Distinguishing Features of Mis-C to Other Paediatric Febrile Diseases in the Acute Setting

Background Multisystem inflammatory syndrome in children (MIS-C) presents with fever, shock, rash, abdominal pain and raised inflammatory markers, as well as common features of inflammatory childhood illnesses. In the acute setting, especially in countries where infectious diseases are common differential diagnoses, it is challenging to diagnose MIS-C. Therefore, data differentiating MIS-C from other inflammatory and/or febrile diseases at presentation is needed. Methods Prospective data was collected from children admitted to the Red Cross War Memorial Children's Hospital in Cape Town, South Africa from May 2020 to end November 2021 where MIS-C was part of their differential diagnoses. Clinical features on the day of admission were compared between children with confirmed MIS-C (MIS-C+) and those with alternate diagnoses (MIS-C-). Results In this time period, 60 children were MIS-C+ and 34 were MIS-C-. There was no significant difference in age (p = 0.321), sex (p = 0.525), ethnicity (p = 0.279), or in the frequency of comorbidities (p = 0.151) between the two groups. The presence of conjunctivitis (OR=8.12), rash (OR=8.67), tachycardia (OR=2.8) and oral mucositis (OR=3.75) was associated with MIS-C+ while abdominal pain and hypotension were not. MIS-C+ had statistically higher median C-reactive protein (CRP), pro-brain natriuretic protein (pro-BNP) and ferritin, and lower median lymphocyte count, platelet count and sodium levels than MISC-. Ferritin discriminated MIS-C+ well (AUC=0.86) with a 94% sensitivity and 60% specificity at a cut off of>195ng/l. Sodium had an AUC of 0.72, with a 70% sensitivity and 71% specificity at a cut off of<132.5 mmol/l. CRP did not distinguish MIS-C well (AUC=0.52) and although they had good AUC, platelet count and pro-BNP had cut off values in the normal range decreasing clinical utility. Conclusion We provide evidence for the use of accessible clinical and laboratory variables for the diagnosis of MIS-C in diverse settings. Implications These data will aid clinicians to do a rapid diagnosis (and ultimately treat earlier) of patients with MIS-C in the acute setting, especially those in under-resourced settings.

Estimation of incidence of MIS-C in Cape Town, South Africa

Background: Multisystem inflammatory syndrome is a severe manifestation of SARS-CoV-2 in children. The incidence of MIS-C after SARS-CoV-2 infection is poorly understood. There are very few cohorts describing MIS-C in Africa despite MIS-C being more common in Black children worldwide. Methods: A cohort of children with MIS-C and healthy children was recruited from May 2020 to May 2021 from the two main paediatric hospitals in Cape Town, South Africa. Clinical and demographic data were collected, and serum was tested for SARS-CoV-2 antibodies. The incidence of MIS-C was calculated using an estimation of population exposure from seroprevalence in the healthy group. Summary data, non-parametric comparisons and logistic regression analyses were performed. Results: Sixty-eight children with MIS-C were recruited with a median age of 7 years and 97 healthy children were recruited with a 30% seroprevalence. The estimated incidence of MIS-C was 22/100 000 SARS-COV-2 infections in children under 14 years old in the city at that time. Black children were over-represented in the MIS-C group (62% vs 37%, p = 0.002). The most common clinical features in MIS-C were fever (100%), tachycardia (98.5%), rash (85.3%), conjunctivitis (77.9%), abdominal pain (60.3%) and hypotension (60.3%). Median levels of haemoglobin, sodium, CRP, ferritin, cardiac (pro-BNP, trop-T) and coagulation markers (D-dimer and fibrinogen), neutrophil and white cell count were markedly deranged in MIS-C. Cardiac, pulmonary, central nervous and renal organ systems were involved in 71%, 29.4%, 27.9% and 27.9% respectively. Ninety-four point one per cent patient received intravenous immune globulin, 64.7% received methylprednisolone and 61.7% received both. ICU admission was required in 39.7% patient while 38.2% required inotropic support, 38.2% required oxygen therapy, 11.8% required invasive ventilation and 6% required peritoneal dialysis. The median hospital stay duration was 7 days with no deaths. Conclusion: The lack of reports from Southern Africa does not reflect a lack of cases of MIS-C. The clinical manifestations and outcomes of MIS-C in this region highlight the need for improved surveillance, reporting and data to inform diagnosis and treatment. Implications: To our knowledge, these are the first data on MIS-C in Africa. This shows that children in Africa are indeed presenting with MIS-C which will increase surveillance around the continent.

