Journal of the American Heart Association ORIGINAL RESEARCH Determinants of First-E ver Stroke Severity in West Africans: Evidence From the SIREN Study Oladimeji Adebayo , MMed; Onoja Akpa , PhD; Osahon J. Asowata , MSc; Adekunle Fakunle , PhD; Fred S. Sarfo , PhD; Albert Akpalu , MD; Kolawole Wahab , PhD; Reginald Obiako , PhD; Morenikeji Komolafe , MD; Lukman Owolabi, PhD; Godwin O. Osaigbovo , MBBS; Akinkunmi Paul Okekunle , PhD; Taofiki Sunmonu, MBBS; Hemant K. Tiwari, PhD; Carolyn Jenkins , DrPH; Oyedunni Arulogun , PhD; Lambert Appiah, MD; Joshua Akinyemi , PhD; Abiodun M. Adeoye , MSc; Godwin Ogbole , MD; Joseph Yaria , MBBS; Donna Arnett , PhD; Philip Adebayo , MBBS; Benedict Calys-T agoe , MPH; Okechukwu S. Ogah , PhD; Olayemi Balogun , MSc; Luqman Ogunjimi , MBBS; Yaw Mensah , MD; Obiageli U. Agbogu- Ike , MBBS; Rufus Akinyemi , PhD; Bruce Ovbiagele , MD; Mayowa O. Owolabi , MD; SIREN BACKGROUND: Baseline stroke severity is probably partly responsible for poor stroke outcomes in sub- Saharan Africa. However, there is a paucity of information on determinants of stroke severity among indigenous Africans. We sought to identify the factors associated with stroke severity among West Africans in the SIREN (Stroke Investigative Research and Educational Networks) study. METHODS AND RESULTS: Stroke was diagnosed clinically and confirmed with brain neuroimaging. Severe stroke was defined as a Stroke Levity Scale score of ≤5. A multivariate logistic regression model was constructed to identify factors associated with stroke severity at 95% CI and a nominal cutoff of 5% type 1 error. A total of 3660 stroke cases were included. Overall, 50.7%% had severe stroke, including 47.6% of all ischemic strokes and 56.1% of intracerebral hemorrhage. Factors independently associated with se- vere stroke were meat consumption (adjusted odds ratio [aOR], 1.97 [95% CI, 1.43– 2.73]), low vegetable consumption (aOR, 2.45 [95% CI, 1.93–3 .12]), and lesion volume, with an aOR of 1.67 (95% CI, 1.03– 2.72) for lesion volume of 10 to 30 cm3 and aOR of 3.88 (95% CI, 1.93–7 .81) for lesion volume >30 cm3. Severe ischemic stroke was independently associated with total anterior circulation infarction (aOR, 3.1 [95% CI, 1.5–6 .9]), posterior circulation infarction (aOR, 2.2 [95% CI, 1.1– 4.2]), and partial anterior circulation infarction (aOR, 2.0 [95% CI, 1.2– 3.3]) compared with lacunar stroke. Increasing age (aOR, 2.6 [95% CI, 1.3–5 .2]) and lesion volume >30 cm3 (aOR, 6.2 [95% CI, 2.0– 19.3]) were independently associated with severe intracerebral hemorrhage. CONCLUSIONS: Severe stroke is common among indigenous West Africans, where modifiable dietary factors are independently associated with it. These factors could be targeted to reduce the burden of severe stroke. Key Words: determinant ■ SIREN ■ stroke severity ■ West Africa Stroke has a huge burden in Africa, with an annual stroke among Africans have been attributed to the high incidence rate of 316 per 100 000, a prevalence rate prevalence of undiagnosed and untreated cardiovascu-of up to 1.4 per 1000, and a 3-y ear fatality rate of lar risk factors, greater severity of risk factors or higher up to 84%.1,2 The higher burden and poor outcomes of sensitivity to the risk factors, and lack of access to care.3 Correspondence to: Mayowa O. Owolabi, MBBS, MSc, DrM, FAAN, FANA, FRCP, FAS, Center for Genomic and Precision Medicine, College of Medicine, University of Ibadan, Ibadan, Nigeria. Email: mayowaowolabi@yahoo.com This paper was sent to Meng Lee, Guest Editor, for review by expert referees, editorial decision, and final disposition. Supplemental Material is available at https://www.ahajo urnals.org/doi/suppl/ 10.1161/JAHA.122.027888 For Sources of Funding and Disclosures, see page 11. © 2023 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. JAHA is available at: www.ahajournals.org/journal/jaha J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 1 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans there is an urgent need to identify the determinants of CLINICAL PERSPECTIVE stroke severity in the African context. We investigated the sociodemographic data, vascular risk factors, and What Is New? radiological features that influence stroke severity in • We demonstrated enormous burden of severe West Africans. stroke in this largest study on stroke in Africa. • It is the broadest exploration of factors (sociode- mographic, clinical, laboratory, and radiological) METHODS for severe stroke among indigenous Africans. The data for this study can be accessed upon request • Our study is the first to demonstrate the pro- from the data access committee (sibs2017@gmail. tective effects of vegetable consumption and com). reduced meat consumption against severe stroke. Design What Are the Clinical Implications? The SIREN (Stroke Investigative Research and • Modifiable risk factors such as reduced meat Educational Networks) study is a case–c ontrol study consumption and increased vegetable con- conducted across 15 hospitals and adjoining com- sumption could be targeted for prevention of munities.15 For this study, we focused on patients with severe stroke among Africans. stroke who were recruited in the SIREN stroke project • Other key independent risk factors for severe stroke include lesion volume, while risk factors between January 1, 2013, and July 30, 2018. for severe ischemic stroke include total anterior circulation infarction, posterior circulation infarc- Study Population tion, and partial anterior circulation infarction. The SIREN project encompassed 2 West African • Patients with large lesion volume and specific countries, Nigeria and Ghana, with the 2 countries stroke subtypes associated with severe stroke constituting 58.68% of the subregion population.16,17 could benefit from more intensive acute care. Stroke cases were adults aged >18 years with clinico- radiological diagnoses of stroke including cranial com- puterized tomographic scan/magnetic resonance imaging within 10 days of symptom onset to aid in radi- ographic differentiation of ischemic from intracerebral Nonstandard Abbreviations and Acronyms hemorrhages. The eligibility criteria for participants are SIREN Stroke Investigative Research and available in Data S1. Educational Networks TACI total anterior circulation infarction Data Collection All eligible adult patients fulfilling the case definition for stroke confirmed by neuroimaging were recruited into Similarly, managing severe stroke places a tremendous the study from medical wards, stroke units, intensive strain on the West African subregion’s underresourced care units, emergency rooms, and outpatient stroke health care system and health resources.4– 7 It is not clinics of the participating centers by neurologists/ only a difficult, life- threatening condition in most cases, stroke physicians.15,18 Methods for stroke evaluation but it also imposes a significant socioeconomic bur- and phenotyping are available in Data S1. den on a low-r esource setting with limited government After documenting their sociodemographic and intervention.8 clinical parameters (such as blood pressure, waist Little is known about the profile and sociodemo- and hip circumferences, and height), the neurologists/ graphic, vascular, and radiological determinants of stroke physicians assessed the stroke severity using stroke severity among indigenous Africans.2,9 Although the Stroke Levity Scale. Other cardiovascular risk fac- various studies have explored the determinants of se- tors were assessed through laboratory investigations vere stroke in non- African populations, such studies are such as fasting lipid profile, glycated hemoglobin level, mostly among White individuals and have established fasting blood glucose,  electrocardiography  (ECG), associations between congestive cardiac failure and echocardiography, and carotid Doppler ultrasound myocardial infarction, female sex, age at stroke onset, according to the published SIREN protocol.15,18–2 0 small- vessel strokes, intracerebral hemorrhages, left- Data were collected by trained research assistants. sided brain infarcts, massive infarction, and the pres- Sociodemographic data (such as age; sex; occupa- ence of atrial fibrillation and severe strokes.10– 14 Given tion; income; educational attainment; ethnicity; na- the current increase in the region’s stroke burden, tionality; family history of stroke, cardiovascular, and J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 2 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans metabolic diseases; dietary history; history of alcohol, and lesion volume) were found to be independently as- substance, and tobacco use; physical activity and sociated with stroke severity, overall and in subgroup psychosocial stress) were obtained from patients or analysis of stroke primary types, and were used in caregivers (when patients were unable to answer the 2- way interaction analysis with selected established questions) in line with our published protocol.15,18 The stroke risk factors. The Hosmer- Lemeshow test was detailed definitions of the vascular risk factors and pro- used to assess the goodness of fit of each multivariate cedures for assessing them are available in Data S1 logistic regression model. All statistical tests of hypoth- and Table S1.15,18 Stroke lesion volume was determined eses were 2-s ided, with a P  value <0.05 considered using the ellipsoid equation.21 significant. Statistical analyses and graphics were pro- duced with SPSS version 20 (IBM, Armonk, NY) and Assessment of Stroke Severity R statistical program version 3.4.2 (R Foundation for The validated Stroke Levity Scale was used to assess Statistical Computing, Vienna, Austria). stroke severity at enrollment, with lower scores imply- ing more severe strokes, and stratified as mild, moder- Ethical Approval ate, or severe, with only minor predictive information The SIREN study is a multicenter study, and institu- lost.22– 27 The Stroke Levity Scale was classified as >5 tional review boards at all study sites provided ethi- (nonsevere stroke) or ≤5 (severe stroke).23 The Stroke cal approval for the study. Informed consent was Levity Scale was validated with the National Institutes obtained from all participants before enrollment. The of Health Stroke Scale score with a strong correlation overall coordinating institutional review board for the (rho=−0.79; P<0.0001).22– 27 SIREN study was the University of Ibadan/University College Hospital Ibadan, Nigeria (Approval No.: UI/ Statistical Analysis EC/13/0105). Bivariate analyses were performed to determine the relationship between stroke severity (as stratified by stroke type) and associated participants’ risk factors, RESULTS such as clinical presentation, prior vascular risk fac- Characteristics of All Participants tors, neuroradiological, carotid/vertebral artery Doppler A total of 3660 participants (severe stroke=1854 studies, ECG, and echocardiographic determinant [50.7%] and nonsevere stroke=1806 [49.3%]) were in- factors. Bivariate associations between risk factors cluded in the study, with 70.1% (2292/3268) confirmed and stroke severity were evaluated within stroke types as ischemic stroke and 29.9% (976/3268) as hemor- (ischemic and hemorrhagic) using the chi- square (or rhagic stroke on the basis of brain imaging. The mean Fisher’s exact) test for categorical outcomes and the age of participants was 59.8±14.4 years, with 43.7% independent t test for comparing continuous data. men and 1772 (51.4%) aged ≥60 years. We used unconditional multivariate logistic regres- sion models to determine the adjusted associations of risk factors with stroke severity in the combined Characteristics of Participants With stroke samples and stratified by stroke types (ischemic Severe Stroke and hemorrhagic). Here, we adjusted for sociodemo- Overall, 1854 (50.6%) had severe stroke, with 1090 of graphic, vascular, and lifestyle risk factors. In general, 2292 (47.6%) severe ischemic stroke and 548 of 976 covariates were selected for inclusion in adjusted mod- (56.1%) severe intracerebral hemorrhage (P≤0.001; els after literature review and empirical evidence on the Table 1). Those with severe stroke had less formal edu- basis of significant