key: cord-0948164-v1335dzd authors: Keddie, Stephen; Pakpoor, Julia; Mousele, Christina; Pipis, Menelaos; Machado, Pedro M; Foster, Mark; Record, Christopher J; Keh, Ryan Y S; Fehmi, Janev; Paterson, Ross W; Bharambe, Viraj; Clayton, Lisa M; Allen, Claire; Price, Olivia; Wall, Jasmine; Kiss-Csenki, Annamaria; Rathnasabapathi, Dipa P; Geraldes, Ruth; Yermakova, Tatyana; King-Robson, Joshua; Zosmer, Maya; Rajakulendran, Sanjeev; Sumaria, Sheetal; Farmer, Simon F; Nortley, Ross; Marshall, Charles R; Newman, Edward J; Nirmalananthan, Niranjanan; Kumar, Guru; Pinto, Aswin A; Holt, James; Lavin, Tim M; Brennan, Kathryn M; Zandi, Michael S; Jayaseelan, Dipa L; Pritchard, Jane; Hadden, Robert D M; Manji, Hadi; Willison, Hugh J; Rinaldi, Simon; Carr, Aisling S; Lunn, Michael P title: Epidemiological and cohort study finds no association between COVID-19 and Guillain-Barré syndrome date: 2020-12-14 journal: Brain DOI: 10.1093/brain/awaa433 sha: 0cba03b959e9b22bdc0cfb923a501959f10c5b3f doc_id: 948164 cord_uid: v1335dzd Reports of Guillain-Barré syndrome (GBS) have emerged during the Coronavirus disease 2019 (COVID-19) pandemic. This epidemiological and cohort study sought to investigate any causative association between COVID-19 infection and GBS. The epidemiology of GBS cases reported to the UK National Immunoglobulin Database was studied from 2016 to 2019 and compared to cases reported during the COVID-19 pandemic. Data were stratified by hospital trust and region, with numbers of reported cases per month. UK population data for COVID-19 infection were collated from UK public health bodies. In parallel, but separately, members of the British Peripheral Nerve Society prospectively reported incident cases of GBS during the pandemic at their hospitals to a central register. The clinical features, investigation findings and outcomes of COVID-19 (definite or probable) and non-COVID-19 associated GBS cases in his cohort were compared. The incidence of GBS treated in UK hospitals from 2016 to 2019 was 1.65–1.88 per 100 000 individuals per year. In 2020, GBS and COVID-19 incidences varied between regions and did not correlate with one another (r = 0.06, 95% confidence interval: −0.56 to 0.63, P = 0.86). GBS incidence fell between March and May 2020 compared to the same months of 2016–19. In an independent cohort study, 47 GBS cases were reported (COVID-19 status: 13 definite, 12 probable, 22 non-COVID-19). There were no significant differences in the pattern of weakness, time to nadir, neurophysiology, CSF findings or outcome between these groups. Intubation was more frequent in the COVID-19 affected cohort (7/13, 54% versus 5/22, 23% in COVID-19-negative) likely related to COVID-19 pulmonary involvement. Although it is not possible to entirely rule out the possibility of a link this study finds no epidemiological or phenotypic clues of SARS-CoV-2 being causative of GBS. GBS incidence has fallen during the pandemic, which may be the influence of lockdown measures reducing transmission of GBS inducing pathogens such as Campylobacter jejuni and respiratory viruses. The first cases of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were reported to the WHO in late 2019, and by March 2020 COVID-19 was pandemic. 1 SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV) were associated with neurological sequelae. 2 Early reports identified neurological symptoms of COVID-19 infection as fever, headache, anosmia and dysgeusia. 3 Subsequently COVID-19 infection has been associated with stroke, meningoencephalitis, acute disseminated encephalomyelitis and Guillain-Barré syndrome (GBS). [4] [5] [6] The first reported case of GBS questioned a possible link with COVID-19 and occurred in late January 2020 in a COVID-19 asymptomatic patient who that to patients hospitalized with GBS. Separately, but in parallel, we characterized a large cohort of the incident UK GBS cases presenting both with, and without, COVID-19 to explore timing of onset, and any identifying phenotypic characteristics that might hint at a specific mechanistic link (as for example in sensory GBS associated with CMV). Finally, we explored any homology between SARS-CoV-2 and the human genome and proteome that would support a molecular mimicry mechanism. Incident hospitalized cases of GBS were retrospectively ascertained from the UK National Immunoglobulin Database from the 1 January to 31 May 2020, demonstrating the frequency of GBS cases across the year, pre and during the COVID-19 pandemic. NHS England (NHSE) procures the total intravenous immunoglobulin (IVIg) supply for England, Scotland and Northern Ireland. NHSE mandates that every IVIg prescription is approved by a clinical panel and is reported onto the database within 90 days. Recording compliance is almost 100% as hospital trusts are only reimbursed once records of dispensed volumes are submitted; these are retrospectively cross-checked against supply and returned stocks. 