Response by Yalkun et al to Letter Regarding Article, "Impact of Infection on the Risk of Recurrent Stroke Among Patients With Acute Ischemic Stroke".

Abstract

HomeStrokeVol. 51, No. 12Response by Yalkun et al to Letter Regarding Article, “Impact of Infection on the Risk of Recurrent Stroke Among Patients With Acute Ischemic Stroke” Free AccessLetterPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyRedditDiggEmail Jump toFree AccessLetterPDF/EPUBResponse by Yalkun et al to Letter Regarding Article, “Impact of Infection on the Risk of Recurrent Stroke Among Patients With Acute Ischemic Stroke” Gulbahram Yalkun, MD, Jie Xu, MD, PhD and Yongjun Wang, MD Gulbahram YalkunGulbahram Yalkun Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, China. China National Clinical Research Center for Neurological Diseases, Beijing. Search for more papers by this author , Jie XuJie Xu Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, China. China National Clinical Research Center for Neurological Diseases, Beijing. Search for more papers by this author and Yongjun WangYongjun Wang https://orcid.org/0000-0002-9976-2341 Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, China. China National Clinical Research Center for Neurological Diseases, Beijing. Search for more papers by this author Originally published23 Nov 2020https://doi.org/10.1161/STROKEAHA.120.032319Stroke. 2020;51:e366In Response:We thank Zongliang Xu et al for their comment on our recent work regarding the association of infection with the risk of recurrent stroke in patients with acute ischemic stroke.1 In the letter to the editor, they raised 2 questions: (1) impact of prestroke infection on the risk of recurrent stroke during hospitalization according to stroke subtype (especially those with large artery atherosclerosis [LAA] subtype) and (2) association between prestroke infection burden and recurrent stroke during hospitalization in our study.In our study, prestroke infection was found to independently predict recurrent stroke during hospitalization. To evaluate whether infection had a stronger impact on recurrent events among those with LAA-related stroke, we analyzed the association of infection (prestroke infection and infection during hospitalization, respectively) with recurrent stroke according to stroke subtype via the same statistical method as the original study.Regarding the question about the subtype of recurrent acute ischemic stroke, these data were unavailable in our study. However, previous data showed that 50% to 80% of the recurrent acute ischemic stroke had the same etiology subtype as the index stroke,2–4 and thus, we used subtype (categorized by Trial of ORG 10172 in Acute Stroke Treatment criteria)5 of the index stroke as a substitute.We found that in the LAA subgroup, the absolute event rate of recurrent stroke was higher among those with prestroke infection (with prestroke infection, 4 of 61 [6.6%]; without prestroke infection, 99 of 2014 [4.9%]), though the odds ratio did not reach a significant level (adjusted odds ratio, 1.17 [95% CI, 0.41–3.36]; P=0.77). Reduced statistical power owing to very small sample size of the prestroke infection group might be an important reason. Meanwhile, patients with infection during hospitalization were found to have a higher risk of recurrent stroke in LAA subtype (with infection, 29 of 274 [10.6%]; without infection, 161 of 3216 [5.0%]; adjusted odds ratio, 1.75 [95% CI, 1.11–2.75]; P=0.02). In fact, the main exposure in our study was infection during hospitalization, while we considered associations of prestroke infection with acute ischemic stroke outcomes as ancillary tests to support the main analysis. Furthermore, infection during hospitalization was judged and recorded by local clinicians, while prestroke infection was reported by patients, which might cause recall bias, and thus the relevant results should be interpreted with caution.For the second issue mentioned in the letter, infection burden might be a more sensitive predictor of recurrent stroke, but data on pathogen of infection were unavailable in our study. Impact of infection burden on LAA-related recurrent stroke and underlying mechanisms remain to be elucidated.DisclosuresNone.FootnotesFor Disclosures, see page e366.