TYPE Original Research PUBLISHED 09 August 2022 DOI 10.3389/fneur.2022.944779 Automated scoring of OPEN ACCESS collaterals, blood pressure, and clinical outcome after EDITED BY Steffen Tiedt, LMU Munich University Hospital, Germany REVIEWED BY endovascular treatment in Youngran Kim, University of Texas Health Science Center at Houston, United States patients with acute ischemic Clara M. Barreira, University of São Paulo, Brazil stroke and large-vessel occlusion *CORRESPONDENCE Pol Camps-Renom pcamps@santpau.cat SPECIALTY SECTION This article was submitted to Daniel Guisado-Alonso1 , Pol Camps-Renom1*, Stroke, a section of the journal Raquel Delgado-Mederos1 , Esther Granell2 , Frontiers in Neurology Luis Prats-Sánchez1 , Alejandro Martínez-Domeño1 , RECEIVED 15 May 2022 ACCEPTED 11 July 2022 Marina Guasch-Jiménez1 , M. Victoria Acosta1 , PUBLISHED 09 August 2022 Anna Ramos-Pachón1 and Joan Martí-Fàbregas1 CITATION 1 Guisado-Alonso D, Camps-Renom P, Stroke Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Biomedical Research Delgado-Mederos R, Granell E, Institute Sant Pau, Universitat Autònoma de Barcelona (Department of Medicine), Barcelona, Spain, 2 Prats-Sánchez L, Martínez-Domeño A, Department of Radiology, UDIAT Corporació Sanitària Parc Taulí, Sabadell, Spain Guasch-Jiménez M, Acosta MV, Ramos-Pachón A and Martí-Fàbregas J (2022) Automated scoring of collaterals, blood pressure, and clinical Introduction: We aimed to determine whether the degree of collateral outcome after endovascular treatment circulation is associated with blood pressure at admission in acute ischemic in patients with acute ischemic stroke stroke patients treated with endovascular treatment and to determine its and large-vessel occlusion. Front. Neurol. 13:944779. prognostic value. doi: 10.3389/fneur.2022.944779 Methods: We evaluated patients with anterior large vessel occlusion treated COPYRIGHT with endovascular treatment in a single-center prospective registry. We © 2022 Guisado-Alonso, Camps-Renom, Delgado-Mederos, collected clinical and radiological data. Automated and validated software Granell, Prats-Sánchez, (Brainomix Ltd., Oxford, UK) was used to generate the collateral score (CS) Martínez-Domeño, Guasch-Jiménez, from the baseline single-phase CT angiography: 0, filling of ≤10% of the Acosta, Ramos-Pachón and Martí-Fàbregas. This is an open-access occluded MCA territory; 1, 11–50%; 2, 51–90%; 3, >90%. When dichotomized, article distributed under the terms of we considered that CS was good (CS = 2–3), or poor (CS = 0–1). We performed the Creative Commons Attribution License (CC BY). The use, distribution bivariate and multivariable ordinal logistic regression analysis to predict CS or reproduction in other forums is categories in our population. The secondary outcome was to determine the permitted, provided the original influence of automated CS on functional outcome at 3 months. We defined author(s) and the copyright owner(s) are credited and that the original favorable functional outcomes as mRS 0–2 at 3 months. publication in this journal is cited, in Results: We included 101 patients with a mean age of 72.1 ± 13.1 years and 57 accordance with accepted academic practice. No use, distribution or (56.4%) of them were women. We classified patients into 4 groups according reproduction is permitted which does to the CS: 7 patients (6.9%) as CS = 0, 15 (14.9%) as CS = 1, 43 (42.6%) as CS not comply with these terms. = 2 and 36 (35.6%) as CS = 3. Admission systolic blood pressure [aOR per 10 mmHg increase 0.79 (95% CI 0.68–0.92)] and higher baseline NIHSS [aOR 0.90 (95% CI, 0.84–0.96)] were associated with a worse CS. The OR of improving 1 point on the 3-month mRS was 1.63 (95% CI, 1.10–2.44) favoring a better CS (p = 0.016). Frontiers in Neurology 01 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 Conclusion: In acute ischemic stroke patients with anterior large vessel occlusion treated with endovascular treatment, admission systolic blood pressure was inversely associated with the automated scoring of CS on baseline CT angiography. Moreover, a good CS was associated with a favorable outcome. KEYWORDS collateral circulation, acute stroke, endovascular