Characteristics and Outcome of Children Admitted with Sars- Cov-2 Infection: Experiences from a Pediatric Public Hospital in Western India

Background SARS-CoV-2 infection in children is asymptomatic or mildly symptomatic. Clinical characteristics and outcome of children admitted with COVID 19, especially with underlying illnesses, has not been studied. Objective To study the clinical characteristics and outcome of children admitted, with SARS-CoV-2 infection, to a paediatric multispecialty hospital in Mumbai, the epicentre of the COVID19 pandemic in India. Design and Setting Retrospective observational study of medical records of 969 children admitted between 19 March and 7 August 2020. Participants Clinico-demographic characteristics and outcome of COVID 19 positive admitted during the study period. Variables compared between children who were previously healthy (Group I) and children with co-morbidity (Group II).


Introduction
Regression model was applied to identify predictors of mortality accounting for the role of other factors, wherein adjusted odds ratio (OR) and 95% Con dence Intervals (CI) were estimated. A p-value of <0.05 was considered statistically signi cant for all the comparisons.

Results
From19 March to 7 August 2020, 969 children were admitted and RT-PCR for SARS-CoV-2 was done in 964.
While compiling the study, 4 children were still admitted. The median duration of PCR negativity was 5 days (range, 3-15 days). Most patients were discharged (n = 105/119; 88%) with a median length of hospital stay of 9 days (range, 4-17 days), which did not differ signi cantly between Groups I and II. (Table II) There were 14 (11.4%) deaths of which 3 (21.5%) were neonates and 5 (36%) were more than 8 years of age.  6; p = 0.04) were mortality predictors. On logistic regression, pulseox saturation <94% at admission (OR, 9.1; 95% CI, 1.04-99.1) and hospital stay of less than 9 days (OR, 35.9; 95% CI, 1.5-856.0) were predictors of mortality. (Table III) Challenges and Experiences Lack of de nite pediatric guidelines necessitated the treatment and care strategies to be adapted from adult , ICMR/GOI guidelines [4]. It was a challenge to devise an institutional protocol of care not only for children but also their caretakers and health care workers (HCW) that needed frequent updating and scrutiny. Segregating care areas, logistic support, reallocating manpower, redistributing medical equipments, and redesigning the services to provide a dedicated, high standard of care at affordable price. Each specialty services had to improvise care and protocol to include home monitoring, telephonic consultation, clustering of care during hospital visit. Cross specialty consults and investigations were coordinated so as to minimize intervention and hospital visits.