20 To ensure complete reporting of cases, NHSE specifically mandated all users of the National Immunoglobulin Database to log any outstanding GBS cases by 30 June 2020 by email on 9 June 2020. Data retrieval was then performed on 7 July 2020 to allow time for reporting delay. Current UK guidance for GBS treatment indicates IVIg or plasma exchange (PLEX) as first line therapy, 21 but IVIg is, in practice, first line in most UK hospitals as PLEX is normally not as available. IVIg is also only authorized in the UK for patients with Hughes Grade ≥4, progressing towards intubation and ventilation, with a high likelihood of respiratory support (mEGRIS score ≥ 3) or a predicted poor prognosis (mEGOS ≥ 4). The patients usually treated with IVIg in the UK are those who require admission, and although this under-ascertains the true incidence of GBS, it reduces the effects of attendance bias in a pandemic. IVIg is given to nearly all presenting GBS patients in Europe as illustrated by 86% (612/715) of European cases treated with IVIg in the International GBS Outcome Study (IGOS), 22 and 88% (37/42) of 'COVID-19 GBS' in the literature until July. 23 We also searched the NHSE Immunglobulin Database for GBS cases from 1 January to 31 May in each of the years 2016 to 2019) to determine the incidence of non-COVID-19 reported cases of GBS to compare to the 2020 pandemic data. Data were stratified by hospital trust and region. UK population data for COVID-19 PCR confirmed infection were collated from Public Health England, Health Protection Scotland and the Public Health Agency of Northern Ireland. Because of the lack of testing available testing early in the pandemic, COVID-19 PCR confirmed cases were significantly fewer than the true incidence of infection across the UK. 24 Thus in addition, we obtained data from the NHS Blood Transfusion Service (NHSBT) 25 of antibody seroprevalence across the UK to SARS-CoV-2, and used the London data to study the number of GBS cases that occurred compared both to the number of PCR confirmed cases and the number of seroconverted COVID-19 cases during the pandemic months. In parallel to the epidemiological study, we conducted a prospective cohort study to compare the demographic, phenotypic and infective associations of COVID-19 associated GBS (definite and probable) to COVID-19 negative GBS reported during the same study period. At the time of this study relatively little is known of the epitope presentation and immunobiology of SARS-CoV-2. We searched for evidence of molecular mimicry between any SARS-CoV2 proteins and human nerve axonal or myelin proteins and glycoproteins, recognizing that epitopes are not all protein and not necessarily all linear. We searched for human homologues of proteins encoded by the SARS-CoV-2 genome using the National Centre for Biotechnology Information (NCBI's) Basic Local Alignment Search Tool (BLAST) to identify common amino acid sequences in the human Reference Sequence Database (refseq_protein). The NCBI BLAST was also used to query the SARS-CoV-2 genome against the human genome for any significant alignments at specific genomic loci. The expect value (E-value) quantifies the number of times a specific alignment can be 'expected' to occur in a database by chance. As the E-value decreases the significance of the alignment in the specified database increases. Any alignment with an E-value of ≤1 × 10 −4 was considered homologous to a human protein (error rate <0.01%). The incidence rates of GBS and COVID-19 (95% confidence intervals by Byar's approximation method) 28 were calculated by dividing regional cases and time period by the relevant mid-year population estimate. Mid-year population estimates at both regional and national level were obtained from the UK Office for National Statistics. 29 We explored any association between the incidence of GBS and the incidence of COVID-19 in UK regions in 2020 (January to May) using Pearson's correlation coefficient. The Shapiro-Wilk test was used to determine suitability of parametric tests. COVID-19 definite and probable cases in the cohort were statistically compared against non-COVID-19 associated GBS. In addition, to determine whether characteristics differed between pandemic and non-pandemic GBS phenotypes, clinical characteristics of our study cohort were also compared to published IGOS study participants. We used Mann-Whitney U to test non- The UK Health Research Authority was consulted and advised the study did not require review by an NHS Research Ethics Committee as an analysis of previously collected non-identifiable information. The project was submitted as a 'Service Evaluation' to the Clinical Audit and Quality Improvement Subcommittee (CAQISC). Data are available upon the request to the corresponding author. The The clinical characteristics of the patients with GBS are shown in Table 1 and Supplementary There were no differences in the treatment of GBS subgroups. IVIg was the first therapy in 83% cases, and only one patient received PLEX as second line therapy. One patient received more than one course of IVIg. One patient (COVID-19 definite) died. Death was attributed to pulmonary complications rather than neuromuscular weakness. Although it is profoundly difficult to prove no link in a rare disease, this retrospective epidemiological and prospective cohort study does not support any significant causal link between COVID-19 infection and