key: cord-0769028-uocs65sl
authors: Alders, N.; Penner, J.; Grant, K.; Patterson, C.; Hassell, J.; MacDermott, N.; Pincott, S.; Bamford, A.; du Pre, P.; Johnson, M.; Moshal, K.
title: COVID-19 Pandemic Preparedness in a United Kingdom Tertiary and Quaternary Children`s Hospital: Tales of the Unexpected
date: 2020-08-22
journal: nan
DOI: 10.1101/2020.08.20.20178541
sha: 20b76bbe80062cb2d64a68b8fe2d3bb5ba2c7adc
doc_id: 769028
cord_uid: uocs65sl

Background: The paucity of data describing SARS-CoV-2 in the paediatric population necessitated a broad-arching approach to pandemic planning, with preparations put in place to manage a heterogeneous cohort. We describe a diverse group of SARS-CoV-2 positive paediatric patients treated at a large tertiary/quaternary children`s hospital in the United Kingdom and the adaptive coping strategies required. Methods: All paediatric patients with positive RT-PCR on a respiratory sample and/or serology for SARS-CoV-2 up to 19th May 2020 were included. Results: 57 children met the inclusion criteria. 70% were of non-Caucasian ethnicity with a median age of 9.3 years (IQR 5.16-13.48). Four distinct groups were identified: paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 (PIMS-TS) (54%), primary respiratory (18%), incidental (7%), and non-specific febrile illnesses with or without extra-pulmonary organ dysfunction (21%). These groups presented in distinct chronological blocks as the pandemic unfolded. Discussion: The diverse range of presentations of SARS-CoV-2 infection in this population exemplified the importance of preparedness for the unknown in the midst of a novel infectious pandemic. Descriptions of paediatric patients during the initial phase of the pandemic from other parts of the globe and extrapolation from adult data did not serve as an accurate representation of paediatric COVID-19 in our centre. An adaptive, multidisciplinary approach was paramount. Expanded laboratory testing and incorporation of technology platforms to facilitate remote collaboration in response to strict infection control precautions were both indispensable. Lessons learned during the preparation process will be essential in planning for a potential second wave of SARS-CoV-2.

The relative paucity of data describing SARS-CoV-2 in the paediatric population mandates a broad-arching approach to pandemic planning with preparations put in place to manage a heterogeneous population of patients presenting with a range of single and multi-organ pathology of varying severity. We describe a diverse group of SARS-CoV-2 infected paediatric patients treated at Great Ormond Street Hospital, a tertiary and quaternary paediatrics hospital in London, UK with 383 inpatient beds and approximately 50 specialties, whom exemplified the importance of preparedness for the paediatric COVID-19 unknown. We further illustrate four distinct temporal waves of SARS-CoV-2 clinical phenotypes at our centre beginning with our first case 25 th March 2020.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint Methods All patients aged ≤ 18 years with positive respiratory or nasal SARS-CoV-2 RT-PCR and/or serum IgG (Epitope Diagnostics Inc. TM ) up to 19 th May 2020. This time interval was chosen as it corresponds to the first two months of local paediatric cases when preparation measures remained in flux. It also represents the onset of the local epidemic when community seroprevalence remained low, thus, both diagnostic methods are likely to represent recent infections. A search of electronic medical records for clinical, laboratory, and radiographic data was performed. False-positive results due to lab contamination were excluded.

Anonymised data was collected and stored in a secure Excel® database. The project was registered with the local research department (approval #2857). Real-time laboratory data was collected via the microbiology tracking system of the electronic patient records. Results 2,194 SARS-CoV-2 RT-PCR tests had been undertaken on 933 patients. Antibody tests were also introduced on-site assessing IgG response to SARS-CoV-2 nucleocapsid (sensitivity 92%, specificity 96%). 126 patient antibody tests had been completed.

