Muralidhar Reddy   Y

OBJECTIVES /BACKGROUND:
Stroke associated with COVID-19 has been characterised in several multicentre retrospective studies and meta-analyses. However, they did not distinguish first-ever ischemic strokes (F-AIS). Therefore, we aimed to study the incidence, clinical characteristics, and outcomes of a cohort of F-AIS associated with COVID-19 during the first wave of the pandemic and compare this cohort with those of F-AIS without COVID-19, COVID-19 without stroke. We also sought to compare the stroke admissions and mechanisms during the pandemic and immediate prepandemic periods.

MATERIAL and METHOD:
This study is an Institute Stroke Registry-based study and a retrospective review of prospectively collected data. The study period was between October 2019 and September 2020, and the Institutional Ethics Committee approved the study (Reference number - IEC/CARE/20837/2020/IIS). The study was conducted following the Declaration of Helsinki Ethical principles and Good Clinical Practices. Working definitions Time periods were defined as follows - 1) Pandemic period (first wave) April 2020 – September 2020; and 2) Pre-pandemic period October 2019 – March 2020. F-AIS was defined as focal cerebral infarction based on clinical and imaging evidence of ischemia in a defined vascular distribution occurring for the first time in life(8). Patients were categorised into five groups (Supplementary figure 1) as per the following criteria - a) Group I: All consecutive patients with F-AIS admitted during the pandemic and tested reverse transcriptase-polymerase chain reaction (RT-PCR) positive for SARS-CoV2 b) Group II: All consecutive patients with F-AIS admitted during the pandemic and tested negative for SARS-CoV2 c) Group III: All consecutive patients with F-AIS admitted in pre- pandemic period d) Group IV: -Consecutive COVID-19 patients, without stroke admitted during the first wave of COVID-19; and e) Group V: Consecutive patients admitted for neurological diseases other than stroke and tested negative for SARS-CoV2, during the pandemic. Data elements In patients with F-AIS (Group I, II and III), we collected data pertaining to demographics, vascular risk factors, clinical features, National Institutes of Health Stroke Scale (NIHSS) at admission, NIHSS stroke severity, time interval between onset of COVID-19 and stroke (in group I), vascular territory, mechanism as per Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification and duration of hospitalisation. Modified Rankin Scale (mRS) at 90 days and long-term outcomes were assessed by either video or physical consultation. In COVID-19 patients (groups I and IV), we collected details of disease severity, peak values of total leukocyte count (TLC), C-reactive protein (CRP), interleukin-6 (IL-6), procalcitonin (PCTQ), lactate dehydrogenase, serum ferritin, prothrombin time, and activated partial thromboplastin time, d-dimer and the lowest values of haemoglobin, absolute lymphocyte count and platelet count. In non-stroke cohorts (group IV and V), we collected details of demographics, vascular risk factors and comorbidities. Statistical Analysis We performed statistical analysis using SPSS version 21.0. (Armonk, NY: IBM Corp.). We used the unpaired student's t-test and Chi-square test as a test of significance for continuous and categorical data, respectively. We used the Mann-Whitney U test to compare differences between two groups when the dependent variable was not normally distributed. Adjustment for multiple comparisons was made using Bonferroni correction. A P-value ≤ 0.05 was considered significant. Multiple logistic regression (I and II) was done to identify the association between COVID-19 and 90-day mortality. Multiple logistic regression (I, II, IV and V) was done to identify the association between COVID-19 and the occurrence of F-AIS.