GBS. 36 We have used several reliable sources of data to collate the best evidence from each to demonstrate the lack of likelihood of a significant causative link. The population-based data find no plausible temporal relationship between COVID-19 and GBS (Fig. 1) , a reduction in cases of GBS in comparison to preceding years ( Fig. 2) and no correlation between COVID and GBS incidence at regional level (Fig. 3 ). There are in addition no identifiable COVID-19 associated GBS features that differentiate it from GBS in non-pandemic circumstances, in this, the largest cohort reported to date. There are also no scientific data to support a molecular mimicry link of SARS-CoV-2 to GBS at the nucleic acid or protein level, other than a presumptive analogy to other known bacterial and viral GBScausing pathogens. The lack of even a short, linear homology between the SARS-CoV-2 structure proteins and any axonal or myelin surface proteins reduces the likelihood that molecular mimicry with SARS-CoV-2 might be a putative mechanistic link of SARS-CoV2 to GBS. The UK has a single highly regulated IVIg supply. IVIg is routinely available for all patients with GBS, but every vial given for GBS is logged under a mandatory NHS-based system linked directly to clinicians and to subsequent payment. The indication for IVIg treatment in GBS is for non-ambulant patients, and therefore it is unlikely such patients remained at home or would not be admitted to hospital. Even in 2016-19 only cases with significant disability and meeting criteria for treatment were recorded. This significantly reduces the likelihood of a disparity resulting from mild disease attendance bias, as a result of COVID-19 explaining the decline. Furthermore, within our cohort study 83% of cases were treated with IVIg, providing cross validation of high treatment rates, but also the fact that milder patients continued to attend to some extent. We hypothesize that the lockdown measures introduced to prevent COVID-19 transmission have had secondary effects of reducing other common transmissible infective GBS triggers such as upper respiratory tract infections through social distancing and mask wearing, and gastrointestinal illnesses as fewer people dined out and stricter hand hygiene was adhered to. In our cohort of 47, only 1/47 (2%) reported diarrhoea preceding their GBS, significantly fewer than in the European IGOS patients at 25%. 22 This is speculative but consistent with our hypothesis. Other studies have reported significant reductions in airborne or faeco-oral transmissible infectious diseases during lockdown, supporting this assertion. 41 Although the true impact of hygiene measures is unknown, the avoidance of C. jejuni and respiratory pathogens could conceivably reduce the incidence of GBS, and may explain the pandemicrelated reduction of GBS cases. Successful interventions to lower Campylobacter contamination of fresh poultry meat have previously been reported to reduce hospitalizations for GBS by 13%, 42 and so this assertion is not impossible. The true COVID-19 incidence in the UK is known to have been significantly under-reported. in which SARS-CoV-2 specific T-cell immune responses may occur. 44 The COVID-19 seroprevalence estimated GBS incidence is 0.016 per 1000 COVID-19 infections (1.6 per 100 000, and equivalent to the usual incidence of GBS). Although it is difficult to entirely rule out a causative link, these data provide further evidence for a lack of strong relationship between COVID-19 and GBS compared to other recognized GBS-associated infective pathogens, and potential over-reporting of an association when using PCR confirmed cases only. Whilst at an epidemiological level we found no increase in GBS linked to the COVID-19 epidemic, our data do not exclude the possibility that SARS-Cov-2 might be a driver of GBS in very rare cases, or that a significant reduction in non-COVID-19 GBS could mask a smaller spike of COVID GBS cases. However, other infective causes of GBS have been identified through demonstrating a peak in incidence temporally related to rises in the causative infective pathogen. With SARS-CoV-2 being one of the most prevalent infective pathogens in the last century, it is more conceivable that the absence of any increase in GBS cases during the pandemic is more likely due to a lack of causation between COVID-19 and GBS. We have also shown that there is no significant homology between any SARS-CoV-2 genetic or linear protein structure and human linear protein structures, making a molecular mimicry causation less likely. The lack of homology does not exclude immunological similarity entirely as antibody epitopes are often non-linear. Furthermore, post-translational modification of viral proteins by their host cells can occur, which theoretically could result in the generation of immunogenic surface glycomolecules so far unknown. 45 Although molecular mimicry is the only fully proven pathogenic GBS mechanism, we acknowledge others could exist. More research is required to determine whether a causal relationship exists between SARS-CoV-2 and GBS. Some small early series of COVID-19 associated GBS have been reported. 