treatment, blood pressure, outcome Introduction Materials and methods Endovascular treatment (EVT) is the standard of Study population care for acute ischemic stroke (AIS) in selected patients with large intracranial vessel occlusion (LVO) (1, 2). We conducted a retrospective study of all consecutive AIS However, only around 46% of the patients treated with patients included in our single-center prospective registry from EVT achieve functional independence at 3 months (3). January 2018 to December 2019. Study inclusion criteria were: Therefore, there is still room for improvement in EVT (1) age ≥18 years, (2) intracranial anterior circulation LVO clinical outcomes. (M1, M2, or terminal internal carotid artery (TICA) with or Pre-treatment degree of collateral circulation (CC) has without occlusion of extracranial internal carotid artery) treated been reported as an important determinant for successful with EVT. We excluded patients with either: (1) those with reperfusion (4) and clinical outcome (3) after EVT. The an occluded artery in addition to those mentioned above, (2) effects of the CC are crucial in maintaining perfusion to presence of intracranial hemorrhage on baseline CT, (3) previous penumbral regions and also in facilitating the clearance of anterior circulation territorial infarct, or (4) extreme artifacts on fragmented thrombus (5). In some studies, a higher admission baseline CT or technical issues precluding image analysis. blood pressure (BP) (6) and BP drops during EVT (7) The following variables were recorded for all of the patients: are associated with a poorer clinical outcome after EVT. demographic data: age and sex; vascular risk factors: previous However, there are scarce and contradictory data evaluating stroke, ischemic stroke or transient ischemic attack within the the effect of admission BP on CC in AIS (8, 9). A better week before the index stroke, arterial hypertension, diabetes, understanding of this relationship could lead to an optimisation hypercholesterolemia; drug treatment at admission (statins, of CC by a better management of pre-procedural and intra- antiplatelet, anti-coagulants, antihypertensives, beta-blockers); procedural BP. previous mRS score; clinical data: admission BP, National Multiple scores are available to measure CC in AIS, Institutes of Health Stroke Scale (NIHSS) score at admission; but the intra- and inter-observer agreement for all of blood test parameters: serum urea and creatinine, hemoglobin, them is modest (10, 11). Automated quantitative CC haematocrit, glycaemia, platelet count, haemostasis; logistics scoring in patients with AIS is a reliable, quick, and and metrics: time from stroke onset or last known well to user-independent measure of the CC degree on baseline admission at the Emergency department <6 h, door-to-groin Computed Tomography Angiography (CTA) (12). Although puncture time. Etiologic classification of stroke by the TOAST there is no gold standard, one of the most widely used criteria (15). Admission BP was the first value recorded as part of CC scales on CTA is the one described by Tan et al. (13). routine clinical care on admission to the emergency department For this scale, there is a validated software to get a fully with a non-invasive blood pressure cuff (GE Dinamap R and GE automated collateral score (CS) (14), that provides an Critikon R cuff). The time of admission to BP is the same as the objective quantification that is much more reproducible by time of admission to the emergency department. other researchers. The aim of the current study was to determine the association of admission BP with the degree of Imaging analysis CC using an automated CS and to determine the prognostic value of CC in patients with AIS treated CT and single-phase CTA were performed on a Philips with EVT. Brilliance iCT 256-slice scanner. Intracranial LVO identified Frontiers in Neurology 02 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 in the CTA was recorded. Automated and validated (14, 16) predict CS categories in our population. From an initial model software was used to generate the ASPECTS (Alberta Stroke including all the variables with p < 0.1 in the