2025+共识声明：大血管闭塞伴轻度神经功能障碍患者血管内取栓决策

Introduction

Acute stroke patients with mild deficits(baseline National In-stitutes of Health Stroke Scale[NIHSS] of 0-5) but proven large vessel occlusion(LVO) present a clinical treatment dilemma.The evidence on benefit from endovascular thrombectomy(EVT) is very limited; the Multicenter Randomized Clinical Trial of En-dovascular Treatment for Acute Ischemic Stroke in the Nether-lands(MR CLEAN) and Extending the Time for Thrombolysis in Emergency Neurological Deficits-Intra-Arterial(EXTEND-IA)trials included a total of 14 patients with NIHSS scores of 0-5.12 No clear guideline recommendations for EVT in patients with low NIHSS exist, and EVT decisions are made on a case-by-case basis, with high variability among practitioners and institutions.3 Two randomized trials are currently ongoing to investigate the effect of EVT in anterior circulation LVO stroke with low NIHSS:Endovascular Therapy for Low NIHSS Ischemic Strokes(EN-DOLOW, NCT04167527) and Minor Stroke Therapy Evaluation(MOSTE, NCT03796468).

With up to 30% of stroke patients with LVO reported to pres-ent with an NIHSS<6,4 this minor stroke subset of patients with LVO is not small. It is estimated that 30%-40% will remain dis-abled at 90 days,5,6 and this proportion doubles in the 20% of

patients in whom early neurological deterioration (END) occurs. 7 Observational data on EVT for minor stroke remain inconclusive,with reports ranging from clinical benefit to harm, including el-evated rates of symptomatic intracranial hemorrhage.8-11 The risk/benefit ratio seems to differ based on the risk of END, the occlusion location, and size of the ischemic territory at risk. Cur-rently, no randomized trials are specifically testing the safety and efficacy of EVT in stroke due to vertebrobasilar occlusion and medium vessel occlusion (MeVO) with low NIHSS, although sev-eral ongoing and recently completed EVT trials for stroke due to MeVO do include patients with low NIHSS.12

Even when the results of these trials are available, the ques-tion of EVT benefit in the minor stroke(low NIHSS) population remains complex. Heterogeneity in outcome may be influenced by within-NIHSS score variation: a total NIHSS score of 3 can be caused by severe aphasia as well as facial droop and loss of sensation-but the former will be more likely to lead to a much higher disability burden than the latter. Heterogeneity may be further influenced by the modified Rankin Scale(mRS) score it-self, which is traditionally used as the primary outcome in EVT trials, but may not be sensitive enough to detect small but mean-ingful differences in long-term post-EVT outcomes in minor stroke patients.The individual impact of any neurological loss of func-

JoS

Boshart et al.EVT in LVO Strokes With Low NIHSS

tion will differ from patient to patient-for example, hemianopia in a truck driver is much more detrimental compared to hemi-anopia in a retiree whose vision was impaired to begin with.13 The inflection point in the risk/benefit ratio may vary according to the arterial lesion location: patients with a proximal occlu-sion and a large territory at risk may have more potential bene-fit with revascularization. In contrast, the predicted benefit may be smaller in the setting of an MeVO, where there will be a smaller territory at risk, or when extensive ischemic changes are already present at baseline. Even a slight increase in complica-tion rates could tip the balance against EVT.

A simple total NIHSS score threshold for EVT decision-making does not take the anticipated complexity of the endovascular procedure into account. EVT decisions in this setting are there-fore more complex and require a more nuanced approach as compared to major stroke due to LVO with more severe deficits.

During the 2024 5T (Teamwork,Training,Technology,Tech-nique,Transport) Think Tank conference, this topic was discussed following a live modified Delphi consensus approach. In prepa-ration for the live session, a systematic literature search had been performed. We summarized the results of the literature search and synthesized the expert opinions of the 31 participat-ing neurointerventionalists, neurologists, and neurosurgeons on the topic of EVT for low NIHSS LVO stroke.