associations found in our initial bi- cation, used less alcohol, consumed fewer vegetables variate analyses. The odds ratios and 95% CIs in our and more meat before stroke occurrence, had less models were estimated. family history of cardiovascular diseases, had bigger Specifically, in model 1, we first assessed the as- lesion volumes, and had higher mean systolic blood sociation of stroke severity with established stroke pressure than those with nonsevere strokes (Table 1). risk factors on the basis of our previous investigations Participants with severe ischemic stroke had less for- and bivariate analyses of the present study. In model mal education; less family history of cardiovascular 2, we assessed the association of stroke severity with diseases; consumed less alcohol, fewer vegetables, sociodemographics, lifestyles, clinical characteristics, and more meat before the stroke; and had bigger le- comorbidities, radiological features, and echocardio- sion volumes, higher systolic blood pressure, and graphic variables. The predictor variables were selected higher diastolic blood pressure compared with those on the basis of bivariate significant association with with nonsevere ischemic stroke (Table 2). Participants stroke severity in the present study or evidence in the with severe intracerebral hemorrhage had less formal literature. Two risk factors (low vegetable consumption education; had less family history of cardiovascular J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 3 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans Table 1. Distribution of Vascular Risk Factors for Stroke Stratified by Stroke Severity Status Severe Severe All severe All nonsevere ischemic hemorrhagic stroke stroke stroke stroke Characteristics N=1854 N=1806 P value N=1090 N=548 P value Country, Ghana, n (%) 501 (27.0) 699 (38.7) <0.001 292 (26.8) 209 (38.1) <0.001 Age, y, n (%) ≤59 789 (42.6) 799 (44.2) 0.579 425 (39.0) 364 (66.4) <0.001 ≥60 848 (45.7) 826 (45.7) 664 (60.9) 184 (33.6) Sex, male, n (%) 900 (48.5) 936 (51.8) 0.159 545 (50.0) 355 (64.8) <0.001 Domicile, n (%) Rural 161 (8.6) 133 (7.4) 0.244 110 (10.1) 51 (9.3) 0.679 Semiurban 462 (24.9) 467 (25.9) 313 (28.7) 149 (27.2) Urban 1009 (54.4) 1026 (56.8) 664 (60.9) 345 (63.0) Marital status, n (%) Never married/single 58 (3.1) 68 (3.8) 0.362 33 (3.0) 25 (4.6) 0.284 Married 1211 (65.3) 1193 (66.1) 773 (6.7) 438 (79.9) Monthly income >$100, n (%) 873 (47.1) 920 (50.9) 0.097 566 (51.9) 307 (56.0) 0.120 Education, some, n (%) 1281 (69.01) 1378 (76.3) <0.001 819 (75.1) 462 (84.3) <0.001 Living situation, n (%) Loneliness 78 (4.2) 99 (5.5) 0.101 53 (4.9) 25 (4.6) 0.794 Living with others 1547 (83.4) 1522 (84.3) 1029 (94.4) 518 (94.5) Risk factors, n (%) Hypertension 1575 (85.0) 1552 (85.9) 0.241 1036 (95.1) 539 (98.4) <0.001 Dyslipidemia 1339 (72.2) 1401 (77.6) 0.001 921 (84.5) 418 (76.3) <0.001 Diabetes 592 (31.9) 637 (35.3) 0.083 146 (13.4) 446 (81.4) <0.001 Cardiac disease 203 (11.0) 185 (10.2) 0.363 162 (14.9) 41 (7.5) <0.001 Waist-t o- hip ratio raised, n (%) 1273 (68.7) 1254 (69.4) 0.035 857 (78.6) 416 (75.9) 0.045 BMI, kg/m2, mean±SD 26.6±5.3 26.7±5.2 0.441 26.7±5.3 26.5±5.2 0.308 BMI >30 kg/m2, n (%) 246 (13.3) 311 (17.2) 0.481 173 (15.9) 73 (13.3) 0.145 Physical inactivity, n (%) 85 (4.6) 58 (3.2) 0.021 58 (5.3) 27 (4.93) 0.721 Tobacco, any use, n (%) 150 (8.1) 171 (9.5) 0.186 95 (8.7) 55 (10.0) 0.385 Alcohol use categories, n (%) Never use 1151 (62.1) 999 (55.3) <0.001 810 (74.3) 341 (62.2) <0.001 Ever low use 262 (14.1) 340 (18.8) 150 (13.8) 112 (20.4) Ever high use 43 (2.5) 43 (2.4) 24 (2.2) 19 (3.5) Stress, n (%) 267 (14.4) 341 (18.9) 0.002 164 (15.1) 103 (18.8) 0.177 Cancer, n (%) 6 (0.3) 12 (0.7) 0.044 5 (0.5) 1 (0.2) 0.139 Depression, n (%) 107 (5.7) 138 (7.6) 0.065 69 (6.3) 38 (6.9) 0.904 Family history of CVD, n (%) 578 (31.2) 695 (38.5) <0.001 372 (34.1) 206 (37.6) 0.166 Adding salt at table, n (%) 115 (6.2) 117 (6.5) 0.927 63 (5.8) 52 (9.5) 0.007 Low vegetable consumption*, n (%) 460 (24.8) 352 (19.5) <0.001 289 (26.5) 171 (31.2) 0.059 Whole grain consumption, n (%) 1301 (70.2) 1258 (69.7) 0.061 859 (78.8) 442 (80.7) 0.620 Legume consumption, n (%) 1002 (54.1) 1024 (56.7) 0.500 667 (61.2) 335 (61.1) 0.745 Fruit consumption, n (%) 1262 (68.1) 1267 (70.2) 0.803 840 (77.1) 422 (77.0) 0.796 Sugar consumption or otherwise, n (%) 474 (25.6) 413 (22.9) 0.022 300 (27.5) 174 (31.8) 0.126 Regular meat consumption,† n (%) 1152 (62.1) 1046 (57.9) <0.001 747 (68.5) 405 (73.9) 0.040 Fish consumption or otherwise, n (%) 1391 (75.0) 1386 (76.7) 0.730 915 (83.94) 476 (86.9) 0.266 (Continued) J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 4 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First-S troke Severity in West Africans Table 1. (Continued) Severe Severe All severe All nonsevere ischemic hemorrhagic stroke stroke stroke stroke Characteristics N=1854 N=1806 P value N=1090 N=548 P value Lesion volume, cm3, n (%) <10 717 (38.7) 972 (53.8) <0.001 545 (50.0) 172 (31.4) <0.001 10– 30 327 (17.6) 297 (16.4) 153 (14.0) 174 (31.8) >30 401 (21.6) 185 (10.2) 237 (21.7) 164 (29.9) Blood pressure at presentation, mm Hg, mean±SD Systolic 162.9±32.2 153.8±29.1 <0.001 157.6±31.3 173.3±32.1 <0.001 Diastolic 97.7±19.2 95.4±119.7 0.421 93.85±17.9 105.33±20.0 <0.001 Fasting glucose, mean±SD 121.0±48.94 116.2±50.4 0.069 124.2±54.5 116.8±39.8 0.088 Neutrophil:lymphocyte ratio, mean±SD 7.4±30.5 5.2±22.0 0.053 7.3±38.7 8.4±13.9 0.624 Subgroup (ischemic and hemorrhagic stroke) analyses were based on individuals with data on stroke primary types confirmed by brain scan. BMI indicates body mass index; and CVD, cardiovascular disease. *Low vegetable consumption was defined as a self- reported frequency of vegetable consumption less than once per month; 12 months before stroke. †Regular meat consumption was defined as a self- reported frequency of meat intake more than once (including daily) per month; 12 months before stroke occurrence. diseases; consumed less alcohol, fewer vegetables, progressively associated with severe stroke, with an and more meat before the stroke; and had bigger le- aOR of 1.67 (95% CI, 1.03– 2.72) for lesions 10– 30 cm3 sion volumes and higher systolic blood pressure com- and 3.88 (95% CI, 1.93–7 .81) for lesions >30 cm3. For pared with those with nonsevere hemorrhagic stroke ischemic strokes, other clinical subtypes were inde- (Table 2). pendently associated with severe stroke relative to la- The relationship between stroke subtype and stroke cunar stroke, while increasing age (years) (aOR, 2.56 severity is presented in Table 3 for ischemic stroke and [95% CI, 1.25– 5.24]) and lesion volume >30 cm3 (aOR, Table 4 for hemorrhagic stroke. Further characteriza- 6.16 [95% CI, 1.97– 19.25]) were independently asso- tion of the participants is available in Tables S2 through ciated with severe intracerebral hemorrhage (Table 6, S10. Figures S1). Meat consumption, low vegetable intake, and lesion Factors Associated With Severe Stroke volume were the significant risk factors independently Phenotypes associated with severity. Therefore, interactions be- tween these factors and with the key candidate vascu- Total or partial anterior circulation infarcts were more lar risk factors were examined (Tables S10– S15). There severe than lacunar stroke (Table 3), while hypertensive were significant interactions between lesion volume or nonlobar hemorrhagic strokes were more severe and hypertension, vegetable consumption and hy- than other hemorrhagic stroke phenotypes (Table 4). pertension, and lesion volume and meat consumption Reduced ejection fraction was associated with severe (Tables S11– S15). hemorrhagic stroke (Table 5). Factors Independently Associated With DISCUSSION Severe Stroke This is the largest study of stroke severity in sub- Meat consumption (adjusted odds ratio [aOR], 1.97 Saharan Africa to date, and it examines a broad range [95% CI, 1.43– 2.73]) and low vegetable consumption of candidate variables spanning the sociodemo- (aOR, 2.45 [95% CI, 1.93–3 .12]) were independently graphic, vascular risk factors, radiological, electrocar- associated with all severe strokes after adjusting for diographic, and echocardiographic categories. The age, hypertension, dyslipidemia, diabetes, obesity, key findings of this study are the high proportion of se- cigarette smoking, stress, cardiac disease, alcohol, vere stroke and the independent role of lesion volume depression, physical inactivity, salt intake, and atrial in all stroke types. Furthermore, predictors for stroke enlargement (Table  6, Figure  1). Similarly, meat con- severity were lesion volume in the hemorrhagic stroke sumption and low vegetable consumption were inde- subtype and lesion location for ischemic stroke. There pendently associated with severe ischemic stroke and was also the interesting finding of independent effect severe hemorrhagic stroke (Table  6).After including of meat consumption and low vegetable intake on se- more covariates (Table 6, Figure 2), lesion volume was vere stroke. J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 5 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans Table 2. Vascular Risk Factors for Stroke Severity Stratified by Primary Stroke Type Ischemic stroke* Hemorrhagic stroke* Nonsevere P value Nonsevere P value Characteristics Severe N=1090 N=1202 (exact test) Severe N=548 N=428 (exact test) Country, Ghana, n (%) 292 (26.8) 431 (35.9) <0.001† 209 (38.1) 268 (62.6) <0.001 Age, n (%) ≤59 y 425 (39.0) 507 (42.2) 0.117 364 (66.4) 292 (68.2) 0.451 ≥60 y 664 (60.9) 693 (57.7) 184 (33.6) 133 (31.1) Sex, male, n (%) 545 (50.0) 673 (56.0) 0.004† 355 (64.8) 263 (61.5) 0.284 Living situation, n (%) Loneliness 53 (4.9) 58 (4.8) 0.957 25 (4.6) 41 (9.6) 0.002† Living with others 1029 (94.4) 1138 (94.7) 518 (94.5) 384 (89.7) Domicile, n(%) Rural 110 (10.1) 101 (8.4) 0.372 51 (9.3) 32 (7.5) 0.570 Semiurban 313 (28.7) 351 (29.2) 149 (27.2) 116 (27.1) Urban 664 (60.9) 748 (62.2) 345 (63.0) 278 (65.0) Marital status, n (%) Never married/single 33 (30.2) 32 (26.6) 0.517 25 (45.6) 36 (84.1) 0.008 Married 25 (22.9) 36 (29.9) 438 (79.9) 310 (72.4) Monthly income >$100, n (%) 566 (51.9) 681 (56.7) 0.030† 307 (56.0) 239 (55.8) 0.860 Education (some), n (%) 819 (75.1) 995 (82.8) <0.001† 462 (84.3) 383 (89.5) 0.022† Hypertension, n (%) 1036 (95.1) 1135 (94.4) 0.618 539 (98.4) 417 (97.4) 0.312 Dyslipidemia, n (%) 921 (84.5) 1053 (87.6) 0.028† 418 (76.3) 348 (81.3) 0.065 Diabetes, n (%) 146 (13.4) 506 (42.1) 0.567 446 (81.4) 131 (30.6) 0.171 Cardiac disease, n (%) 162 (14.9) 159 (13.2) 0.262 41 (7.5) 26 (6.1) 0.395 Waist- to-h ip ratio raised, n (%) 857 (78.6) 943 (78.5) 0.107 416 (75.9) 311 (72.7) 0.066 BMI, kg/m2, mean±SD 26.6±5.5 26.8±5.2 0.380 26.6±5.3 26.5±5.1 0.740 BMI >30 kg/m2, n (%) 173 (15.9) 243 (20.2) 0.631 73 (13.3) 68 (15.9) 0.851 Physical inactivity, n (%) 58 (5.3) 45 (3.7) 0.066 27 (4.9) 13 (3.0) 0.129 Tobacco, any use, n (%) 95 (8.7) 128 (10.7) 0.109 55 (10.0) 43 (10.1) 0.987 Alcohol use categories, n (%) Never use 810 (74.3) 778 (64.7) 0.001† 341 (62.2) 221 (51.6) 0.009† Ever low use 150 (13.8) 223 (18.6) 112 (20.4) 117 (27.3) Ever high use 24 (2.2) 27 (2.3) 19 (3.47) 20 (4.7) Stress, n (%) 164 (15.1) 249 (20.7) 0.002† 103 (18.8) 92 (21.5) 0.408 Cancer, n (%) 5 (0.5) 9 (0.8) 0.272 1 (0.2) 3 (0.7) 0.147 Depression, n (%) 69 (6.3) 102 (8.5) 0.138 38 (6.9) 36 (8.4) 0.290 Family history of CVD, n (%) 372 (34.1) 490 (40.8) 0.001† 206 (37.6) 205 (47.9) 0.001† Adding salt at table, n (%) 63 (5.8) 75 (6.2) 0.681 52 (9.5) 42 (9.8) 0.940 Low vegetable consumption, 289 (26.5) 261 (21.7) 0.009† 171 (31.2) 91 (21.3) <0.001† n (%) Whole grains consumption, 859 (78.8) 181 (15.1) 0.199 442 (80.7) 69 (16.1) 0.155 n (%) Legumes consumption, n (%) 667 (61.2) 736 (61.2) 0.988 335 (61.1) 288 (67.3) 0.203 Fruit consumption, n (%) 840 (77.1) 929 (77.3) 0.705 422 (77.0) 338 (79.0) 0.835 Sugar consumption or 300 (27.5) 295 (24.5) 0.117 174 (31.8) 118 (27.6) 0.125 otherwise, n (%) Regular meat 747 (68.5) 767 (63.8) 0.025† 405 (73.9) 279 (65.2) <0.001† consumption, n (%) Fish consumption or 915 (83.9) 1006 (83.7) 0.647 476 (86.9) 380 (88.8) 0.921 otherwise, n (%) (Continued) J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 6 