bivariate analysis, Program Early CT Score) and Collateral Score (CS) (Brainomix we performed a stepwise backward regression modeling to select Ltd., Oxford, UK). Processed images had a maximum slice variables independently associated with CS. The final model was thickness of 1 mm. adjusted for the CTA acquisition phase. The CS described by Tan et al. (13) for the evaluation of CC We performed a shift analysis using the unpaired Kruskal– comprises a score ranging from 0 to 3: 0, no collateral filling; Wallis test to demonstrate differences in the mRS score 1, ≤50% but >0 of the occluded MCA territory; 2, >50% but distribution at 3-months of follow-up between patients with <100% of the occluded MCA; 3, 100% collateral supply to the different CS. In the shift analysis, we calculated the common occluded MCA territory. Because the software gives an exact odds ratio (OR) of worsening by 1 point on the mRS according quantitative value, it is very unlikely to get 0% or 100% accurate. to the CS (unadjusted and adjusted for age) using ordinal logistic Thus, the automatically obtained CS in our study was graded regression. We performed a subgroup analysis considering only as follows: 0, collateral filling of ≤10% of the occluded MCA those patients who reached successful reperfusion according to territory; 1, 11–50%; 2, 51–90%; 3, >90%. When dichotomized, the TICI 2b/2c/3. we considered that CS was good for a CS of 2 or 3, and poor for Finally, to evaluate the functional outcomes, we a CS of 0 or 1 (see Figure 1). The acquisition phase of the CTA dichotomized the study population as presenting favorable is given by the software (early arterial, peak arterial, equilibrium, (mRS = 0–2) or unfavorable functional outcomes (mRS = and peak venous and late venous). The acquisition phase of the 3–6). We performed bivariate analysis and multivariable study may influence the CS, an earlier acquisition phase can ordinal logistic regression analysis as indicated in the previous underestimate the CS. paragraph and multivariable logistic regression analysis The automated analysis was performed after patient care and after dichotomising CS. We considered the final TICI as for research purposes only; it was not used for the selection of an intermediate outcome in the pathway to functional patients to be treated with EVT. outcomes at 3 months, we have excluded it from the final EVT success was measured using the thrombolysis in multivariable models. cerebral infarction (TICI) score. Statistical significance for all the analyses was set at 0.05 (two-sided). All the analyses were performed using Stata v.15 (Texas, USA). Outcomes The primary end-point was to determine whether the Results admission BP values may be related to the degree of CC as classified by the automated CS. The secondary outcome was Study population to determine the influence of automated CS on functional outcome at 3 months. At 90 days, a stroke neurologist evaluated Over a period of two years, we treated 130 AIS patients with the functional outcome with the mRS. We defined favorable EVT in our center. We included 101 patients (see Flowchart in functional outcomes as mRS 0–2 at 3 months for patients with Supplementary Figure I) with a mean age of 72.1 (SD = 13.1), 57 a previous mRS score of 0–2. For patients with a previous mRS (56.4%) of them were women, and 98 (97%) had a previous mRS score of 3, the outcome was favorable when the score at discharge = 0–2. We treated 41 patients (40.6%) with recombinant tissue- was 3. type plasminogen activator (r-tPA). We found no significant differences in the proportion of patients treated with r-tPA between CS subgroups (p = 0.081). Characteristics of the Statistical analysis population are detailed in Table 1. Continuous variables were reported as means and standard deviations or medians and interquartile range (IQR) if they Primary outcome were not-normally distributed, as tested by the Shapiro–Wilk normality test. Categorical variables were expressed as counts We classified patients into groups based on their category in and percentages. the CS: 7 patients (6.9%) were classified as CS = 0, 15 (14.9%) We classified the study population according to the four