Terrence Thomas

Aim: To develop the Pediatric Autoimmune encephalitis Severity Scale (PASS) using expert consensus and the Delphi process, and validate it in children with autoimmune encephalitis. Method: This prospective observational study enrolled children who underwent serial rating with PASS, the Clinical Assessment Scale in Autoimmune Encephalitis (CASE), and the modified Rankin Scale, at eight time points until 6 months of presentation. Clinical outcomes included the Pediatric Quality of Life Inventory (PedsQL) score (parent‐rated) and developmental and intellectual outcome at 6 months. Interrater agreement, internal consistency, and correlations of PASS and CASE with clinical outcomes were compared. Results: There were 361 assessments in 27 children (11 females, 16 males) with autoimmune encephalitis (15 with antibody‐negative autoimmune encephalitis, 11 with N ‐methyl‐ D ‐aspartate receptor antibody encephalitis, and one with steroid‐responsive encephalopathy associated with autoimmune thyroiditis) with a median age at onset of 4 years (interquartile range = 3 years to 6 years 5 months). There was excellent interrater agreement for PASS and CASE with comparable intraclass correlation coefficients (0.984 vs 0.983). The worst PASS score was inversely correlated with the final PedsQL score ( r = −0.42, p = 0.047). Factor analysis revealed distinct ‘cognitive and behavioural’ and ‘disinhibition and excitability’ dimensions in the PASS items. Interpretation: PASS is a severity assessment tool for autoimmune encephalitis in children and offers an alternative to adult scales, which may be difficult to use in children.

Aim: To evaluate the timing and safety of tocilizumab in acute necrotizing encephalopathy (ANE). Method: This was a multicentre retrospective study evaluating outcomes (modified Rankin Scale [mRS]) in children with ANE treated with tocilizumab. A logistic regression model determined a cut‐off for outcomes (binary good [mRS 0–2] vs poor [mRS 3–5]) according to day of tocilizumab administration, enabling a dichotomy of ‘early’ and ‘late’ treatment groups. Expected outcomes according to ANE severity score (ANE‐SS) in early‐treatment and late‐treatment groups were compared against reference data (conventional therapy, no tocilizumab) for outcomes according to risk category (medium risk: ANE‐SS score 2–4; high risk: ANE‐SS score ≥5). Results: From 2018 to 2023, 63 children (mean age 6 years 0 months [range: 8 months–15 years 4 months]; 33 males, 30 females) were treated with tocilizumab, alongside corticosteroids in all and intravenous immunoglobulin in 33 (52.4%). Logistic regression defined an optimal treatment window of ≤3.3 days. Compared to the reference data, early tocilizumab (≤ 72 hours after onset, n = 49) was associated with good outcomes in both high‐risk (OR = 11.43, 95% confidence interval [CI] = 1.64–127.4, p = 0.01) and medium‐risk (OR = 12.80, 95% CI = 2.15–67.8; p = 0.006) ANE‐SS categories. No tocilizumab‐related adverse events were reported. Interpretation: Early tocilizumab treatment was associated with improved outcomes in severe ANE, with no adverse events, compared to conventional therapy alone.