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans Table 2. (Continued) Ischemic stroke* Hemorrhagic stroke* Nonsevere P value Nonsevere P value Characteristics Severe N=1090 N=1202 (exact test) Severe N=548 N=428 (exact test) Lesion volume, cm3, n (%) <10 cm3 545 (50.0) 779 (64.8) <0.001† 172 (31.4) 193 (45.1) <0.001† 10– 30 cm3 153 (14.0) 145 (12.1) 174 (31.8) 152 (35.5) >30 cm3 237 (21.7) 123 (10.2) 164 (29.9) 62 (14.5) Blood pressure at presentation, mm Hg, mean±SD Systolic 157.31±31.3 150.0±27.0 <0.001† 173.3±32.1 163.57±32.3 <0.001† Diastolic 93.85±17.9 90.1±15.7 <0.001† 105.33±20.0 111.09±243.3 0.583 Fasting glucose, mg/dL, 124.24±54.5 118.1±52.9 0.084 116.75±39.8 113.1±43.9 0.386 mean±SD Neutrophil:lymphocyte ratio, 7.3±38.7 4.48±22.4 0.101 8.36±13.9 7.2±24.5 0.452 mean±SD *Subgroup (ischemic and hemorrhagic stroke) analyses were based on individuals with data on primary stroke types confirmed by brain scan. BMI indicates body mass index; and CVD, cardiovascular disease. †P value <0.05 was considered significant. There was a high proportion of severe stroke among Our findings reinforce the observed racial disparity, participants, with approximately half of all strokes with a high proportion of severe stroke among Black classified as severe, and more severe strokes among individuals compared with other racial groups.31,32 This hemorrhagic than ischemic. A similar high burden has is partly explained by high hypertension and diabetes been demonstrated in smaller populations derived burden among Black individuals31 as observed in this from a single centers in Nigeria and Ethiopia.12,28 A study, with hypertension in 85% and diabetes in 32% similar high stroke severity in hemorrhagic stroke was of the patients with stroke. Other factors may play a found in the Copenhagen Stroke Study.29 Aside from role in such disparity including the burden of comor- the high-f atality severe stroke causes, it also portends bid vascular risk, increased likelihood of motor defi- a high level of impairment among survivors, thereby cits, increased likelihood of hemorrhagic stroke type, predisposing to the high burden of care.30 This finding and poor access to the usage of preventative therapy highlights potential enormous strain on the stroke care such as carotid endarterectomy among Black individ- system in the subregion and undermines the already uals and underserved populations such as the sub- fragile health care system.30 Saharan region.33 Table 3. Ischemic Stroke Subtypes Stratified by Severity Status Ischemic stroke, n (%) Variables Subvariables Severe N=1090 Nonsevere N=1202 P value TOAST classification* Large- artery arteriosclerosis 394 (36.2) 318 (26.5) <0.001 Cardioembolism 80 (7.4) 75 (6.3) Small- vessel occlusion 233 (21.4) 492 (41) Other determined etiology 3 (0.3) 6 (0.5) Undetermined etiology 237 (21.8) 230 (19.2) OCSP subtype† Total anterior circulation infarct 179 (16.5) 117 (9.8) <0.001 Partial anterior circulation infarct 359 (33) 327 (27.3) Posterior circulation infarct 112 (10.3) 94 (7.9) Lacunar infarct 296 (27.2) 560 (46.6) ASCO subtype‡ Atherosclerosis 199 (18.3) 197 (16.4) <0.001 Small- vessel disease 299 (27.5) 538 (44.8) Cardioembolism 125 (11.4) 124 (10.4) Other causes 22 (2.1) 21 (1.8) *Trial of Org 10172 in acute stroke treatment. †Oxfordshire Community Stroke Project classification. ‡A for atherosclerosis, S for small vessel disease, C for cardiac source, O for other cause (phenotypic) classification of stroke. J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 7 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First-S troke Severity in West Africans Table 4. Hemorrhagic Stroke Subtypes Stratified by Severity Status Hemorrhagic stroke, n (%) Variables Subvariables Severe N=548 Nonsevere N=428 P value Location of lesion Lobar 107 (9.6) 117 (27.4) 0.004 Nonlobar 441 (80.5) 311 (72.7) SMASH- U Structural 16 (3.0) 20 (4.7) 0.027 Medication related 0 (0) 4 (1) Amyloid angiopathy 2 (0.4) 7 (1.7) Systemic disease 1 (0.2) 3 (0.8) Hypertensive 436 (79.6) 337 (78.8) Undetermined 12 (2.2) 8 (1.9) While numerous vascular risk factors are implicated in vivo homocysteine level and stroke severity.18,34– 36 in stroke etiology, many were not found to be signifi- Other important factors in vegetables, especially cant in all stroke or stroke subtype severe outcomes. green leafy ones, are micronutrients and vitamins. This is particularly interesting for hypertension and The nitrate–n itrite– nitric oxide pathway has also been dyslipidemia, which played a vital role in stroke occur- suggested.18,34–3 6 The pathophysiologic mechanism rence but not severity. Our previous report demon- associated with excessive meat consumption include strated that hypertension had an aOR of 19.36 (95% the high heme iron in red meat.37 Iron is a redox- active CI, 12.11– 30.93) for stroke occurrence.18 Although, metal and catalyzes the formation of hydroxyl free radi- dyslipidemia was found in 4 of 5 participants, it did not cals in the Fenton reaction.37 Iron may lead to oxidative play a significant independent role for severe stroke. stress, a state with increased peroxidation of lipids, The overarching implication of this observation is that protein modification, and DNA damage.37 severity predictors are not necessarily etiological fac- Depending on the classification, there was a sig- tors. Nevertheless, we observed the interactive effect nificant association of ischemic stroke with large- of hypertension on vegetable and meat consumption artery atherosclerosis in the Trial of Org 10172 in Acute and lesion volume. It would be interesting to unravel Stroke Treatment classification and partial anterior these interactions further in future studies. circulation infarcts in Oxfordshire Community Stroke The previous SIREN report found vegetable intake Project subtype.38,39 A previous study suggested that as protective against stroke.18 It is, however, inter- total anterior circulation infarction (TACI) is associ- esting that it also plays a key role in all stroke types ated with worse stroke outcome.40 This is similar to and the subtypes in addition to meat consumption our study, which demonstrated a 3- fold increased risk with low vegetable consumption independently asso- of severe outcome with TACI.41 On the other hand, ciated with severe stroke. While the evidence of this nonlobar and hypertensive hemorrhagic stroke were role has been established, the mechanism is poorly more severe, none was independently associated understood.18,34,35 The likely link for the vegetable con- with severe hemorrhagic stroke. Nevertheless, lesion sumption may be the dose– response relationship of volume ≥30 cm3 was independently associated with Table 5. Echocardiographic and Electrocardiographic Determinants of Severe Stroke Patients by Stroke Levity Scale Severe Nonsevere ischaemic Nonsevere ischemic stroke stroke; f (%) P value hemorrhagic Severe hemorrhagic P value Variable N=1202 N=1090 (exact test) stroke N=428 stroke; f (%) N=548 (exact test) Sinus arrhythmia 26 (2.2) 23 (2.2) 0.382 11 (2.6) 14 (2.6) 0.550 Atrial fibrillation 38 (3.2) 44 (4.1) 0.495 2 (0.5) 8 (1.5) 0.200 Atrial flutter, ECG 3 (0.2) 7 (0.7) 0.223 0 (0.0) 1 (0.2) 1.000 determined Atrial enlargement, 197 (16.4) 170 (15.6) 0.026 92 (21.5) 92 (16.8) 0.099 right/left Ejection fraction ≤40% 58 (4.8) 53 (4.9) 0.078 5 (1.2) 15 (2.8) 0.037 41%– 50% 61 (5.1) 55 (5.1) 14 (3.3) 8 (1.5) ≥51% 394 (32.8) 252 (23.2) 93 (21.7) 93 (17.0) J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 8 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First-S troke Severity in West Africans Table 6. Factors Associated With Stroke Severity by Stroke Levity Scale Model 1 All severe stroke Severe ischemic stroke Severe hemorrhagic stroke Risk factor aOR (95% CI) aOR (95% CI) aOR (95% CI) Age, y 1.21 (0.93– 1.58) 1.09 (0.75– 1.56) 1.81 (1.13–2 .89)† Hypertension 1.00 (0.56–1 .79) 1.03 (0.52– 2.02) 1.47 (0.29– 7.28) Dyslipidemia 0.96 (0.70– 1.31) 0.90 (0.58– 1.37) 1.11 (0.63–1 .95) Diabetes 0.78 (0.62–1 .00) 0.89 (0.67– 1.19) 0.59 (0.36–0 .98) Obesity 0.93 (0.71– 1.22) 1.07 (0.77– 1.50) 0.62 (0.35–1 .10) Cigarette smoking 1.01 (0.52– 1.93) 1.11 (0.43– 2.87) 0.89 (0.29– 2.72) Stress 0.99 (0.75– 1.32) 0.84 (0.59– 1.20) 1.34 (0.76– 2.36) Cardiac disease 0.93 (0.66–1 .30) 0.83 (0.56– 1.22) 1.37 (0.57–3 .27) Alcohol 0.88 (0.67– 1.16) 0.72 (0.50– 1.03) 0.95 (0.57–1 .60) Meat consumption 1.97 (1.43– 2.73)† 1.50 (1.03– 2.20)† 6.16 (2.88– 13.18)† Low vegetable consumption 2.45 (1.93– 3.12)† 2.23 (1.66– 3.00)† 4.34 (2.60– 7.23)† Depression 0.98 (0.64– 1.48) 1.17 (0.70– 1.96) 0.60 (0.26– 1.36) Physical inactivity 1.11 (0.59– 2.08) 0.95 (0.44– 2.04) 2.91 (0.78– 10.88) Salt intake 1.05 (0.71–1 .56) 1.01 (0.61– 1.69) 0.79 (0.39– 1.61) Right/left atrial enlargement 1.00 (0.78–1 .27) 0.96 (0.70–1 .31) 0.97 (0.60–1 .58) Model 2 Age, y 1.18 (0.73– 1.91) 0.92 (0.52– 1.64) 2.56 (1.25– 5.24)† Dyslipidemia 0.77 (0.38– 1.55) 1.22 (0.52– 2.84) 0.63 (0.24– 1.65) Obesity 0.93 (0.56– 1.55) 0.82 (0.48–1 .39) 0.92 (0.39– 2.17) Low vegetable consumption 1.05 (0.65– 1.69) 1.22 (0.72– 2.07) 1.14 (0.45– 2.90) Physical inactivity 0.70 (0.27–1 .78) 1.04 (0.39– 2.79) 0.18 (0.02– 1.72) Right/left atrial enlargement 0.89 (0.55– 1.45) 0.92 (0.56– 1.51) 0.69 (0.33–1 .45) Lesion volume, cm3 ≥10 1.67 (1.03– 2.72)† 1.01 (0.53– 1.95) 1.86 (0.92– 3.76) ≥30 3.88 (1.93–7 .81)† 1.80 (0.83–3 .89) 6.16 (1.97– 19.25)† Systolic blood pressure at 1.00 (1.00– 1.01) 1.00 (0.99– 1.01) 1.00 (0.99– 1.01) presentation Fasting glucose 1.00 (0.99–1 .01) 1.00 (1.00– 1.01) 0.99 (0.98–1 .01) Neutrophil:lymphocyte ratio 1.00 (0.99– 1.01) NA NA Stroke type, ICH vs ischemic 0.80 (0.50– 1.27) NA NA Nonlobar vs lobar NA NA 1.01 (0.46– 2.21) Causes of ICH, hypertensive vs NA NA 1.38 (0.41–1 3.87) others Ischemic stroke subtypes NA NA OCSP subtypes NA NA LACI (ref) NA NA TACI NA 3.15 (1.45–6 .82)† NA PACI NA 1.97 (1.19– 3.26)† NA POCI NA 2.16 (1.10– 4.21)† NA aOR indicates adjusted odds ratio; ICH, intracerebral hemorrhage; LACI, lacunar infarct; NA, not applicable; OCSP, Oxfordshire Community Stroke Project; PACI, partial anterior circulation infarct; POCI, posterior circulation infarct; and TACI, total anterior circulation infarct. †P value <0.05 was considered significant. severe stroke among all stroke types and patients with meat consumption, and hypertension and low vege- hemorrhagic stroke, while ≥10 cm3 lesion volume was table consumption. associated with severity of all strokes combined.42 Conversely, for ischemic stroke the location of the Furthermore, there was strong interaction between lesion was more relevant,43 with TACI, partial anterior lesion volume and hypertension, lesion volume and circulation infarcts, and posterior circulation infarcts J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 9 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans Figure 1. Forest plot of the factors associated with stroke severity by Stroke Levity Scale (SLS) (model 1). being more severe. Although the severity of TACI Our study has several strengths. First, the charac- could be attributable to its larger size,40 TACI can re- terization of the burden of severe stroke in the largest sult in fatal complications like transtentorial herniation, study of stroke in Africa will facilitate planning of the whereas posterior circulation infarcts can interfere with treatment protocol in this population, particularly in vital brain stem structures.40,43 the early phase of care, which is associated with high Figure 2. Forest plot of the factors associated with stroke severity by Stroke Levity Scale (SLS) (model 2). BP indicates blood pressure; and ICH, intracerebral hemorrhage. J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 10 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans mortality.12,41,44–4 6 Moreover, the identified novel modi- (R.O., O.B., O.U.A.