as CS = 1, 43 (42.6%) as CS = 2 and 36 (35.6%) as CS = 3. different categories of CS. Bivariate analyses were performed? As shown in Table 1, we found no significant differences in age, between these groups using ANOVA or the Kruskal–Wallis sex, or stroke etiology among the four groups. In the bivariate test (when a non-parametric test was required) for continuous analysis, we found that patients with a higher CS score had a variables, and the χ 2 test for categorical variables. Thereafter, we lower baseline NIHSS score, were taking fewer antihypertensive performed a multivariable ordinal logistic regression analysis to drugs prior to admission, had lower serum creatinine levels, and Frontiers in Neurology 03 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 FIGURE 1 Selected images of our patients processed with e-Stroke (Brainomix Ltd.) are shown as illustrative examples for each CS: 0, collateral filling of ≤10% of the occluded MCA territory; 1, 11–50%; 2, 51–90%; 3, >90%. The first row shows the number of patients in each group in our study. The orange area indicates the area without collateral circulation, the red circle indicates the location of the arterial occlusion. had lower admission systolic and mean BP. Patients classified 3 group (p = 0.024) (see Figure 2). The OR of improving 1 point as CS = 0 had a non-significant (p = 0.263) lower admission on the 3-month mRS was 1.63 (95% CI, 1.10–2.44) favoring a systolic and median BP than the CS = 1 group. Among patients better CS (p = 0.016). After adjusting by age, the common odds with a CS = 0, there was a higher percentage of TICA occlusion. ratio was 1.58 (95% CI, 1.05–2.39, p = 0.027). When considering Because mean BP is calculated from systolic BP and diastolic only patients with successful recanalization (TICI 2b/2c/3) the BP, we conducted two multivariable ordinal logistic regression CS category was also associated with the probability of achieving analyses, model 1 with systolic BP and model 2 with mean a better mRS score at three months with a common OR of 1.59 BP (see details in Table 2). In model 1, higher systolic BP (95% CI, 1.01–2.49, p = 0.044). CS was no longer associated with [aOR for 10 mmHg increase 0.79 (95% CI 0.68–0.92)] and mRS score when adjusted by baseline NIHSS. higher baseline NIHSS [aOR 0.90 (95% CI, 0.84–0.96)] were At 3 months of follow-up 57 (57.6%) patients presented a independent variables associated with a lower CS. Similarly, in favorable outcome. A better CS category was related to a better model 2, higher mean BP [aOR for 10 mmHg increase 0.70 (95% clinical outcome at 3 months in the bivariate analysis (p = CI 0.54–0.90)] and higher baseline NIHSS [aOR 0.90 (95% CI, 0.026). However, the significance was lost in the multivariable 0.83–0.95)] were associated with a lower category of CS. logistic regression analysis (Models 1 and 2). Details of bivariate analyses are shown in Supplementary Table I and of the multivariable logistic regression analyses are shown in Table 3. Secondary outcome Discussion The shift analysis showed a median of mRS at 3 months of 4 (IQR 2–6) in the CS = 0 group, 3 (IQR 2–5) in the CS = 1 group, In patients with AIS secondary to LVO who were treated 2 (IQR 0–3) in the CS = 2 group, and 2 (IQR 1–3) in the CS = with EVT, we found that a higher BP at admission was associated Frontiers in Neurology 04 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 TABLE 1 Bivariate analyses of predictors of collaterals classified by CS categories. All CS = 0 CS = 1 CS = 2 CS = 3 p (n = 101) (n = 7) (n = 15) (n = 43) (n = 36) Age, mean (SD) 72.1 (13.1) 74.9 (8.4) 77.7 (10.6) 71.8 (13.3) 69.5 (14.2) 0.217 Sex (woman), n (%) 57 (56.4) 4 (57.1) 7 (46.7) 24 (55.8) 22 (61.1) 0.823 Baseline NIHSS, median (IQR) 16 (9–20) 20 (17–22) 21 (18–23) 13 (8 (20) 11 (9–19) <0.001 Stroke onset–to–door <6h, n (%) 76 (75.25) 7 (100) 12 (80) 33 (76.7) 24 (66.7) 0.266 Door–to–groin puncture time, median (IQR) 62 (55–76) 51 (48–59) 62 (55–76) 65 (57–79) 62 (52–78) 0.108 ASPECTS, median (IQR) 9 (8–10) 8 (8–9) 8 (7–10) 9 (9–10) 9 (9–10) 0.098 M1 occlusion, n (%) 52 (51.5) 4 (57.1) 10 (66.7) 22 (51.1) 16 (44.4) 0.534 M2 occlusion, n (%) 28 (27.7) 0 (0.0) 