Aim: We report the epidemiology, treatment and outcomes of acute flaccid myelitis (AFM), including the first two cases of enterovirus (EV) D68 myelitis, in Singapore.Methods: Retrospective observational study from a paediatric hospital in Singapore, from January 2012 to December 2022. Clinical, laboratory, neuroimaging and outcome data were analysed.Results: Eight patients were identified (all male). Median age at presentation was 3 (interquartile range (IQR) 2.2–10.0) years. Seven (87.5%) patients were clustered in two time periods (July 2018–January 2019 and April–August 2022). Presenting features were a preceding febrile respiratory illness (n = 6) or hand‐foot‐and‐mouth disease (n = 2), upper limb weakness (n = 5, 62.5%) and neurogenic bladder (3, 37.5%), while the spinal cord lesions were predominantly in the cervical region (7, 87.5%). Five (62.5%) and six (75.0%) patients had elevated cerebrospinal fluid (CSF) white cells (median 7.5/mm³ (IQR 2.8–40.3)) and protein (median 0.6 g/L (IQR 0.4–0.7)), respectively. CSF was negative for pathogens. Two (50%) of four patients with EV detected in respiratory/rectal swabs had EVD68 (2022 cluster). All received intravenous methylprednisolone, six (75%) had additional intravenous immunoglobulin and either plasma exchange therapy (n = 1) or intravenous tocilizumab (n = 1). Median modified Rankin Scale (mRS) at acute illness was 4 (IQR grades 3–5), with an improvement (median 2 (IQR 1.8–2.3) mRS grades) on follow‐up (median duration 3.7 (IQR 1.4–4.1) years). One patient (12.5%) had a full recovery and seven (87.5%) have moderate disability (mRS 2–3).Conclusion: Disability risk in AFM is high despite aggressive immunotherapy. We report the first two confirmed cases of EV D68 AFM in Singapore and Southeast Asia.

We characterize clinical and neuroimaging features of SARS-CoV-2-related acute necrotizing encephalopathy (ANE). Systematic review of English language publications in PubMed and reference lists between January 1, 2020, and June 30, 2023, in accordance with PRISMA guidelines. Patients with SARS-CoV-2 infection who fulfilled diagnostic criteria for sporadic and genetic ANE were included. From 899 articles, 20 cases (17 single case reports and 3 additional cases) were curated for review (50% female; 8 were children). Associated COVID-19 illnesses were febrile upper respiratory tract infections in children while adults had pneumonia (45.6%) and myocarditis (8.2%). Children had early neurologic deterioration (median day 2 in children vs day 4 in adults), seizures (5 (62.5%) children vs 3 of 9 (33.3%) adults), and motor abnormalities (6 of 7 (85.7%) children vs 3 of 7 (42.9%) adults). Eight of 12 (66.7%) adults and 4 (50.0%) children had high-risk ANE scores. Five (62.5%) children and 12 (66.7%) adults had brain lesions bilaterally and symmetrically in the putamina, external capsules, insula cortex, or medial temporal lobes, in addition to typical thalamic lesions of ANE. Hypotension was only seen in adults (30%). Hematologic derangements were common: lymphopenia (66.7%), coagulopathy (60.0%), or elevated D-dimers (100%), C-reactive protein (91.7%), and ferritin (62.5%). A pathogenic heterozygous c/.1754 C>T variant in RANBP2 was present in 2 children: one known to have this before SARS-CoV-2 infection, and a patient tested because the SARS-CoV-2 infection was the second encephalopathic illness. Three other children with no prior encephalopathy or family history of encephalopathy were negative for this variant. Fifteen (75%) received immunotherapy (with IV methylprednisolone, immunoglobulins, tocilizumab, or plasma exchange): 6 (40.0%) with monotherapy and 9 (60.0%) had combination therapy. Deaths were in 8 of 17 with data (47.1%): a 2-month-old male infant and 7 adults (87.5%) of median age 56 years (33-70 years), 4 of whom did not receive immunotherapy. Children and adults with SARS-CoV-2 ANE have similar clinical features and neuroimaging characteristics. Mortality is high, predominantly in patients not receiving immunotherapy and at the extremes of age.