-I.); Department of Medicine, Obafemi Awolowo University fiable risk factors for severe stroke can be targeted for Teaching Hospital, Ile- Ife, Nigeria (M.K.); Department of Medicine, Aminu Kano Teaching Hospital, Kano, Nigeria (L. Owolabi); Jos University Teaching prevention of severe stroke. Hospital Jos, Jos, Nigeria (G.O.O.); Department of Medicine, Federal Medical Our study has some limitations. The burden of se- Centre, Owo, Ondo State, Nigeria (T.S.); University of Alabama at Birmingham, vere stroke may be exaggerated because of missing Birmingham, AL, USA (H.K.T.); Medical University of South Carolina, Mount Pleasant, SC, USA (C.J.); College of Medicine (O.Arulogun) and Department cases as patients with mild strokes may not visit hospi- of Radiology (G.O.), University of Ibadan, Ibadan, Nigeria; College of Public tals because of poor resources. Nevertheless, we had Health, University of Kentucky, Lexington, KY, USA (D.A.); Ladoke Akintola an active community engagement strategy to facili- University of Technology (LAUTECH) and LAUTECH Teaching Hospital, Ogbomoso, Oyo State, Nigeria (P.A.); Aga-Khan University, Dar es Salaam, tate presentation of stroke cases from the catchment Tanzania (P.A.); Department of Pharmacology and Therapeutics, Olabisi population for enrollment into the study and mitigate Onabanjo University, Abeokuta, Nigeria (L. Ogunjimi); Neuroscience and presentation bias. There was also disparity in severe Ageing Research Unit, Institute for Advanced Medical Research and Training, College of Medicine (R.A.), and Center for Genomic and Precision stroke burden in the 2 countries in this study. This may Medicine, College of Medicine (R.A., M.O.O.), University of Ibadan, Ibadan, be attributable to diversity in the variable mix of urban/ Nigeria; Weill Institute for Neurosciences, School of Medicine, University of suburban domicile, sex distribution, vascular risk factor California San FranciscoCA, USA (B.O.); and Lebanese American University, Beirut, Lebanon (M.O.O.). burden, and other factors. Sources of Funding The study and investigators are supported by the National Institutes of CONCLUSIONS Health grants Stroke Investigative Research and Educational Network (SIREN) (U54HG007479), Systematic Investigation of Blacks with Stroke- In this study, we described an enormous burden of se- Genomics (SIBS Genomics) (R01NS107900), African Neurobiobank for Precision Stroke Medicine (U01HG010273), Facilitating Implementation vere stroke among all stroke types and subtypes, with Science within SIBS Genomics Study (SIBS-G en-G en) (R01NS107900- significant implications for the stroke care system in 02S1), ARISES (R01NS115944- 01), H3Africa Cardiovacular Diseases (CVD) West Africa. We also discovered dietary and radiologi- Supplement (3U24HG009780- 03S5), CaNVAS (1R01NS114045-0 1), sub- Saharan Africa Conference on Stroke Conference 1R13NS115395-0 1A1, cal factors independently associated with stroke se- and Training Africans to Lead and Execute Neurological Trials & Studies verity. Reduced meat consumption and high vegetable (TALENTS) D43TW012030. The funding bodies played no role in the design consumption could reduce the likelihood of developing of the study or the collection, analysis, or interpretation of data or in writing the manuscript. severe stroke in this population. Disclosures None. APPENDIX Supplemental Material SIREN Investigators Data S1Tables S1– S15 Oladimeji Adebayo, Onoja Akpa, Osahon J. Asowata, Figures S1– S4 Adekunle Fakunle, Fred S. Sarfo, Albert Akpalu, References [47– 58] Kolawole Wahab, Reginald Obiako, Morenikeji Komolafe, Lukman Owolabi, Godwin O. Osaigbovo, Akinkunmi Paul Okekunle, Taofiki Sunmonu, Hemant REFERENCES K. Tiwari, Carolyn Jenkins, Oyedunni Arulogun, 1. Owolabi M. Taming the burgeoning stroke epidemic in Africa: stroke Lambert Appiah, Joshua Akinyemi, Abiodun M. quadrangle to the rescue. West Indian Med J. 2011;60:412– 421. Adeoye, Godwin Ogbole, Joseph Yaria, Donna Arnett, 2. Krishnamurthi RV, Feigin VL, Forouzanfar MH, Mensah GA, Connor M, Bennett DA, Moran AE, Sacco RL, Anderson LM, Truelsen T. Philip Adebayo, Benedict Calys- Tagoe, Okechukwu Global and regional burden of first-e ver ischaemic and haemorrhagic S. Ogah, Olayemi Balogun, Luqman Ogunjimi, Yaw stroke during 1990– 2010: findings from the Global Burden of Disease Mensah, Obiageli U. Agbogu-I ke, Rufus Akinyemi, Study 2010. Lancet Global Health. 2013;1:e259– e281. doi: 10.1016/S2214- 109X(13)70089- 5 Bruce Ovbiagele, Mayowa Owolabi 3. Gorelick PB. Cerebrovascular disease in African Americans. Stroke. 1998;29:2656– 2664. doi: 10.1161/01.STR.29.12.2656 4. Wolfe CD. The impact of stroke. Br Med Bull. 2000;56:275– 286. doi: ARTICLE INFORMATION 10.1258/0007142001903120 5. Bwala S. Stroke in a subsaharan Nigerian hospital— a retrospective Received August 22, 2022; accepted January 12, 2023. study. Trop Doct. 1989;19:11– 14. doi: 10.1177/004947558901900104 6. Cox AM, McKevitt C, Rudd AG, Wolfe CD. Socioeconomic sta- Affiliations tus and stroke. Lancet Neurol. 2006;5:181– 188. doi: 10.1016/ Department of Medicine, University College Hospital, Ibadan, Nigeria S1474- 4422(06)70351-9 (O. Adebayo, A.M.A., J.Y., O.S.O., M.O.O.); Department of Epidemiology 7. Schlegel D, Kolb SJ, Luciano JM, Tovar JM, Cucchiara BL, and Medical Statistics (O. Akpa, O.J.A., A.P.O., J.A.), and Institute of Liebeskind DS, Kasner SE. Utility of the NIH Stroke Scale as a pre- Cardiovascular Diseases (O. Akpa), University of Ibadan, Ibadan, Nigeria; dictor of hospital disposition. Stroke. 2003;34:134– 137. doi: 10.1161/01. Department of Public Health, Osun State University, Osogbo, Nigeria STR.0000048217.44714.02 (A.F.); Department of Medicine, Kwame Nkrumah University of Science 8. Bonita R, Truelsen T. Stroke in sub- Saharan Africa: a neglected chronic dis- and Technology, Kumasi, Ghana (F.S.S., L.A., Y.M.); Department of ease. Lancet Neurol. 2003;2:592. doi: 10.1016/S1474- 4422(03)00524- 6 Medicine, University of Ghana Medical School, Accra, Ghana (A.A., B.C.-T.); 9. Connor MD, Walker R, Modi G, Warlow CP. Burden of stroke in black Department of Medicine, University of Ilorin Teaching Hospital, Ilorin, Nigeria populations in sub-S aharan Africa. Lancet Neurol. 2007;6:269–2 78. (K.W.); Department of Medicine, Ahmadu Bello University, Zaria, Nigeria doi: 10.1016/S1474-4 422(07)70002-9 J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 11 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First-S troke Severity in West Africans 10. Garbusinski JM, van der Sande MA, Bartholome EJ, Dramaix M, Gaye Stroke Treatment (TOAST). Neurology. 1999;53:126.  doi:  10.1212/ A, Coleman R, Nyan OA, Walker RW, McAdam KP, Walraven GE. wnl.53.1.126 Stroke presentation and outcome in developing countries a prospec- 31. Kuhlemeier K, Stiens S. Racial disparities in severity of cerebrovascular tive study in the Gambia. Stroke. 2005;36:1388– 1393. doi: 10.1161/01. events. Stroke. 1994;25:2126– 2131. doi: 10.1161/01.STR.25.11.2126 STR.0000170717.91591.7d 32. Stansbury JP, Jia H, Williams LS, Vogel WB, Duncan PW. Ethnic dis- 11. Ween JE, Alexander MP, D’Esposito M, Roberts M. Factors predictive of parities in stroke: epidemiology, acute care, and postacute outcomes. stroke outcome in a rehabilitation setting. Neurology. 1996;47:388–3 92. Stroke. 2005;36:374– 386. doi: 10.1161/01.STR.0000153065.39325.fd doi: 10.1212/WNL.47.2.388 33. Jones MR, Horner RD, Edwards LJ, Hoff J, Armstrong SB, Smith- 12. Ekeh B, Ogunniyi A, Isamade E, Ekrikpo U. Stroke mortality and its pre- Hammond CA, Matchar DB, Oddone EZ. Racial variation in initial stroke dictors in a Nigerian teaching hospital. African Health Sci. 2015;15:74– severity. Stroke. 2000;31:563– 567. doi: 10.1161/01.STR.31.3.563 80. doi: 10.4314/ahs.v15i1.10 3 4. Larsson SC, Virtamo J, Wolk A. Total and specific fruit and vegetable 13. Broderick JP, Phillips SJ, O’Fallon WM, Frye RL, Whisnant JP. consumption and risk of stroke: a prospective study. Atherosclerosis. Relationship of cardiac disease to stroke occurrence, recurrence, and 2013;227:147– 152. doi: 10.1016/j.atherosclerosis.2012.12.022 mortality. Stroke. 1992;23:1250– 1256. doi: 10.1161/01.STR.23.9.1250 3 5. Gariballa SE. Nutritional factors in stroke. Nutrition. 2000;84:5– 17. doi: 14. Kimura K, Minematsu K, Yamaguchi T. Atrial fibrillation as a predictive 10.1017/S0007114500001173 factor for severe stroke and early death in 15 831 patients with acute 3 6. Wu X- Q, Ding J, Ge A- Y, Liu F- F, Wang X, Fan W. Acute phase homocys- ischaemic stroke. J Neurol Neurosurg Psychiatry. 2005;76:679– 683. teine related to severity and outcome of atherothrombotic stroke. Eur J doi: 10.1136/jnnp.2004.048827 Intern Med. 2013;24:362– 367. doi: 10.1016/j.ejim.2013.01.015 15. Akpalu A, Sarfo FS, Ovbiagele B, Akinyemi R, Gebregziabher M, Obiako 37. Kaluza J, Wolk A, Larsson SC. Red meat consumption and risk of R, Owolabi L, Sagoe K, Jenkins C, Arulogun O, et al. Phenotyping stroke: a meta- analysis of prospective studies. Stroke. 2012;43:2556– stroke in sub-s aharan Africa: Stroke Investigative Research and 2560. doi: 10.1161/STROKEAHA.112.663286 Education Network (SIREN) phenomics protocol. Neuroepidemiology. 38. Tei H, Uchiyama S, Ohara K, Kobayashi M, Uchiyama Y, Fukuzawa M. 2015;45:73– 82. doi: 10.1159/000437372 Deteriorating ischemic stroke in 4 clinical categories classified by the 16. ECOWAS. Economic community of West African states. 2016;2016. Oxfordshire Community Stroke Project. Stroke. 2000;31:2049–2 054. Accessed October 1, 2020. https://www.ecowas.int/member-states / doi: 10.1161/01.STR.31.9.2049 17. ECOWAS. Economic community of West African states 3 9. Paci M, Nannetti L, D’ippolito P, Lombardi B. Outcomes from isch- (ECOWAS):Nigeria. 2020;2020. Accessed October 1, 2020. https:// emic stroke subtypes classified by the Oxfordshire Community Stroke www.ecowas.int/member-state s/ Project: a systematic review. Eur J Phys Rehabil Med. 2011;47:19– 23. 18. Owolabi MO, Sarfo F, Akinyemi R, Gebregziabher M, Akpa O, Akpalu 4 0. Bamford J, Sandercock P, Dennis M, Warlow C, Burn J. Classification A, Wahab K, Obiako R, Owolabi L, Ovbiagele B. Dominant modi- and natural history of clinically identifiable subtypes of cerebral infarc- fiable risk factors for stroke in Ghana and Nigeria (SIREN): a case- tion. Lancet. 1991;337:1521– 1526. doi: 10.1016/0140- 6736(91)93206-O control study. Lancet Global Health. 2018;6:e436– e446. doi: 10.1016/ 41. Donkor ES. Stroke in the century: a snapshot of the burden, epide- S2214- 109X(18)30002- 0 miology, and quality of life. Stroke Res Treat. 2018;2018:1–1 0. doi: 19. Sarfo FS, Ovbiagele B, Gebregziabher M, Wahab K, Akinyemi R, Akpalu 10.1155/2018/3238165 A, Akpa O, Obiako R, Owolabi L, Jenkins C. Stroke among young West 42. Chen C-L , Tang F- T, Chen H-C , Chung C- Y, Wong M- K. Brain lesion Africans: evidence from the SIREN (Stroke Investigative Research size and location: effects on motor recovery and functional outcome in and Educational Network) large multisite case– control study. Stroke. stroke patients. Arch Phys Med Rehabil. 2000;81:447– 452. doi: 10.1053/ 2018;49:1116–1 122. doi: 10.1161/STROKEAHA.118.020783 mr.2000.3837 2 0. Fischer U, Cooney MT, Bull LM, Silver LE, Chalmers J, Anderson CS, 4 3. Payabvash S, Taleb S, Benson J, McKinney A. Acute ischemic stroke Mehta Z, Rothwell PM. Acute post- stroke blood pressure relative to infarct topology: association with lesion volume and severity of symp- premorbid levels in intracerebral haemorrhage versus major ischaemic toms at admission and discharge. Am J Neuroradiol. 2017;38:58– 63. stroke: a population- based study. Lancet Neurol. 2014;13:374– 384. doi: doi: 10.3174/ajnr.A4970 10.1016/S1474- 4422(14)70031-6 4 4. Njoku C, Aduloju A. Stroke in Sokoto, Nigeria: a five year retrospective 21. Terminology  and Diagnostic Criteria Committee, Japan  Society  of study. Eur J Phys Rehabil Med. 2004;3:73– 76. Ultrasonics in Medicine. Standard method for ultrasound evaluation of 45. Chang K-C , Lee H-C , Tseng M-C , Huang Y- C. Three-y ear survival carotid artery lesions. J Med Ultrason. 2009;36:219–2 26. doi: 10.1007/ after first- ever ischemic stroke is predicted by initial stroke severity: a s10396- 009- 0238- y hospital- based study. Clin Neurol Neurosurg. 2010;112:296– 301. doi: 22. Govan L, Langhorne P, Weir CJ. Categorizing stroke prognosis using 10.1016/j.clineuro.2009.12.016 different stroke scales. Stroke. 2009;40:3396– 3399. doi: 10.1161/ 46. Chang K-C , Tan T- Y, Liou C-W , Tseng M- C. Predicting 3- month mortality STROKEAHA.109.557645 among patients hospitalized for first-e ver acute ischemic stroke. J Formos 23. Owolabi M, Platz T. Proposing the Stroke Levity Scale: a valid, reli- Med Assoc. 