Ramos-Araque et al. BMC Neurology (2021) 21:43 https://doi.org/10.1186/s12883-021-02075-1 RESEARCH ARTICLE Open Access Stroke etiologies in patients with COVID-19: the SVIN COVID-19 multinational registry María E. Ramos-Araque1,2, James E. Siegler3*, Marc Ribo4,5, Manuel Requena4,5, Cristina López2, Mercedes de Lera2, Juan F. Arenillas2, Isabel Hernández Pérez2, Beatriz Gómez-Vicente2, Blanca Talavera2, Pere Cardona Portela6, Ana Nuñez Guillen6, Xabier Urra7, Laura Llull7, Arturo Renú7, Thanh N. Nguyen8, Dinesh Jillella9, Fadi Nahab9, Raul Nogueira10, Diogo Haussen10, Ryna Then3, Jesse M. Thon3, Luis Rodríguez Esparragoza11, Maria Hernández-Pérez11, Alejandro Bustamante11, Ossama Yassin Mansour12, Mohammed Megahed13, Tamer Hassan14, David S. Liebeskind15, Ameer Hassan16,17, Saif Bushnaq18, Mohamed Osman18, Alejandro Rodriguez Vazquez7 and SVIN Multinational Registry and Task Force Abstract Background and purpose: Coronavirus disease 2019 (COVID-19) is associated with a small but clinically significant risk of stroke, the cause of which is frequently cryptogenic. In a large multinational cohort of consecutive COVID-19 patients with stroke, we evaluated clinical predictors of cryptogenic stroke, short-term functional outcomes and in- hospital mortality among patients according to stroke etiology. Methods: We explored clinical characteristics and short-term outcomes of consecutively evaluated patients 18 years of age or older with acute ischemic stroke (AIS) and laboratory-confirmed COVID-19 from 31 hospitals in 4 countries (3/1/20–6/16/20). Results: Of the 14.483 laboratory-confirmed patients with COVID-19, 156 (1.1%) were diagnosed with AIS. Sixty-one (39.4%) were female, 84 (67.2%) white, and 88 (61.5%) were between 60 and 79 years of age. The most frequently reported etiology of AIS was cryptogenic (55/129, 42.6%), which was associated with significantly higher white blood cell count, c-reactive protein, and D-dimer levels than non-cryptogenic AIS patients (p/= 18 years of age hospitalized or evaluated in Normality of continuous data was assessed histographi- the emergency department (ED) with AIS and confirmed cally and confirmed using the Shapiro-Wilk test. Non- diagnosis of COVID-19 by oropharyngeal polymerase normally distributed continuous variables are summa- chain reaction (PCR) or IgG and/or IgM antibody in rized as medians with interquartile range. Categorical sera, using local commercial assays, in 31 hospitals variables are presented as frequencies (%). Descriptive across 4 countries. A more detailed summary of the statistics were used to compare groups, with the χ2 test Methods has been previously described [7]. Informed for discrete variables (or Fisher’s exact test when cell consent was waived by the local institutional review counts were 5 or less), and the Wilcoxon Rank-sum test board at each participating center. No sample size calcu- or Kruskal-Wallis equality of populations rank test for lations were made for this study as analyses were ex- non-normally distributed continuous variables, as appro- ploratory and hypothesis-generating. The results of this priate. Bivariate analyses were performed to associate investigation are reported in accordance with the baseline variables with the occurrence of endpoints. STrengthening the Reporting of OBservational studies in Backward stepwise logistic regression was performed to Epidemiology (STROBE) statement. estimate the association between clinical, laboratory, and radiographic findings with the primary and secondary Data collection outcomes. Estimates of association are presented as odds Demographic, clinical, and laboratory findings (including ratios (OR) and respective 95% confidence intervals (CI), method of COVID-19 diagnosis), as well as stroke treat- with adjustment for all variables significant to p < 0.2 on ment (including intravenous thrombolysis, endovascular the respective bivariate analysis. Variables were elimi- thrombectomy, and antithrombotic therapy), and short- nated from each adjusted model if they were no longer term functional outcomes at time of discharge were col- significant to p < 0.2. All adjusted regression models lected. Stroke etiology was determined by local site in- were clustered by site. Using the remaining variables vestigators (vascular neurologists) using a modified Trial from the final backward stepwise regression model for of ORG 10172 in Acute Stroke Treatment (TOAST) mortality, we created a simple index assigning 1-point to classification [15]. Cryptogenic stroke was defined as each variable (age, male sex, history of diabetes, NIHSS, stroke of undetermined etiology after exclusion of and cryptogenic stroke; Supplementary Table 1). In Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 3 of 11 order to reduce overfitting, NIHSS was stratified such irrespective of SARS- CoV-2 infection during a similar that one point was assigned for values ≥10 and age period the year prior (3/1/19–05/31/19), the prevalence stratified such that one point was assigned if the patient of cryptogenic (unspecified or undetermined) stroke was were ≥ 60 years old based on visual association of these 19.7% (260/1319 patients) [19]. Using a two-sample test thresholds with respect to associated risk of mortality of proportions, the rate of cryptogenic stroke was signifi- (Supplementary Figs. 1 and 2). The mortality rate of the cantly higher in this population of COVID-19-associated cohort was assessed for each point value. A receiver op- stroke (42.6% vs. 19.6%, p < 0.001). Furthermore, the re- erating characteristic curve was generated to test the as- ported in-hospital mortality rate among these patients sociation between the mortality score and in-hospital was also significantly greater than in this historic control mortality. Model