3 (20.0) 15 (34.9) 10 (27.8) 0.237 TICA occlusion, n (%) 12 (11.9) 3 (42.9) 1 (6.7) 2 (4.7) 6 (16.7) 0.021 Tandem occlusion, n (%) 16 (15.9) 1 (14.3) 3 (20.0) 6 (14.0) 6 (16.7) 0.952 Cardioembolic, n (%) 49 (48.5) 4 (57.1) 7 (46.7) 22 (51.1) 16 (44.4) 0.899 Large–artery atheromatosis, n (%) 13 (12.9) 0 (0.0) 3 (20.0) 5 (11.6) 5 (13.9) 0.614 Hypertension*, n (%) 70 (70.7) 7 (100) 9 (60.0) 32 (76.2) 22 (62.9) 0.146 Diabetes**, n (%) 19 (19.3) 2 (28.6) 4 (26.7) 6 (14.7) 7 (20.0) 0.685 Dyslipidemia**, n (%) 47 (48.0) 5 (71.4) 7 (50.0) 22 (52.3) 13 (37.1) 0.316 Statins*, n (%) No statin 68 (68.7) 4 (57.1) 8 (53.3) 30 (71.4) 26 (74.3) 0.546 Low intensity 4 (4.0) 1 (14.3) 0 (0.0) 2 (4.8) 1 (2.9) Moderate intensity 18 (18.2) 1 (14.3) 5 (33.3) 8 (19.1) 4 (11.4) High intensity 9 (9.1) 1 (14.3) 2 (13.3) 2 (4.8) 4 (11.4) Antihypertensive*, n (%) No drugs 37 (37.4) 0 (0.0) 6 (40.0) 11 (26.2) 20 (57.1) 0.037 1 drug 15 (15.1) 2 (28.6) 2 (13.3) 6 (14.3) 5 (14.3) 2 or more drugs 47 (47.5) 5 (71.4) 7 (46.7) 25 (59.5) 10 (25.6) Previous stroke*, n (%) 6 (6.1) 1 (14.3) 0 (0.0) 1 (2.4) 4 (11.43) 0.206 Stroke or TIA previous week*, n (%) 8 (8.1) 0 (0.0) 0 (0.0) 4 (9.52) 4 (11.43) 0.461 Hemoglobin*, median (IQR) g/L 134 (121–144) 138 (124–153) 136 (128–140) 133 (120–147) 130 (115–141) 0.438 Hematocrit***, median (IQR) L/L 0.4 (0.37–0.42) 0.4 (0.38–0.46) 0.4 (0.39–0.41) 0.4 (0.36–0.43) 0.4 (0.35–0.42) 0.545 Glycaemia****, median (IQR) mg/100mL 121 (103–146) 145 (132–161) 125 (106–153) 115 (102–134) 120 (99–131) 0.1361 Creatinine*, median (IQR) micromol 77 (67–91) 77 (75–112) 83 (60–108) 84 (72–96) 70 (63–82) 0.048 Admission systolic blood pressure, mean (SD) 155 (26) 154 (30) 169 (29) 160 (25) 143 (22) 0.003 Admission mean blood pressure, mean (SD) 105 (15) 105 (12) 111 (17) 109 (14) 99 (14) 0.020 Admission diastolic blood pressure, mean (SD) 81 (13) 81 (5) 81 (13) 83 (12) 78 (13) 0.252 Missing information for *2, ** 3, *** 4 and **** 24 patients. CS (Collateral Score: 0, collateral filling of ≤10% of the occluded middle cerebral artery (MCA) territory; 1, 11–50%; 2, 51–90%; 3, >90%), NIHSS (National Institutes of Health Stroke Scale), Stroke onset–to–door < 6h (time from stroke onset or last know well to admission at the Emergency department < 6h), Door–to–groin puncture time is expressed in minutes, ASPECTS (Alberta Stroke Program Early CT Score), TICA (Terminal Internal Carotid Artery), TIA (Transient Ischemic Attack), U/C ratio (Urea mmol /Creatinine mmol ratio). with poorer automated CS on baseline CTA. In addition, a better that moderately elevated systolic BP was associated with good automated CS was associated with a better functional outcome at CC but with an inverse association of systolic BP and 3-month 3 months of follow-up. clinical outcome (8); another study found that lower systolic BP Prior studies reported that higher admission BP is associated was associated with good CC but in patients with ICA occlusion with a higher final infarct volume, a lower probability of during the first 3 days from stroke onset (9). In this regard, our successful reperfusion, and with a lower likelihood of favorable study adds relevant information confirming that both systolic clinical outcomes in AIS patients treated with EVT (6, 17). BP and mean BP at admission are inversely associated with However, the relationship between baseline BP and CC has the CS. been barely described. One study of AIS patients treated with Pathophysiological knowledge suggests that the direction intravenous thrombolytic therapy (with or without LVO) found of this independent relationship is that poor CC leads to Frontiers in Neurology 05 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 increased BP. LVO causes a drop in perfusion pressure, which direction and amount of blood flow through CC rely on leads to recruitment and vasodilatation of the leptomeningeal systemic perfusion, which can be estimated by BP (18). Until collaterals and, due to impaired cerebral autoregulation, the successful reperfusion, the survival of hypoxemic brain tissue depends on CC flow, and for this reason guidelines do not recommend lowering BP unless it is extremely high (19). In TABLE 2 Results of the