2006;105:310– 317. doi: 10.1016/S0929- 6646(09)60122-4 able, simple, and time- saving measure of stroke severity. Eur J Neurol. 47. Adeoye AM, Ogah OS, Ovbiagele B, Akinyemi R, Shidali V, Agyekum F, Aje A, 2008;15:627– 633. doi: 10.1111/j.1468- 1331.2008.02140.x Adebayo O, Akinyemi JO, Kolo P, et al. Prevalence and prognostic features 24. De Haan R, Horn J, Limburg M, Van Der Meulen J, Bossuyt P. A com- of ECG abnormalities in acute stroke: findings from the SIREN study among parison of five stroke scales with measures of disability, handicap, and Africans. Glob Heart. 2017;12:99–1 05. doi: 10.1016/j.gheart.2017.01.002 quality of life. Stroke. 1993;24:1178–1 181. doi: 10.1161/01.STR.24.8.1178 48. Adeoye AM, Ovbiagele B, Kolo P, Appiah L, Aje A, Adebayo O, Sarfo F, 25. Kasner SE. Clinical interpretation and use of stroke scales. Lancet Akinyemi J, Adekunle G, Agyekum F. Exploring overlaps between the Neurol. 2006;5:603– 612. doi: 10.1016/S1474-4 422(06)70495-1 genomic and environmental determinants of LVH and stroke: a multi- 26. Hilbrich L, Truelsen T, Yusuf S. Stroke and cardiovascular diseases: the center study in West Africa. Glob Heart. 2017;12:107– 113. need for a global approach for prevention and drug development. Int J 4 9. Kolominsky- Rabas PL, Weber M, Gefeller O, Neundoerfer B, Stroke. 2007;2:104–1 08. doi: 10.1111/j.1747-4 949.2007.00118.x Heuschmann PU. Epidemiology of ischemic stroke subtypes ac- 27. Meyer BC, Hemmen TM, Jackson CM, Lyden PD. Modified National cording to toast criteria: incidence, recurrence, and long- term sur- Institutes of Health Stroke Scale for use in stroke clinical trials prospec- vival in ischemic stroke subtypes: a population-b ased study. Stroke. tive reliability and validity. Stroke. 2002;33:1261–1 266. doi: 10.1161/01. 2001;32:2735– 2740. doi: 10.1161/hs1201.100209 STR.0000015625.87603.A7 50. Meretoja A, Strbian D, Putaala J, Curtze S, Haapaniemi E, Mustanoja 28. Deresse B, Shaweno D. Epidemiology and in- hospital outcome of S, Sairanen T, Satopää J, Silvennoinen H, Niemelä M. SMASH-U : a stroke in South Ethiopia. J Neurol Sci. 2015;355:138– 142. doi: 10.1016/j. proposal for etiologic classification of intracerebral hemorrhage. Stroke. jns.2015.06.001 2012;43:2592–2 597. doi: 10.1161/STROKEAHA.112.661603 29. Reith J, Jørgensen HS, Nakayama H, Raaschou HO, Olsen TS. 51. Expert Panel on Detection, Evaluation,  and  Treatment  of  High  Blo Seizures in acute stroke: predictors and prognostic significance. Stroke. od  Cholesterol  in  Adults. Executive summary of the third report of the 1997;28:1585– 1589. doi: 10.1161/01.STR.28.8.1585 National Cholesterol Education Program (NCEP) Expert Panel on 30. Adams HP, Davis PH, Leira EC, Chang K- C, Bendixen BH, Clarke WR, Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults Woolson RF, Hansen MD. Baseline NIH Stroke Scale score strongly (Adult Treatment Panel III). JAMA. 2001;285:2486– 2497. doi: 10.1001/ predicts outcome after stroke: a report of the Trial of Org 10172 in Acute jama.285.19.2486 J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 12 Downloaded from http://ahajournals.org by on August 8, 2023 Adebayo et al First- Stroke Severity in West Africans 5 2. Yusuf S, Hawken S, Ôunpuu S, Dans T, Avezum A, Lanas F, McQueen 55. Herold D, Boyd J, Bruns D, Emerson J, Burns K, Bray R, Vandenhoff M, Budaj A, Pais P, Varigos J. Effect of potentially modifiable risk factors G, Freedlender A, Fortier G, Pohl S. Measurement of glycosylated he- associated with myocardial infarction in 52 countries (the INTERHEART moglobins using boronate affinity chromatography. Ann Clin Lab Sci. study): case-c ontrol study. Lancet. 2004;364:937–9 52. doi: 10.1016/ 1983;13:482–4 88. S0140- 6736(04)17018-9 56. Johnson R, McNutt P, MacMahon S, Robson R. Use of the Friedewald 53. O’Donnell MJ, Chin SL, Rangarajan S, Xavier D, Liu L, Zhang H, Rao- formula to estimate LDL- cholesterol in patients with chronic renal Melacini P, Zhang X, Pais P, Agapay S. Global and regional effects of failure on dialysis. Clin Chem. 1997;43:2183– 2184. doi: 10.1093/ potentially modifiable risk factors associated with acute stroke in 32 clinchem/43.11.2183 countries (INTERHEART): a case- control study. Lancet. 2016;388:761– 57. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymatic deter- 775. doi: 10.1016/S0140- 6736(16)30506-2 mination of total serum cholesterol. Clin Chem. 1974;20:470– 475. doi: 5 4. Nathan DM, Singer DE, Hurxthal K, Goodson JD. The clinical infor- 10.1093/clinchem/20.4.470 mation value of the glycosylated hemoglobin assay. N Engl J Med. 58. Albers JJ, Warnick GR, Chenng MC. Quantitation of high density lipo- 1984;310:341–3 46. doi: 10.1056/NEJM198402093100602 proteins. Lipids. 1978;13:926– 932. doi: 10.1007/BF02533852 J Am Heart Assoc. 2023;12:e027888. DOI: 10.1161/JAHA.122.027888 13 Downloaded from http://ahajournals.org by on August 8, 2023 SUPPLEMENTAL MATERIAL Downloaded from http://ahajournals.org by on August 8, 2023 Data S1. Supplemental Methods Eligibility Criteria The patients considered for the study were specifically those who presented to the SIREN project hospital sites in Nigeria and Ghana. Only one admission episode was recorded for each individual. To be included in this study, participants had to be at least 18 years of age with clinical features of acute stroke confirmed by Cranial Computerized Tomography (CT)/Magnetic Resonance Imaging (MRI). For this study, stroke was defined clinically based on American Heart Association (AHA)/American Stroke Association (ASA) Expert Consensus Document.26 Eligible cases were patients with the first stroke admitted within 8 days of current symptoms onset or "last seen without deficit" and CT or MRI scan within ten days of symptom onset. Evaluation of stroke We excluded patients with extra-axial haemorrhage, tumour or brain abscess, primary subarachnoid haemorrhage, current hospitalization for coronary heart disease, or unable to communicate or provide consent with no valid surrogate available. Age, Sex, marital status, type of domicile, level of education, living situation, ethnic group, family history of stroke, smoking, and alcohol consumption were among the socio-demographic variables assessed. Hypertension, diabetes, dyslipidaemia, obesity, transient ischemic attack (TIA), HIV, chronic kidney disease, heart disease, sickle cell anaemia, and atrial fibrillation were among the vascular risk factors evaluated. Other variables assessed included: history of neck injury, recurrent miscarriage, bronchitis, cancer, a dental problem in past one year, use of oral contraceptives, sleep disorders, febrile illness in last four weeks, use of anticoagulants, migraine, obstructive sleep apnoea and neck manipulation. The methods for assessment of the various variables have been described in the previously published protocols and articles from SIREN.15,47,48 Brain imaging variables included: the location of the lesion (cortical, subcortical white matter, subcortical, cerebellar, brainstem, ventricles, circulation type), size and volume of lesion, age of lesion (hyper-acute, acute, sub-acute, chronic), TOAST (Trial of ORG 10172 in acute stroke treatment) classification (large artery atherosclerosis, cardio-embolism, the stroke of other Downloaded from http://ahajournals.org by on August 8, 2023 determined aetiology, small vessel occlusion, the stroke of undetermined origin), and presence or absence of incidental findings.49 Stroke Phenotyping Stroke diagnosis and phenotyping were based on clinical evaluation and brain neuroimaging (computed tomography or magnetic resonance imaging), ECG, transthoracic echocardiography, and carotid Doppler ultrasound performed according to standardized protocols at each site as have been previous been described.15 Presumed etiologic subtypes of ischemic stroke were defined using the TOAST (Trial of ORG 10172 in Acute Stroke Treatment)49 and intracerebral hemorrhage was classified etiologically into SMASH-U causes (Structural, Medication-Related, Amyloid Angiopathy, Systemic/Other Disease, Hypertension and Undetermined).19,20,50 The lesion volume was measured using the ellipsoid formula (A x B x C/ 2) after the axial slice with the largest lesion is selected by visual inspection.21 On the selected slice, A is the longest dimension of the infarct lesion, B is perpendicular to A at the widest dimension while C is the slice thickness multiplied by the number of axial slices on which the infarct lesion is seen.21 Definition of Stroke Severity Both SLS and NIHSS tool was used to defined stroke severity in SIREN study. Severe stroke was defined as Stroke Levity Scale score of ≤ 5 or NIHSS score of > 20. Risk Factor Definitions and Measurements Definition of Risk Factors Basic demographic and lifestyle data including, socioeconomic status, cardiovascular risk profile, dietary patterns, routine physical activity, stress using a validated INTERSTROKE instrument, depression, cigarette smoking, and alcohol use.15,51,52 See Table S1 for more details. Measurement of risk factors Hypertension We measured blood pressure using a standard sphygmomanometer (Omron or Accoson England mercury sphygmomanometer). Systolic blood pressure was determined by Korotkoff phase 1 Downloaded from http://ahajournals.org by on August 8, 2023 while diastolic pressure was recorded at Korotkoff phase V. Subject was resting for ≥5 minutes, and had not smoked for at least 30 minutes before the measurement. We ensured an adequate cuff size with bladder encircling and covering 2/3 of length of arm with the bladder over the brachial artery and the lower border should be 1 inch (2-3cm) above the antecubital space. The bladder was deflated slowly and exact values to the nearest 2mmHg were recorded. Blood pressure (average of three measurements used) was recorded at time of admission (from patient’s medical notes), the morning after admission (from patient’s medical notes) and daily for 7 days or until death. At time of interview blood pressure was again measured by research personnel using an automated blood pressure monitor. Adjustments to systolic BP based on reported associations between pre-morbid BP and acute post-stroke pressure in the Oxford Vascular Study (OXVASC) were applied. Typically stroke subjects present for care late after about 72 hours of stroke onset during which time the acute rise in blood pressure in response to stroke may have started to subside. Weight The scales were standardized to 0 before each use. Weight was measured in undergarments using a platform scale to the nearest 0.2kg. We recorded the participant’s weight twice in kilogram (kg). Height: We recorded the participant’s height in meters (cm). If the participant was able to stand, standing height was measured with the subject bare footed, back square against the bed and eyes looking straight ahead. Supine height was measured with the subject in bare feet, lying on their back square against the bed and eyes looking straight upward. Height was measured to the nearest 0.5cm. Body Mass Index (BMI) This was calculated by dividing weight (in kg) by square of the height (in meters). Waist and hip circumferences were measured in the standing and supine positions. Where cases were unable to stand due to disability, these measurements were conducted in the supine position only. Standing waist and hip measurements were used in the present analysis where available. Waist circumference This was measured to the nearest 0.1cm using a non-stretchable standard tape measure attached to a spring balance exerting a force of 750gm over the unclothed abdomen at the midway between Downloaded from http://ahajournals.org by on August 8, 2023 the costal margin and the iliac crest. The tape measure was kept horizontal for standing measurement and vertical for supine measurement with the subject relaxed with arms held loosely at sides.51 Hip circumference This was measured to the nearest 0.1cm using a non-stretchable standard tape measure attached to a spring balance exerting a force of 750gm. Measurements were taken over light clothing at the level of the greater trochanters (usually the widest diameter around the buttocks). The tape measure