performance was assessed using the c- group (38.1% vs. 7.8%, p < 0.001). statistic with 95% CI. The statistical significance level of On admission, neurological deficits according to base- all tests was defined as a p-value ≤0.05. As this was an line NIHSS were generally moderate-to-severe among all exploratory study, no adjustments were made for mul- patients, with more severe deficits noted in patients with tiple comparisons. P-values are provided for conven- strokes due to a cryptogenic mechanism, cardioembo- tional purposes only and should be interpreted with lism, or other identifiable mechanism, compared to pa- caution. All tests were performed at the two-sided level tients with large or small vessel disease (p = 0.02; using STATA (College Station, TX) version 15.0. Miss- Table 2). ing data were not imputed. Fully de-identified data can Compared to non-cryptogenic stroke patients, those be made available upon reasonable request of the corre- with cryptogenic strokes had no difference in baseline sponding author. NIHSS (p = 0.53). Patients with cardioembolic stroke had more frequently identified intracranial arterial oc- Results clusions and cortical involvement compared to other Clinical features stroke mechanisms (p < 0.05 for both findings), followed Of the 14,483 laboratory-confirmed SARS-CoV-2 pa- by cryptogenic stroke. Admission white blood cell count, tients, 156 (1.1%) were diagnosed with a clinical and/or C-reactive protein and D-dimer levels were significantly radiographic acute ischemic stroke. Antecedent systemic greater for patients with cryptogenic stroke when com- symptoms were reported with variable frequency but in- pared to all other stroke groups, and after multivariable cluded fever in 69/134 patients (51%), 77/133 (58%) with regression with adjustment for all candidate variables cough, 60/128 (47%) dyspnea, 21/112 (19%) chest pain, significant to p < 0.2 (Hispanic ethnicity, congestive heart 18/107 (17%) myalgias, 12/110 (11%) headache, and < failure, cortical infarction, white blood cell count and D- 10% with symptoms of congestion, dizziness, odynopha- dimer), only congestive heart failure was statistically sig- gia, hyposmia, or hypogeusia. Among stroke patients, nificantly and inversely associated with cryptogenic 61/155 (39.4%) were female, 84/125 (67.2%) White, and stroke mechanism (p < 0.01; Table 3). more than half of patients (n = 88/143, 61.5%) were be- Patients with strokes due to cardioembolism and a tween 60 and 79 years of age. Seventeen of 145 patients cryptogenic mechanism had a significantly higher risk of (11.7%) received intravenous thrombolysis, and 14/156 any hemorrhagic transformation when compared to (9.0%) were reported to experience any intracerebral strokes of all other classifications, although event rates hemorrhage—1 patient experienced a NIHSS worsening for strokes due to ‘other’ causes and small vessel disease of 4 points attributed to a parenchymal hematoma, how- were low (p = 0.05; Table 4). Patients with cryptogenic ever this patient did not receive intravenous thromboly- stroke were non-significantly more likely to be dis- sis or thrombectomy. charged with greater disability than strokes of other The etiology of stroke was reported in 129 of 156 mechanisms (p = 0.07), despite having a similar pre- stroke patients (82.7%), with cryptogenic being com- morbid functional status. Furthermore, cryptogenic monly reported (n = 55, 42.6%). Compared to patients stroke patients had a significantly higher risk of in- with any other defined stroke mechanism (or multiple hospital mortality when compared to strokes of all other possible mechanisms), patients with cryptogenic stroke mechanisms (OR 2.27, 95%CI 1.01–5.08, p = 0.05). In the were more frequently Hispanic (51.0% vs. 33.3%, p = backward stepwise regression model for in-hospital mor- 0.05), had less atrial fibrillation (p < 0.01) and congestive tality, cryptogenic stroke remained independently associ- heart failure (p = 0.05; Table 1). ated with in-hospital mortality (aOR 5.16, 95%CI 1.41– There was no difference with respect to age, sex, race, 18.87, p = 0.01). In that model, older age (aOR 2.05 per or history of pre-morbid disability among patients with decade, 95%CI 1.35–3.11, p < 0.01), higher baseline cryptogenic versus known stroke mechanisms. NIHSS (aOR 1.12, 95%CI 1.02–1.21, p = 0.01), and his- In a separate analysis of 11 sites included in this study tory of diabetes (aOR 6.89, 95%CI 1.02–46.76, p = 0.05) which submitted data of ischemic stroke patients remained independently associated with mortality, Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 4 of 11 Table 1 Demographic data All patients acute Cryptogenic Cardioembolic Large vessel Small vessel Other p-valueb p-valuec ischemic strokea stroke stroke stroke stroke (n = 11) (n = 156) (n = 55) (n = 35) (n = 15) (n = 4) Age, no. (%) 0.10 0.97 < 30 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 30–39 3 (2.5) 1 (1.8) 2 (5.7) 0 (0) 0 (0) 2 (18.2 40–49 12 (10.0) 6 (10.9) 2 (5.7) 0 (0) 0 (0) 4 (36.4) 50–59 15 (12.5) 6 (10.9) 5 (14.3) 2 (13.3) 0 (0) 2 (18.2) 60–69 36 (30.0) 19 (34.6) 8 (22.9) 6 (40.0) 1 (25.0) 2 (18.2) 70–79 37 (30.8) 16 (29.1) 13 (37.1) 5 (33.3) 2 (50.0) 1 (9.1) 80–89 12 (10.0) 5 (9.1) 4 (11.4) 2 (13.3) 1 (25.0) 0 (0) > 89 5 (4.2) 2 (3.6) 3 (8.6) 0 (0) 0 (0) 0 (0) Sex, no. female (%) 61/155 (39.4) 19/55 (34.6) 15/35 (42.9) 5/14 (35.7) 0/4 (0) 8/11 (72.7) 0.08 0.45 Race, no. (%) 0.52 > 0.9 White 84/125 (67.2) 31/45 (68.9) 21/28 (75.0) 9/12 (75.0) 2/3 (66.7) 5/10 (50.0) Black 35/125 (28.0) 12/45 (26.7) 7/28 (25.0) 2/12 (16.7) 1/3 (33.3) 4/10 (40.0) Asian 2/125 (1.6) 1/45 (2.2) 0/28 (0) 0/12 (0) 0/3 (0) 0/10 (0) More than one race 2/125 (1.6) 1/45 (2.2) 0/28 (0) 1/12 (8.3) 0/3 (0) 0/10 (0) Other 2/125 (1.6) 0/45 (0) 0/28 (0) 0/12 (0) 0/3 (0) 1/10 (0) Hispanic ethnicity, no. (%) 58/140 (41.4) 25/49 (51.0) 10/33 (30.3) 4/12 (33.3) 4/4 (100) 10/11 (90.9) 0.02 0.05 d Diagnosis of COVID-19 , no. (%) 0.11 0.83 Nasopharyngeal PCR 153 (98.1) 55/55 (100) 35/35 (100) 15/15 (100) 4/4 (100) 10/11 (90.9) Serum IgM and/or IgG 6 (3.8) 2/55 (3.6) 3/35 (8.6) 0/15 (0) 1/4 (25.0) 1/11 (9.1) Known COVID-19 exposure, no. (%) 19 (12.2) 8/55 (14.6) 5/35 (14.3) 1/15 (6.7) 0/4 (0) 2/11 (18.2) 0.82 0.69 Medical history, no. (%) Hypertension 111/154 (72.1) 38/53 (71.7) 26/35 (74.3) 12/15 (80.0) 4/4 (100) 4/11 (36.4) 0.10 0.87 Diabetes mellitus 65/152 (42.8) 22/51 (43.1) 13/35 (37.1) 5/15 (33.3) 2/4 (50.0) 7/11 (63.6) 0.56 0.87 Dyslipidemia 58/142 (40.9) 18/53 (34.0) 17/35 (48.9) 7/15 (46.7) 3/4 (75.0) 0/11 (0) 0.01 0.33 Atrial fibrillation 22/148 (14.9) 0/52 (0) 16/35 (45.7) 0/15 (0) 0/4 (0) 0/11 (0) < 0.01 < 0.01 Congestive heart failure 27/154 (17.5) 5/53 (9.4) 9/35 (25.7) 4/15 (26.7) 0/4 (0) 1/11 (9.1) 0.17 0.05 Active tobacco use 15/145 (10.3) 4/52 (7.7) 5/33 (15.2) 0/15 (0) 1/4 (25.0) 0/11 (0) 0.21 0.68 Prior stroke 15/140 (10.7) 6/42 (12.5) 2/33 (6.1) 0/13 (0) 0/4 (0) 0/11 (0) 0.59 0.33 Chronic renal insufficiency 16/144 (11.1) 3/47 (6.4) 1/33 (3.0) 2/13 (15.4) 0/4 (0) 3/11 (27.3) 0.11 0.49 (stage III/IV or dialysis-dependent) Chronic obstructive pulmonary 11/135 (8.2) 2/48 (4.2) 3/33 (9.1) 0/13 (0) 1/4 (25.0) 0/11 (0) 0.28 0.24 disease and/or asthma Cancer 8/134 (6.0) 4/47 (8.5) 1/33 (2.9) 0/13 (0) 1/4 (25.0) 0/11 (0) 0.29 0.35 Premorbid mRS, median (IQR) 0 (0–1) 0 (0–0) 0 (0–1) 0 (0–0) 0 (0–1) 0 (0–0) 0.54 > 0.9 (n = 124) (n = 49) (n = 33) (n = 13) (n = 4) (n = 10) Of the 11 patients with ‘Other’ etiologies, 2 had hypercoagulable states due to malignancy, 1 with hypercoagulability on laboratory testing unrelated to malignancy, 2 with arterial dissection, 1 with vasculitis due to mucormycosis, 3 with radiographic findings of posterior reversible encephalopathy with infarction, and 2 with etiologies not otherwise specified a Acute ischemic stroke includes suspected or radiographically-confirmed cases, and in patients with multiple possible stroke mechanisms (n = 9). Patients with multiple stroke mechanisms were not included in the subsequent columns in order to minimize heterogeneity b P-values indicate categorical comparisons between all stroke subtypes c P-values indicate comparisons between cryptogenic and non-cryptogenic stroke subtypes (including strokes due to multiple possible etiologies) d 3 patients with an acute ischemic stroke were diagnosed with COVID-19 using both serum antibodies and nasopharyngeal PCR. COVID-19 denotes coronavirus disease 2019, PCR polymerase chain reaction whereas male sex was slightly but non-significantly asso- were associated with mortality in univariate modeling. A ciated with mortality (aOR 2.39, 95%CI 0.69–8.28, p = simple risk score was derived from the five variables 0.17; supplementary Table 1). Neither intravenous from the backward stepwise regression model by assign- thrombolysis (OR 0.82, 95%CI 0.27–2.52, p = 0.73) nor ing 1-point each in order to estimate the risk of in- thrombectomy (OR 0.97, 95%CI 0.40–2.34, p = 0.95) hospital mortality. Additional points showed an Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 5 of 11 Table 2 Clinical, laboratory, and radiographic measures and management All patients acute Cryptogenic Cardioembolic Large vessel Small vessel Other p-valueb p-valuec ischemic strokea stroke stroke stroke stroke (n = 11) (n = 156) (n = 55) (n = 35) (n = 15) (n = 4) Clinical and radiographic findings Baseline NIHSS median, (IQR) 13 (5–21) 12 (6–23) 14 (6–21) 8 (4–11) 2 (1–2) 21 (9–22) 0.02 0.53 (n = 130) (n = 48) (n = 32) (n = 14) (n = 4) (n = 10) Head CT performed, no. (%) 135/151 (89.4) 48/55 (87.3) 34/35 (97.1) 14/15 (93.3) 4/4 (100) 10/11 0.56 0.25 (90.9) First CT indicating acute stroke, no. (%) 76/132 (57.6) 33/48 (68.8) 21/34 (61.8) 10/14 (71.4) 0/4 (0) 4/10 (40.0) 0.04 0.10 CT angiogram performed, no. (%) 99/151 (65.6) 36/55 (65.5) 23/35 (65.7) 14/15 (93.3) 4/4 (100) 10/11 0.07 0.36 (90.9) Intracranial occlusion, no. (%) 53/107 (49.5) 17/37 (46.0) 17/24 (70.8) 6/14 (42.9) 0/4 (0) 3/10 (30.0) 0.04 0.70 MRI brain performed, no. (%) 55/151 (36.4) 21/55 (38.2) 11/35 (31.4) 5/15 (33.3) 1/4 (25.0) 8/11 (72.7) 0.17 > 0.9 MRI evidence of acute stroke, no. (%) 45/55 (81.8) 15/21 (71.4) 9/11 (81.8) 5/5 (100) 1/1 (100) 8/8 (100) 0.42 0.05 CT or MRI location of infarction, no. (%) Cortical 81/101 (80.2) 29/39 (74.4) 26/28 (92.9) 11/12 (91.7) 0/2 (0) 10/10 < 0.01 0.09 (100) Subcortical supratentorial 54/101 (53.5) 18/39 (46.2) 15/28 (53.6) 5/12 (41.7) 2/2 (0) 7/10 (70.0) 0.43 0.38 Infratentorial 12/101 (11.9) 7/39 (18.0) 0/28 (0) 1/12 (8.3) 0/2 (0) 1/10 (10.0) 0.12 0.11 Laboratory testing D-dimer (mcg/mL), median (IQR) 0.78 (0.13–4.9) 1.05 0.14 (0.05–0.21) 0.61 0.07, 0.49 0.26 0.04 0.02 (n = 87) (0.21–9.15) (n = 18) (0.14–10.08) (n = 2) (0.13–1.3) (n = 36) (n = 8) (n = 6) Admission platelet count (cells/μL), median 223 (183–313) 242 221 (183–289) 259 198 212 0.34 0.25 (IQR) (n = 143) (201–321) (n = 34) (188–353) (106–326) (150–281) (n = 53) (n = 15) (n = 4) (n = 9) C-reactive protein peak (mg/dL), median 11 (3–25) 15.8 4.2 (2.4–12.7) 12.9 6.2 (0.7–49.9) 11.6 0.14 0.05 (IQR) (n = 96) (6.2–52.7) (n = 23) (0.8–74.9) (n = 3) (9.5–58.9) (n = 38) (n = 11) (n = 4) Admission WBC (cells/μL), median (IQR) 8.4 (6.5–11.9) 10.7 8.0 (6.5–10.8) 6.8 (5.8–9.5) 7.7 (5.7–9.3) 6.4 0.05 0.01 (n = 139) (6.8–14.4) (n = 35) (n = 15) (n = 4) (5.5–6.7) (n = 53) (n = 9) Lymphocyte count (cells/μL), median (IQR) 1.2 (0.7–1.7) 1.3 (0.7–1.8) 1.4 (1.0–1.9) 0.7 (0.7–1.1) 1.7 (1.2–1.8) 1.4 0.05 0.60 (n = 132) (n = 47) (n = 32) (n = 15) (n = 4) (0.7–1.5) (n = 9) Treatment Intravenous thrombolysis, no. (%) 17/145 (11.7) 