multivariable ordinal logistic regression analysis of predictors of good collaterals classified by collateral score (CS) categories. aOR 95% CI p TABLE 3 Results of the multivariable logistic regression analysis of predictors of good functional outcome at 90 days of follow–up. Model 1 OR 95% CI p Systolic blood pressure 0.98 0.96–0.99 0.003 (x1 mmHg increase) 0.79 0.68–0.92 Model 1 (x10 mmHg increase) Collateral Score (CS) 1.39 0.80–2.42 0.240 Baseline NIHSS 0.90 0.84–0.96 0.002 Baseline NIHSS 0.90 0.83–0.97 0.009 Model 2 Previous mRS 0.50 0.30–0.80 0.005 Mean blood pressure (x1 0.96 0.94–0.99 0.006 Model 2 mmHg increase) 0.70 0.54–0.90 Good vs. poor Collaterals 2.64 0.78–9.00– 0.120 (x10 mmHg increase) Baseline NIHSS 091 0.83–0.99 0.021 Baseline NIHSS 0.90 0.83–0.95 0.001 Previous mRS 0.50 0.31–0.81 0.005 NIHSS (National Institutes of Health Stroke Scale). Poor Functional Outcome was defined as mRS = 3–6, mRS (modified Rankin Scale), Adjusted for Computed Tomography Angiography acquisition phase. OR for worsening Good Collaterals included CS = 2–3, Poor Collaterals included CS = 0–1, TICI score in one CS category. (thrombolysis in cerebral infarction score). FIGURE 2 Distribution of the mRS score at 3 months according to the baseline automated determined Collateral Score (CTA-CS). Frontiers in Neurology 06 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 this regard, in the bivariate analysis we found a relationship Data availability statement between a history of taking a greater number of antihypertensive drugs before admission and a worse CS in baseline CTA. The data that support the findings of this study are available It should be noted that in the CS = 0 subgroup (which from the corresponding author upon reasonable request. had a lower BP than the others although not significant probably because of its small size) all patients were taking Ethics statement antihypertensive drugs. We also found an independent association in the The Ethics Committee of Hospital de la Santa Creu i Sant multivariable analysis between more severe strokes (as Pau reviewed and approved this study (IIBSP-COL-2019-120). defined by a higher NIHSS score) and worse CS. Probably the Written informed consent for participation was not required for direction of this relationship is probably poorer CS implying this study in accordance with the national legislation and the lower blood flow to the penumbra area (5), a larger ischemic institutional requirements. core (20), and thus greater clinical severity. Our study confirms that a better CC status is associated with a better clinical outcome following EVT. A neuroimaging Author contributions meta-analysis of pivotal clinical trials of EVT, reported that a better CC status was associated with a better NIHSS within the DG-A and JM-F designed and conceptualized the study. first 24 h and a better clinical outcomes at 3 months (21). Our DG-A and PC-R major role in the acquisition of data. DG-A, study confirms that CS is an independent prognostic factor after JM-F, PC-R, and RD-M analyzed and interpreted the data. DG- adjusting by age and when considering only those achieving a A wrote the first draft of the manuscript. DG-A, PC-R, RD- successful recanalization. When adjusting for NIHSS, CS lost M, EG, AM-D, LP-S, MG-J, MA, AR-P, and JM-F reviewed significance as a prognostic factor, but we believe this is due to and edited the manuscript and approved the final version of the relationship we have already discussed: the worse the CC, the the manuscript. higher the NIHSS. Assessment of CS can be time-consuming, subjective, and Funding the challenge for the manual scoring system is the low agreement between evaluators, making patient characterization This work was supported by Redes de Investigación Con less consistent. Several studies have shown that fully automated Objetivos de Resultados en Salud (RICORS) RD21/0006/0006, or semi-automated methods are a quantitative, reliable, time- Instituto de Salud Carlos III, Ministry of Science and Innovation saving, and user-independent measure of CS on baseline CTA (Government of Spain). DG-A (CM18/00065) and MG-J (12, 14, 22). Semiautomated CS evaluation has been associated (CM20/00056) received a Río Hortega Research grant from the with clinical outcome previously in AIS patients treated with Instituto de Salud Carlos III, Ministry of Science and Innovation EVT (22), however, to the best of our knowledge the relationship (Government of Spain). Collateral Circulation (collaterome) in between admission BP and automatically measured CS had not Acute Ischemic Stroke with Large Vessel Occlusion: A Study of been previously described. Clinical, Radiological, Plasma and Genetic Factors, Proyecto de Our study has some limitations. Patients treated with EVT investigación en salut (PI19/00859), Instituto de Salud Carlos III, may have a selection bias by causing the number of patients Ministry of Science and Innovation (Government of Spain). This with good CS to be higher in this study than in the general study has been co-funded by the European Union through the population. Collateral circulation assessed by using perfusion Fondo Europeo de Desarrollo Regional (FEDER). CT or multiphase CTA was significantly better than those assessed by CTA, however, the results obtained by CTA are much more reproducible due to their availability. Neither BP nor CC Acknowledgments are static, and in this study, we have only one measurement of We thank Jaume Llopis for his wise advice and for his each variable, without being able to analyse what happens in the statistics website: https://jllopisperez.com. rest of the acute process. In conclusion, in patients with an AIS due to anterior LVO, admission BP was independently and inversely associated with Conflict of interest automated CS on baseline CTA. Moreover, automated CS was associated with clinical outcomes at 3 months. Prospective data Brainomix has provided its software free of charge to the and clinical trials to determine the best management of BP Stroke Unit of the Hospital de la Santa Creu i Sant Pau, in and other modifiable factors to maintain optimal CC peri-EVT accordance with the regulations of the Biomedical Research are needed. Institute Sant Pau and for research purposes only. Frontiers in Neurology 07 frontiersin.org Guisado-Alonso et al. 10.3389/fneur.2022.944779 The authors declare that the research was conducted in the editors and the reviewers. Any product that may be the absence of any commercial or financial relationships evaluated in this article, or claim that may be made that could be construed as a potential conflict by its manufacturer, is not guaranteed or endorsed by of interest. the publisher. Publisher’s note Supplementary material All claims expressed in this article are solely those The Supplementary Material for this article can be of the authors and do not necessarily represent those found online at: https://www.frontiersin.org/articles/10.3389/ of their affiliated organizations, or those of the publisher, fneur.2022.944779/full#supplementary-material References 1. Turc G, Bhogal P, Fischer U, Khatri P, Lobotesis K, Mazighi M, et al. European 12. Boers AMM, Sales Barros R, Jansen IGH, Berkhemer OA, Beenen LFM, Stroke Organisation (ESO) - European Society for Minimally Invasive Neurological Menon BK, et al. Value of quantitative collateral scoring on CT angiography in Therapy (ESMINT) guidelines on mechanical thrombectomy in acute ischemic patients with acute ischemic stroke. AJNR Am J Neuroradiol. (2018) 39:1074– stroke. J Neurointerv Surg. (2019) 6:535–8. doi: 10.1136/neurintsurg-2018-014568 82. doi: 10.3174/ajnr.A5623 2. Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, 13. Tan IYL, Demchuk AM, Hopyan J, Zhang L, Gladstone D, Wong K, et al. Becker K, et al. 2018 guidelines for the early management of patients CT angiography clot burden score and collateral score: correlation with clinical with acute ischemic stroke: a guideline for healthcare professionals from the and radiologic outcomes in acute middle cerebral artery infarct. Am J Neuroradiol. American Heart Association/American Stroke Association. Stroke. (2018) 49:46– (2009) 30:525–31. doi: 10.3174/ajnr.A1408 110. doi: 10.1161/STR.0000000000000158 14. Grunwald IQ, Kulikovski J, Reith W, Gerry S, Namias R, Politi M, et al. 3. Goyal M, Menon BK, Van Zwam WH, Dippel DWJ, Mitchell PJ, Demchuk