Of 65 paediatric patients with positive samples (n=28 RT-PCR, n=27 serology, n=10 both), 57 patients were included in the final analysis. Three patients were excluded for unconfirmed positive nasopharyngeal aspirates (NPA) reported prior to hospital transfer with negative All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint admission screening. One serology-positive infant with negative RT-PCR was excluded because of uncertainty in the significance of the result. The mother had symptoms consistent with COVID-19 late in the third trimester, therefore passive maternal antibody transfer was suspected. Three patients were excluded due to lack of data as they did not require hospital admission

The cohort characteristics are presented in Table 1 . The median age was 9.3 years (IQR 5.16-13.48) and 70% were of non-Caucasian or mixed ethnicity. Four distinct clinical groups were identified: paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 (PIMS-TS) (54%), primary respiratory (18%), incidental (7%), and non-specific febrile/viral illness with or without single organ dysfunction (21%). These groups presented in distinct chronological blocks (Figure 1 ) with respiratory and other febrile illnesses predominating in the first three weeks after the first positive case was admitted. This was followed almost exclusively by PIMS-TS cases in the latter third of the study period. Compared to those with a primary respiratory phenotype, PIMS-TS patients were generally older (median 10. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint tract symptoms were present in 30% and 40% of patients respectively, predominantly amongst primary respiratory phenotypes, though still present in all symptomatic groups. Systemic inflammatory signs often seen in conditions such as Kawasaki's disease were common, as were central nervous system symptoms (55% of symptomatic patients ≥ 1 neurological sign/symptom) such as: headache (38%), encephalopathy (32%), weakness (25%), and meningism (9%). Abdominal and neurologic symptoms were mostly seen in PIMS-TS where conjunctivitis and rash were almost exclusive to this group. 43% were overweight (weight-forage >85%) or obese (weight-for-age >95%) and 9% were severely obese (weight-for-age >99 th percentile). Obesity was equally common amongst PIMS-TS and respiratory phenotypes and largely restricted to these two groups. For newly admitted patients with community acquisition (n=50), median length of stay was 9 days (IQR 6-15.5). This was longest for respiratory perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint Two patients were found to have detectable SARS-CoV-2 RT-PCR in stool. SARS-CoV-2 was not found in urine or blood on RT-PCR though this was not routinely requested. In addition to standard bacterial cultures (blood, urine), concurrent viral infections were routinely screened for in all patients. Anti-streptolysin-O titres and specific bacterial blood PCRs were performed (meningococcus, staphylococcus aureus, group A streptococcus, and pneumococcus) in those presenting with shock. Epstein-Barr virus was the most common co-infection (n=8).

Moreover, 30 cytokine profiles (BD FACSCanto TM ) were completed on 13 patients. Interferon gamma, IL4, IL2, and tumour necrosis factor alpha were universally <50 pg/ml. Elevated IL10 was observed in one patient with a primary respiratory presentation [110pg/ml] and one PIMS-TS (55pg/ml). Elevated IL-6 was demonstrated in six patients, four PIMS-TS (range: 59-443 pg/ml) and two primary respiratory (range: 71-218 pg/ml).

In those who underwent abdominal imaging and echocardiograms, gut inflammatory pathology on ultrasound (61%) and inflammatory heart complications (31%) were almost exclusive to the PIMS-TS group. Brain MRI (n=14) and EEGs (n=17) were performed more routinely in the latter part of the local epidemic when subtle neurological complications were recognised as common.

Abnormal findings were seen in 79% and 82% of those conducted respectively, again almost exclusive to the PIMS-TS group. However, these tests were conducted less frequently in the early stages of the local epidemic thus leading to few MRIs and EEGs in respiratory and other presentations.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint Treatment modalities are outlined in Table 3 . Four patients were treated off-label with remdesivir via a compassionate access programme of which pK studies were conducted in three (sample analysis outstanding). All four were admitted to paediatric/cardiac intensive care (PICU/CICU) with primary respiratory disease meeting the 2015 PALICC criteria for severe paediatric acute respiratory distress syndrome (pARDS) (2) . Twenty-nine (52%) were treated with immunoglobulins all of which were in the PIMS-TS group apart from one 'other group' treated for acute demyelinated encephalomyelitis (ADEM). Twenty-nine (52%) were treated with steroids, all were of PIMS-TS phenotype apart from two in the respiratory group where steroids were used for blood pressure support and the patient treated for ADEM. Anakinra (11%) was used off-label exclusively in PIMS-TS. Prophylactic low molecular weight heparin (LMWH) was prescribed in 56% of patients, more routinely in the later part of our local paediatric epidemic when clotting complications (4/57) were increasingly recognised.