RESULT:
A total of 1369 patients were admitted to our hospital with COVID-19 during the first wave of pandemic. Of these, 25 (1.8%) developed acute stroke out of which 19 (1.38%) were ischemic and 6 (0.44%) were haemorrhagic strokes. 17 patients had (1.24%) had F-AIS and these were characterised and analysed further (Supplementary table 1 & Table 1). Mean age was 62.29 years (range 42-78), 13 (76.5%) were aged > 50 years and 11 (64.7%) males. Hypertension and diabetes was seen in 14 (82.4%) and 13 (76.5%), respectively. COVID-19 was severe in 11 (64.7%) and in 4 (23.5%) the infection was asymptomatic at presentation. The median duration between the onset of COVID-19 and stroke was 9 days (range 1–45). Five (29.4%) developed stroke during in-hospital stay. The stroke was severe in 5 (29.4%). Admission-NIHSS was 12.59±8.33 (range 2-25). The stroke mechanism was undetermined (cryptogenic) in 10 (58.8%) and lacunar in one (5.9%). Stroke was due to large artery occlusion (LAO) in 10 (58.8%) subjects (Figure 1&2) with involvement of anterior circulation in 14 (82.4%) and multiple territories in 2 (11.8%). Comparative analysis between the cohorts of F-AIS with and without COVID-19 is shown in Table Comparative analysis of seventy-five consecutive COVID-19 patients without stroke (admitted in April and May 2020) with COVID-19-associated F-AIS patients is shown in Table There was a significant drop of 40% in the total number of admissions into the neurology department during the pandemic compared to the immediate pre-pandemic period (Supplementary table 2). There were no differences in the proportion of stroke admission between the two periods (48.23% vs. 44.74%; p=0.38) [ischemic 36.86% vs. 30.56%; p=0.09; haemorrhagic 7.84% vs. 10.02%; p=0.34 and; F-AIS 23.9% vs. 27.9%; p=0.26]. When compared to prepandemic period, cryptogenic strokes were seen frequently in the pandemic (19.4% vs 33.8%; P=0.03). Multiple logistic regression was performed to find independent risk factors of F-AIS using the following variables - age, male gender, hypertension, diabetes, coronary artery disease, chronic kidney disease and COVID-19. It showed that age > 50 years (RR, 0.36 [95% CI, 0.20–0.67]; P<0.01), diabetes (RR, 4.03 [95% CI, 2.1-7.5]; P<0.01) and COVID-19 (RR, 0.35 [95% CI, 0.18–0.68]; P<0.01) were the independent risk factors. Among the patients with COVID-associated F-AIS, the mean duration of hospital stay was 13.6±11.7 (range 1-40) days. The 90-day mortality rate was 41.2%, and the median 90-day mRS in the remaining 10 (58.8%) patients was 2 (range 0-5). Multiple logistic regression showed that COVID-19 (RR, 4.81 [95% CI, 1.14–20.18] P=0.03) and admission-NIHSS (RR, 1.23 [95% CI, 1.06-1.42] P<0.01) were independently associated with 90-day mortality. There were significant differences in the outcome variables between COVID-19-associated and non-COVID-19 stroke groups: death rate (41.2% vs. 11.3%; p<0.01) and unfavourable outcome (64.7% vs. 31%; p=0.01). Of the remaining ten COVID-19-associated F-AIS subjects, one died due to urosepsis five months after the stroke. The mean long-term follow-up duration of the remaining nine patients was 20.00±1.00 (range 19-22) months, and seven of them showed good outcomes (mRS 0-2).

DISCUSSION:
Acute ischemic stroke is infrequent in patients with COVID-19, and the exact incidence is unknown. Observational studies suggest that 1–6% of hospitalised COVID-19 patients develop stroke(3). The most common type of stroke is ischemic stroke(6,9). In our study, of 1.8% of COVID-19 patients developed acute stroke, out of which 17 (1.24%) had F-AIS. Systematic reviews and meta-analyses suggest that COVID-19-associated ischemic stroke occurs more often in elderly males(6). The reported mean time interval between the COVID-19 and stroke was eight days(4). Similar were the observations in our study. COVID-19- associated ischemic strokes are more often severe. The reported median NIHSS was 15 (range 13-18)(6,10). In our study, the mean admission-NIHSS was 12.59 (range 8-33). COVID-19-associated ischemic stroke usually occurs in the presence of cardiovascular risk factors. They are more likely to have hypertension, diabetes mellitus, and coronary artery disease(4,6,10). Similar were our observations. Cryptogenic mechanism was observed in most case series(6). Small vessel disease was infrequently reported(11). Cryptogenic stroke accounted for 58.8% in our study. Of the 17 patients, 10 (58.8%) had LAO. Only one patient (5.9%) had lacunar stroke. The reported frequency of LAO in patients with COVID-19-associated ischemic stroke is higher(3-7,11). We compared the clinical characteristics of COVID- 19-associated F-AIS, and stroke without COVID-19. Admission-NIHSS and stroke severity was higher in COVID-19-associated stroke. Cryptogenic mechanism and LAO, longer hospital-stay and higher 90-day mortality were seen in