Severe Pediatric Neurological Manifestations With SARS-CoV-2 or MIS-C Hospitalization and New Morbidity

Key Points Question Are severe pediatric neurological manifestations during a SARS-CoV-2–related hospitalization associated with new neurocognitive or functional morbidities? Findings In this cohort study of 3568 patients younger than 18 years, hospitalized with acute SARS-CoV-2 or multisystem inflammatory syndrome in children, severe neurological manifestations, including acute encephalopathy, seizures or status epilepticus, and delirium, were common and were associated with new neurocognitive or functional morbidity at hospital discharge. Meaning These findings suggest that patients under 18 years with a SARS-CoV-2–related hospitalization who experience severe neurological manifestations should be evaluated for new morbidity.


Introduction
Most children infected with SARS-CoV-2 do not require hospitalization, and mortality rates are low. 1,24][5][6][7][8][9] Children with the SARS-CoV-2-related condition multisystem inflammatory syndrome in children (MIS-C) often require critical care due to multisystem organ dysfunction, with high survival rates 10,11 but with risk of post-critical illness sequelae.In particular, neurological manifestations of pediatric SARS-CoV-2-related conditions have been associated with morbidity and mortality in prior studies. 12,133][14][15][16][17] In our group's multicenter preliminary report of neurological manifestations in hospitalized individuals younger than 18 years with SARS-CoV-2related conditions, we found that 44% had at least 1 neurological manifestation (40% in patients with acute SARS-CoV-2 and 66% in patients with MIS-C). 12The most common symptoms were headache (16% in acute SARS-CoV-2 and 47% in MIS-C), which may be nonspecific and challenging to evaluate in younger children, and acute encephalopathy or altered mental status (15% in acute SARS-CoV-2 and 22% in MIS-C).To date, few studies have evaluated the potential association of neurological manifestations with outcomes, and only a limited number of symptoms and conditions have been considered. 17 performed a secondary analysis of an international, multicenter cohort study 12 to describe severe neurological manifestations and analyze their association with new neurocognitive or functional morbidity at hospital discharge.We hypothesized that the occurrence of severe neurological manifestations during acute SARS-CoV-2 or MIS-C hospitalization would be associated with new neurocognitive and/or functional morbidity at hospital discharge compared with individuals under 18 years without severe neurological manifestation.An association between severe neurological manifestations and new neurocognitive and/or functional morbidity may suggest severe neurological manifestations as an independent predictor of poor outcomes in hospitalized young patients with acute SARS-CoV-2 or MIS-C and therefore could be used to guide postdischarge management.

Study Population
GCS-NeuroCOVID was a prospective, multicenter, international cohort study performed in 46 centers across 10 countries.Patients under the age of 18 years who were admitted with acute SARS-CoV-2-related MIS-C were included, and only those previously enrolled were excluded.As described in the initial GCS-NeuroCOVID, 12 12 Patients were enrolled between January 2, 2020, and July 31, 2021.The enrollment period was extended beyond the preliminary report for the original sites from May 1, 2021, to July 31, 2021, and additional sites were added for the entire reporting period. 12Local regulatory approval was obtained at each study site, and the need for informed consent was determined by each site's regulatory authority.The University of Pittsburgh Institutional Review Board was the main study site, and it approved the study.The data coordinating center that received and analyzed data was at the University of Pittsburgh.This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Data Collection
Data were collected from each study site using a case report form with common data elements and a data dictionary as previously described. 12Demographic, illness, laboratory, imaging, and outcomes data were collected.Race and ethnicity data, as defined by each site and reported in the medical records, were collected because of reported associations with COVID-19 outcomes. 18Race categories included Black, White, and other (including American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and all other races), and ethnicity categories included Hispanic or Latino and non-Hispanic or non-Latino.
Neurological manifestations were previously defined and included a broad spectrum of symptoms from headache and malaise to encephalopathy and coma. 12,19For this study, severe neurological manifestations were defined during the design phase and before the primary analysis and were approved by expert opinion including consensus (by C.F., A.M.A., C.L.R., M.S.W., J.D.R., M.E.S., and E.L.F.) and evidence of association with worse outcomes in both SARS-CoV-2-related conditions and other acute pediatric illnesses. 13,20The final definition of severe manifestations included acute encephalopathy (altered mental status, lethargy, or drowsiness), seizures or status epilepticus, meningitis or encephalitis, sympathetic storming or dysautonomia, hypoxic ischemic brain injury secondary to cardiac arrest, coma, delirium, and stroke.These terms were indicated on the primary case report form.Laboratory values, including liver enzymes, inflammatory markers (eg, interleukin 6, C-reactive protein, ferritin, and procalcitonin), complete blood count, and electrolytes, were collected.