was kept horizontal for standing measurement and vertical for supine measurements.51,52 Psychosocial factors assessment We used a combined measure of psychosocial stress employed in INTERHEART52 and INTERSTROKE53, which combined self-report of stress at home or and work, life events and depression. Psychosocial stress at home/work was defined as the experience, in the two weeks prior to the stroke, of irritability, anxiety, or sleep difficulties as a result of conditions at work or home. For life events, respondents were asked to give a ‘yes’ or ‘no’ response to questions about whether, in the two weeks before the stroke, they experienced a stressful life event such as the death of a spouse, death/major illness of a close family member, marital separation/divorce, major personal injury or illness, loss of crop, loss of job/retirement, business failure, major intra-family conflict, violence (including kidnapping, assault, theft, etc.), financial stress, home-related stress, work-related stress, or other major stress. Depression For the assessment of depression, respondents were first screened for the presence of depressed mood in the four weeks before the stroke. Those who answered in the affirmative were next asked if, for at least two weeks during the four- week period before the stroke, they also experienced at least four out of seven other depression symptoms: loss of interest, feeling tired or low on energy, significant changes in weight, trouble falling asleep as usual, difficulty concentrating, thoughts of death, or feelings of worthlessness. Downloaded from http://ahajournals.org by on August 8, 2023 Dietary History We used a food frequency questionnaire to collect data on the frequency of food consumption in the past 12 months preceding a stroke. We evaluated whether or not subjects consumed cooking oil, vegetable intake, sprinkling salt at table, meat consumption, fruits, whole grains, refined grains, dairy products, poultry, eggs, fish and seafood, legumes, prickled food, deep fried foods, salty snacks, confectionary and carbonated beverages. For each of the food items, subjects had to record the number of times it was consumed per month or per week or per day. Regular consumption of a food item was defined as intake of at least once a day, a week, or a month whilst consumption rates less than once a month or never was defined as ‘not regular’. Determination of blood glucose level, HBA1c and lipid profile Blood samples were collected from each case within 10 days of symptom onset, and from each control upon enrolment after an overnight fast and into relevant anticoagulant bottles tubes. All blood samples collected were centrifuged at 3000rpm for 20 minutes (2,500rpm for samples in Sodium citrate tubes) and separated into relevant fractions [serum, plasma, buffy coat and red cell concentrates] within 2 hours of collection. Fractions were stored at -20o C in non-self defrosting freezers at peripheral sites before transfer to central biorepository. A daily temperature chart was kept on every freezer to monitor the freezer temperature in order to maintain the samples’ integrity Spot determination of plasma glucose level was carried out across all study sites using the ACCU- CHEK Active Blood Glucose Monitoring Device (Roche Diagnostics, GmBH, Germany), the principle of which was based on the reaction of blood glucose with glucose dehydrogenase enzyme resulting in colour changes which the meter converted to numerical values. Values obtained in mg/dl were converted to mmol/L.54 Glycated haemoglobin (HbA1c) level was also determined on whole blood from all subjects within 24 hours of sample collection using the Clover A1c Test Catridge System (Infopia Co. Ltd., Korea). The Clover A1c system uses the principle of boronate affinity chromatographic method for the determination of HbA1c in whole blood.55 Reagents in the system lyse red cells and bind haemoglobin, also the boronate resins bind the cis-doils of glycated haemoglobin. These are measured separately within the system and the ratio of glycated haemoglobin to total haemoglobin were expressed as percentage. Fasting lipid profile of subjects was determined by quantitative determination of cholesterol, triglycerides, HDL cholesterol using Downloaded from http://ahajournals.org by on August 8, 2023 commercially available kits (Randox Laboratories Ltd., UK; Biolabo S.A., France) and the LDL cholesterol was calculated using Friedwald equation.56 Cholesterol and Triglycerides were determined using the enzymatic hydrolysis/colorimetric method while HDL-cholesterol was determined by precipitation method and the cholesterol fraction measured as previously described.56,57 Values obtained in mg/dl were converted to mmol/L.51,58 To ensure equivalence across all sites, a standard operating procedure (SOP) was developed on the above laboratory tests and applied across all SIREN sites after a 3- day hands- on-training involving laboratory scientists from across all sites. Refresher trainings were also organized every year. The same brand of test equipment, reagents and test strips were procured and utilized across study sites. Downloaded from http://ahajournals.org by on August 8, 2023 Table S1. Definition of risk factors Variables Definitions Hypertension A sustained elevation of BP ≥140/90 mm Hg >72 hours after stroke, a premorbid history of hypertension, use of antihypertensive drugs before stroke or >72 hours after stroke onset. Diabetes mellitus History of diabetes mellitus, use of medications for diabetes mellitus, an HBA1c >6.5% or a fasting blood glucose levels of >7.0 mmol/L measured after the post-acute phase because of the known acute transient elevation of glucose as a stress response after stroke. Dyslipidemia Fasting total cholesterol ≥5.2 mmol/L, HDL-C (high-density lipoprotein cholesterol) ≤1.03 mmol/L, triglyceride ≥1.7 mmol/L, or LDL-C (low-density lipoprotein cholesterol) ≥3.4 mmol/L or use of statin before stroke onset. Cardiac disease Defined after evaluation by study cardiologists based on history or current diagnosis of atrial fibrillation, cardiomyopathy, heart failure, ischemic heart disease, rheumatic heart disease, or valvular heart diseases. Obesity Cut offs of waist-to-hip ratio of 0.90 (men) and 0.85 (women) ; body mass index of 30 kg/m2 Family history of Family history of cardiovascular risk/diseases was defined based on self- cardiovascular reported history of any of hypertension, diabetes mellitus, dyslipidemia, risk/diseases stroke, cardiac disease, or obesity in participants’ father, mother, sibling, or second-degree relative. Physical inactivity Individuals were classified as physically inactive if they do not regularly involved in moderate exercise (walking, cycling, or gardening) or strenuous exercise (jogging, football, and vigorous swimming) for at least 4 hours or more per week. Alcohol intake Alcohol use was categorized into current users (users of any form of alcoholic drinks) or never/former drinker, whereas alcohol intake was categorized as low drinkers (1–2 drinks per day for female and 1–3 drinks per day for male) and high drinker (>2 drinks per day for female and >3 drinks per day for male; 1 drink or 1 U of alcohol=8 g of alcohol). Smoking status Defined as current smoker (individuals who smoked any tobacco in the past 12 months) or never/former smoker. Downloaded from http://ahajournals.org by on August 8, 2023 Dietary history Dietary history included regularity of intake of food items such as meat, green leafy vegetables, fish, addition of salt at table, nuts, sugar, and other local staple food items. Regular intake was defined as intake on daily, weekly, or at least once monthly versus none in a month and non- regular intake is the reverse. Low vegetable consumption was defined as a self-reported frequency of vegetable consumption less than once per month; 12 months before stroke. Similarly, regular meat consumption was defined as a self-reported frequency of meat intake more than once (including daily etc) per month; 12 months before stroke occurrence. Psychosocial stress Psychosocial stress combined measures of stress at home/work (eg, irritability, anxiety, or sleeping difficulties) and life events, experienced in the 2 weeks preceding the stroke Downloaded from http://ahajournals.org by on August 8, 2023 Table S2. Distribution of vascular risk factors for Stroke by National Institutes of Health Stroke Scale score Characteristics All severe All non-severe P-value Severe Ischemic Severe P-value stroke stroke Stroke Hemorrhagic N=1027 N=1701 N=649 stroke N=378 Country, Ghana, n (%) 412(40.1) 644(37.9) 0.255 242(37.3) 170(45.0) 0.015 Age <0.001 <60 462(45.0) 893(52.5) <0.001 212(32.7) 250(66.1) ≥60 563(54.8) 805(47.3) 437(67.3) 126(33.3) Sex, Male, n (%) 560(54.5) 988(58.1) 0.069 327(50.4) 233(61.6) <0.001 Domicile 89(8.7) 136(8.0) 0.666 56(8.6) 33(8.7) 0.993 Rural, n (%) 282(27.5) 486(28.6) 179(27.6) 103(27.3) Semi-urban, n (%) 656(63.9) 1068(62.8) 414(63.8) 242(64.0) Urban, n (%) Marital status never married/single 35(3.4) 73(4.3) 0.308 14(2.16) 21(5.6) 0.013 married 751(73.13) 1264(74.3) 459(70.7) 292(77.3) 553(53.85) 970(57.0) 0.115 337(51.9) 216(57.1) 0.109 Monthly Income >$100, n (%) 810(78.9) 1439(84.6) <0.001 314(48.38) 496(31.2) 0.013 Education, (some) n (%) Living situation loneliness 47(4.6) 97(5.7) 0.220 27(4.2) 20(5.3) 0.387 living with others 969(94.4) 1600(94.1) 617(95.1) 352(93.1) Comorbidities Downloaded from http://ahajournals.org by on August 8, 2023 996(97.0) 1616(95.0) 0.009 626(96.5) 370(97.9) 0.122 Hypertension, n (%) 370(36.0) 1449(85.2) 0.738 565(87.1) 73(19.3) 0.006 Dyslipidemia, n (%) 375(36.5) 624(36.7) 0.944 271(41.8) 104(27.5) <0.001 Diabetes 103(10.0) 194(11.4) 0.270 86(13.3) 17(4.5) <0.001 Cardiac Disease, n (%) 791(77.0) 1326(78.0) 0.697 510(78.6) 281(74.3) 0.053 Waist-to-hip Ratio raised, n (%) 26.4±5.0 26.8±5.2 0.023 26.7±5.3 26.5±5.2 0.308 BMI*** (kg/m2), mean ± SD 148(14.4) 329(19.3) 0.014 95(16.64) 53(14.0) 0.984 BMI*** >30kg/m2, n (%) 51(5.0) 63(3.7) 0.114 36(5.6) 15(4.0) 0.253 Physical (Inactivity), n (%) 80(7.8) 176(10.4) 0.022 49(7.6) 31(8.2) 0.709 Tobacco (any use), n (%) Alcohol use categories: 698(68.0) 1080(63.5) 0.168 463(71.3) 235(62.2) 0.002 Never Use, n (%) 180(17.5) 339(19.3) 97(15.0) 83(22.0) Ever Low Use, n (%) 31(3.0) 51(3.0) 15(2.3) 16(4.2) Ever High Use, n (%) 175(17.0) 337(19.8) 0.171 102(15.7) 73(19.3) 0.279 Stress, n (%) 6(0.6) 11(0.7) 0.006 6(0.9) 0(0.0) 0.165 Cancer, n (%) 84(8.2) 123(7.2) 0.539 51(7.9) 33(8.7) 0.850 Depression, n (%) 381(37.1) 692(40.7) 0.063 231(35.6) 150(39.7) 0.191 Family history of CVD, n (%) 89(8.7) 115(6.8) 0.059 50(7.7) 39(10.3) 0.155 Adding salt at table, n (%) 353(34.4) 364(21.4) <0.001 227(35.0) 126(33.3) 0.615 Low vegetable consumption, n (%) Downloaded from http://ahajournals.org by on August 8, 2023 821(79.9) 1340(78.8) 0.195 512(78.9) 309(81.8) 0.224 Whole grains consumption, n (%) 549(53.5) 1129(66.4) <0.001 337(51.9) 212(56.1) 0.152 Legumes consumption, n (%) 784(76.3) 1341(78.8) 0.278 493(76.0) 291(77.0) 0.525 Fruit consumption, n (%) 316(30.8) 434(25.5) 0.002 194(29.9) 122(32.2) 0.369 Sugar consumption or otherwise, n (%) 680(66.2) 1134(66.7) 0.881 415(63.9) 84(22.2) 0.037 Regular Meat consumption % 879(87.9) 1452(85.4) 0.503 555(85.5) 324(85.7) 0.941 Fish consumption or otherwise, % Lesion volume <10 431(42.0) 1023(60.1) <0.001 320(49.3) 111(29.4) <0.001 10-30 231(22.5) 292(17.2) 97(15.0) 134(35.5) >30 259(25.2) 195(11.5) 145(22.3) 114(30.2) Blood pressure at presentation Systolic Mean±SD 163.4±32.8 154.3±29.0 <0.001 158.0±32.0 172.9±32.5 <0.001 Diastolic Mean±SD 97.0±19.2 96.1±116.7 <0.001 92.8±17.7 104.0±20.0 <0.001 Fasting Glucose Mean±SD 124.4±56.7 113.1±43.7 <0.001 126.0±61.2 125.1±52.3 0.883 Neutrophil/ lymphocyte ratio 7.1±29.4 5.5±27.2 0.236 7.4±38.1 7.2±7.3 0.950 *Compared to corresponding table for SLS, less variables have significant results in this National Institutes of Health Stroke Scale score results BMI-Body mass index Downloaded from http://ahajournals.org by on August 8, 2023 Table S3. Vascular risk factors for stroke severity stratified by Stroke primary type by National Institutes of Health Stroke Scale score Characteristics All severe Ischemic Stroke Hemorrhagic