Objective To create an international consensus treatment recommendation for pediatric NMDA receptor antibody encephalitis (NMDARE). Methods After selection of a panel of 27 experts with representation from all continents, a 2-step Delphi method was adopted to develop consensus on relevant treatment regimens and statements, along with key definitions in pediatric NMDARE (disease severity, failure to improve, and relapse). Finally, an online face-to-face meeting was held to reach consensus (defined as ≥75% agreement). Results Corticosteroids are recommended in all children with NMDARE (pulsed IV preferred), with additional IV immunoglobulin or plasma exchange in severe patients. Prolonged first-line immunotherapy can be offered for up to 3–12 months (oral corticosteroids or monthly IV corticosteroids/immunoglobulin), dependent on disease severity. Second-line treatments are recommended for cases refractory to first-line therapies (rituximab preferred over cyclophosphamide) and should be considered about 2 weeks after first-line initiation. Further immunotherapies for refractory disease 1-3 months after second-line initiation include another second-line treatment (such as cyclophosphamide) and escalation to tocilizumab. Maintenance immune suppression beyond 6 months (such as rituximab redosing or mycophenolate mofetil) is generally not required, except for patients with a more severe course or prolonged impairments and hospitalization. For patients with relapsing disease, second-line and prolonged maintenance therapy should be considered. The treatment of NMDARE following herpes simplex encephalitis should be similar to idiopathic NMDARE. Broad guidance is provided for the total treatment duration (first line, second line, and maintenance), which is dictated by the severity and clinical course (i.e., median 3, 9 and 18 months in the best, average, and worst responders, respectively). Recommendations on the timing of oncologic searches are provided. Conclusion These international consensus recommendations for the management of pediatric NMDARE aim to standardize the treatment and provide practical guidance for clinicians, rather than absolute rules. A similar recommendation could be applicable to adult patients.

Introduction: Fibrocartilaginous embolism and spinal cord infarction may resemble transverse myelitis with antecedent minor trauma (sporting activity or minor falls) or with hyperacute (<12-hour) presentation. Methods: Diagnostic criteria for fibrocartilaginous embolism and spinal cord infarction were applied to a 10-year (2007-2016) cohort of children aged 1 month to 16 years with transverse myelitis and clinical, laboratory, neuroimaging, and outcome data compared between those with and without antecedent minor trauma. Results: Thirty-two children of median age 8.9 (range 2.7-15.8) years were included; 19 (59%) were female. Falls at home, school, or play (6 children, 60%), swimming (2, 20%), physical education (1, 10%), and caning (1, 10%) were antecedent events in 10 (31%) children. Six (19%) had hyperacute presentations. One patient met spinal cord infarction criteria; none had fibrocartilaginous embolism. Children with transverse myelitis and antecedent minor trauma had single, short spinal cord lesions (median 3 vertebral bodies) but without a specific neuroimaging lesion pattern. None had intervertebral disc abnormalities or brain involvement and were negative for myelin oligodendrocyte and aquaporin 4 antibodies. Twenty-five (86%) of 29 had cerebrospinal fluid inflammation, and 30 (94%) received immunotherapy. Thirty (97%) were followed for a median of 3.6 (0.1-10.2) years, with good outcome (modified Rankin Scale score 0-1) in the majority (80%). Four (75%) with hyperacute presentation had a good outcome (modified Rankin Scale score 0-1), but the patient with spinal cord infarction was the most disabled (modified Rankin Scale score 4). Conclusion: Minor trauma or hyperacute presentations does not always indicate fibrocartilaginous embolism or spinal cord infarction. Children with minor trauma preceding transverse myelitis have a distinct clinicoradiologic syndrome, with good outcome following immunotherapy.

Objective To test the utility and diagnostic yield of a medical-exome gene panel for identifying pathogenic variants in Mendelian disorders. Methods Next-generation sequencing was performed with the TruSight One gene panel (targeting 4813 genes) followed by MiSeq sequencing on 216 patients who presented with suspected genetic disorders as assessed by their attending physicians. Results There were 56 pathogenic and 36 likely pathogenic variants across 57 genes identified in 87 patients. Causal mutations were more likely to be truncating and from patients with a prior clinical diagnosis. Another 18 promising variants need further evaluation for more evidence to meet the requirement for potential upgrade to pathogenic. Forty-five of the 92 clinically significant variants were novel. Conclusion The 40.3% positive yield compares favourably with similar studies using either this panel or whole exome sequencing, demonstrating that large gene panels could be a good alternative to whole exome sequencing for quick genetic confirmation of Mendelian disorders.