The patients in this study were selected based on medical records kept at Henan Provincial People's Hospital of Zhengzhou University in Henan Province, People's Republic of China. Patients who were diagnosed with ADEM between January 2003 and December 2013 were reviewed retrospectively. Approval for the study was obtained from the human research ethics committees of Zhengzhou University.

Clinical records were carefully reviewed by experienced investigators (H.-Q.Y. and W.-M.Y.) in order to ensure accurate diagnoses. Demographic data, the results of clinical presentations, physical examinations, and laboratory studies, neuroimaging findings, and treatment and clinical outcomes were extracted from medical records using a standardized form.

For laboratory examinations, microbiological data were extracted from laboratory reports and progress notes in the individual records. To ascertain the diagnosis of a clinical infectious condition, the patient history, medical history, and results of physical examinations and microbiological tests were carefully reviewed. For neuroimmunology examinations, the results for oligoclonal bands (OCB) were defined as positive when they were present in CSF but absent in serum. The IgG index and 24-hour IgG synthesis rate were also measured in the CSF. Brain magnetic resonance imaging (MRI) was carried out using a uniform protocol that included the acquisition of axial T1-weighted, T2-weighted, sagittal T1-weighted, and fluid-attenuated inversion recovery (FLAIR) images. Spinal cord MRI was performed when necessary. Contrast images were obtained in some patients after administering a single dose of gadolinium. During hospitalization in the acute phase, short-term MRI was repeated when the patient exhibited clinical deterioration. MRI was also repeated whenever possible if the patient exhibited clinical improvement. A senior neuroradiologist (Y.-L.L.) who was blinded to the study design reviewed the MRI findings.

Patients who met the following criteria were included in this study:6 1) polysymptomatic, acute, or subacute onset in the clinical presentation, 2) presence of encephalopathy defined by either alterations in consciousness or behavior changes, 3) focal or multifocal lesions in MRI, predominantly involving white matter, and 4) no MRI evidence of previous destructive white-matter changes.

The following patients were excluded from the investigations performed in the present study: 1) younger than 18 years, 2) suspicious diagnosis of ADEM, 3) no CSF examination, 4) no cranial MRI scan, 5) incomplete clinical data, or 6) loss to follow-up.

An appointment for the follow-up was made prior to discharge from hospital. At the follow-up, any neurological symptoms that had developed during the intervening period were documented, and a clinical evaluation was performed. Follow-up MRI was carried out in some cases. Complete recovery was defined as total resolution of the initial clinical symptoms.

Various types of relapse were defined as described below.6

Of the 82 potential patients initially diagnosed with ADEM, 42 were included in this study after careful clinical evaluation. The 40 patients were excluded for the following reasons: younger than 18 years (n=11), suspicious diagnosis (n=7), incomplete clinical data (n=13), and loss to follow-up (n=9).

The 42 ADEM patients comprised 16 males and 26 females, giving a male:female ratio of approximately 1:1.6. The age at presentation was 33.5±8.6 years (mean±SD), and ranged from 18 to 66 years. About two-thirds (27/42) of patients were admitted in summer and autumn. No prior vaccinations were recorded. Four patients were identified as having prior upper respiratory tract infection, and one had a digestive tract infection. The mean duration of symptoms before admission was 9.6 days (range, 2–40 days), the mean time interval from initial neurological symptom presentation to maximum was 6.3 days (range, 1–15 days), and the mean length of hospital stay was 14.1 days (range, 1–34 days).

Thirty-three patients (79%) presented with alternations in consciousness such as somnolence, lethargy, and coma, while 69% of patients exhibited behavior changes such as confusion, excessive irritability, and impaired orientation. Twenty-seven patients (64%) showed motor deficits including hemiplegia, paraplegia, or single-limb weakness. Half of the patients presented with fever, which lasted from 5 to 26 days. Sensory deficits and cranial-nerve impairment was observed in 42% of patients. Seizures were recorded in nine patients (21%), with one developing status epileptics. Spinal cord involvement was reported in 20 patients (48%); the frequencies of other presenting symptoms are listed in Table 1.