stroke stroke N=1027 Severe Not severe p-value Severe Not severe p-value N=649 N=1283 (Exact N=378 N=418 (Exact test) test) Country, Ghana, n (%) 412(40.1) 242(37.3) 398(31.0) 0.006 170(45.0) 246(58.9) <0.001 Age ≤59 462(45.0) 212(32.7) 596(46.5) <0.001 250(66.1) 297(71.1) 0.150 ≥60 563(54.8) 437(67.3) 437(34.1) 126(33.3) 120(28.7) 560(54.5) 327(50.4) 713(55.6) 0.034 233(61.6) 275(65.8) 0.238 Sex, Male, n (%) Domicile 89(8.7) 56(8.6) 109(8.5) 0.690 33(8.7) 27(6.5) 0.493 Rural, n (%) 282(27.5) 179(27.6) 377(29.4) 103(27.25) 112(26.8) Semi-urban, n (%) 656(63.9) 414(63.8) 794(61.9) 242(64.0) 274(65.6) Urban, n (%) 553(53.9) 337(51.9) 723(56.4) 0.068 216(57.14) 248(59.3) 0.622 Monthly Income >$100, n (%) Living situation 0.197 loneliness 47(4.6) 27(4.2) 65(5.1) 0.388 20(5.3) 32(7.7) living with others 969(94.4) 617(95.1) 1214(94.6) 352(93.12) 386(92.3) 810(78.9) 314(48.4) 1058(82.5) 0.002 496(131.2) 381(91.2) <0.001 Education, (some) n (%) Downloaded from http://ahajournals.org by on August 8, 2023 996(97.0) 626(96.5) 1207(94.1) 0.025 370(97.9) 409(97.9) 0.766 Hypertension, n (%) 370(36.0) 565(87.1) 1120(87.3) 0.882 73(19.3) 329(78.7) 0.488 Dyslipidemia, n (%) 375(36.5) 271(41.8) 508(39.6) 0.360 104(27.5) 116(27.8) 0.959 Diabetes 103(10.0) 86(13.3) 171(13.3) 0.972 17(4.5) 23(5.5) 0.522 Cardiac Disease, n (%) 791(77.0) 510(78.6) 1015(79.1) 0.675 281(74.3) 311(74.4) 0.411 Waist-to-hip Ratio raised, n (%) 26.4±5.0 26.2±5.2 26.9±5.3 0.010 26.44±4.8 26.5±5.2 0.76 BMI*** (kg/m2), mean ± SD 148(14.4) 95(16.6) 261(20.3) 0.014 53(14.0) 68(16.3) 0.695 BMI*** >30kg/m2, n (%) 51(5.0) 36(5.6) 1216(94.8) 0.055 15(4.0) 397(95.0) 0.928 Physical Activity (Inactivity), n (%) 80(7.8) 49(7.6) 131(10.2) 0.050 31(8.2) 45(10.8) 0.207 Tobacco (any use), n (%) Alcohol use categories: 698(68.0) 860(67.0) 0.189 235(62.2) 220(52.6) 0.115 Never Use, n (%) 463(71.3) 83(22.0) 109(26.1) 180(17.5) Ever Low Use, n (%) 97(15.0) 230(17.93) 31(3.0) 16(4.2) 21(5.0) Ever High Use, n (%) 15(2.3) 30(2.3) 175(17.0) 0.050 73(19.3) 81(19.4) 0.959 Stress, n (%) 102(15.7) 256(20.0) 6(0.6) 0.012 0(0.0) 0(0.0) 0.017 Cancer, n (%) 6(0.9) 8(0.6) 84(8.2) 51(7.9) 0.668 33(8.7) 29(6.9) 0.605 Depression, n (%) 94(7.3) 381(37.1) 231(35.6) 495(38.6) 0.200 150(39.7) 197(47.3) 0.034 Family history of CVD, n (%) Downloaded from http://ahajournals.org by on August 8, 2023 89(8.7) 50(7.7) 78(6.1) 0.171 39(10.3) 37(8.9) 0.411 Adding salt at table, n (%) 353(34.4) 227(35.0) 265(20.7) <0.001 126(33.3) 99(23.7) <0.001 Low vegetable consumption, n (%) 821(79.9) 512(78.9) 1000(77.9) 0.712 309(81.8) 340(81.3) 0.097 Whole grains consumption, n (%) 549(53.5) 337(51.9) 831(64.8) <0.001 212(56.1) 298(71.3) 0.001 Legumes consumption, n (%) 784(76.3) 493(76.0) 1002(78.1) 0.156 291(77.0) 339(81.1) 0.859 Fruit consumption, n (%) 316(30.8) 194(29.9) 320(24.9) 0.031 122(32.3) 114(27.3) 0.043 Sugar consumption or otherwise, n (%) 680(66.2) 415(63.9) 846(65.9) 0.311 84(22.22) 288(68.90) 0.122 Regular Meat consumption % 879(85.5) 555(85.5) 1073(83.6) 0.503 324(85.7) 379(90.6) 0.300 Fish consumption or otherwise, % Lesion volume 431(41.97) 320(49.31) 827(64.46) <0.001 111(29.37) 196(46.89) <0.001 <10 231(22.49) 97(14.95) 153(11.93) 134(35.45) 139(33.25) 10-30 259(25.22) 145(22.34) 137(10.68) 114(30.16 58(13.88) >30 Blood pressure at presentation 163.4±32.8 158.0±32.0 150.7±27.1 <0.001 172.9±32.5 164.3±32.8 <0.001 Systolic Mean±SD 97.0±19.2 92.8±17.7 90.8±15.8 0.022 104.0±20.0 11100.6±19.6 0.017 Diastolic Mean±SD 124.4±56.7 126.0±61.2 115.2±46.7 0.005 125.7±42.6 109.1±32.7 0.001 Fasting Glucose Mean±SD 7.1±29.35 7.4±38.1 5.0±29.2 0.240 7.2±7.3 7.8±27.5 0.735 Neutrophil/lymphocyte ratio Severe stroke was defined as National Institute of Health severity scale score >20 BMI-Body mass index Downloaded from http://ahajournals.org by on August 8, 2023 Table S4. Association of Prior Vascular risk factors and comorbidities of severe stroke patients by Stroke Levity Scale Variable All severe Ischaemic p-value Haemorrhagic p-value stroke Stro ke (Exact stroke (Exact N=1854 test) test) Severe Not severe Severe Not severe N=1090 N=1201 N=548 N=428 Hypertension 1575 (85.0%) 1036 (95.1%) 1135(94.4%) 0.618 539 (98.4%) 417(97.4%) 0.312 Diabetes Mellitus 592 (32.0%) 446 (41.0%) 506(42.1%) 0.567 146 (26.7%) 131(30.6%) 0.171 stroke 253 (13.7%) 199 (18.3%) 267(22.2%) 0.055 54 (9.9%) 48(11.2%) 0.760 Neck manipulation 78 (4.3%) 28(2.57%) 30(2.50%) 0.837 12(2.2%) 8(1.8%) 0.831 Dyslipidemia 1339 (72.3%) 921 (84.5%) 1053(87.60) 0.028 418 (76.3%) 348(81.31) 0.065 Obesity 246 (13.3%) 173 (15.9%) 243(20.2%) 0.631 73 (13.4%) 68(15.89) 0.851 History Of TIA 44 (2.4%) 36 (3.4%) 46(3.8%) 0.747 8 (1.5%) 1(0.2%) 0.133 HIV /AIDs 10 (0.6%) 8 (0.8%) 10(0.8%) 0.505 2 (0.4%) 2(0.5%) 0.525 History of Chronic Kidney 14 (0.8%) 9 (0.9%) 11(0.9%) 0.965 5 (1.0%) 13(3.0%) 0.049 Disease Sickle Cell disease 6 (0.4%) 4 (0.4%) 2(0.2%) 0.453 2 (0.4%) 3(0.7%) 0.650 Atrial fibrillation 14 (0.8%) 12 (1.2%) 12(1.0%) 0.899 2 (0.4%) 1(0.2%) 0.725 history of neck injury 13 (0.8%) 9 (0.9%) 10(0.8%) 0.974 4(0.8%) 9(2.1%) 0.104 schizophrenia 5 (0.3%) 5(0.5%) 5(0.4%) 0.372 4(0.7%) 2(0.5%) 0.386 history of recurrent 21 (1.2%) 18 (1.7%) 27(2.3%) 0.370 3 (0.6%) 5(1.2%) 0.180 miscarriage history of chronic bronchitis 3 (0.2%) 8(0.733) 15(1.3%) 0.457 1(0.2%) 1(0.2%) 0.981 history of cancer 6(0.4%) 5 (0.5%) 9(0.8%) 0.272 1 (0.2%) 3(0.7%) 0.147 history of tuberculosis 3 (0.2%) 2(0.2%) 4(0.3%) 0.548 1 (0.2%) 1(0.2%) 0.583 history of dental problem in 83 (4.5%) 64 (5.9%) 98(8.1%) 0.101 19 (3.5%) 37(8.6%) 0.002 last one year history of use of sleep 53 (2.9%) 39 (3.6%) 53(4.4%) 0.761 14 (2.6%) 11(2.6%) 0.721 disorders history of febrile illness in the 141 (7.7%) 102 (9.4%) 129(10.730 0.360 39 (7.2%) 40(9.4%) 0.443 last four weeks History of use of 17 (1.0%) 14 (1.3%) 9(0.8%) 0.446 3 (0.6%) 3(0.7) 251 anticoagulants History of Migraine 37(2.0%) 22(2.1%) 44(3.7%) 0.062 15(2.8%) 20(4.7%) 0.259 Stress in the last 2 weeks 267 (14.5%) 164(15.1%) 249(20.7%) 0.002 103 (18.8%) 92(21.5%) 0.408 Downloaded from http://ahajournals.org by on August 8, 2023 Depression in the last 4 weeks 107 (5.8%) 69 (6.4%) 102(8.5%) 0.138 38 (7%) 36(8.4%) 0.290 History of heart disease 50 (2.7%) 41 (3.8%) 48(4.0%) 0.904 9 (1.7%) 9(2.1%) 0.860 Downloaded from http://ahajournals.org by on August 8, 2023 Table S5. Table showing the clinical presentation of participants by Stroke Levity Scale Clinical Presentation All severe Ischaemic Stroke Haemorrhagic stroke stroke N=1854 Severe Not severe p-value Severe Not severe p-value N=1090 N=1202 (Exact N=548 N=428 (Exact test) test) Headache at first presentation 633(34.1%) 388(35.6%) 430(35.8%) 0.993 245(44.7%) 143(33.4%) 0.001 Vomiting at first presentation 369(19.9%) 170(15.6%) 160(13.3%) 0.107 199(36.3%) 140(32.7%) 0.327 Disturbed consciousness 853(46.0%) 496(45.5%) 335(27.9%) <0.001 357(65.2%) 192(44.9%) <0.001 Neck stiffness 183(9.9%) 98(9.0%) 75(6.2%) 0.012 85(15.5%) 74(17.3%) 0.370 Ictus occurred with activity 615(33.2%) 354(32.5%) 380(31.6%) 0.699 261(47.6%) 183(42.8%) 0.215 Ictus occurred at rest 733(39.5%) 578(53.0%) 674(56.1%) 0.144 155(28.3%) 117(27.3%) 0.895 Focal neurological deficit 1090(58.8%9) 731(67.1%) 377(31.4%) 0.220 359(65.5%) 275(64.3%) 0.844 Disturbed speech 705(38.0) 364(33.4%) 489(40.7%) <0.001 341(62.2%) 220(51.4%) 0.002 Paraesthsiae 981(52.9%) 928(85.1%) 951(79.1%) <0.001 53(9.7%) 57(13.3%) 0.065 Ataxic gait 150(8.1%) 103(9.5%) 160(13.3%) 0.004 47(8.6%) 46(10.8%) 0.212 Seizure 1006(54.3%) 931(85.4%) 405(33.7%) 0.06 75(13.7%) 41(9.6%) 0.071 Stroke affecting right 865(46.7%) 573(52.6%) 452(37.6%) <0.001 292(53.3%) 133(31.1%) <0.001 Stroke affects left 579(31.2%) 391(35.9%) 554(46.1%) <0.001 188(34.3%) 206(48.1%) <0.001 Stroke affecting both 130(7.0%) 84(7.7%) 85(7.1%) <0.001 46(8.4%) 40(9.4%) <0.001 Disability prior to stroke (Yes) 623(33.6%) 367(33.7%) 203(16.9%) <0.001 256(46.7%) 344(80.4%) <0.001 Gait(Normal) 209(11.3%) 130(11.9%) 268(22.3%) <0.001 79(14.4%) 101(23.6%) 0.048 Downloaded from http://ahajournals.org by on August 8, 2023 Previous stroke 164(8.9%) 124(11.4%) 135(11.2%) 0.568 40(7.3%) 27(6.3%) 0.909 Previous Transient Ischemic attack 136(7.3%) 102(9.4%) 120(10.0) 0.102 34(6.2%) 22(5.1%) 0.785 Cranial nerve deficit 603(32.5%) 267(24.5%) 433(36.0%) <0.001 336(61.3%) 226(52.8%) 0.006 Facial nerve deficit 1171(63.2%) 789(72.4%) 671(55.8%) <0.001 382(69.7%) 234(54.7%) <0.001 Language deficit 871(47.0) 569(52.2%) 336(28.0) <0.001 302(55.1%) 105(24.5%) <0.001 Visual field deficit 309(16.7%) 201(18.4%) 112(9.3%) <0.001 108(24.7%) 48(11.2%) <0.001 Speech articulation deficit 732(39.5%) 485(44.5%) 445(37.0%) <0.001 247(45.1%) 161(37.7%) <0.001 Ideomotor apraxia 497(26.8%) 305(28.0%) 189(15.7%) <0.001 192(35.0%) 140(32.7%) <0.001 Constructional apraxia 171(9.2%) 109(10.0%) 77(6.4%) <0.001 62(11.3%) 32(7.48) <0.001 Tactile agnosia 143(7.7%) 86(7.9%) 56(4.7%) <0.001 57(10.4%) 22(5.1%) <0.001 Cortical sensory loss 112(6.0%) 89(8.2%) 42(3.5%) <0.001 23(4.2%) 62(14.5%) <0.001 Hemineglect 147(7.9%) 59(5.4%) 28(2.3%) <0.001 88(16.1%) 64(15.0) <0.001 Abnormal involuntary movement (tremor, 64(3.5%) 45(4.1%) 51(2.2%) <0.001 19(3.5%) 11(2.6) <0.001 dystonia, chorea etc. in affected limbs) Vibration/Joint position deficit 37(2.0%) 22(2.0%) 16(1.3%) <0.001 15(2.7%) 7(1.6%) <0.001 Cerebellar deficit 45(2.4%) 28(2.6%) 55(4.6%) <0.001 17(3.1%) 19(4.4%) <0.001 Pseudobulbar affectation 99(5.3%) 64(5.9%) 56(4.7%) <0.001 35(6.4%) 12(2.8%) <0.001 Intermitent claudication 34(1.8%) 29(2.7%) 37(3.1%) <0.001 5(0.9%) 5(1.2%) 0.005 angina 32(1.7%) 21(1.9%) 25(2.1%) <0.001 11(2.0%) 7(1.6%) 0.004 palpitation 133(7.2%) 93(8.5%) 145(12.1%) <0.001 40(7.3%) 44(10.3%) 0.001 Differential warmth in extremities 58(3.1%) 12(1.1%) 5(0.4%) 0.105 46(8.4%) 32(7.5%) 0.368 Thickened arterial wall 481(25.9%) 336(30.8%) 360(30.0) 0.105 145(26.5%) 94(22.0) 0.073 Downloaded from http://ahajournals.org by on August 8, 2023 Locomotor brachialis 387(20.9%) 278(25.5%) 300(25.0) 0.205 109(19.9%) 69(16.1%) 0.068 Heaving apex 394(21.3%) 261(23.9%) 173(14.4%) <0.001 133(24.3%) 78(18.2%) 0.028 Cardiac murmurs 101(5.5%) 80(7.3%) 36(3.0%) <0.001 21(3.8%) 11(2.6%) 0.164 displaced apex 477(25.7%) 327(30.0%) 288(24.0%) 0.001 150(27.4%) 120(28.0%) 0.598 DVT 37(2.0%) 31(2.8%) 11(0.9%) <0.001 6(1.1%) 3(0.7%) 0.248 Heart failure 52(2.8%) 47(4.3%) 33(2.8%) 0.021 5(0.9%) 3(0.7%) 0.365 Unequal carotid 26(1.4%) 18(1.7%) 16(1.3%) 0.082 8(1.5%) 3(0.7%) 0.059 Downloaded from http://ahajournals.org by on August 8, 2023 Table S6. Table showing the clinical presentation of participants by National Institutes of Health Stroke Scale score Clinical Presentation All severe Ischemic Stroke Hemorrhagic stroke stroke N=1027 Severe Not severe p- Severe Not severe p- N=649 N=1283 value N=378 N=418 value (Exact (Exact test) test) Headache at first presentation 429(41.8%) 232(35.8%) 477(37.2%) 0.546 197(52.1%) 264(63.2%) 0.004 Vomiting at first presentation 258(25.1%) 518(79.8%) 1098(85.6%) 0.001 143(37.8%) 130(31.1%) 0.039 Disturbed consciousness 590(57.5%) 335(51.6%) 337(26.3%) <0.001 255(67.5%) 172(41.2%) <0.001 Neck stiffness 121(11.8%) 50(7.7%) 80(6.2%) 0.213 71(18.8%) 53(12.7%) 0.017 Ictus occurred with activity 427(41.6%) 366(56.4%) 405(31.6%) 0.098 195(51.6%) 163(39.0%) <0.001 Ictus occurred at rest 373(36.3%) 268(41.3%) 517(40.3%) 0.614 105(27.8%) 114(27.3%) 0.837 Focal neurological deficit 641(62.4%) 405(62.4%) 839(65.4%) 0.134 236(62.4%) 267(63.9%) 0.744 Disturbed speech 680(66.2%) 440(67.8%) 705(46.0) <0.001 240(63.5) 216(51.7%) <0.001 Paraesthsiae 119(11.6%) 79(12.2%) 1819(14.1%) 0.235 40(10.6%) 49(11.7%) 0.691 Ataxic gait 103(10.0%) 65(10.0%) 169(13.2%) 0.045 38(10.1%) 46(11.0%) 0.667 Seizure 111(10.8%) 68(10.5%) 93(7.3%) 0.017 43(11.4%) 38(9.1%) 0.266 Stroke affecting right 486(47.3%) 310(47.8%) 544(42.4%) 0.007 176(46.6%) 174(41.6%) 0.471 Stroke affects left 412(40.1%) 261(40.2%) 564(44.0) 0.007 151(40.0) 172(41.2%) 0.471 Stroke affecting both 90(8.8%) 56(8.6%) 69(5.4%) 0.007 34(9.0%) 29(6.9%) 0.471 Disability prior to stroke (Yes) 273(26.6%) 195(30.1%) 168(13.1%) <0.001 78(20.6%) 34(8.1%) <0.001 Gait(Normal) 168(16.4%) 104(16.0%) 254(19.8%) 0.013 64(16.9%) 88(21.1%) 0.353 Previous stroke 114(11.1%) 82(12.6%) 125(19.8%) 0.057 32(8.5%) 26(6.2%) 0.227 