key: cord-0685073-hpb20si4
authors: Au, Sunny Chi Lik
title: Comments on Stroke as a Neurological Complication of COVID-19: A Systematic Review and Meta-Analysis of Incidence, Outcomes and Predictors
date: 2021-05-18
journal: J Stroke Cerebrovasc Dis
DOI: 10.1016/j.jstrokecerebrovasdis.2021.105863
sha: 23fd1491c7f527eb9a1dddc52d3b5165793b5380
doc_id: 685073
cord_uid: hpb20si4

nan

We read with interest the research article entitled "Stroke as a Neurological Complication of COVID-19: A Systematic Review and Meta-Analysis of Incidence, Outcomes and Predictors", [1] it was a systematic review and meta-analysis on stroke incidence and different outcomes over COVID-19 patients. We are particularly concerned about the potential groups of overlapped COVID-19 stroke subjects in the meta-analysis under the worldwide SARS-CoV-2 pandemic.

Firstly, Nalleballe et al analyzed COVID-19 adult patients in the TriNetX database, [2] which was a global health collaborative clinical research platform collecting real-time electronic medical records data from a network of health care organizations from January 20 th to June 10 th , 2020. 3 The real-time updating subjects included in this primary study likely overlapped the other remaining 29 studies in the meta-analysis reporting COVID-19 stroke cases, thus exaggerating the results in Table 1 .

Secondly, Varatharaj et al collected their data via the clinicians (United Kingdom) notified cases on the ABN portal, BASP portal and the RCPsych portal on April 26 th , 2020. [3] Going through the included research sites of these portals, there were potential overlapping of cases from King's College Hospital in London published by Benger et al, [4] and National Hospital for Neurology and Neurosurgery in Queen Square of London by Beyrouti et al. [5] The final percentage result in Table 5 would be increased by this dual reporting.

Thirdly, Belani et al and Kihira et al both conducted their multicenter retrospective study with patient who had "stroke alert" or "code stroke" from six hospitals spread across three New York City boroughs. [6, 7] Although the details of the 3 boroughs were not mentioned in Kihira et al's study, [17] both studies were conducted exactly over the same time from March 16 th to April 5 th , 2020; and both author groups were affiliated to the Mount Sinai Health System in New York of the United States. Going through the two studies, the 139 and 141 patients included were with of extremely similar demographics, characteristics data and stroke incidence. There was potential duplication of stroke subjects, which would exaggerate the total number in Table 1 . in United States. [8, 9] Despite their studies were started at different time points, there was almost a month of overlap study period from March 15 th to April 13 th , 2020. Duplicated subjects could 4 not be neglected within a month time of active local COVID-19 outbreak in New York. Results in Figure 2 , 3, 4 and Table 1 , 2, 5 would potentially be biased.

One of the basics of meta-analysis is that it should not include correlated data. Although duplicate publication bias can potentially skew the final outcome of the meta-analysis, it could be avoided by carefully looking into each primary study's research site and the time period of their subject recruitment. Readers need to be cautious on interpreting the meta-analysis's results when there is potential duplicate publication bias.

(482 words)

Stroke as a Neurological Complication of COVID-19: A Systematic Review and Meta-Analysis of Incidence, Outcomes and Predictors

Spectrum of neuropsychiatric manifestations in COVID-19

Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study

Intracerebral haemorrhage and COVID-19: Clinical characteristics from a case series

Characteristics of ischaemic stroke associated with COVID-19

COVID-19 Is an Independent Risk Factor for Acute Ischemic Stroke

Incidental COVID-19 related lung apical findings on stroke CTA during the COVID-19 pandemic

COVID-19 related neuroimaging findings: A signal of thromboembolic complications and a strong prognostic marker of poor patient outcome

SARS-CoV-2 and Stroke in a New York Healthcare System