4/54 (7.4) 7/35 (20.0) 3/15 (20.0) 0/4 (0) 2/11 (18.2) 0.31 0.12 Endovascular treatment, no. (%) 33/114 (29.0) 8/42 (19.1) 14/26 (53.9) 4/10 (40.0) 0/3 (0) 2/7 (28.9) 0.03 0.03 Endovascular treatment d, no. (%) 33/53 (62.3) 8/17 (47.1) 14/17 (82.4) 4/6 (66.7) n/a 2/3 (66.7) 0.18 0.11 Acute antithrombotic treatment e, no. (%) 0.01 0.17 Single or dual antiplatelet therapy 66/108 (61.1) 31/43 (72.1) 13/30 (43.3) 12/12 (100) 2/3 (66.7) 6/10 (60.0) f Therapeutic anticoagulation 22/108 (20.4) 5/43 (11.6) 11/30 (36.7) 0/12 (0) 1/3 (33.3) 1/10 (10.0) None 20/108 (18.5) 7/43 (16.3) 6/30 (20.0) 0/12 (0) 0/3 (0) 3/10 (30.0) a Acute ischemic stroke includes suspected or radiographically-confirmed cases, and in patients with multiple possible stroke mechanisms (n = 9). Patients with multiple stroke mechanisms were not included in the subsequent columns in order to minimize heterogeneity b P-values indicate categorical comparisons between all stroke subtypes c P-values indicate comparisons between cryptogenic and non-cryptogenic stroke subtypes (including strokes due to multiple possible etiologies) d Analysis limited to patients with any intracranial occlusion on vascular imaging e Acute antithrombotic treatment was defined as use of an antithrombotic within 24 h of ischemic stroke diagnosis, or between 24 and 48 h of ischemic stroke diagnosis in patients who received intravenous thrombolysis f One patient with a stroke due to multiple etiologies was treated with combination antiplatelet (single or dual) with therapeutic anticoagulation NIHSS Denotes National Institutes of Health Stroke Scale, IQR Interquartile range, CT Computed tomography, MRI Magnetic resonance imaging, HI-1 Denotes hemorrhagic infarction grade 1 (petechial hemorrhage), HI-2 Hemorrhagic infarction grade 1 (confluent petechiae), PH-1 Parenchymal hematoma grade 1 (confluent hemorrhage within < 30% of infarct bed), PH-2 Parenchymal hematoma grade 2 (confluent hemorrhage within > 30% of infarct bed and having mass effect), ICU intensive care unit and mRS modified Rankin Scale incrementally greater probability of mortality (Fig. 1), curve 0.79, 95%CI 0.70–0.88). Three or more points in- with the total score having a strong discriminatory dicated an 88.9% sensitivity and 53.0% specificity for the power to predict in-hospital mortality (area under the outcome of in-hospital mortality. Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 6 of 11 Table 3 Multivariable logistic regression model evaluating predictors for cryptogenic stroke in Covid-19 patients Variable Univariate analysis Multivariable analysis OR (95% CI) p-value OR (95% CI) p-value Age, per decade 0.98 (0.75–1.27) 0.86 Male sex 1.32 (0.64–2.73) 0.45 White race 1.06 (0.46–2.42) > 0.9 Hispanic ethnicity 1.44 (0.99–2.10) 0.05 Dropped from model due to non- significance Medical history Hypertension 0.94 (0.43–2.06) 0.87 Diabetes mellitus 0.94 (0.46–1.93) 0.87 Dyslipidemia 0.70 (0.33–1.45) 0.34 Congestive heart failure 0.35 (0.12–1.02) 0.05 0.15 (0.05–0.44) < 0.01 Active tobacco use 0.76 (0.21–2.75) 0.68 Prior stroke 1.86 (0.53–6.47) 0.33 Chronic renal insufficiency (stage III/IV or dialysis- 0.61 (0.15–2.50) 0.50 dependent) Chronic obstructive pulmonary disease and/or asthma 0.39 (0.08–1.97) 0.26 Cancer 2.08 (0.44–9.74) 0.35 a Baseline NIHSS 0.003 (−0.006–0.013) 0.49 Imaging Cortical infarction 0.40 (0.14–1.16) 0.09 Dropped from model due to non- significance Large vessel occlusion 0.95 (0.37–1.95) 0.70 a Laboratory data Admission WBC (× 1000/mL) 0.02 (0.002–0.04) 0.03 Dropped from model due to non- significance Admission lymphocyte count (× 1000/mL) 0.0001 (−0.12–0.12) > 0.9 Admission platelet count (× 1000/mL) 0.0004 (−0.0003– 0.22 0.001) D-dimer (per mcg/mL) 0.009 (0.002–0.017) 0.02 1.73 (0.98–3.04) 0.06 C-reactive protein (per mg/dL) 0.001 (−0.0008–0.003) 0.23 Variables were entered into the multivariable model if they were significant to p < 0.2 in univariate regression (Hispanic ethnicity, congestive heart failure, cortical infarction, elevated d-dimer, elevated admission white blood cell count). Variables were retained if they remained significant to p < 0.2. Laboratory values indicate serologic studies collected closest to the time of stroke onset, unless otherwise noted a Variables in which β with 95% confidence interval were used to estimate the effect on the outcome of cryptogenic stroke in univariate analysis. If included in the multivariable model, this effect is displayed as an odds ratio with 95% confidence interval. Multivariable regression model was clustered by site Discussion evaluation for cardiac and vascular causes of cerebral in- In this observational cohort of more than 14,000 farction) in the present study would certainly strengthen COVID-19 patients treated at 31 hospitals in 4 coun- the observed association implicating COVID-19 as a po- tries, we found a high proportion of strokes among pa- tentially unique stroke mechanism. However, we did not tients with COVID-19 were attributed to a cryptogenic centrally adjudicate vascular, echocardiographic, or par- mechanism (42%). This finding is concordant with the enchymal imaging in order to confirm the diagnosis of growing literature indicating a high prevalence of crypto- cryptogenic infarction. Our determination of cryptogenic genic stroke in COVID-19, [6, 13, 20, 21] with increasing stroke is congruent with many published reports which data suggesting the novel human coronavirus may be a did not report a detailed methodology for determination novel stroke mechanism. In this study, a cryptogenic of cryptogenic mechanism, [22] did not require intracra- stroke diagnosis was made at the judgment of the treat- nial vascular imaging, [23] or included patients for ing vascular neurologist and was inclusive of strokes whom workup could not be completed [20]. Therefore, with incomplete workup. Use of more stringent trial cri- we do not believe the clinical determination of crypto- teria for “cryptogenic stroke” (which requires thorough genic stroke to be a significant limitation of this analysis. Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 7 of 11 Table 4 Outcome measures All patients acute Cryptogenic Cardioembolic Large vessel Small vessel Other p-valueb p-valuec ischemic strokea stroke stroke stroke stroke (n = 11) (n = 156) (n = 55) (n = 35) (n = 15) (n = 4) Hemorrhagic transformation of 17/98 (17.4) 6/42 (14.3) 3/26 (11.5) 0/8 (0) 0/4 (0) 4/7 0.05 0.55 infarction, no. (%) (57.1) Grade of hemorrhage > 0.9 0.82 HI-1 7/17 (41.2) 3/6 (50.0) 1/3 (33.3) n/a n/a 2/4 (50.0) HI-2 2/17 (11.8) 1/6 (16.7) 0/3 (0) n/a n/a 1/4 (25.0) PH-1 4/17 (23.5) 1/6 (16.7) 1/3 (33.3) n/a n/a 1/4 (25.0) PH-2 4/17 (23.5) 1/6 (16.7) 1/3 (33.3) n/a n/a 0/4 (0) Symptomatic ICHd, no. (%) 1/14 (7.1) 1/4 (25.0) 0/3 (0) n/a n/a 0/4 (0) > 0.9 > 0.9 Transfer to ICU, no. (%) 69/145 (47.6) 26/52 (50.0) 11/33 (33.3) 10/15 (66.7) 0/4 (0) 10/11 < 0.01 0.36 (90.9) Intubation, no. (%) 51/139 (36.7) 17/47 (36.2) 7/32 (21.9) 4/15 (26.7) 0/4 (0) 7/11 0.08 0.56 (63.6) Discharge disposition, no. (%) 0.57 0.55 Home 34/136 (25.0) 11/50 (22.0) 9/29 (31.0) 4/14 (28.6) 2/4 (50.0) 5/11 (45.5) Acute inpatient rehabilitation 18/136 (13.2) 8/50 (16.0) 3/29 (10.3) 1/14 (7.1) 1/4 (25.0) 2/11 (18.2) Skilled nursing facility or 20/136 (14.7) 5/50 (10.0) 6/29 (20.7) 4/14 (28.6) 0/4 (0) 2/11 subacute rehabilitation (18.2) Long-term acute care 5/136 (3.7) 2/50 (4.0) 1/29 (3.5) 2/14 (14.3) 0/4 (0) 0/11 (0) Other acute care facility 7/136 (5.2) 3/50 (6.0) 3/29 (10.3) 0/14 (0) 0/4 (0) 0/11 (0) Hospice 2/136 (1.5) 1/50 (2.0) 0/29 (0) 0/14 (0) 0/4 (0) 1/11 (9.1) Expired during hospitalization 53/139 (38.1) 20/50 (40.0) 7/29 (24.1) 3/14 (21.4) 1/4 (25.0) 1/11 (9.1) Modified Rankin Scale at 4 (2–6) 5 (3–6) 4 (2–5) 4 (2–5) 1 (0–4) 3 (2–4) 0.20 0.07 discharge, median (IQR) (n = 139) (n = 51) (n = 32) (n = 14) (n = 4) (n = 10) Discharge mRS 0–2, no. (%) 35/139 (25.2) 11/51 (21.6) 11/32 (34.4) 4/14 (28.6) 3/4 (75.0) 3/10 0.21 0.14 (30.0) In-hospital mortality, no. (%) 53/139 (38.1) 20/50 (40.0) 7/29 (24.1) 3/14 (21.4) 1/4 (25.0) 1/11 0.18 0.05 (9.1) a Acute ischemic stroke includes suspected or radiographically-confirmed cases, and in patients with multiple possible stroke mechanisms (n = 9). Patients with multiple stroke mechanisms were not included in the subsequent columns in order to minimize heterogeneity b P-values indicate categorical comparisons between all stroke subtypes c P-values indicate comparisons between cryptogenic and non-cryptogenic stroke subtypes (including strokes due to multiple possible etiologies) d Symptomatic ICH defined as a PH-2 grade hematoma with worsening of the NIHSS by 4 or more points Perhaps the most important observation in this co- mortality among COVID-19 stroke patients when hort is the relationship between stroke mechanism compared to their contemporary noninfected or his- and in-hospital mortality. When compared to other torical controls [21]. It should be noted that patients discharge dispositions, death was the most common in the present cohort contributed to this meta- singular discharge disposition in this cohort of stroke analysis by Katsanos and colleagues; however the pa- patients. When death was evaluated as a bivariate tients in the present study did not contribute to the outcome, having a cryptogenic stroke mechanism was comparison of in-hospital mortality rates among significantly and independently predictive of in- COVID-19 stroke patients and COVID-19-negative hospital mortality. The high mortality rate is consist- controls. ent with other cohorts [6, 13, 20] and adds to the re- In our study, admission white blood cell count, c- sults of one recent systematic review and meta- reactive protein, and D-dimer levels were significantly analysis demonstrating a 5-fold increase in in-hospital higher for patients later diagnosed with cryptogenic Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 8 of 11 Fig. 1 Mortality rate among patients with COVID-19 and stroke according to mortality score. One point was assigned for each clinical factor up to a total of 5 possible points: age 60 or more, male sex, NIHSS 10 or more, history of diabetes, and cryptogenic stroke stroke. It has been described that SARS-CoV-2 infection hospitalized stroke patients with COVID-19 when mak- is linked to a prothrombotic state with elevated D-dimer ing long-term plans of care. Patients in this cohort who levels [24]. This finding also suggests that hypercoagula- met each of the 5 criteria (age, male sex, diabetes, NIHS bility could be one of the causes underlying this high S 10+, cryptogenic stroke) were at an 80% chance of in- proportion of cryptogenic strokes [25]. The high inci- hospital mortality, whereas those who met at least 3 cri- dence of intracranial occlusion (46%) and cortical teria were still more likely to die than to survive (74.4%) strokes also suggest a disproportionate number hospitalization. However, this score is not without limi- of patients with embolic strokes who suffer from tations, and it certainly warrants external validation. COVID-19. These findings are in keeping with recently First, the NIHSS is not a specific indicator of stroke se- published observations from a New York hospital system verity in critically ill patients with COVID-19, as it can [20]. Together, these observations validate the relation- be confounded by sedation. Second, a patient’s history of ship between a significant inflammatory and/or pro- diabetes does not reflect how well or poorly controlled thrombotic state and clinically significant stroke that is the medical