Outcomes
The primary outcome was new neurocognitive and/or functional morbidity at hospital discharge.
Neurocognitive morbidity was as defined as a Pediatric Cerebral Performance Category (PCPC) score of 3 or more with a change of 1 or more points from the prehospitalization baseline. 21Scores for the PCPC range from 1 to 6 (1 indicates normal; 2, mild disability; 3, moderate disability; 4, severe disability; and 5, for vegetative state or coma-6 indicates death, which was not included in the baseline scores but was possible after hospitalization and was used to define unfavorable outcomes).
Functional morbidity was defined as a change in the Functional Status Scale score of 1 or more between baseline and hospital discharge scores. 22The Functional Status Scale is based on 6 domains of functioning, and each domain ranges from 1 to 5. The final score is a cumulative score of all 6 domains, with higher scores indicating more severe dysfunction.Secondary outcomes included hospital mortality and new neurocognitive morbidity alone or new functional morbidity alone (rather than a combined outcome).

JAMA Network Open | Pediatrics
Severe Pediatric Neurological Manifestations With Acute SARS-CoV-2 Hospitalization and Morbidities

Statistical Analysis
Data were analyzed using Stata, version 18 (StataCorp LLC).Analyses were performed overall and by acute SARS-CoV-2 or MIS-C group, given the distinct differences in pathophysiology between these disease presentations. 11Categorical data were summarized as frequencies using χ 2 tests and continuous data as median with IQR using Mann-Whitney tests.Multivariable logistic regression modeling was performed to evaluate the independent association of severe neurological manifestations during hospital admission with the presence of new neurocognitive and/or functional morbidity at hospital discharge for survivors.This regression analysis was stratified on and conducted within SARS-CoV-2 and MIS-C cohorts, and potential risk factors were considered separately in these populations.
We included age and sex a priori as potential risk factors for model building.The remaining demographic and clinical variables associated with new neurocognitive or functional morbidity at hospital discharge on univariable logistic regression with a P value <.20 were included as potential risk factors in model building.The final model was determined using backward stepwise elimination with a P value cutoff of .20 for removal from the model.A 2-sided P < .05 was considered statistically significant.Missing data were not imputed.

Patient Characteristics by Acute SARS-CoV-2 and MIS-C Group
Most patients included in this study had acute SARS-CoV-2 (83.5% [n = 2980]), and the remainder had MIS-C (588 [16.5%]).Among the patients with acute SARS-CoV-2, 536 (18.0%) had a severe neurological manifestation during hospitalization, as did 146 (24.8%) of the patients with MIS-C.In patients with acute SARS-CoV-2, those with severe neurological manifestations were less likely to be Hispanic or Latino (26.7% vs 30.4%), more likely to have more than 1 preexisting comorbidity (46.6% vs 32.5%), and more likely to have a neurological preexisting condition (45.7% vs 16.3%) compared with those without severe neurological manifestations (eTable 1 in Supplement 1).In patients with MIS-C, those with severe neurological manifestations were more likely to have a preexisting neurological condition (11.6% vs 5.9%) compared with those without severe neurological manifestations (Table 1).