Methods

5T Think Tank attendees and sponsoring

This Delphi consensus was conducted as part of the 5T Think Tank that was held in September 2024 in Banff,Alberta,Canada.The 5T Think Tank is an annual 3-day in-person meeting of in-

ternational experts in the field of neurology and neurointerven-tion(neurologists, radiologists, and neurosurgeons). Every year,the latest research and future developments in the field of in-terventional stroke therapy are discussed. Focus is laid on areas with lacking or unclear evidence. The meeting is by invitation only.The selection of 5T attendees focused on representing a di-verse spectrum of specialists regarding geography, sub-special-ties, and gender.The 5T Think Tank is sponsored by Medtronic,Stryker,Terumo,Cerenovus,and Penumbra in equal parts.How-ever, the industry did not control the guest list, the choice of topics, or the content of the presentations. The preparation and analysis of the Delphi rounds were conducted by 5T Think Tank attendees unaffiliated with and independent from the sponsors.

Delphi methodology

The Delphi method is a structured, iterative forecasting technique incorporating feedback loops.14 In the medical field, it has been used to develop interim treatment guidelines and standardize patient care in the absence of strong evidence, serving as a pro-viousional approach until sufficient data supports evidence-based practices.15 The Delphi method relies on a panel of individuals with content expertise that answers a series of questions with controlled opinion feedback, the goal being to assess expert opinions and achieve consensus among the respondents. In this study, the Delphi technique was applied to identify aspects in-fluencing decision-making to undergo EVT in patients with low NIHSS strokes due to LVO or MeVO. Figure 1 shows the key el-ements of the Delphi methodology that were used in this study.

Literature search

In preparation for the Delphi consensus, a scoping literature

 Figure 1. Delphi methodology and processes used for this manuscript. Arrows indicate the chronological order of the process elements. Light orange boxes represent pre-5T meeting activities, medium orange boxes denote in-person meeting steps, and dark orange boxes indicate post-meeting tasks.5T,Teamwork,Training,Technology,Technique,Transport.

Vol.27/ No.3/ September 2025

JoS

 Figure 2. Flowchart of the scoping literature search in preparation of the meeting. LVO, large vessel occlusion; NIHSS, National Institutes of Health Stroke Scale;5T,Teamwork,Training,Technology,Technique,Transport.

search was performed to identify key publications on EVT in LVO and MeVO in the anterior circulation with low NIHSS(see search strategy in Figure 2). Neurointerventional treatment of posteri-or circulation occlusions comes with different challenges com-pared to the anterior circulation, which can significantly affect how treatment decisions are made. We chose to exclude pos-terior circulation occlusions to keep the questions simpler and get clearer answers during the Delphi process. The purpose of this literature search was to identify high-quality prospective and retrospective cohort studies and ongoing trials focused on EVT for LVO or MeVO with low NIHSS. One of the coauthors(MK) summarized and presented the results of this search dur-ing the live session. A list of the publications selected for pre-sentation during the meeting can be found in Supplementary Table 1.

Round 1

After a scoping review of literature,5T attendees were asked the following five open-ended questions via email prior to the in-person meeting, inquiring about decision-making on EVT treat-ment and suitable outcome measures in patients with low NI-HSS (0-5) due to LVO or MeVO stroke:

(1) Would you, and if yes, how would you account for differ-ent levels of disability within one NIHSS score(e.g., NIHSS 3 aphasia vs. NIHSS 3 sensory+facial movement) when considering EVT?

(2) Would you, and if yes, how would you account for differ-ent levels of disability impact based on individual patient factors/professions(e.g.,hemianopia in a truck driver, arm

<EndOfPara>drift in a professional musician) when considering EVT?

(3) How would you identify low NIHSS patients with a high risk of early neurological deterioration without EVT?

(4) What are your thoughts on the risk-benefit balance of EVT in low NIHSS patients?

(5) Which outcome measure(s) would be the most suitable for assessing intravenous thrombolysis(IVT)/EVT benefit in patients with baseline NIHSS 0-5?

These questions were intentionally broad, and phrased as open-ended questions, to encourage discussion and brainstorming,and ensure that all relevant points for future rounds are cap-tured. All questions referred to stroke patients with a low NIHSS(0-5) due to LVO or MeVO.

Rounds 2 and 3

Rounds 2 and 3 consisted of 15 closed-ended questions that were drafted based on the round 1 open-ended questions.They aimed to examine the group's attitudes towards EVT decision-making in low NIHSS MeVO and LVO stroke in more concrete,practical terms. The questions in rounds 2 and 3 were identical.Round 2 questions were sent out to participants via email prior to the person-meeting. Prior to round 3, the results of round 2 were presented and discussed at the meeting. All original questions from rounds 2 and 3 can be found in Supplementary Materials.

