To the Editor: Acute necrotizing encephalopathy (ANE) is a rare but often devastating neurologic disorder in children.[1] Hereditary susceptibility and cytokine storm play important roles in ANE pathogenesis,[2,3] but the specific molecular biological mechanism is unclear. The mortality rate among children with ANE is approximately 30%, and most surviving patients are left with various degrees of neurological sequelae.[1,4]
The optimum treatment regimen for this disease has not been established. It is important for pediatricians to establish effective therapy for ANE of childhood. To date, the efficacy of intravenous immunoglobulin (IVIG) and plasma exchange has not been confirmed.[3,5] However, steroid has been used in most studies on ANE treatment. Although early steroid pulse therapy has been recommended for ANE,[5] the current evidence is insufficient. The most commonly used steroid for ANE is methylprednisolone, but varying dosages and courses are proposed in different studies. We hypothesized that different initial dosages of methylprednisolone lead to different outcomes for ANE. Accordingly, a 6-year retrospective study was conducted in four pediatric intensive care units (PICUs). We aimed to identify the appropriate initial dosage of methylprednisolone for ANE treatment to improve outcomes and prevent severe neurological sequelae.
This multicenter retrospective study was conducted from December 2014 to December 2020 at PICUs of Beijing Children's Hospital affiliated to Capital Medical University (Beijing, China), Shengjing Hospital of China Medical University (Shenyang, China), Hebei Children's Hospital affiliated to Hebei Medical University (Shijiazhuang, China), and Bao’an Maternal and Child Health Hospital (Shenzhen, China). The study was approved by the Ethics Committee of Beijing Children's Hospital affiliated to Capital Medical University (approval number: 2020-Z-145), with waiver of the requirement for informed consent.
The diagnostic criteria for ANE were as follows:[6] (1) acute encephalopathy followed by a febrile disease, with rapid deterioration of consciousness and convulsions; (2) commonly increased cerebrospinal fluid (CSF) protein concentration without pleocytosis; (3) elevated serum aminotransferases with variable degrees but no hyperammonemia; (4) cranial imaging showing symmetric and multifocal brain lesions with involvement of bilateral thalami and possible involvement of cerebral periventricular white matter, internal capsule, putamen, upper brain stem tegmentum, and cerebellar medulla, but not of other central nervous system regions; and (5) exclusion of diseases that resemble ANE. All patients aged from 1 month to 18 years diagnosed with ANE were included. Owing to the rapid progression of ANE, lumbar puncture was not conducted in a few patients. Patients who had no routine CSF results but met the other criteria, and had respiratory specimens containing influenza virus nucleic acid or antigen (ie, influenza virus was the pathogen underlying the prodromal infection) were also included. Patients who were not treated with methylprednisolone were excluded.
Data were collected by trained pediatricians using standardized data collection forms, including demographics, clinical manifestations, laboratory examinations on admission, etiological results (detection of respiratory virus antigen or nucleic acid from nasopharyngeal swab), positive initial cranial imaging results, early empirical treatment, and clinical outcomes. The primary outcome was whether the patients survived by discharge. The secondary outcomes were length of PICU stay and length of hospital stay. Follow-up was performed by telephone interview or at face-to-face outpatient appointments to evaluate the neurological function of the surviving children by using the Pediatric Cerebral Performance Category (PCPC) and Pediatric Overall Performance Category (POPC) scales.
Continuous variables are expressed as mean ± standard deviation or median (interquartile range) depending on whether the data are parametric or non-parametric. The initial dose of methylprednisolone cut-off value for low-dose and high-dose groups was determined by receiver operation characteristic curve for primary end-point. Categorical variables were expressed as frequency and percentage. Student's t test was used for parametric continuous variables and Mann–Whitney U test for non-parametric continuous variables. Categorical variables were analyzed by Fisher's exact test. Logistic regression was used to determine whether indicators that differed between the low- and high-dose groups were risk factors for death in children with ANE. A two-sided P value <0.05 was considered statistically significant. Statistical analysis was conducted with SPSS statistical software version 24 (IBM Corp., Armonk, NY, USA).
Fifty-one patients were diagnosed with ANE. Among them, nine patients without typical changes on cranial imaging, three patients without CSF results or negative for influenza virus, and three patients without using methylprednisolone were excluded. A total of 36 children were enrolled. The cut-off value of the initial dose of methylprednisolone was 12 mg·kg−1·day−1 (the area under the receiver operation characteristic curve was 0.686), but the actual initial dose of methylprednisolone – in ANE children was 20 mg · kg−1 · day−1. Thus, low- and high-dose groups were divided by 20 mg · kg−1 · day−1, namely 11 cases in high-dose group (≥20 mg · kg−1 · day−1) and 25 cases in low-dose group (<20 mg · kg−1 · day−1).