condition is, and we did not evaluate the unrelated to other traditional stroke risk factors (e.g., risk of mortality based on markers of diabetic control cervical artery atherosclerosis or atrial fibrillation). Fur- (e.g., hemoglobin A1c) or duration of disease. Further, thermore, the higher mortality rate observed in patients the diagnosis of a stroke as ‘cryptogenic’ may be contin- with cryptogenic stroke speaks to the severity of gent upon which testing was or was not pursued to COVID-19 experienced by these patients. On the other evaluate the proximate cause of stroke in a critically ill hand, diagnostic workup for patients in this study (and patient. As in previously published reports of crypto- in other studies) [20, 22, 23] may not have been com- genic stroke in COVID-19 patients [20], patients with a pleted for some patients with COVID-19 in whom care high probability of dying after a severe stroke may not was being withdrawn or who expired soon before have undergone a more comprehensive workup, and workup could have been completed. This might have therefore could have been prematurely classified as confounded our observation of a high rate of crypto- ‘cryptogenic’. Determination of a stroke being crypto- genic stroke and associated mortality, as reported in one genic on the basis of incomplete workup due to high prior study [11]. probability of imminent death or pursuit of comfort Due to the strong relationship between cryptogenic measures may have falsely contributed to its association stroke and early mortality, we derived a simple risk score with in-hospital mortality. That said, as we have previ- for mortality that could serve as a useful tool in ously reported [7], the median delay from stroke onset Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 9 of 11 to death was 4 days in this cohort, (IQR 1–10 days) 16. Mark Heslin3 which suggests that echocardiographic and vascular im- 17. Chris Higham3 18. Ashutosh P. Jadhav21,22 aging could have been performed in advance of 19. Tudor G. Jovin3 expiration. 20. Artem Kaliaev8 While this study is one of the largest observational co- 21. Priyank Khandelwal26 22. Rakeshh Khatri27 horts of consecutive COVID-19 patients with cerebro- 23. Amy Krueger18 vascular complications, it is limited by its retrospective 24. Carlos Laredo7 nature and the completeness of data elements that were 25. Italo Linfante28 26. Antonio López7 abstracted from the medical record. The imaging find- 27. Racheal McCoy3 ings were not centrally adjudicated. Selection and inter- 28. Mònica Millàn11 pretation of diagnostic tests were made at the discretion 29. Mahmoud H. Mohammaden10 30. Leigh Moore18 of the treating physician, and the derivation of a simpli- 31. Isaac Nuño Ruiz11 fied mortality score warrants external validation. That 32. Víctor Obach7 said, our data reflect a large cohort of consecutive pa- 33. Darko Quispe Orozco23 34. Santiago Ortega-Gutierrez23 tients evaluated at many sites and reflect the diverse ex- 35. Pratit Patel26 periences in care of patients with stroke and COVID-19. 36. Mary S. Patterson18 37. Gonzalo Valle Peñacoba2 38. Leonardo Pisani10 Conclusions 39. Laurie Preston16 Although the overall incidence rate of acute ischemic 40. Razvan Alexandru Radu29 41. Vivek Rai30 stroke in hospitalized patients with COVID-19 is small, 42. Anna Ramos-Pachón11 there is a higher than expected proportion of patients 43. Ankit Rana3 with cryptogenic stroke. Furthermore, patients with 44. Srikant Rangaraju10 45. Jose Rafael Romero8 cryptogenic stroke and COVID-19 are at a significantly 46. Salvatore Rudilosso7 greater risk of early mortality when compared to patients 47. Emma Sanborn3 with COVID-19 and known, traditional stroke mecha- 48. Sunil Sheth20 49. Julie G. Shulman8 nisms. The relationship between cryptogenic stroke in 50. Amit Singla27 COVID-19 and mortality may be driven by more severe 51. Ainsley Smith3 inflammatory or prothrombotic disease in COVID-19. A 52. Amy Starosciak3 53. Lauren Thau3 simple 5-point score may be useful in clinical decision 54. Ephrem Teklemariam3 making for patients at a low risk of short-term survival, 55. Elena Oana Terecoasa29,32 however this score requires external validation. 56. Cristina Tiu29,32 57. Vlad Eugen Tiu29 58. Israr Ul Haq18 Supplementary Information 59. Martha Vargas7 The online version contains supplementary material available at https://doi. 60. Víctor Vera7 org/10.1186/s12883-021-02075-1. 61. Osama Zaidat18 62. Cynthia Zevallos23 63. Alicia M. Zha20 Additional file 1 Supplementary Table 1. Supplementary Figs. 1 Affiliations: and 2 1 Institute of Biomedical Research of Salamanca, Department of Neurology, Hospital Universitario de Salamanca, Salamanca, Spain. 2 Department of Neurology, Hospital Clínico Universitario de Valladolid, Acknowledgements Valladolid, Spain. This work would not have been possible without the coordination and 3 Cooper Neurologic Institute, Cooper University Hospital, Camden, New collaborators of the SVIN COVID-19 Multinational Cohort. Jersey, USA. Consortia Author list 4 Stroke Unit, Department of Neurology, Vall d'Hebron Research Institute, The consortium name is: SVIN Multinational Registry and Task Force3 Barcelona, Spain. 1. Patricia Feineigle3 5 Departament de Medicina, Universitat Autónoma de Barcelona, Barcelona, 2. Mohamad Abdalkader8 Spain. 24 3. Sergio Amaro7 6 Department of Neurology, Hospital Universitari, Bellvitge, Barcelona, Spain. 4. Hugo Aparicio8 7 Department of Neurology, Hospital Clínic, Barcelona, Spain. 