Data collection and analysis

An online survey tool(Qualtrics.com) was used for data collec-tion. Respondent's email addresses were recorded to prevent duplicate responses. The first two rounds were conducted prior to the meeting.

During the 5T Think Tank in-person meeting, the current evi-dence on the topic was presented as well as a summary of the answers from round 1. Thereafter followed a 90-minute discus-sion. The second round was closed after this session, and the answers of round 2 were summarized and distributed to all par-ticipants. All participants then reviewed the results of the sec-ond round in detail before answering the questions of round 3.

Frequent reminders were sent out shortly after the meeting to achieve the same response rate in all rounds. Thereafter, the database was locked, and data analysis was performed by two think tank attendees(SLB and AS), who did not participate in the consensus themselves.

Consensus in multiple-choice questions was defined based on the type of response allowed. For questions permitting mul-tiple answers, consensus required at least 70% agreement among participants. When only a single answer could be selected, a min-imum of 50% agreement on one option was considered suffi-

JoS

Bosshart et al.EVT in LVO Strokes With Low NIHSS

cient. For binary questions, such as yes/no, consensus was de-fined as at least 70% agreement. This approach is similar to the previous Delphi consensus in the field of neurointervention. ${}^{16-18}$

No approval by an institutional review board was necessary for this study. By participating in the survey, respondents have agreed to their data being analyzed in anonymized form.

Results

Response data were collected from September 18 to October 9,2024. Of the 40 5T Think Tank attendees with expertise in stroke medicine,31(78%) responded to rounds 1 and 2, and 29(73%)responded to round 3. A list of the Delphi participants can be found in the Supplementary Material. Participants had the op-tion to skip questions if they felt the topic was outside their area of expertise. As a result, the number of responses varies across questions.

Delphi results round 1

Question 1:Would you, and if yes, how would you account for different levels of disability within one NIHSS score(e.g., NIHSS 3 aphasia vs. NIHSS 3 sensory+facial movement) when consid-ering EVT?

All respondents indicated that they use no fixed NIHSS cut-off in clinical practice to decide whether a patient should receive EVT. In patients with low NIHSS, debilitating symptoms lead to more aggressive management, while less severely disabling symp-toms may be managed conservatively. There was agreement that not all NIHSS categories represent the same level of disability per NIHSS point. Categories such as vision loss and aphasia were considered more disabling deficits, while facial droop and sen-sory deficits were generally considered less disabling and, thus,less likely to trigger a decision in favor of EVT.

Question 2:Would you, and if yes, how would you account for different levels of disability impact based on individual patient factors/professions when considering EVT?

While many respondents stated that they include occupation,age, baseline functional status, quality of life(QoL), and level of independence as factors in their EVT decision-making, oth-ers were critical of this individualized approach: they felt that this approach is often based upon assumptions, as history-tak-ing is typically crude in a hyper-acute setting. Therefore, many times this level of patient history is not readily available. It was also noted that the inherent desire to help a patient when we have more specific details about how a stroke will affect them can also bias decision-making towards overly aggressive or con-

servative treatment decisions.19 Some felt that individual pa-tient factors should only be accounted for after discussing with the patient(or a proxy) how they would manage and adapt if they had to continue living with the present disability in their daily life. Other respondents perceived EVT to be much safer today than in the past, so they tend to offer it to most patients if technically feasible.Finally, some participants stated they treat all patients with LVO who are still working due to the risk of END.

Question 3:How would you identify low NIHSS patients with a high risk of early neurological deterioration without EVT?

Respondents listed the following indicators for END: worsen-ing upon orthostatic challenge, large perfusion deficit on imag-ing that does not match the clinical symptoms, low collateral flow reserve, fluctuating NIHSS score/symptoms, long clot length,proximal occlusion of a major intracranial vessel, potential em-bolic stroke origin/multiple artery occlusions, incomplete circle of Willis, high blood pressure, diabetes, and elderly patients. Five(16%) respondents felt that any low NIHSS patient with LVO was at risk of secondary neurological deterioration, and as this often leads to significant disability, all low NIHSS individuals with LVO should be treated. One of these respondents mentioned they treat all LVOs that are not determined to be from intracranial atherosclerotic disease(ICAD) because ICAD markedly increases the risk of the intervention.