The baseline parameters including age, gender, season of onset, clinical symptoms, Glasgow Coma Scale score, and ANE severity score on admission, which were similar in the two groups [Supplementary Table 1, https://links.lww.com/CM9/B389]. The presence of an underlying disease and family history of encephalopathy occurred in zero patients in the high-dose group, and only one patient each in the low-dose group. The median duration from the onset of consciousness disturbance to admission was 24 h in both groups. The clinical features and severity of ANE on admission were similar in the two groups.
The median blood inflammatory biomarkers on admission, including C-reactive protein (CRP), procalcitonin (PCT), white blood cell count, and neutrophil count, were higher in the low-dose group than in the high-dose group, and only CRP was significant (19.0 vs. 8.0 mg/L, P = 0.017) [Supplementary Table 2, https://links.lww.com/CM9/B389]. However, most patients had elevated CRP (64%, 23/36; reference: ≤8 mg/L) and elevated PCT (86%, 31/36; reference: ≤0.25 ng/mL) on admission. The level of PCT on admission was ≥5 ng/mL for 25 patients (70%, 25/36). Platelet count, blood ammonia, serum transaminases, and CSF cell count and protein concentration on admission did not significantly differ between the two groups [Supplementary Table 2, https://links.lww.com/CM9/B389]. With regard to the discriminatory power of CRP, no significant difference existed between the patients who died or survived at discharge, as indicated by logistic regression (odds ratio [OR] = 0.991, 95% CI: 0.969–1.013, P = 0.397).
Only 20 patients had completed the examination results at admission and 3 days after treatment (13 in the low-dose group and seven in the high-dose group). No significant differences existed in PCT, white blood cell count, or neutrophil count between the two checking time points in either. The CRP level was significantly lower at 3 days after treatment than on admission in the low-dose group (26.0 vs. 5.0 mg/L, P = 0.023). CRP and PCT showed a downward trend by day 3 after treatment in both groups [Supplementary Table 3, https://links.lww.com/CM9/B389].
A pathogen was detected in the majority of cases (64%, 23/36). In most cases, it was influenza virus (83%, 19/23; 9 influenza A and 10 influenza B virus cases). Cranial imaging revealed symmetrical bilateral thalamus involvement in all patients, and no significant difference existed in brainstem injury between the two groups [Supplementary Table 2, https://links.lww.com/CM9/B389].
The major treatments for ANE are shown in Supplementary Table 2, https://links.lww.com/CM9/B389. The first dose of methylprednisolone was administered within 24 h after PICU admission, and the median duration of the initial dose of methylprednisolone was 3 days in both groups. Most patients (80.6%, 29/36) required mechanical ventilation. The median duration of mechanical ventilation in the low-dose group was higher than that in the high-dose group, but the difference was not significant (48.0 h vs. 30.0 h, P = 0.248). Antiviral agents (94%, 34/36), IVIG (92%, 33/36), and plasma exchange (33%, 12/36) were also used to treat ANE, but no significant differences were found between the two groups. The antiviral agents were peramivir or oseltamivir (68%, 23/34), acyclovir (24%, 8/34), and vidarabine (9%, 3/34). The first doses of antiviral agents and IVIG (total 2 g/kg for 2–5 days) were administered within 24 h after PICU admission, and the first application of plasma exchange ≤24 h after admission in 10 cases. However, two cases received plasma exchange at 48 to 72 h after admission because they responded poorly to immunomodulation therapy. Among the 12 patients who received plasma exchange, nine of them also received methylprednisolone and IVIG, and the other three patients who did not receive IVIG only received methylprednisolone.
The mortality rate of ANE at discharge was 36% (13/36), which was significantly lower in the high-dose group than the low-dose group (9% vs. 48%, P = 0.031) [Supplementary Table 2, https://links.lww.com/CM9/B389]. Logistic regression indicated that high-dose methylprednisolone may reduce the risk of mortality rate at discharge in ANE (OR = 9.231, 95% CI: 1.023–83.331, P = 0.048). The median lengths of PICU and hospital stay were longer in the high-dose group than the low-dose group, but no significant differences were found.