Corticosteroid treatment was applied in 35 patients (83%), and immunoglobulin was injected intravenously in 7 patients. About 62% of these patients experienced full recovery, and residual focal neurological deficits occurred in five patients. The neurological sequelae were hemiparesis in two patients, and mental handicap, numbness, and visual impairment in one patient each.

Infection complication was reported in 10 patients, with the diagnoses confirmed by sputum smear and serum examinations. These infections involved the respiratory and urinary tracts in eight and two patients, respectively.

CSF findings were normal in 13 patients. Increased protein levels were recorded in 20 patients (48%), with a highest concentration of 1.39 g/L. Eighteen patients (43%) were found to have elevated levels of white blood cells, with lymphocytes being the predominant cell type. The glucose level in CSF varied from 2.16 to 7.58 mmol/L, with mean of 4.2 mmol/L. Eleven patients (26%) showed positive OCB results, while 21 patients (50%) showed an increased IgG index and 24 patients (57%) showed an increased 24-hour IgG synthesis rate. The IgG index indicated disturbance of the blood-brain barrier in 19 patients (45%). Mean serum and CSF levels of myelin basic protein and neuron-specific enolase were within the normal ranges.

Thirty-five patients (83%) presented with white-matter lesions, which were distributed in the frontal (48%), temporal (31%), parietal (24%), and occipital (24%) lobes. The lesions are frequently identified using T2-weighted and FLAIR sequences as large and patchy lesions with poorly defined margins and increased signal intensities. Most of patients showed multiple lesions and they were usually distributed asymmetrically, with variable extension to the subcortical white matter. Other frequently involved areas are listed in Table 2, and representative MRI findings are shown in Fig. 1.

The administration of gadolinium enhanced the T1-weighted sequences in 9 of 20 patients (45%), with different patterns of enhancement. The lesions were completely enhanced in five patients and partially enhanced in the other four patients. Patchy, nodular, ring-shaped, and linear enhancement types were observed in seven, six, six, and five patients, respectively.

A follow-up MRI scan was performed in 15 patients, which revealed 6 patients with completely normal findings and 9 patients with only partially resolved lesions.

The mean follow-up period was 3.4 years (range, 0.8–5.6 years), and two patients exhibited recurrence and one patient exhibited a multiphasic course. One patient was finally diagnosed with MS.

One of the patients with recurrence initially presented with somnolence and hemiplegia in the left limbs as a result of lesions in the brainstem and right temporal and occipital lobes. Six months later he presented with hemiplegia and hypoesthesia, with MRI indicating that the lesions were in the same areas as when he was first admitted.

The other patient with recurrence initially presented with fever, disorientation, urinary and fecal incontinence, and left hemiparesis. MRI showed multiple lesions in the cerebellum and the frontal and parietal lobes. Thirteen months later she visited the emergency department with lethargy, left hemiparesis, and urinary incontinence. Repeated MRI again showed multiple lesions in the cerebellum and the frontal and parietal lobes.

The patient with the multiphasic form of ADEM initially presented with fever, headache, somnolence, disorientation, and urinary incontinence. MRI showed multiple lesions of various sizes in the cerebellum, brainstem, cerebral cortex, subcortex, and upper-cervical spinal cord. The OCB results were positive. Four months later, just 40 days after discontinuing steroids therapy, she presented with ataxia, diplopia, and vision loss in the right eye. At this presentation she was slightly lethargic, confused, and denied symptoms of dysautonomia. MRI showed new lesions in the temporal and occipital lobes bilaterally. Both the size and distribution of these lesions differed from those present at the first attack. The OCB results were negative, and neuromyelitis optica (NMO)/IgG measurements were negative. She fulfilled the definitions of multiphasic disseminated encephalomyelitis.

A 34-year-old woman presented with slurred speech, dysphagia, salivation, and unresponsiveness. She also showed positive OCB results and increased intrathecal IgG synthesis. MRI showed multiple lesions in white matter of the frontal and temporal lobes bilaterally. Nine months later she presented with unilateral optic neuritis. Repeated MRI showed multiple lesions with enhancement. A CSF examination showed positive OCB results. The diagnosis of MS was suspected and she gradually recovered after the administration of steroids. At a 2-year follow-up she was relapse-free.