Question 4: What are your thoughts on the risk-benefit bal-ance of EVT in low NIHSS patients?

The responses again reflected a wide range of perspectives,highlighting the complexity and nuance of decision-making in this subgroup. Factors felt to decrease the benefit of EVT in low NIHSS stroke patients with LVOs were ICAD, distal location of ar-terial vessel occlusion, simultaneous treatment with intravenous tenetepclase or alteplase, and low degree of disability. Factors felt to increase the benefit of EVT in low NIHSS stroke patients were young patient age, aphasia as a symptom, an experienced interventionalist performing the EVT, and any indicators of a high risk of secondary neurological deterioration or if the latter could not be reliably detected due to, for example, organizational struc-tures of the hospital. Two respondents(6%) felt that there was nearly always a benefit of EVT.

Question 5: Which outcome measure(s) would be the most suitable for assessing IV/T/EVT benefit in patients with baseline NIHSS 0-5?

Some respondents felt that the mRS traditionally used in stroke trials in ordinal or dichotomized form was also appropriate for low NIHSS patients. However, the majority of experts suggested

Vol.27/ No.3/ September 2025

JoS

other scales that could be combined with the mRS if necessary,such as a QoL scale(e.g., EuroQol 5-Dimension 5-Level[EQ-5D-5L]) or early neurological improvement on the NIHSS scale.

Delphi results rounds 2 and 3

Detailed results of rounds 2 and 3 are presented in Table 1. There was consensus that a hard NIHSS cut-off alone is not a good basis for deciding on EVT treatment for low NIHSS patients. Nine-ty-six percent of respondents felt that EVT decision should rather be based on the degree of disability that the symptoms cause.Ninety percent thought that the degree of disability should al-ways be discussed with the patient and/or family despite the ad-ditional time lost for discussion. When deciding on the thresh-old of disability to qualify for EVT, 18/30(60%) in round 2 and 19/29(66%) in round 3 favored any decrease in baseline func-tional status that prevents the patient from exerting their main daily activity, which is favorably determined by the patient.

Disabling symptoms

The expert panel agreed that aphasia and vision loss were the symptoms on the NIHSS scale with the largest impact on QoL and the functional status of a patient.

There was consensus(16/29[55%] strongly agree, 11/29[38%]somewhat agree) that these symptoms should have more weight on EVT decision-making than their actual NIHSS score. For pa-tients with aphasia, the expert panel reached consensus that this symptom alone suffices to justify the indication for EVT.

Study measures of disability in low NIHSS stroke patients The panelists reached a consensus that the Potential of rtPA for Ischemic Strokes With Mild Symptoms(PRISMS) trial's opera-tional definition of the presence or absence of disabling deficits offered the best measure of disability in prospective studies in-volving low NIHSS stroke patients.

After an extensive discussion of possible outcome measures of long-term functional outcome, the expert panel chose the 9-level mRS(currently under development in the StrokeNet Thrombectomy Endovascular Platform [STEP] trial, https://www.nihstrokenet.org/trials/step-trial/master) as the most appropri-ate outcome measure of trials involving low NIHSS LVO/MeVO stroke patients. This was followed by a combination of the ordi-nal mRS and another scale(rank 2), dichotomous mRS of 0(rank 3), and a QoL scale(rank 4). The panel felt that the ordi-nal mRS alone was the least appropriate outcome measure for this patient population(Supplementary Table 2).

Imaging features

The expert panel agreed that the anticipated technical difficul-

ty of the EVT and the occlusion location should be accounted for when deciding on EVT in low NIHSS patients.There was dis-agreement among the group on whether or not to take ASPECTS into account for decision-making, as it is virtually always high in low NIHSS patients.

Patient characteristics/treatment goals

Among the mentioned patient characteristics that potentially influence decision-making for EVT in low NIHSS stroke patients≤65 years, QoL and independence without treatment emerged as the most important factors for EVT decisions in patients with low NIHSS. Ability for social interaction holds moderate impor-tance. Interestingly, 9/29(31%) respondents rated the discus-sion with patient or family without fixed parameters highest,while 10/29(34%) chose it as the least important among the presented characteristics. Detailed results are presented in Table 1.