Discussion
COVID19 is a global health crisis. To our knowledge, this is the largest, in-patient pediatric COVID19 study from pediatric multispecialty public hospital in India. The study highlights the demographic features, clinical characteristics, disease progression, and outcome of 123 children admitted with COVID19. As this study enrolled children who were admitted to the hospital, the data likely represents individuals from the moderate to severe end of the disease spectrum.
As soon as the rst pediatric COVID19 case was reported in March 2020, in Mumbai, a dedicated COVID care area, personnel, equipment, and protocol were organised on an emergency mode. Global data suggested that children were infected early during community transmission phase and hence a low threshold of suspicion was followed for COVID 19 testing. As the pandemic rapidly evolved and emerging evidence suggested that children were largely asymptomatic or mildly symptomatic, we adopted screening for SARS-CoV-2 in all admissions as the entire city had become a hotspot. In the initial few months, COVID19 cases were only from Mumbai. As the lockdown was slightly relaxed, more children from the Mumbai Metropolitan Region were admitted.
Of 969 children admitted, RT-PCR for SARS-CoV-2 was performed in 964. Of these, 123 tested positive, a positivity rate of 12.7% lesser than the reported overall positive rate of 20.8% until 7 August 2020.
There were 76 (62%) cases in Group I comprising of previously healthy children and 47 (38%) in Group II who had underlying illness. (Figure 2) In an earlier study from Columbia Pediatric COVID19 management group comorbidities were de ned as Obesity, Asthma, Infancy or Immune suppression were studied [7].
Median age of presentation was 3 years, older children (>10 years of age) were more in Group II. Twenty seven (21.7%) children were asymptomatic. Initial studies from China reported 4.1-50% cases to be asymptomatic, while 58% were asymptomatic in a study from Pune [2][3][4]. The wide variation could be attributed to the difference in COVID19 testing protocol.
Atypical presentations like seizures (10.6%) and gastrointestinal symptoms (12.2%) were more common as compared to other studies. [7,9,12,14] Seizure and diarrhea as presenting symptoms was more common in Group I. COVID19 disease severity characterization revealed mildly symptomatic children were signi cantly more in Group I (n = 50/76, 66%) than Group II (n = 4/47, 8.5%; p = 0.0001) and moderate to severe COVID19 was signi cantly more in Group II (n = 22/47, 47%) than Group I (n = 4/76, 5.3%; p = 0.0001). Children with an underlying illness had severe disease. Interestingly, the immunological consequence of COVID19, the MIS-C/KD (n = 11/123; 9%) was found more in Group I (n = 8/76, 10.5%) than Group II (n = 3/47, 6.4%). Interestingly, presence of co morbidity, dysregulates or blunts the immunological host responses causing severe infection but is unable to mount a hyperin ammatory immune response like MIS-C/KD. Though chest radiograph is not considered the best modality to diagnose COVID19 pneumonia and unilateral or bilateral peripheral shadows and/or ground glass opacities have been described but pleural effusion is rare [15]. In this series, 12 cases had consolidation/bilateral haziness and 7 had pleural effusion.
Thirty-nine (32%) cases needed intensive care. Severe COVID19 pneumonia, circulatory collapse, MIS-C/KD, and worsening of underlying disease were the common indications. Need for intensive care in our series is higher than reported in literature [16]. This could be because we had more vulnerable children with underlying illness and severe COVID19 disease requiring intensive care. Although adult studies suggest presence of comorbidities as an important predictor of need for intensive care [17,18], this was not found in our study. Children requiring mechanical ventilation (15.5%) were fewer than those in the cohort from USA [7,19] which could be due to more children non respiratory presentations. There was no signi cant difference between the two groups with regard to length of hospital stay or disease outcome. (Table II) A systematic review in adults concluded that co-morbidities like Hypertension, Cardiovascular disease, Diabetes, and chronic renal diseases were signi cantly associated with mortality [20]. A study of children from the European cohort concluded that neonates, male sex, pre-existing medical conditions, fever, lower respiratory tract infection, radiological changes of pneumonia or ARDS, and viral co-infection were associated with more severe course on univariate analysis; however, the study did not compare these parameters to mortality [14]. In our cohort, male sex, hypoxia (SpO2 <94% ) on admission, need for respiratory support, inotropes, intensive care, length of hospital stay <10 days was signi cantly associated with mortality on univariate analysis. Male gender has been associated with a higher risk of severe disease and mortality because of higher ACE-2 receptor expression [21]. On regression analysis ,SpO2 <94% on admission and length of hospital stay of <10 days were predictors of mortality and not the presence of co-morbidities. Similar experience from adult studies has shown mortality within 1 to 2 weeks of ICU admission [17]. To our knowledge, no pediatric study mentioning predictors of mortality has been conducted to date.
As a retrospective study, certain important parameters like onset of symptoms from day of contact, source of infection, and exact duration of COVID19 RT-PCR positivity in all children could not be assessed.

Conclusion
Contrary to belief that pediatric COVID19 is a mild illness, children with co morbidity are more vulnerable and manifest with severe disease. Immunologic manifestations (MIS-C/KD) are more in previously well children.
Male sex, hypoxia on admission, need for ventilator support, inotrope, intensive care, hospital stay of less than 10 days are predictors of risk of mortality. We suggest universal testing with RT PCR for SARS-CoV-2 in all children admitted to hospitals to identify and segregate the cases, provide protocol based care, characterize the severity, initiate prompt treatment and improve outcome.   Tables   Table I BAI JERBAI WADIA HOSPITAL   Median (IQR) -13) 10) Numbers shown in parenthesis are percentages