At the follow-up point in December 2020, seven of the patients (30%, 7/23) who survived by discharge and received palliative treatment finally died [Supplementary Table 4, https://links.lww.com/CM9/B389]. Four patients were lost to follow-up (three in the low-dose group and one in the high-dose group). The mortality rate by December 2020 was 36% (4/11) in the high-dose group, which was much lower than in the low-dose group (64%, 16/25). The follow-up duration was 2 to 60 months, and the median was 27 months. Although the median follow-up duration of patients was non-significantly lower in the high-dose group than in the low-dose group (24 months vs. 42 months) and the median POPC/PCPC was non-significantly higher (3.0 vs. 2.5), the POPC/PCPC tended to be lower [Supplementary Table 4, https://links.lww.com/CM9/B389]. According to the PCPC and POPC, two patients (18%, 2/11) in the high-dose group and one (4%, 1/25) in the low-dose group fully recovered.
Steroids are remarkable contributors to the treatment of ANE. In the present study, we found that high-dose methylprednisolone pulse therapy (initial dose ≥20 mg · kg−1 · day−1) may improve the outcomes of ANE. As Bashiri et al[4] reported, 12 ANE patients from five tertiary centers reported that early use of methylprednisolone pulse therapy (30 mg · kg−1 · day−1 for 3 days) and IVIG may affect outcomes. As recommended,[5] corticosteroids are expected to improve the prognosis of ANE. But none of these reports determined the appropriate steroid pulse dosage. In this study, the median duration of the initial dose was 3 days, and the initial dose of methylprednisolone pulse therapy was lower than those in previous reports. However, the mortality rate of ANE patients at discharge was similar to that in other studies. Thus, high-dose methylprednisolone pulse therapy may improve ANE outcomes.
The mechanism underlying the effectiveness of steroids for treating ANE remains unclear. Proinflammatory cytokines are related to ANE pathogenesis. Although no significant difference in PCT was found between the two groups, most patients had elevated PCT. For the patients with influenza-related neurological complications, serum PCT >4.25 ng/mL might be useful as an early indicator of ANE.[7] Various pro-inflammatory cytokines can promote PCT expression. Glucocorticoids bind to their receptors and regulate specific genes to affect the expression of several cytokines, but further investigations are needed.
In the future, randomized control trials are the most suitable method to obtain robust evidence of the therapeutic effects of methylprednisolone pulse therapy. In the present study, the small sample size and confounding factors may have some impact on the results. Four patients were lost to follow-up, and the follow-up of surviving patients is still ongoing.
In conclusion, this study showed that high-dose methylprednisolone pulse therapy (initial dose ≥20 mg · kg−1 · day−1) may improve the outcomes of ANE.
Funding
This study was supported by a grant from Capital's Funds for Health Improvement and Research of Beijing Children's Hospital, Capital Medical University (No. 2020-2-2094).
Conflicts of interest
None.
References
1. Mizuguchi M, Abe J, Mikkaichi K, Noma S, Yoshida K, Yamanaka T, et al. Acute necrotising encephalopathy of childhood: a new syndrome presenting with multifocal, symmetric brain lesions. J Neurol Neurosurg Psychiatry 1995;58:555–561. doi: 10.1136/jnnp.58.5.555.
2. Neilson DE, Adams MD, Orr CM, Schelling DK, Eiben RM, Kerr DS, et al. Infection-triggered familial or recurrent cases of acute necrotizing encephalopathy caused by mutations in a component of the nuclear pore, RANBP2. Am J Hum Genet 2009;84:44–51. doi: 10.1016/j.ajhg.2008.12.009.
3. Li KC, Zhang T, Liu G, Jin P, Wang YQ, Wang LJ, et al. Plasma exchange therapy for acute necrotizing encephalopathy of childhood. Pediatr Investig 2021;5:99–105. doi: 10.1002/ped4.12280.
4. Bashiri FA, Al Johani S, Hamad MH, Kentab AY, Alwadei AH, Hundallah K, et al. Acute necrotizing encephalopathy of childhood: a multicenter experience in Saudi Arabia. Front Pediatr 2020;8:526. doi: 10.3389/fped.2020.00526.
5. Mizuguchi M, Ichiyama T, Imataka G, Okumura A, Goto T, Sakuma H, et al. Guidelines for the diagnosis and treatment of acute encephalopathy in childhood. Brain Dev 2021;43:2–31. doi: 10.1016/j.braindev.2020.08.001.
6. Hoshino A, Saitoh M, Oka A, Okumura A, Kubota M, Saito Y, et al. Epidemiology of acute encephalopathy in Japan, with emphasis on the association of viruses and syndromes. Brain Dev 2012;34:337–343. doi: 10.1016/j.braindev.2011.07.012.
7. Song YL, Li SY, Xiao WQ, Shen J, Ma WC, Wang Q, et al. Influenza-associated encephalopathy and acute necrotizing encephalopathy in children: a retrospective single-center study. Med Sci Monit 2021;27:e928374. doi: 10.12659/MSM.928374.