In patients>65 years, level of independence without treat-ment and QoL without treatment remained the most important parameters, similar to the general population, but with stron-ger emphasis.

Elaborating reasons not to take individual patient factors into account when deciding on EVT in low NIHSS stroke patients,there was only consensus on the statement that"history-tak-ing was typically crude in a hyper-acute setting, therefore such decisions are based on assumptions that could be wrong."

Early neurological deterioration

The risk for END in low NIHSS LVO/MeVO stroke was generally perceived as moderate by 18/28(64%) and high by 6/28(21%)of the experts in round 3. The panel agreed that no single clini-cal or imaging factor was extremely useful as an indicator for END.However, by round 3 there was consensus that a large pen-umbra, low collateral flow reserve, a proximal occlusion location,and fluctuating symptoms were the most reliable predictors of END. Additionally, orthostatic neurological deterioration, high blood pressure, multiple occlusions, an incomplete circle of Willis,and a long clot length were felt to be somewhat useful as indi-cators of an increased risk of END. There was agreement that di-abetes and potential embolic origin of the stroke were not use-ful predictors of an increased END risk. Detailed results are provided in Supplementary Table 3.

Risk-benefit balance of EVT

There was consensus that some patients with low NIHSS despite LVO benefit more from EVT than others. The expert panel agreed that patients with aphasia, poor collaterals, or with a proximal clot have a great net benefit of EVT. Combining round 3 votes for a small and great perceived benefit of EVT, there was consen-

Bosshart et al.EVT in LVO Strokes With Low NIHSS

Table 1. Summary of round 2 and round 3 questions and answers

|Question|Answer options|Answers|Answers|Final consensus achieved?| |---|---|---|---|---| |Question|Answer options|Round 2|Round 3|Final consensus achieved?| |Do you agree that the decision for or against EVT in low NIHSS stroke patients should not be based on a hard NIHSS cut-off?|-Strongly agree -Somewhat agree -Strongly disagree|18/30(60%) 6/30(20%) 4/30(13%) 2/30(7%)|20/29(69%) 6/29(21%) 1/29(3%) 2/29(7%)|Yes| |Do you agree that the decision for or against EVT in low NIHSS stroke patients should be based on the degree of disability that the symptoms cause?|-Strongly agree -Somewhat agree -Somewhat agree -Strongly disagree|16/30(53%) 9/30(30%) 2/30(7%) 3/30(10%)|23/29(79%) 5/29(17%) 0 1/29(3%)|Yes| |To determine if the disability is high enough to qualify for EVT, which of the following should be the measure?|(A) Decreased baseline functional status such that main daily activity cannot be exerted. (B) Significantly impaired Ool(a per patient).|(A) 18/30(60%) (B) 12/30(40%)|(A) 19/29(66%) (B) 10/29(34%)|Yes| |To determine if the disability is high enough to qualify for EVT, which of the following applies:|(A) Should only be an expert judgement. (B) Has to be discussed with patient and/or family.|(A) 5/29(17%) (B) 24/29(83%)|(A) 3/29(10%) (B) 26/29(90%)|Yes| |Rate your degree of agreement with the following statements about disabling symptoms: (A) Aphasia should always be treated with EVT (if LVO/MeVO) if a benefit of the treatment can be expected. (B) For treatment decision, aphasia should get more weight/points on the NIHSS score. (C)For treatment decision, vision loss/field cut should get more weight/points on the NIHSS score.|-Strongly agree -Somewhat agree -Somewhat disagree -Strongly disagree|-Technical difficulty occurred, thus the results are not valid for the 2nd round -A) 29/29(7%),B(7%),C(4)/29(14%) -A) 0/29(0%),(B) 0/29 (0%),(C) 2/29(7%)|(A) 13/29(45%),(B) 11/29 (38%), (C) 17/29(59%) -A) 2/29(7%),(B) 2/29 (7%), (C) 4/29(14%) -A) 0/29(0%),(B) 0/29 (0%),(C) 2/29(7%)|| |In a prospective study on EVT in patients with low NIHSS, how should disability be assessed and weighed?|(A) Use a simple description of the deficit and give it a number of NIHSS points reached in this category(e.g., aphasia; motor deficit/unable to walk; sensory deficit; visual deficit). (B) Don't necessarily differentiate different levels of disability, but just differentiate presence/absence of disability. (C) Use the PRISMS trial operational definition of the presence or absence of disabling deficits: Disabling is"defined as either inability to return to work or perform basic activities of daily living based on current deficits."|(A) 10/30(33%) (B) 3/30(10%) (C) 17/30(57%)|(A) 5/29(17%) (B) 1/29 (B) 1/29 (C) 23/29(79%)|Yes| |Which of the following factors should be accounted for when deciding over EVT in low NIHSS patients?|(A) Anticipated technical difficulty of EVT. (B) Level of occlusion. (C) ASPECTS in the territory responsible for the(disabling) symptoms.|Strongly agree: (A) 16/29(55%),(B) 10/28 (36%), (C) 10/29(34%) Somewhat agree: (A) 9/29(31%),(B) 12/28 (43%), (C) 10/29(34%) Somewhat disagree: (A) 3/29(10%),(B) 5/28 (18%), (C) 5/29(17%) Strongly disagree: (A) 1/29(3%),(B) 1/28 (49%),(C) 4/29(14%)|Strongly agree: (A) 8/29(28%),(B) 12/29 (41%),(C) 10/29(34%) Somewhat agree: (A) 18/29(62%),(B) 15/29 (52%),(C) 12/29(41%) Somewhat disagree: (A) 1/29(3%),(B) 1/29 (3%),(C) 1/29(3%)|(A) Yes (B) No (C) No|

