We read with interest the recent systematic review entitled “Relation between COVID-19 and Guillain-Barré syndrome in adults. Systematic review” by Trujillo-Gittermann et al.1 However, although the authors reported that they conducted the review following recommendations of PRISMA guidelines,2 there are some important departures, which have likely introduced biases and inconsistencies we would like to draw attention.
Our first concern is the questionable way to developed the Systematic Review (SR), the authors mention in the objective: “to analyze the evidence on the associations between COVID-19 and Guillan-Barré Syndrome (GBS)” and in your selection studies only included case reports, design with a lower level of evidence and that does not allow to establish associations between outcomes measures.3 Furthermore, they do not register the protocol in PROSPERO and do not take as reference the Cochrane Handbook to avoid selective reporting bias.4,5
Secondly, the authors not reported the risk of bias or methodological quality of studies included, introducing another bias in the quality of the SR, since it is suggested, evaluating and reporting these aspects in a complete way to see the quality of the studies in a SR.4,6
Regarding to the search strategy, draws attention to the flow chart diagram shown, is too general, the number of studies found by each database and the reasons for excluding studies are not reported, which makes a confusing interpretation. Furthermore, it does not report a sensitive search strategy that is required in a SR, not does it report how the [MesH] terms or “free text terms” were combined in all the databases.7 Elements required by the PRISMA guidelines and Cochrane Handbook to prevent publication bias.2,4,8,9
On close inspection of the raw data used, it can be seen that there is only a descriptive analysis of the studies, which does not allow to demonstrate categorical conclusions such as “We found a strong association between both conditions…” like in the SR.1 Moreover, if you want to show an association in a SR, you most select observational studies with effect estimators for these designs and should be performed a meta-regression analysis.10,11
Due to this, error may be observed in the selection, results and reports. Therefore, the results in this SR can influence the conclusion and findings of the study, Finally, the decisions or associations on its use need to be based upon unbiased summaries of the available evidence, therefore, your results should be interpreted with caution
FundingThe authors received no financial support for the research, authorship, and/or publication of this article.
Conflict of interestThe author (s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.