Collateral automation for triage in stroke: evaluating automated scoring of AM, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: A meta- collaterals in acute stroke on computed tomography scans. Cerebrovasc Dis. (2019) analysis of individual patient data from five randomised trials. Lancet. (2016) 47:217–22. doi: 10.1159/000500076 387:1723–31. doi: 10.1016/S0140-6736(16)00163-X 15. Adams HPJ, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, 4. Leng X, Fang H, Leung TWH, Mao C, Xu Y, Miao Z, et al. et al. Classification of subtype of acute ischemic stroke. definitions for use in a Impact of collateral status on successful revascularization in endovascular multicenter clinical trial. TOAST. Trial of Org 10172 in acute stroke treatment. treatment: A systematic review and meta-analysis. Cerebrovasc Dis. (2016) 41:27– Stroke. (1993) 24:35–41. doi: 10.1161/01.STR.24.1.35 34. doi: 10.1159/000441803 16. Brinjikji W, Abbasi M, Arnold C, Benson JC, Braksick SA, Campeau 5. Liebeskind DS. Collateral circulation. Stroke. (2003) 34:2279– N, et al. e-ASPECTS software improves interobserver agreement and 84. doi: 10.1161/01.STR.0000086465.41263.06 accuracy of interpretation of aspects score. Interv Neuroradiol. (2021) 27:781–7. doi: 10.1177/15910199211011861 6. Van Den Berg SA, Uniken Venema SM, Mulder MJHL, Treurniet KM, Samuels N, Lingsma HF, et al. Admission blood pressure in relation to clinical 17. Goyal N, Tsivgoulis G, Iftikhar S, Khorchid Y, Fawad Ishfaq M, Doss VT, outcomes and successful reperfusion after endovascular stroke treatment. Stroke. et al. Admission systolic blood pressure and outcomes in large vessel occlusion (2020) 51:3205–14. strokes treated with endovascular treatment. J Neurointerv Surg. (2017) 9:451– 4. doi: 10.1136/neurintsurg-2016-012386 7. Maïer B, Fahed R, Khoury N, Guenego A, Labreuche J, Taylor G, et al. Association of blood pressure during thrombectomy for acute ischemic 18. Kim BJ, Singh N, Menon BK. Hemodynamics of leptomeningeal collaterals stroke with functional outcome a systematic review. Stroke. (2019) 50:2805– after large vessel occlusion and blood pressure management with endovascular 12. doi: 10.1161/STROKEAHA.119.024915 treatment. J Stroke. (2021) 23:343–57. doi: 10.5853/jos.2021.02446 8. Rusanen H, Saarinen JT, Sillanpää N. The association of blood 19. Sandset EC, Anderson CS, Bath PM, Christensen H, Fischer U, Gasecki pressure and collateral circulation in hyperacute ischemic stroke D, et al. European Stroke Organisation (ESO) guidelines on blood pressure patients treated with intravenous thrombolysis. Cerebrovasc Dis. (2015) management in acute ischaemic stroke and intracerebral haemorrhage. Eur Stroke 39:130–7. doi: 10.1159/000371339 J. (2021) 6:XLVIII–LXXXIX. doi: 10.1177/23969873211012133 9. Wufuer A, Mijiti P, Abudusalamu R, Dengfeng H, Jian C, Jianhua M, et al. 20. Lin L, Yang J, Chen C, Tian H, Bivard A, Spratt NJ, et al. Association of Blood pressure and collateral circulation in acute ischemic stroke. Herz. (2019) collateral status and ischemic core growth in patients with acute ischemic stroke. 44:455–9. doi: 10.1007/s00059-018-4691-5 Neurology. (2021) 12:e161–70. doi: 10.1212/WNL.0000000000011258 10. Seker F, Potreck A, Möhlenbruch M, Bendszus M, Pham M. Comparison of 21. Roman LS, Menon BK. Blasco J, Hernandez-Perez M, Davalos A, Majoie four different collateral scores in acute ischemic stroke by CT angiography. CBLM, et al. Imaging features and safety and efficacy of endovascular stroke J Neurointerv Surg. (2016) 8:1116–8. doi: 10.1136/neurintsurg-2015- treatment: a meta-analysis of individual patient-level data. Lancet Neurol. 012101 (2018) 17:895–904. 11. Liu L, Ding J, Leng X, Pu Y, Huang LA, Xu A, et al. Guidelines 22. Cao R, Qi P, Jiang Y, Hu S, Ye G, Zhu Y, et al. Preliminary application for evaluation and management of cerebral collateral circulation in ischaemic of a quantitative collateral assessment method in acute ischemic stroke patients stroke 2017. Stroke Vasc Neurol. (2018) 3:117–30. doi: 10.1136/svn-2017- with endovascular treatments: a single-center study. Front Neurol. (2021) 12:1– 000135 9. doi: 10.3389/fneur.2021.714313 Frontiers in Neurology 08 frontiersin.org