Clinical spectrum and short-term outcome of post-COVID multisystem inflammatory syndrome in children after the 2019 pandemic: a longitudinal cohort study

Study design and setting

This was a prospective longitudinal cohort study conducted at MGM Medical College and LSK Hospital, Bihar, India. From 1^st June 2021 to 30^th April 2022, the study included recruitment, follow-up, and data collection. Informed consent was taken from participant attendants and ethical clearance was obtained by Ethical committee of institute. IEC No. MGM/PRI-880/22.

Study population

Children up to the age of 18 years admitted to MGM Medical College & Hospital with a clinical diagnosis of MIS-C, in accordance with the WHO criteria, were included in the study.¹⁷ Infections including dengue, scrub typhus, and bacterial sepsis and active covid infection (diagnosed by nasopharyngeal rapid antigen test) were excluded by appropriate investigations. COVID-19 antibody IgG and IgM qualitative testing was done in all cases (method: Enzyme Linked Fluorescent Assay Technique on instrument of the VIDAS family). Data collection was done by trained resident doctors in a preformed abstraction form during the hospital stay and on follow-up after six weeks of discharge.

Procedure

A detailed clinical history and examination were conducted to determine precipitating factors for the illness and reach a clinical diagnosis. Necessary investigations for the diagnosis of MIS-C and evaluation of organ impairment at the time of discharge & follow-up were conducted. Laboratory markers including erythrocyte sedimentation rate, C-reactive protein, neutrophil/lymphocyte ratio, serum ferritin, D-dimer, liver and renal function tests, prothrombin time, and activated partial thromboplastin time (APTT) were done at the time of admission before starting immune-modulator therapy. Creatinine phosphokinase (CPK-MB) and troponin-1 were not included as they were not done routinely at this hospital. Chest X-rays, high-resolution chest computed tomography (CT) scans, electrocardiogram (ECG), and echocardiography were performed where indicated at the time of admission and at the time of follow-up. After six weeks, the children were reassessed with a clinical evaluation and lab tests.

Operational definitions

Demographics, clinical signs and symptoms, laboratory parameters, echocardiography, and coronary findings (at admission and at six-week follow-up), duration of ICU stay, medication (with special emphasis on immunomodulators and inotropes), mechanical ventilation, and mortality were all collected using a pre-designed proforma.

The choice of immunomodulators for treatment was decided by the treating team based on affordability for the patient and guidelines for the management of COVID-19 provided by the Ministry of Health and Family Welfare, Government of India.¹⁸ Echocardiography was done in all patients with shock at admission and again six weeks after discharge in case of residual cardiac impairment. Left ventricular ejection fraction (LVEF), coronary artery abnormality with a Z-score of >2.5, and other features like myocarditis, dilated cardiomyopathy, pericardial effusion, and arrhythmia were assessed.¹⁹ A person was thought to be in shock if they needed more than 20 mL/kg of IV fluid or an inotrope to keep their blood pressure above the 5th centile.

Patients who were treated with methylprednisolone received a pulse dose of 30 mg/kg/day for three days, followed by 2 mg/kg/day for a week or until their CRP value normalised. Steroids were tapered off over the next 2–3 weeks. Children treated with IVIG received 2 gm/kg as a continuous infusion over 8–12 hours. Recovery with residual organ impairment was defined on the basis of clinical features and laboratory findings or echocardiographic findings at the time of discharge.

Categorisation of children with MIS-C

The patients were categorized into three groups according to age: <5 years, 5–10 years, and >10 years for subgroup comparison. All patients were further categorised based on hospital stay (duration) and organ impairment (fully recovered or partial recovery).

Sampling method

Samples were taken using a simple random sampling method technique.

Sample size

Based on a 95% confidence interval, a 5% alpha error, and the fact that MIS-C occurs in 2% of cases,² the sample size was calculated to be 32 by online Epi Info software.

Statistical analysis

We used R software, version 3.4 (RRID:SCR_001905), for statistical analysis. All continuous variables were summarised using the mean (SD) or median (IQR). Categorical variables were expressed in counts (%). Age group and laboratory findings and complication mean were compared by ANOVA test and categorical frequency were compared by Chi squared test. Logistic regression analysis was used to assess the significant association between complications and laboratory findings. P value less than 0.05 was considered significant.