SARS-CoV-2 Group
Of patients with severe neurological manifestations in acute SARS-CoV-2, 24 (4.8%) died within the hospitalization compared with 7 (0.3%) of those without severe neurological manifestations.and language therapy, and rehabilitation consults during hospitalization compared with those without severe neurological manifestations (Table 2).

MIS-C Group
Among patients with MIS-C, those with severe neurological manifestations had higher in-hospital mortality (4.9% [n = 7]) compared with those without severe neurological manifestations (0.5% [n = 2]) (P < .001).Among survivors with MIS-C, 28.0% (n = 39) of those with severe neurological manifestations had new neurocognitive and/or functional morbidity at hospital discharge compared with 15.5% (n = 68) of those without severe neurological manifestations (P = .002).New neurocognitive morbidity was identified in 16.5% (n = 23) vs 7.5% (n = 33) (P = .003)and new functional morbidity in 23.0% (n = 32) vs 14.3% (n = 63) (P = .03) of those with severe neurological manifestations compared with those without severe neurological manifestations.Patients with MIS-C who had severe neurological manifestations were more likely to receive physical therapy, occupational therapy, speech and language therapy, and rehabilitation consults during hospitalization compared with those without (Table 2).

Laboratory Values and Neurodiagnostics by Acute SARS-CoV-2 and MIS-C Group
Patients with severe neurological manifestations had lower median (IQR) platelet counts on admission than did those without severe neurological manifestations for both the acute SARS-

Association Between Severe Neurological Manifestations and New Neurocognitive and/or Functional Morbidity at Discharge
For patients admitted with acute SARS-CoV-2 who survived the admission, those with severe neurological manifestations had higher odds of having new neurocognitive and/or functional morbidity at hospital discharge (odds ratio, 1.85 [95% CI, 1.27-2.70];P = .001),adjusting for race, neurological comorbidity, corticosteroid administration, pandemic epoch, lymphocyte counts, and initial sodium levels (Table 3).Similarly, for surviving patients with MIS-C, those with severe neurological manifestations had higher odds of having new neurocognitive and/or functional morbidity at hospital discharge (odds ratio, 2.18 [95% CI, 1.22-3.89];P = .009),adjusting for race, baseline PCPC score, IVIG administration, and initial sodium levels (Table 4).