5. Ivo Bach19 8 Department of Neurology, Department of Radiology, Department of 6. Jordi Blasco7 Neurosurgery, Boston Medical Center, Boston University School of Medicine, 7. Ángel Chamorro7 MA, USA. 8. Judith Clark8 9 Department of Neurology, Emory University School of Medicine, Atlanta, 9. Alexandra Czap20 USA. 10. Natalia Perez de la Ossa11 10 Department of Neurology, Grady Memorial Hospital, Atlanta, GA, USA. 11. Shashvat Desai21,22 11 Stroke Unit, Neuroscience Department, Hospital Universitari Germans Trias i 12. Laura Dorado11 Pujol, Badalona, Barcelona, Spain. 13. Denise Evans16 12 Neurology Department, Stroke and neurointervention unit, Alexandria 14. Mudassir Farooqui23 University, Egypt. 15. Meritxell Gomis11 Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 10 of 11 13 Critical care Department, Alexandria University, Egypt Competing interests 14 Neurosurgery Department, Stroke and Neurointervention unit, Alexandria No competing financial interests exist for any of the authors or collaborators. University, Egypt. 15 Department of Neurology, Ronald Reagan UCLA Medical Center, Los Author details 1 Angeles, USA Institute of Biomedical Research of Salamanca, Department of Neurology, 16 Department of Clinical Neuroscience Research, Valley Baptist Medical Hospital Universitario de Salamanca, Salamanca, Spain. 2Department of Center, Harlingen TX, USA. Neurology, Hospital Clínico Universitario de Valladolid, Valladolid, Spain. 17 3 Department of Neurology, University of Texas Rio Grande Valley, Harlingen, Cooper Neurologic Institute, Cooper University Hospital, Camden, NJ, USA. 4 TX, USA. Stroke Unit, Department of Neurology, Vall d’Hebron Research Institute, 18 Neuroscience Institute, Bon Secours Mercy Health St. Vincent Hospital, Barcelona, Spain. 5Departament of Medicina, Universitat Autónoma de Toledo, OH, USA. Barcelona, Barcelona, Spain. 6Department of Neurology, Hospital Universitari, 19 New Jersey Medical School, Newark, NJ, 07103, USA Bellvitge, Barcelona, Spain. 7Department of Neurology, Hospital Clínic, 20 Department of Neurology, McGovern Medical School, University of Texas Barcelona, Spain. 8Department of Neurology, Department of Radiology, Health Science Center, Houston, TX, 77030, USA Department of Neurosurgery, Boston Medical Center, Boston University 21 University of Pittsburgh Medical Center Mercy Hospital, Pittsburgh, PA, School of Medicine, Boston, MA, USA. 9Department of Neurology, Emory 15219, USA University School of Medicine, Atlanta, USA. 10Department of Neurology, 22 University of Pittsburgh Medical Center Presbyterian Medical Center, Grady Memorial Hospital, Atlanta, GA, USA. 11Stroke Unit, Neuroscience Pittsburgh, PA 15213, USA Department, Hospital Universitari Germans Trias i Pujol, Badalona, Barcelona, 23 Department of Neurology, University of Iowa Hospitals and Clinics, Iowa Spain. 12Neurology Department, Stroke and neurointervention unit, City, IA, 52242, USA Alexandria University, Alexandria, Egypt. 13Critical care Department, 26 Department of Neurology, Robert Wood Johnson University Hospital, New Alexandria University, Alexandria, Egypt. 14Neurosurgery Department, Stroke Brunswick, NJ, 08901, USA and Neurointervention unit, Alexandria University, Alexandria, Egypt. 27 15 Department of Neurosurgery, Robert Wood Johnson University Hospital, Department of Neurology, Ronald Reagan UCLA Medical Center, Los New Brunswick, NJ, 08901, USA Angeles, USA. 16Department of Clinical Neuroscience Research, Valley Baptist 27 Division of Neurointerventional Surgery, Department of Neurology, Medical Center, Harlingen, TX, USA. 17Department of Neurology, University of Lutheran Health, Fort Wayne, IN, 46804, USA Texas Rio Grande Valley, Harlingen, TX, USA. 18Neuroscience Institute, Bon 28 Department of Interventional Neuroradiology and Endovascular Secours Mercy Health St. Vincent Hospital, Toledo, OH, USA. Neurosurgery, Baptist Health South Florida, Coral Gables, FL, 33146, USA 29 Department of Neurology, University Emergency Hospital Bucharest, Received: 18 November 2020 Accepted: 19 January 2021 Bucharest, Romania 30 OhioHealth Neuroscience Center, Riverside Methodist Hospital, Columbus, OH, 43214, USA 31 Center for Outcomes Research, Baptist Health South Florida, Coral Gables, References Florida 33146, USA 1. Dong E, Du H, Gardner L. An interactive web-based dashboard to track 32 Carol Davila" University of Medicine and Pharmacy, Bucharest, Romania COVID-19 in real time. Lancet Infect Dis 2020; 20:533–534. https://doi.org/ https://doi.org/10.1016/S1473-3099(20)30120-1. 2. Mao L, Jin H, Wang M, et al. Neurologic manifestations of hospitalized Authors’ contributions patients with coronavirus disease 2019 in Wuhan, China. JAMA Neurol. MERA drafted the manuscript, JES conceived the study drafted the 2020;77:683–90. https://doi.org/10.1001/jamaneurol.2020.1127. manuscript and performed statistical analysis. MR, MR, CL, MDL, JFA, IHP, 3. Ellul M, Benjamin L, Singh B, et al. Neurological associations of COVID-19. BGV, BT, PCP, ANG, XU, LL, AR, TNN, DJ, FN, RN, DH, RT, JMT, LRE, MH-P, AB, [published online ahead of print, 2020 Jul 2]. Lancet Neurol. 2020;S1474- OYM, MM, TH, DSL, AH, SB, MO, ARV acquired data and critically reviewed 4422(20):30221. https://doi.org/10.1016/S1474-4422(20)30221-0. the manuscript. All authors contributed to the article and approved the sub- 4. Romero-Sánchez C, Díaz-Maroto I, Fernández-Díaz E, et al. Neurologic mitted version. manifestations in hospitalized patients with COVID-19: The ALBACOVID registry. [published online ahead of print, 2020 Jun1]. Neurology. 2020. Funding https://doi.org/10.1212/WNL.0000000000009937. MER-A was funded by The Instituto de Salud Carlos III (JR19/00020), Spain. 5. Qureshi AI, Abd-Allah F, Alsenani F, et al. Management of acute ischemic stroke in patients with COVID-19 infection: Report of an international panel. Availability of data and materials [published online ahead of print, 2020 Jun 22]. Int J Stroke. 2020: The datasets used and/or analyzed during the current study are available 1747493020935396. https://doi.org/10.1177/17474930209353966. from the corresponding author on reasonable request. 