Underlying comorbidities were common in the respiratory group (9/10), 60% demonstrated more than one co-morbidity. In the incidental group, all but one had an underlying immunecompromising condition (malignancy, metabolic disease, sickle cell anaemia). The remainder of the immune-compromised patients generally had mild disease phenotypes. Conversely, PIMS-TS cases were largely healthy at baseline (61% no comorbidities) with obesity making up the majority of the comorbid conditions. Full characterisation of co-morbid states per SARS-CoV-2 disease phenotype are presented in Supplementary Table A. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint At discharge, 73% of patients were functionally back to baseline, whereas 18% were functionally independent but not back to pre-admission status and/or had minor organ sequelae requiring ongoing follow-up but no active management. Seven percent of patients had major sequelae requiring assistance with activities of daily living and/or requiring active management of end organ damage. One patient with serious complications of PIMS-TS remained an inpatient. No directly-related COVID-19 deaths have been reported. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint prophylactic anticoagulation as part of a modification to our normal practice. 81% of PIMS-TS and 60% of respiratory patients were anticoagulated with LMWH, although anticoagulation was not exclusive to the PICU setting. All children survived discharge from PICU.

Our cohort illustrates four distinct SARS-CoV-2 disease groups, with distinct demographic differences who required different management approaches. Dissimilarities in presentation, management, and follow-up of paediatric versus adult cohorts must be considered in both the anticipation of a second wave of COVID-19 and for future pandemic planning.

Compared to adult cohorts, to date, there remains a paucity of data relevant to COVID-19 in paediatric populations even at the conclusion of the initial wave of the pandemic. Notably, there was minimal understanding of paediatric SARS-CoV-2 effects prior to the influx of cases in Europe and America. The largest (n =2143) and earliest review of paediatric COVID-19 patients in China described a relatively unaffected cohort with 5.6% and 0.6% suffering from hypoxia and multi-organ/pARDS respectively; 4.4% of paediatric patients were completely asymptomatic and 89.7% had only mild/moderate disease (3). Likewise, a Centers for Disease Control and Prevention (CDC) report of 150,000 cases in the United States comprised of only 1.7% paediatric cases, suggesting low incidence in children with less severe pathology (4). Similar rates had been described in Europe with paediatric cases in Italy encompassing only 1.2% of initial COVID-19 patients (5).

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint Our patient characteristics are similar to a separate UK single centre experience although in this cohort serologic positivity was not an inclusion criterion and children were not grouped by phenotypic presentation (6) . Of significance, there was a parallel preponderance for affected children from ethnic minorities. Additionally, a case series of patients admitted to a New York Children's hospital demonstrated obesity as a similarly prominent risk factor for admission with severe disease, with traditionally at risk paediatric populations relatively spared (7).

As a result of reassuring paediatric data at the start of the pandemic, initial planned patient pathways included transfer of SARS-CoV-2 patients to regional High Consequence Infectious Diseases Units. In order to facilitate expansion of adult inpatient capacity as cases surged in the UK, paediatric secondary care services from the region were relocated to our centre. Limited initial adaptation was made for severe cases of COVID-19 requiring intensive care as, based on the Chinese and Italian data the assumption was that more would not be necessary. However, crucially, protocols were put in place to activate and expand quickly if necessary. At the onset of the unexpected paediatric surge, a separate dedicated 16 bed COVID-19 PICU was created within 48 hours to accommodate increased patient numbers.