Discussion
This international, multicenter cohort study of hospitalized individuals younger than 18 years with SARS-CoV-2-related conditions found that severe neurological manifestations were common in those admitted with acute SARS-CoV-2 (18.0%) or MIS-C (24.8%).Acute encephalopathy, seizure or status epilepticus, and delirium were among the most common severe manifestations in patients with both acute SARS-CoV-2 and MIS-C.For those with acute SARS-CoV-2 or MIS-C, the occurrence of a severe manifestation was associated with increased odds of new neurocognitive and/or functional morbidity at hospital discharge.The individuals with severe manifestations were also more likely to receive inpatient services including physical and occupational therapy and rehabilitation consults.These patients may be at risk for ongoing postdischarge sequelae.
Central nervous system involvement in acute SARS-CoV-2 or MIS-C carries potential for both shortterm and long-term morbidities.Young persons who are hospitalized and critically ill can have significant morbidity after hospital discharge, [23][24][25] including physical 26 and neurocognitive and/or psychosocial dysfunctions. 25In our study, patients with severe neurological manifestations were more likely to have new neurocognitive and functional morbidities at hospital discharge.These patients were also more likely to receive physical therapy, occupational therapy, speech and language therapy, and rehabilitation consults during hospitalization.The finding that patients with severe neurologic manifesations were more likely to have new neurocognitive and functional morbidities at hospital discharge parallels observations in other pediatric critical illnesses.Among children with sepsis, the presence of neurological dysfunction is associated with greater odds of death or new moderate disability at discharge in survivors. 27Identifying the patients at the highest risk of new and persistent impairment is essential to direct them to the follow-up care necessary to best support their recovery and adaptation.Low platelet count, for example, found in both of the groups with severe neurological manifestations, may be a marker for disease severity in both acute SARS-CoV-2 28,29 and MIS-C 30,31 and therefore may serve as a flag for patients deserving scrutiny acutely and in follow-up.Despite strong agreement about the importance of postdischarge follow-up care of children who are critically ill, few systematic, multidisciplinary post-pediatric intensive care unit (ICU) follow-up programs exist, 32 often due to logistical and reimbursement challenges.By identifying cohorts of patients at high risk of postdischarge sequelae, future research could demonstrate the value of these programs and their potential effects on improving long-term outcomes.
For both acute SARS-CoV-2 and MIS-C, patients with severe neurological manifestations were more likely to have a preexisting neurological condition.These findings are in agreement with a national study in the United Kingdom which found that children with medical comorbidities and/or neurodisability were at higher risk of pediatric ICU admission with severe COVID-19. 33Similarly, there is an association between short-term and long-term outcomes in patients with preexisting neurological conditions following ICU discharge for other illnesses.For example, children with neurodisability are at increased risk of ICU delirium, [34][35][36] and children with ICU delirium are at increased risk of postdischarge decline in health-related quality of life. 37,38Preexisting neurological conditions are also a risk factor for the development of severe neurological manifestations of other virus-induced acute pediatric disease processes.0][41] Interestingly, in our MIS-C cohort, it was more common for patients without known comorbidities to have severe neurological manifestations.Some data do show, however, that despite initial severe illness, patients younger than 18 years with MIS-C have minimal functional impairment at 6-month follow-up. 42 acute SARS-CoV-2 was classified as either confirmed (a positive SARS-CoV-2 test result) or presumed (a clinical diagnosis in the setting of clinical suspicion and/or close contact) without a diagnosis of MIS-C.An MIS-C diagnosis was classified by the Centers for Disease Control and Prevention definition.

JAMA Network Open | Pediatrics Severe
Survivors with acute SARS-CoV-2 who experienced severe neurological manifestations were more likely to have new neurocognitive and/or functional morbidity at hospital discharge compared with Pediatric Neurological Manifestations With Acute SARS-CoV-2 Hospitalization and Morbidities

Table 1 .
Clinical Characteristics of Patients With and Without Severe Neurological Manifestations by Acute SARS-CoV-2 and MIS-C Group a c Other includes American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and all other races.

Table 2 .
Hospital Outcomes for Patients With and Without Severe Neurological Manifestations by Acute SARS-CoV-2 and MIS-C Group

Table 3 .
Multivariable Logistic Regression Model for New Neurocognitive and/or Functional Morbidity Based on the Presence of Severe Neurological Manifestations With Acute SARS-CoV-2

Table 4 .
Multivariable Logistic Regression Model for New Neurocognitive or Functional Morbidity Based on the Presence of Severe Neurological Manifestations With MIS-C Our findings highlight the risk of new neurological manifestations among individuals in this age group with preexisting neurological conditions who present with either acute SARS-CoV-2 or MIS-C.Future studies of patients with acute SARS-CoV-2 and MIS-C, and likely those with other acute viral illnesses as well, should include and comprehensively evaluate those with preexisting neurological conditions for both in-hospital and postdischarge outcomes.Severe Pediatric Neurological Manifestations With Acute SARS-CoV-2 Hospitalization and Morbidities submitted work.Dr Chang reported receiving grants from the NIH during the conduct of the study.Dr Fink reported receiving grants from the NIH and personal fees from the American Board of Pediatrics outside the submitted work.eTable 2. Laboratory, Neurotesting, and Treatment Variables of Patients With and Without Severe Neurologic Manifestations by Acute SARS-CoV-2 and MIS-C Groups