SARS-COV2 antibody phenotype and immune gene expression in MIS-C

Introduction: Multisystem Inflammatory Syndrome in Children (MIS-C) is a severe disease that affects a small proportion of children exposed to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). Differences in SARS-CoV-2 antibody responses and immune gene expression between SARS-CoV-2-infected children who develop MIS-C and those who do not may provide insight into the mechanism of MIS-C. Objectives: To determine the difference in SARS-CoV2 antibody responses and immune gene expression in children with MIS-C and healthy children with evidence of previous SARS-CoV2 infection. Methods: Healthy children presenting for elective surgery and those with MIS-C were recruited between 22 June 2020 and 5 November 2020 from a single paediatric hospital during the first wave of SARSCoV- 2 in the region. Clinical data, whole blood RNA and serum were collected. Titres of SARS-CoV-2 spike-specific antibody (SAb) and their capacity to perform neutralization, antibody-dependent cellular phagocytosis (ADCP) and antibody dependant cellular cytotoxicity (ADCC) were measured. Whole blood RNA gene expression was measured using multiplex Fluidigm quantitative Polymerase Chain Reaction (qPCR) with a panel of 84 immune genes. Principal component analysis was performed to assess for differences in gene expression. A linear regression model was developed with a forward stepwise model selection method to assess which genes associated with Creactive protein (CRP) in MIS-C after controlling for the neutrophil to lymphocyte ratio (NLR). Results: Twenty-three children with MIS-C and 25 healthy children were recruited. Nine healthy children had detectable SARS-CoV-2 serum antibodies (healthy exposed). No children had preceding clinical disease related to SARS-CoV-2 infection. Comparing children with MIS-C and healthy exposed children showed no difference in SAb binding responses (p=0.372) or ADCC (p=0.992). Increased neutralisation titre (p=0.084) and ADCP (p=0.086) in children with MIS-C was observed although was non-significant. Antibody function or titre did not change over time or with treatment in MIS-C. There was a clear distinction in immune gene expression between healthy children and those with MIS-C. Immune gene expression in MIS-C resolved to become indistinct from healthy children with time. Whole blood immune gene expression associated with an abundance of neutrophils in MIS-C. In a model that accounted for 66% of the variance in CRP (adjusted R2 = 0.66) the expression of IL27 accounted for 64% of the model effect (B=35;p<0.001) followed by NLR (15%, B=6.6, p=0.002) and the expression of MCP2 (11%, B=-14.59, p=0.008). Conclusion: Comparing children infected with SARS-COV-2 from the same time period and region with or without MIS-C provides unique mechanistic insight into the disease. A trend towards higher SAb titres and ADCP implies a distinct humoral immune response to SARSCOV- 2 in children with MIS-C, although further studies are required to validate this observation. The resolution of the abnormal immune gene expression in MIS-C implies a monophasic immune perturbation. The association of IL27 and MCP2 with CRP suggests that these may be important targets in future studies for possible pathogenicity and as potential biomarkers in MIS-C.

MIS-C at two tertiary hospitals in Cape Town, South Africa: clinical phenotype and distinguishing features from similar acute inflammatory conditions

Introduction: Distinguishing Multisystem Inflammatory Syndrome in Children (MIS-C) associated with Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV2) from acute, pyrexial childhood illness can be challenging. We present a case series from two tertiary centres in Cape Town, South Africa and compare the clinical phenotype of MISC with mimicking systemic inflammatory disorders. Objectives: 1. Describe the clinical characteristics of children with MIS-C in the region. 2. Compare the clinical features of children with confirmed MIS-C to those who presented during the same period with suspected MIS-C and ultimately an alternative diagnosis of inflammatory or infective conditions (inflammatory controls). Methods: Children with MIS-C admitted to the Red Cross War Memorial Children's Hospital (RXH) and Tygerberg Hospital (TBH) between 22 June 2020 and 5 March 2021 were recruited. At RXH only, children with suspected MIS-C with an ultimate alternate diagnosis (inflammatory controls) were also recruited. Clinical data were collected. Results: During the time period, 70 children had confirmed MIS-C and 27 suspected MIS-C cases had an alternate diagnosis including typhoid, tuberculosis, sepsis and appendicitis among others. Sixty five percent of children with MIS-C had no SARS-CoV2 contact but all had evidence of SARS-CoV2 exposure by antibody (90%) or Polymerase Chain Reaction (PCR) tests (14%). There was no difference in age, sex or ethnic distribution between children with MIS-C and inflammatory controls (Table 1). The most common presenting features of MIS-C were fever (100%), tachycardia (99%), rash (86%), conjunctivitis (79%), and abdominal pain (60%). Compared to inflammatory controls, the presence of tachycardia, abdominal pain and conjunctivitis resulted in 96%;93% and 91% respectively increased odds of a diagnosis of MIS-C after controlling for all other presenting features. Compared to inflammatory controls, children with MIS-C had lower platelets, sodium and albumin and higher troponin-T and pro-brain natriuretic peptide (pro-BNP) (Table 1). The median minimum ejection fraction in MIS-C was lower than inflammatory controls (52% vs 63%, p=0.048). Ninety four percent of MIS-C patients received at least one dose of intravenous immunoglobulin (IVIG), 63% required methylprednisolone and 6% received IL-6 inhibition. Children with MIS-C were more commonly admitted to ICU compared to inflammatory controls (38% vs 12.5%, p=0.013) although there was no difference in mean hospital stay which was 8.2 days in MIS-C. There was no difference in requirement for inotropes (p=0.142) or ventilation (p=0.493). No children died. Conclusion: Distinguishing MIS-C from acute infectious or inflammatory causes of childhood fever may be challenging. The presence of conjunctivitis, tachycardia or abdominal pain associates with higher odds of MIS-C in this population. Differences in widely available blood tests like sodium, albumin and platelets may be useful to differentiate MIS-C in the acute setting.