6. Ntaios G, Michel P, Georgiopoulos G, et al. Characteristics and outcomes in patients with COVID-19 and acute ischemic stroke: the global COVID-19 Ethics approval and consent to participate stroke registry. [published online ahead of print, 2020 Jul 9]. Stroke. 2020; The research was carried out in accordance with the Declaration of Helsinki. STROKEAHA120031208. https://doi.org/10.1161/STROKEAHA.120.031208. The Cooper University Hospital Institutional Review Board (IRB) has approved 7. Siegler JE, Cardona P, Arenillas JF, et al. EXPRESS: cerebrovascular events this multicenter study. IRB #20–365. Informed consent was waived by the and outcomes in hospitalized patients with COVID-19: the SVIN COVID-19 local institutional review board at each participating center: Hospital Clínico multinational registry. Int J Stroke. 2020. https://doi.org/10.1177/ Universitario de Valladolid, Valladolid, Spain, Cooper Neurologic Institute, 1747493020959216. Cooper University Hospital, New Jersey, USA, Vall d’Hebron Hospital, 8. Zubair AS, McAlpine LS, Gardin T, et al. Neuropathogenesis and neurologic Barcelona, Spain, Hospital Universitari, Bellvitge, Barcelona, Spain, Hospital manifestations of the coronaviruses in the age of coronavirus disease 2019: Clínic, Barcelona, Spain, Boston Medical Center, Boston University School of a review. JAMA Neurol. 2020;77:1–10. Medicine, MA, USA, Grady Memorial Hospital, Atlanta, GA, USA, Hospital 9. Paterson R, Brown R, Benjamin L, et al. The emerging spectrum of COVID-19 Universitari Germans Trias i Pujol, Barcelona, Spain, Alexandria University, neurology: clinical, radiological and laboratory findings. [published online Egypt, Ronald Reagan UCLA Medical Center, Los Angeles, USA, Valley Baptist ahead of print, 2020 Jul 8]. Brain. 2020; awaa240. doi:https://doi.org/10.1093/ Medical Center, Harlingen TX, USA, Bon Secours Mercy Health St. Vincent brain/awaa240. Hospital, Toledo, OH, USA. 10. Joly BS, Siguret V, Veyradier A. Understanding pathophysiology of The results of this investigation are reported in accordance with the hemostasis disorders in critically ill patients with COVID-19. Intensive Care STrengthening the Reporting of OBservational studies in Epidemiology Med 2020;1–4. https://doi.org/https://doi.org/10.1007/s00134-020-06088-1. (STROBE) statement. 11. Majidi S, Fifi J, Ladner T, et al. Emergent large vessel occlusion stroke during New York City's COVID-19 Outbreak: Clinical Characteristics and Paraclinical Consent for publication Findings. [published online ahead of print, 2020 Jul 31]. Stroke. 2020; Not applicable. STROKEAHA120030397. https://doi.org/10.1161/STROKEAHA.120.030397. Ramos-Araque et al. BMC Neurology (2021) 21:43 Page 11 of 11 12. Spence JD, de Freitas GR, Pettigrew LC, et al. Mechanisms of Stroke in COVID-19. [published online ahead of print, 2020 Jul 20]. Cerebrovasc Dis. 2020:1–8. https://doi.org/10.1159/000509581. 13. Katz JM, Libman RB, Wang JJ, et al. Cerebrovascular Complications of COVID-19. [published online ahead of print, 2020 Aug 6]. Stroke. 2020; STROKEAHA120031265. https://doi.org/10.1161/STROKEAHA.120.031265. 14. Sweid A, Hammoud B, Bekelis K, et al. Cerebral ischemic and hemorrhagic complications of coronavirus disease 2019. [published online ahead of print, 2020 Jun 26]. Int J Stroke. 2020:1747493020937189. https://doi.org/10.1177/ 1747493020937189. 15. Adams HP Jr, Bendixen BH, Kappelle LJ, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of org 10172 in acute stroke treatment. Stroke. 1993;24:35–41. https://doi.org/10.1161/01.str.24.1.35. 16. Rankin J. Cerebral vascular accidents in patients over the age of 60. II Prognosis. Scott Med J. 1957;2:200–15. https://doi.org/10.1177/ 003693305700200504. 17. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19:604–7. https://doi.org/10.1161/01.str.19.5.604. 18. Harris PA, Taylor R, Minor BL, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inform. 2019;95:103208. https://doi.org/10.1016/j.jbi.2019.103208. 19. Ortega-Gutierrez S, Farooqui M, Zha A, et al. Decline in mild stroke presentations and intravenous thrombolysis during the COVID-19 pandemic: the Society of Vascular and Interventional Neurology Multicenter Collaboration [published online ahead of print, 2020 Dec 15]. Clin Neurol Neurosurg. 2020;201:106436. https://doi.org/10.1016/j.clineuro.2020.106436. 20. Yaghi S, Ishida K, Torres J, et al. SARS-CoV-2 and stroke in a New York healthcare system. Stroke. 2020;51:2002–11. https://doi.org/10.1161/ STROKEAHA.120.03033513. 21. Katsanos AH, Palaiodimou L, Zand R, et al. The impact of SARS-CoV-2 on stroke epidemiology and care: a meta-analysis [published online ahead of print, 2020 Nov 21]. Ann Neurol. 2020;10.1002/ana.25967. https://doi.org/10. 1002/ana.25967. 22. Rothstein A, Oldridge O, Schwennesen H, Do D, Cucchiara BL. Acute cerebrovascular events in hospitalized COVID-19 patients. Stroke. 2020;51(9): e219–22. https://doi.org/10.1161/STROKEAHA.120.030995. 23. Grewal P, Pinna P, Hall JP, et al. Acute ischemic stroke and COVID-19: experience from a comprehensive stroke center in Midwest US. Front Neurol. 2020;11:910. Published 2020 Aug 20. https://doi.org/10.3389/fneur. 2020.00910. 24. Beyrouti R, Adams ME, Benjamin L, et al. Characteristics of ischaemic stroke associated with covid-19. J Neurol Neurosurg Psychiatry. 2020;91:889–91. https://doi.org/10.1136/jnnp-2020-323586. 25. Hernández-Fernández F, Valencia H, Barbella-Aponte R, et al. Cerebrovascular disease in patients with COVID-19: neuroimaging, histological and clinical description. [published online ahead of print, 2020 Jul 9]. Brain. 2020:awaa239. https://doi.org/10.1093/brain/awaa239. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.