Demographic characteristics

A total of 42 participants were enrolled in the study, but 10 individuals had active cases of COVID-19 (rapid antigen test positive) and 5 were serologically negative and hence were excluded. Among the 27 patients with confirmed MIS-C, 14 (51.9%) were male and 13 (48.1%) were female. The median age of patients was 8 years (range 2–16 yrs.). The most common age group of MIS-C was 5–10 years (59.28%). Three (11.11%) patients had comorbidities such as cerebral palsy, seizure disorder, and pulmonary tuberculosis. A temporal association with COVID-19 infection was identified with serological testing of COVID-19 IgG and IgM. All patients were serologically IgG positive and four (14.81%) were IgM positive.

Clinical characteristics

A fever was present in 26 (96.29%) cases, and one patient was afebrile (Table 1). The most common system involved was the respiratory system, with 13 patients displaying complications of this kind (48.14%). Dyspnea (100%) and cough (38.4%) were the most common symptoms of the respiratory system. Table 1 gives the demographic, clinical characteristics, and the outcome of MIS-C cases across age groups.

The second most common organ involvement was the cardiac system, which was affected in 11 patients (40.74%). Shock and hypotension were noted in all cases and one patient suffered from cardiac arrhythmia in the form of sinus tachycardia. The incidence of shock was highest in the age group of 5–10 years (81.18%). Mucocutaneous involvement was seen in seven (25.92%) children, among whom a macula-papular rash over the trunk and extremities was evident in six of them (85.71%). Two patients (28.57%) had non-purulent conjunctivitis, and one each had petechia and oropharyngeal congestion. Cervical lymphadenopathy was seen in one child. Seven (25.92%) patients had CNS affection, including six (85.71%) with altered sensorium, four (57.14%) with seizures, and two (28.57%) with headaches. Gastrointestinal involvement was seen in five (18.51%) children, of which diarrhoea was reported in four (80%), followed by abdominal pain in three (60%), and nausea and vomiting in two (40%). Two patients (7.4%) suffered from hepatitis C.

Laboratory investigations and cardiac assessment

At the time of admission, anaemia (below 2 standard deviations for age and sex) was found in 14 (51.85%) children. Four (14.8%) had thrombocytopenia (1.5 lakh/mm³) and four (14.8%) had leucopenia (5000 mm³).

Among inflammatory markers, C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) were raised in all cases, with an average mean value of 15.89 and 25.59, respectively. Serum ferritin (mean value of 942.04 ng/mL) and dimer (3255 ng/mL) levels were also raised in all cases. Prothrombin time/activated partial thromboplastin time (PT/APTT) was abnormal in two (7.4%) cases and transaminase values were raised in three (11.11%) patients (Table 2).

Echocardiography (ECHO) was done in 11 patients in which myocarditis (54.54%), global hypokinesia (18.18%), severely decreased left ventricular ejection fraction (LVEF) less than 30% was 18.18%, moderately decreased LVEF [30–55%] was 45.45%, and one patient pulmonary embolism were noted.

Clinical course, treatment, and outcome

Twenty-four (88.88%) patients required intensive care, and the median duration of PICU stay was 6.5 days (range 2–15). Four patients (16.66%) required mechanical ventilation and 11 (45.83%) required inotropic support.

Twenty-three patients (85.18%) were treated with methylprednisolone, and four (14.8%) patients were treated with IVIG. Of the patients treated with IVIG, one patient had a complete recovery and three had residual organ impairment at the time of discharge (95% CI, 25–75%; p=0.239). One patient who suffered from a pulmonary embolism was given anticoagulant therapy.

At the time of discharge, 14 (51.85%) children had fully recovered, and 13 (48.14%) patients suffered from organ impairment, of which 10 had cardiovascular involvement (76.9%), two had CNS involvement (15.38%), and one had respiratory involvement (7.6%). The average median hospital stay duration was 13.5 days (range 7–34). No patients had expired.

Follow-up at six weeks after discharge

All discharged patients remained clinically stable, and no new symptoms or deterioration were noted. In patients with partial recovery, improvement in terms of increased LVEF, Z scores, and respiratory findings was noted. However, hemiparesis and facial palsy persisted in children with CNS involvement. Multiple logistic regression was done to predict the effect of laboratory markers on the likelihood of occurrence of organ impairment and prolongation of hospital stay in which only serum ferritin level [Coefficient 0.00674; p = 0.0041] was significant for prolongation of hospitalization possibly due to its positive association with greater organ impairment (Tables 3, 4, 5 and Figure 1).

Figure 1. Incidence of systemic complications related to hospital stay.

CVS (cardiovascular system), CNS (central nervous system), RESP (respiratory system).

Limitations

What this study adds to our knowledge