COVID-19 F-AIS. We also compared features of COVID-19 subjects with and without stroke. Diabetes and hypertension were seen more frequently in the stroke. Severity of COVID-19 was more in stroke patients. Inflammatory and coagulation markers were significantly elevated in patients with COVID-19- associated stroke. We studied the differences between patients of non-COVID-19 F-AIS during the pandemic and those with F-AIS registered in the Institutional Stroke Register in the immediate six months before the pandemic. In-hospital stroke, cryptogenic mechanism and LAO stroke were higher in pandemic. Published data suggest that patients with severe COVID-19 had an increased risk of acute ischemic stroke than patients with non-severe COVID-19(3,6). In a study from Wuhan, stroke occurred in 5.7% of critically-ill patients compared to 0.8% with milder COVID-19(12).In our study, 64.7% of patients with COVID-19-associated stroke had severe COVID-19. Growing comparisons of COVID-19-positive and COVID-19-negative stroke cohorts support the association between COVID-19 and ischemic stroke(13). Multivariate logistic regression analysis of our data suggests that COVID-19 is an independent risk factor for F-AIS, in addition to age (>50 years) and diabetes. The exact mechanism by which COVID-19 increases the risk of ischemic stroke is unknown. Presumed hypothesis include coagulopathy, inflammation, platelet activation, and endotheliopathy(14-15). COVID-19 is associated with the activation of coagulation and inflammatory pathways. Activation of the coagulation pathway with elevated D-dimer and fibrinogen is a common feature in severe COVID-19. A recent review found consistently elevated D-dimer in severe COVID-19(16). 64.7% of COVID-19-associated stroke had severe infection in our study. D-Dimer was significantly higher in COVID-19-associated stroke. The other possible mechanism for initiating pathological thrombosis in COVID-19 is a hyper-inflammatory response with a resultant “cytokine storm”(17). Patients with COVID-19 have elevated IL-6 and CRP(18). In our study, patients with COVID-19 had higher IL-6 levels than patients without stroke. The CRP titres had not reached statistical significance. COVID-19-associated stroke is associated with high mortality. The reported pooled mortality rate was 31.7%, and the estimate is much higher in severely ill patients 84.8%(5,19). Short-term mortality rate in patients with COVID-associated stroke was 41.2% in our study. Compared to individuals of stroke without COVID-19, patients with infection and stroke had a higher mortality(6). COVID-19 was found to be an independent risk factor for 90-day mortality of stroke in present study. In other studies, the cryptogenic stroke subtype was an independent predictor of mortality, and diabetes has been shown to increase the mortality and severity of COVID-19(20-21). There was no difference in the mortality between patients with ischemic stroke who tested negative for COVID-19 during the pandemic and patients with ischemic stroke in the six months immediately before the pandemic in our study (17.0 vs 17.3, p=0.96). In our study, the mean in-hospital stay was longer (I3.6±11.7 vs 6.7±4.1, p<0.01) in COVID-19-associated stroke cohort than COVID-19 negative stroke cohort. Long-term functional outcome was favourable in most of the remaining nine patients (mRS 0-2 in 7; mRS 3-4 in 2). We followed up with the patients for a mean duration of 20.00±1.00 (range 19-22) months. Probably ours is the only study that studied the long-term functional outcome. The present study is unique in many aspects. First, this is perhaps the first study to characterise the F-AIS associated with COVID-19. Second, we studied the long-term outcomes. Third, we showed COVID-19 to be an independent risk factor of ischemic stroke and mortality. Notwithstanding, there are a few limitations. First, this is a retrospective study. Second, inflammatory and coagulation markers were not available in all COVID-19 patients. However, this accounted only for small proportion. Third, data pertaining to dyslipidaemia, hyperhomocysteinemia, tobacco and alcohol consumption was lacking in COVID-19 patients without stroke. Fourth, COVID-19 patients without stroke who were recruited were predominantly non-severe. However, recruiting consecutively admitted patients for analysis eliminated the bias. Fifth, the study is a single-centre study. In conclusion, F-AIS accounted for 1.24%. Compared with stroke patients without COVID-19, those with COVID-19 had higher stroke severity and five-time higher risk of death. They are mainly cryptogenic and showed LAO. Compared with COVID-19 without stroke, those with stroke had a higher incidence of diabetes, hypertension and chronic kidney disease, severe infection and elevated inflammatory and coagulation parameters. SARS-CoV2 is an independent risk factor for F-AIS.