lower in adults with MS than in the CSF from adults with ADEM or in normal controls.

CSF Th1/Th2 cytokine concentrations were not significantly different in adults with MS, those with ADEM, or in normal healthy controls. No significant differences in serum concentrations of cytokines or chemokines were noted in the 3 adult groups. These findings raise the possibility that elevated chemokine concentrations might serve as biomarkers for ADEM and that they may provide keys to understanding the nature of and differences in the pathogenesis of ADEM and MS.

Disturbance of the blood-brain barrier is likely to be an important event. The elaboration of antibodies occurs but remains of uncertain significance. In particular, multiple researchers have demonstrated the presence of serum IgG antibodies to myelin oligodendrocyte glycoprotein (MOG) in up to 40% of children with ADEM, [51, 52, 53] though these antibodies do not appear to be specific to ADEM. Still the presence of anti-MOG antibodies in ADEM may affect the nature and course of the disease. A recent study has demonstrated that MOG-positive ADEM patients are more likely to have large, bilateral and widespread lesions and longitudinally extensive transverse myelitis on MRI and are more likely to have a favorable clinical outcome when compared to MOG-negative ADEM patients. [49]




United States

The Canadian Pediatric Surveillance Program for acute demyelinating syndromes (ADS) in individuals younger than 18 years found that in 219 identified Canadian cases, ADEM represented 22% of diagnoses. This study generated an estimate of Canadian disease incidence of 0.9 per 100,000 for ADS, while cases of ADEM manifested an annual incidence of 0.2/100,000. [5] Higher incidence rates have been reported in San Diego County at 0.4/100,000 per year. [55]

Whether the increasing incidence of MS at increasing distance from the equator is also true of ADEM is unknown. The seasonal incidence of ADEM within North America peaks in the winter and spring months. [55, 54] Some severe forms of ADEM, such as those that occur in the wake of measles and the severe hemorrhagic variant called acute hemorrhagic leukoencephalopathy (AHLE) are probably less commonly encountered than they were prior to widespread immunization against measles and other formerly common and potentially serious illnesses that may serve as triggers for ADEM/AHLE.


Few studies have provided incidence data from other countries, thus little is known about occurrence throughout the world. Data from Germany quote an incidence rate of 0.07 per 100,000, [56] while data from Japan show an incidence of 0.64 per 100,000 per year. [57] Genetic factors, prevalence of infectious pathogens, immunization status, degree of skin pigmentation, diet, and other factors may influence risk.


Although older studies suggest a 10% mortality rate, the data upon which such estimates were based were obtained in epochs during which measles was prevalent, techniques for intensive care were comparatively primitive, and anti-inflammatory therapies were inadequate. Formerly, deaths occurred in patients with AHLE, a severe ADEM variant, which has become less common since children have received immunization to many common childhood illnesses.

Current acute mortality rates are probably less than 2%, typically consisting of cases with fulminant cervical transverse myelitis or brain swelling. Children younger than 2 years are particularly subject to such severe presentations. [37] See the image below.

Fatal case of ADEM involving the brainstem in a 13 Fatal case of ADEM involving the brainstem in a 13-month-old.

Morbidity chiefly includes visual, motor, autonomic, and behavioral/intellectual deficits and epilepsy. Based upon existing data, full recovery occurs in approximately 57-92% of patients. Residual focal neurologic deficits remain in 4-30%. [54, 47, 58, 60, 61, 62, 55, 63, 85] Neurocognitive deficits after acute demyelinating syndromes are receiving more attention, as even children having an apparent full recovery after ADEM have demonstrated subtle deficits on formal neuropsychological testing years after the sentinel event. [65] Additionally, these intellectual and behavioral impairments are variable depending on age of the child at the time of disease onset. There may be a greater impact upon behavior and intelligence in those with an ADEM-onset of younger than 5 years of age. [64]

Though uncommon, adult-onset ADEM, when correctly diagnosed, appears to have similar outcomes and a typically favorable prognosis. [39]


The scientifically imprecise concept of race does not lend itself readily to discussions of ADEM. In the author's (RSR) series of more than 150 cases, the ratio of light-skinned to dark-skinned individuals who have some contribution of genetic material from individuals who have left Africa in the past 5 centuries is approximately 6:1. In the former group, the element of African heritage from the past 5 centuries is presumed small but is in fact unknown. ADEM is found in all ethnic groups and races; referral bias complicates any assessment of relative prevalence.

Regardless of race, the degree of skin pigmentation directly influences vitamin D status in any given individual. [66] Several studies have implicated vitamin D deficiency as a contributing risk factor for multiple sclerosis, [70, 68, 67] though these studies have not been performed within ADEM cohorts.


Though no clear gender predominance has been identified, a handful of ADEM cohorts have reported a slightly male predominance. [5, 54, 85] In the author's (RSR) series of more than 150 cases, the ratio of boys to girls is 1.3:1. These data are in opposition to the strong female preponderance noted within multiple sclerosis.


More than 80% of childhood cases occur in patients younger than 10 years, with a mean age range of 5 to 8 years. [47, 63, 71] . Somewhat less than 20% of cases occur in the second decade of life. Incidence in adulthood is unclear, accounting for less than 3% of the reported cases; however, diagnostic overlap with MS may lead to underestimation of the prevalence in adults. [39] Adult-onset cases of particular severity are recognized upon the basis of biopsy. These cases may manifest very large white matter lesions that involve, as is the case with childhood-onset cases, the gray-white junction of forebrain.