Downloaded from http://ahajournals.org by on August 8, 2023 Previous Transient Ischemic attack 82(8.0) 54(8.32) 143(11.2%) 0.138 28(7.4%) 22(5.3%) 0.043 Cranial nerve deficit 687(66.9%) 451(69.5%) 726(56.6%) <0.001 236(62.4%) 219(52.4%) 0.001 Facial nerve deficit 739(72.0%) 475(73.2%) 723(56.4%) <0.001 264(69.8%) 229(54.8%) <0.001 Language deficit 605(58.9%) 386(59.5%) 345(26.9%) <0.001 219(57.9%) 105(25.1%) <0.001 Visual field deficit 238(23.2%) 150(23.1%) 100(7.8%) <0.001 88(23.3%) 36(8.6%) <0.001 Speech articulation deficit 542(52.8%) 349(53.8%) 461(35.9%) <0.001 193(51.1%) 162(38.8%) <0.001 Ideomotor apraxia 230(22.4%) 144(22.2%) 140(10.9%) <0.001 86(22.8%) 49(11.7%) <0.001 Constructional apraxia 144(14.0%) 93(14.3%) 70(5.5%) <0.001 51(13.5%) 23(5.5%) <0.001 Tactile agnosia 122(11.9%) 74(11.4%) 46(3.6%) <0.001 48(12.7%) 11(2.6%) <0.001 Cortical sensory loss 122(11.9%) 69(10.6%) 35(2.7%) <0.001 53(14.0%) 15(3.6%) <0.001 Hemineglect 122(11.9%) 70(10.8%) 55(4.3%) <0.001 52(13.8%) 21(5.0%) <0.001 Abnormal involuntary movement 50(4.9%) 32(4.9%) 40(3.1%) 0.007 18(4.8%) 6(1.4%) 0.001 (tremor, dystonia, chorea etc. in affected limbs) Vibration/Joint position deficit 27(2.6%) 14(2.2%) 17(1.3%) <0.001 13(3.4%) 4(1.0%) <0.001 Cerebellar deficit 42(4.1) 21(3.2%) 52(4.1%) <0.001 21(5.6%) 13(3.1%) <0.001 Pseudobulbar affectation 86(8.4%) 52(8.0%) 39(3.0%) <0.001 34(9.0) 3(0.7%) <0.001 Intermitent claudication 20(2.0) 16(2.5%) 40(3.1%) <0.001 4(1.1%) 7(1.7%) 0.001 angina 17(1.7%) 12(1.9%) 26(2.0%) <0.001 5(1.3%) 5(1.2%) 0.055 palpitation 87(8.5%) 50(7.7%) 153(11.9%) <0.001 37(9.8%) 32(7.7%) 0.003 Differential warmth in extremities 29(2.8%) 23(3.5%) 34(2.7%) 0.489 6(1.6%) 3(0.7%) 0.248 Thickened arterial wall 280(27.3%) 177(27.3%) 388(30.2%) 0.075 103(27.3%) 81(19.4%) 0.015 Locomotor brachialis 203(19.8%) 127(19.6%) 332(25.9%) 0.038 76(20.1%) 58(13.9%) 0.001 Downloaded from http://ahajournals.org by on August 8, 2023 Heaving apex 229(22.3%) 137(21.1%) 193(15.0%) 0.002 92(24.3%) 70(16.8%) 0.013 Cardiac murmurs 48(4.7%) 39(6.0%) 49(3.8%) 0.030 9(2.4%) 13(3.1%) 0.469 displaced apex 289(28.1%) 188(29.0%) 300(23.4%) 0.025 101(26.7%) 103(24.6%) 0.463 DVT 17(1.7%) 14(2.2%) 15(1.18%) 0.076 3(0.8%) 3(0.7%) 0.693 Heart failure 30(2.9%) 26(4.0%) 35(2.7%) 0.146 4(1.1%) 3(0.7%) 0.544 Unequal carotid 17(1.7%) 12(1.9%) 17(1.3%) 0.050 5(1.3%) 3(0.7%) 0.657 Downloaded from http://ahajournals.org by on August 8, 2023 Table S7. Radiological determinants of severe Ischemic stroke patients by National Institutes of Health Stroke Scale score Variables Sub-variables Ischemic stroke p-value Severe Non-severe N=649 N=1283 TOAST classification Large artery arteriosclerosis 230 (35.5%) 359 (28%) <0.001 Cardio-embolism 61 (9.4%) 76 (6%) small vessel occlusion 168 (25.9%) 478 (37.3%) Other determined etiology 1 (0.2%) 4 (0.4%) Undetermined etiology 120 (18.5%) 260 (20.3%) OCSP Subtype TACI 125 (19.3%) 110 (8.6%) <0.001 PACI 205 (31.6%) 372 (29%) POCI 54 (8.4%) 108 (8.5%) LACI 189 (29.2%) 572 (44.6%) ASCO Subtype Atherosclerosis 107 (16.5%) 212 (16.6%) 0.010 Small vessel disease 203 (31.3%) 538 (42%) Cardio embolism 82 (12.7) 130 (10.2%) Other causes 12 (1.9%) 25 (2%) Downloaded from http://ahajournals.org by on August 8, 2023 Table S8. Echocardiographic determinant of severe stroke patients by National Institutes of Health Stroke Scale score Variable Non-severe Severe Ischaemic p-value Non-severe Severe p-value ischemic Stroke; f (%) (Exact Hemorrhagic Haemorrhagic (Exact test) stroke N=649 test) stroke stroke; f (%) N=1283 N=418 N=428 N=378 Sinus Arrythmia 25 (1.9) 18 (2.8%) 0.418 10 (2.3) 13 (3.5%) 0.392 Atrial Fibrillation 38 (3.0) 24 (3.7%) 0.588 1 (0.2) 1(0.3%) 1.000 Atrial Flutter(ECG 4 (0.3) 6 (1.0%)` 0.180 0 (0.0) 0 (0%) - determined ) Atrial enlargement 38 (3.0) 24 (5.7) 0.580 1 (0.2) 1 (0.3) 0.236 (Right/Left Ejection fraction ≤ 40% 57 (4.4) 33 (5.1%) 7 (14.0) 9 (2.4%) 41-50% 71 (5.5) 26 (4.1%) 0.046 14 (3.3) 7 (1.9%) 0.254 ≥51% 429 (32.8) 138 (21.3%) 95 (22.2) 53 (14.1%) Compared to corresponding table for SLS, less variables have significant results in this NIHSS results Downloaded from http://ahajournals.org by on August 8, 2023 Table S9. Assessment of factors associated with stroke severity by National Institutes of Health Stroke Scale score Risk Factor All Severe Stroke Severe Ischemic Severe Hemorrhagic Stroke Stroke Age 1.26(0.95 – 1.68) 1.42(0.94 – 2.16) 1.28(0.79 – 2.07) Hypertension 1.35(0.69 - 2.62) 1.44(0.64 – 3.23) 1.76(0.29 – 10.55) Dyslipidemia 1.09(0.77 – 1.53) 0.91(0.57 – 1.45) 1.72(0.95 – 3.11) Diabetes mellitus 0.76(0.59 – 0.98) 0.89(0.65 – 1.22) 0.65(0.37 – 1.14) Obesity 0.74(0.55 – 0.98) 0.75(0.50 – 1.10) 0.67(0.36 – 1.23) Cigarette smoking 1.10(0.53 – 2.26) 1.50(0.5 – 0.45) 0.69(0.23 – 2.11) Stress 0.83(0.60 – 1.15) 0.71(0.47 – 1.08) 1.19(0.64 – 2.18) Cardiac disease 0.68(0.45 – 1.01) 0.75(0.48 – 1.19) 0.55(0.18 – 1.70) Alcohol 0.76(0.56 – 1.03) 0.61(0.40 – 0.93) 0.74(0.43 – 1.27) Meat consumption 1.31(0.92 – 1.86) 1.00(0.66 – 1.52) 3.65(1.66 – 8.00) Low vegetable consumption 2.54(1.97 – 3.28) 2.52(1.83 – 3.47) 3.38(2.02 – 5.67) Depression 1.34(0.84 – 2.12) 1.32(0.73 – 2.38) 1.63(0.68 – 3.85) Physical inactivity 1.54(0.78 – 3.02) 1.40(0.63 – 3.14) 2.12(0.53 – 8.38) Salt intake 1.03(0.67 – 1.57) 0.96(0.56 – 1.67) 0.87(0.41 – 1.84) Right/Left Atrial 0.90(0.68 – 1.18) 0.88(0.61 – 1.25) 0.95(0.56 – 1.60) Downloaded from http://ahajournals.org by on August 8, 2023 Table S10. Radiological determinants of severe stroke patients by National Institutes of Health Stroke Scale score Variables Sub-variables Hemorrhagic stroke p-value Severe Non-severe N=378 N=418 Location of lesion Lobal 73 (19.4%) 108(25.9%) 0.028 Non-lobal 305 (80.7%) 310(74.2%) SMASH-U Structural 8 (2.2%) 21 (5.1%) Medication-related 0 (0%) 3 (0.8%) 0.054 Amyloid Angiopathy 3 (0.8%) 1 (0.3%) Systemic Disease 1 (0.3%) 3 (0.8%) Hypertensive 307 (81.3%) 331 (79.2%) Undetermined 10 (2.7%) 6 (1.5%) Downloaded from http://ahajournals.org by on August 8, 2023 Table S11. Assessment of factors associated with stroke severity by National Institutes of Health Stroke Scale score ALL STROKE ISCHEMIC STROKE ICH Adjusted OR (95% Adjusted OR (95% Adjusted OR (95% CI) CI) CI) Right/Left Atrial 0.94(0.59 – 1.52) 1.03(0.63 – 1.68) 0.92(0.45 – 1.88) Lesion Volume ≥10 2.00(1.29 – 3.11) 1.45(0.79 – 2.68) 1.52(0.79 – 2.89) ≥30 3.47(1.80 – 6.68) 1.98(0.95 – 4.13) 3.91(1.32 – 11.61) Systolic BP at presentation 1.00(0.99 – 1.01) 1.00(0.99 – 1.01) 1.00(0.99 – 1.01) Fasting Glucose 1.00(1.00 – 1.01) 1.00(1.00 – 1.01) 1.01(1.00 – 1.02) Neutrophil/lymphocyte 1.00(0.99 – 1.01) NA NA ratio Stroke type ICH vs 0.83(0.54 – 1.28) NA NA Ischemic Non-lobar vs lobar NA NA 1.13(0.55 – 2.29) Causes of ICH, NA NA 0.88(0.20 – 3.79) Hypertensive vs others Ischemic stroke subtypes NA NA NA OCSP] subtypes NA NA LACI(Ref) NA 1 NA TACI NA 2.98(1.43 – 6.21) NA PACI NA 1.39(0.84 – 2.31) NA POCI NA 1.44(0.73 – 2.82) NA Downloaded from http://ahajournals.org by on August 8, 2023 Table S12. Two-way interaction factors and low consumption of vegetables with Stroke severity by stroke levity scale ALL STROKE ISCHEMIC ICH STROKE Adjusted OR (95% Adjusted OR (95% CI) CI) Adjusted OR (95% CI) Age 1.03(0.84 – 1.27) 0.88(0.66 – 1.18) 1.47(1.04 – 2.08) Hypertension 0.92(0.61 – 1.39) 0.87(0.54 – 1.41) 1.73(0.55 – 5.48) Dyslipidemia 0.77(0.61 – 0.97) 0.73(0.53 – 1.01) 0.85(0.55 – 1.29) Meat consumption 1.41(1.08 – 1.83) 1.35(0.97 - 1.88) 1.72(1.05 – 2.81) Low vegetable consumption 0.42(0.14 – 1.25) 0.32(0.09 – 1.17) # Depression 0.72(0.54 – 0.97) 0.68(0.47 – 0.98) 0.75(0.44 – 1.29) Age * Low vegetable consumption 1.18(0.81 – 1.73) 1.76(1.01 – 3.08) 0.75(0.38 – 1.46) Hypertension* Low vegetable consumption 1.97(0.78 – 5.00) 1.84(0.64 – 5.24) # Dyslipidemia* Low vegetable consumption 1.10(0.72 – 1.68) 1.23(0.68 – 2.21) 1.02(0.46 – 2.25) Meat consumption* Low vegetable 1.48(0.95 – 2.32) 1.26(0.73 – 2.17) 2.88 (1.16 – 7.16) consumption Depression* Low vegetable consumption 1.59(0.75 – 3.38) 1.60(0.67 – 3.80) 1.69(0.37 – 10.38) # Unstable odds ratio Downloaded from http://ahajournals.org by on August 8, 2023 Table S13. Two-way interaction factors and low consumption of vegetables with Stroke severity by National Institutes of Health Stroke Scale score ALL STROKE ISCHEMIC ICH STROKE Adjusted OR (95% Adjusted OR (95% CI) Adjusted OR (95% CI) CI) Age 1.28(1.01 – 1.63) 1.58(1.09 - 2.28) 1.44(0.97 – 2.13) Hypertension 0.91(0.56 – 1.48) 0.76(0.42 – 1.37) 1.57(0.40 – 6.16) Dyslipidemia 1.20(0.90 – 1.61) 1.07(0.70 – 1.63) 1.75(1.06 – 2.91) Meat consumption 0.88(0.66 – 1.18) 0.70(0.49 – 1.00) 1.55(0.89 – 2.70) Low vegetable consumption 0.44(0.12 – 1.64) 0.25(0.04 – 1.29) # Depression 1.30(0.94 – 1.81) 1.35(0.88 – 2.05) 1.36(0.75 – 2.47) Age * Lesion Volume 1.11(0.73 – 1.69) 1.35(0.71 – 2.56) 0.80(0.40 – 1.59) Hypertension* Lesion Volume 3.40(1.11 – 10.45) 5.35(1.32 – 21.68) # Dyslipidemia* Lesion Volume 0.64(0.40 – 1.03) 0.65(0.33 – 1.27) 0.61(0.26 – 1.41) Meat consumption* Lesion Volume 1.98(1.20 – 3.25) 2.37(1.29 – 4.32) 2.13(0.77 – 5.93) Depression* Lesion Volume 0.59(0.27 – 1.30) 0.48(0.19 – 1.21) 0.56(0.09 – 3.30) # Unstable odds ratio Downloaded from http://ahajournals.org by on August 8, 2023 Table S14. Two-way interaction factors and lesion volume of vegetables with Stroke severity by stroke levity scale ALL STROKE ISCHEMIC ICH STROKE Adjusted OR (95% Adjusted OR (95% CI) Adjusted OR (95% CI) CI) Age 0.86(0.54 – 1.35) 0.78(0.43 – 1.39) 1.12(0.51 – 2.48) Hypertension 0.51(0.24 – 1.07) 0.56(0.24 – 1.30) 0.35(0.05 – 2.34) Dyslipidemia 0.72(0.44 – 1.18) 0.60(0.33 – 1.08) 1.17(0.46 – 2.97) Meat consumption 2.32(1.41 – 3.83) 1.89(1.06 – 3.37) 4.95(1.67 – 14.65) Low vegetable consumption 1.19(0.78 – 1.81) 1.38(0.85 – 2.25) 0.90(0.39 – 2.10) Depression 0.74(0.36 – 1.51) 0.75(0.32 – 1.74) 0.79(0.20 – 3.13) Age * Lesion Volume 1.13(0.89 – 1.44) 1.18(0.85 – 1.63) 1.09(0.73 – 1.62) Hypertension* Low vegetable consumption 1.56(1.09 – 2.24) 1.38(0.89 – 2.13) 2.14(1.03 – 4.41) Dyslipidemia* Lesion Volume 1.09(0.83 – 1.42) 1.23(0.87 – 1.73) 0.87(0.54 – 1.40) Meat consumption* Lesion Volume 0.83(0.63 – 1.10) 0.85(0.61 – 1.18) 0.67(0.38 – 1.17) Low Vegetable consumption* Lesion Volume 1.16(0.92 – 1.46) 1.00(0.75 – 1.32) 1.45(0.94 – 2.25) Depression* Lesion Volume 1.02(0.69 – 1.51) 0.99(0.61 – 1.62) 1.03(0.51 – 2.07) Downloaded from http://ahajournals.org by on August 8, 2023 Table S15. Two-way interaction factors and lesion volume of vegetables with Stroke severity by National Institutes of Health Stroke Scale score ALL STROKE ISCHEMIC ICH STROKE Adjusted OR (95% Adjusted OR (95% CI) Adjusted OR (95% CI) CI) Age 1.40(0.83 – 2.35) 1.63(0.81 – 3.29) 1.32(0.55 – 3.18) Hypertension 0.50(0.21 – 1.19) 0.42(0.15 – 1.13) 0.51(0.06 – 4.26) Dyslipidemia 0.92(0.52 – 1.63) 0.87(0.43 – 1.76) 1.24(0.43 – 3.51) Meat consumption 1.49(0.84 – 2.62) 1.33(0.68 – 2.58) 2.67(0.82 – 8.75) Low vegetable consumption 1.98(1.26 – 3.11) 2.20(1.29 – 3.75) 1.32(0.54 – 3.22) Depression 1.36(0.61 – 3.02) 1.32(0.51 – 3.41) 1.61(0.34 – 7.50) Age* Lesion Volume 0.96(0.74 – 1.26) 1.00(0.69 – 1.45) 1.02(0.66 – 1.58) Hypertension* Low vegetable consumption 1.90(1.26 – 2.88) 1.98(1.18 – 3.32) 1.68(0.73 – 3.85) Dyslipidemia* Lesion Volume 1.07(0.79 – 1.44) 1.01(0.68 – 1.50) 1.13(0.67 – 1.88) Meat consumption* Lesion Volume 0.86(0.62 – 1.18) 0.81(0.55 – 1.18) 0.80(0.42 – 1.51) Low Vegetable consumption* Lesion Volume 1.00(0.78 – 1.27) 0.95(0.70 – 1.28) 1.21(0.77 – 1.90) Depression* Lesion Volume 1.01(0.64 – 1.58) 1.01(0.58 – 1.75) 0.98(0.42 – 2.28) Downloaded from http://ahajournals.org by on August 8, 2023 Figure S1. Forest plot of the factors associated with ischaemic stroke severity by Stroke Levity Scale Downloaded from http://ahajournals.org by on August 8, 2023 Figure S2. Forest plot of the factors associated with hemorrhagic stroke severity by Stroke Levity Scale Downloaded from http://ahajournals.org by on August 8, 2023 Figure S3. Forest plot of the factors associated with ischemic stroke severity by Stroke Levity Scale Downloaded from http://ahajournals.org by on August 8, 2023 Figure S4. Forest plot of the factors associated with haemorrhagic stroke severity by Stroke Levity Scale Downloaded from http://ahajournals.org by on August 8, 2023