Treatment guidelines for all evolving paediatric SARS-CoV-2 phenotypes, based on available evidence, were drafted prior to arrival of the first patients and were given expedited approval by the hospital drug and therapeutics committee. A system for rapid approval of investigational drugs was set up in partnership with the trust bioethics committee with decision-making support from an established multidisciplinary team (MDT) of specialists, inclusive of external All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint specialist support as an impartial presence. In light of the lack of evidence for treatments of this patient group, pooling expertise in the MDT has provided a basis for the development of a standardised treatment protocol upon which an evidenced based approach could be built. The MDT embraced videoconferencing technology to uphold infection-prevention-and-control (IPC) precautions, could be convened urgently by any member ad hoc (sometimes within minutes) to discuss critical cases, or when rapid approval of investigational treatments was necessary.

The complexity of paediatric pandemic preparedness in our centre can be summarised in four distinct ways: PICUs with a specific COVID-19 general paediatric step-down ward.

(ii) Unforeseen Disease Pathology: The emergence of PIMS-TS not previously described in the initial Asian epidemic. Timely collaborative efforts with other paediatric centres capitalising on intra-and inter-institutional multidisciplinary input was essential in the early recognition and management of this novel paediatric phenomenon. As long-term effects remain unknown, the MDT approach will equally be essential in patient follow-up.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint also allowed for more widespread and timely assessment of cytokine responses and immunological effects of SARS-CoV-2 in the paediatric population.

Our local demographic population may not be generalisable to other settings. The referral nature of our centre did not allow us to capture data prior to transfer or after discharge back to local hospitals.

Supported by an ability to adapt quickly in all phases of a pandemic, our COVID-19 paediatric outcomes were reassuringly good. Our centre needed to evolve throughout the SARS-CoV-2 pandemic as new phenotypes arose, those not previously described in other regions, and adjust when new at risk populations were identified. An adaptive, MDT approach was paramount.

Expanded laboratory capacity and incorporation of technology platforms to facilitate remote collaboration in response to strict infection control precautions were both indispensable.

Paediatric-specific planning must not be static and evolution of preparedness endeavours must continue, particularly in the face of a potential second wave of SARS-CoV-2. Of utmost importance, the distinction between paediatric and adult populations must not be overlooked.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint *patients admitted long term to Great Ormond Street hospital testing positive during admission and those asymptomatic whom were already admitted were not included in the calculation of length of stay ** Self-identified ethnicity as extracted from paediatric electronic medical records †contact with household member with COVID-19 compatible symptoms or proven SARS-CoV-2 infection, hospital contact with laboratory proven SARS-CoV-2 infection ‡ pharyngitis, coryza BMI: Body mass index, COVID-19: Coronavirus disease 2019; RT-PCR: real time polymerase chain reaction; PIMS-TS: paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2; SARS-CoV-2: Severe acute respiratory syndrome coronavirus-2, URTI: Upper respiratory tract infection All rights reserved. No reuse allowed without permission.

perpetuity.

preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in

The copyright holder for this this version posted August 22, 2020. . https://doi.org/10.1101/2020.08.20.20178541 doi: medRxiv preprint 

Cornonavirus (COVID-19) in the UK 2020

Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Epidemiology of COVID-19 Among Children in China

MMWR Morbidity and mortality weekly report

Coronavirus Disease 2019 (COVID-19) in Italy

COVID-19: A UK Children's Hospital Experience. Hospital pediatrics

Clinical Features, and Disease Severity in Patients With Coronavirus Disease 2019 (COVID-19) in a Children's Hospital in

The clinical care provided by all junior doctors, nurses, allied health professionals, health care assistants, and technicians at GOSH was essential in this a novel pandemic. Important members 

Incidental cases (n=4) did not receive any management interventions *High dose is defined as 10-30mg/kg/day, max 1g; low dose as 0.4mg/kg/day-4mg/kg/day **High dose is defined as 30-50mg/kg/day, Low dose is defined as 3-5 mg/kg/day †Treatment dose defined as dalteparin 100-200 units/kg BD (goal anti-Xa level 0.5-1unit/ml); low dose defined as dalteparin 50 units/kg BD (anti-Xa levels unmonitored) ECMO: extracorporeal membrane oxygenation; PIMS-TS: paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2; SARS-CoV-2: severe acute respiratory syndrome coronavirus.