8, 46 The series of Gigli et al. 46 reported eight patients, all of whom were swab negative, one seropositive only and only four with COVID symptoms. Another small cohort of five reported specific disease characteristics suggesting differences from typical AIDP. 8 This population based epidemiological study was not fully prospective but has been able to demonstrate no relationship between GBS and COVID-19 infections across the UK through the interrogation of several complementary data sources. As we explore potential COVID-19 associated neurological disease, a measured analysis of the statistical probability of rare disease occurrence in the context of a pandemic is required to investigate causation appropriately, and continue to manage non-COVID-19 neurology with the associated challenges on healthcare resources. This epidemiological and cohort study contradicts a growing number of reports postulating causation between SARS-CoV-2 and GBS, and indeed demonstrates a reduction of GBS cases. This paper alone cannot be considered definitive in ruling out SARS-CoV-2 as a cause of GBS, but further prospective data collection of COVID-19 associated GBS cases and laboratory research are required. Although prompt reporting of disease manifestations and potential associations of COVID-19 is important to inform public health decisions, robust scientific assessment to establish causality versus association is essential to evolve our understanding of this novel viral pathogen and its sequelae. The authors report no competing interests. Timeline of WHO's response to COVID-19. World Health Organisation The neuroinvasive potential of SARS-CoV2 may play a role in the respiratory failure of COVID-19 patients Neurologic Manifestations of Hospitalized Patients with Coronavirus Disease Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study. The Lancet Psychiatry Neurological involvement of coronavirus disease 2019: a systematic review The emerging spectrum of COVID-19 neurology: clinical, radiological and laboratory findings Guillain-Barré syndrome associated with SARS-CoV-2 infection: causality or coincidence? Guillain-Barré Syndrome Associated with SARS-CoV-2 Guillain-Barré syndrome after exposure to influenza virus Guillain-Barré Syndrome outbreak associated with Zika virus infection in French Polynesia: A case-control study Guillain-Barré syndrome and preceding infection with Campylobacter, influenza and Epstein-Barr virus in the General Practice Research Database Peripheral neuropathies and anti-glycolipid antibodies Guillain Barré syndrome (GBS): New insights in the molecular mimicry between C. jejuni and human peripheral nerve (HPN) proteins Antiganglioside antibodies and their pathophysiological effects on Guillain-Barré syndrome and related disorders -A review Antibodies to heteromeric glycolipid complexes in Guillain-Barré syndrome Nodal proteins are target antigens in Guillain-Barré syndrome Accessory proteins of SARS-CoV and other coronaviruses European Centre for Disease Prevention and Control. COVID-19 Situation Update Worldwide, as of 29th Department of Health. Clinical Guidelines for Immunoglobulin Use: Update to Second Edition Regional variation of Guillain-Barré syndrome Guillain-Barré syndrome in SARS-CoV-2 infection: an instant systematic review of the first six months of pandemic COVID-19) Infection Survey -Office for National Statistics Sero-surveillance of COVID-19 European center for disease prevention and control. Case Definition for Coronavirus Disease 2019 (COVID-19), as of 29 Neurological associations of COVID-19 Office for National Statistics. Estimates of the population for the UK Population incidence of Guillain-Barré syndrome: A systematic review and meta-analysis Coronavirus (COVID-19) in the UK Guillain-Barré syndrome and Fisher syndrome: Case definitions and guidelines for collection, analysis, and presentation of immunization safety data Diagnosis of Guillain-Barré syndrome and validation of Brighton criteria OpenSAFELY: factors associated with COVID-19 death in 17 million patients Features of 16,749 hospitalised UK patients with COVID-19 using the ISARIC WHO Clinical Characterisation Protocol Public Health Classics Association or Causation: Evaluating Links between 'Environment and Disease The Curious Case of the Missing Strokes During the COVID-19 Pandemic What has caused the fall in stroke admissions during the COVID-19 pandemic? COVID-19 and stroke-A global World Stroke Organization perspective Emergency Department Syndromic Surveillance System Week 23 COVID-19 pandemic: Impact caused by school closure and national lockdown on pediatric visits and admissions for viral and non-viral infections, a time series analysis Declining Guillain-Barré syndrome after campylobacteriosis control Guillain-barré syndrome following primary cytomegalovirus infection: A prospective cohort study Robust T cell immunity in convalescent individuals with asymptomatic or mild COVID-19. bioRxiv Post-translational modifications in capsid proteins of recombinant adeno-associated virus (AAV) 1-rh10 serotypes Guillain-Barré syndrome in the COVID-19 era: just an occasional cluster? Supplementary material is available at Brain online.