Vol.27/ No.3/ September 2025

Table 1. Continued

|Question|Answer options|Answers|Answers|Final consensus achieved?| |---|---|---|---|---| |Question|Answer options|Round 2|Round 3|Final consensus achieved?| |For EVT decision in low NIHSS stroke patients which of the following parameters is the most important in patients<65 years?|Profession alone; profession+hobbies; profession+age; QoL without treatment; level of independence without treatment; no fixed parameters, always discuss with patient or proxy.|The 3 top choices: - No fixed parameters: 11/29(38%) - QoL without treatment: 8/29(28%) - Level of independence without treatment: 4/29 (14%)|The top 3 ranks: 1. Level of independence without treatment, 2. No fixed parameters, 3. QoL without treatment|Yes| |For EVT decision in low NIHSS stroke patients which of the following parameters is the most important in patients>65 years?|Ability for social interaction; QoL without treatment; level of independence without treatment; no fixed parameters, always discuss with patient or proxy.|The 3 top choices: - No fixed parameters: 11/27(41%) - QoL without treatment: 7/27(26%) - Level of independence without treatment: 6/27 (22%)|The top 3 ranks: 1. Level of independence without treatment, 2. No fixed parameters, 3. QoL without treatment|Yes| |Mark your agreement with the following statements: individual patient factors should NOT be taken into account when deciding on EVT in low NIHSS stroke patient because... (A) History-taking is typically crude in a hyper-acute setting, therefore such decisions are based on assumptions that could be wrong. (B) The risk of EVT is generally low and should be done anyway if possible. (C) This information could lead us to be overly aggressive or conservative in our treatment decision. (D) Any patient who is still working/of working age should be treated.|-Somewhat agree - Somewhat disagree|- (A) 3/26 (12%), (B) 2/26 (8%), (D) 3/27 (11%) -Somewhat disagree|- (A) 3/28 (11%),(B) 1/28 (4%), (C) 4/28 (14%), (D) 4/28 (14%) - (A) 17/28 (61%),(B) 11/28 (39%), (C) 11/28 (39%), (D) 14/28 (50%) - (A) 5/26 (19%),(B) 10/26 (37%) (D) 10/27 (37%) - (A) 6/26 (23%), (B) 6/26 (3%), (C) 4/28 (14%), (D) 3/27 (11%) - (A) 4/28 (14%),(B) 5/28 (18%), (C) 2/28 (13%) (D) 3/28 (13%)|Yes| |How high would you estimate the risk of neurological deterioration in low NIHSS HVL/OMeVO stroke patients without EVT? -High risk|- Low risk - Moderate risk -High risk|4/28 (14%) 17/28 (61%) 7/28 (25%)|4/28 (14%) 18/28 (64%) 6/28 (21%)|Yes| |Do you think the benefit of EVT is always given in low NIHSS LVO/MeVO stroke patients?|-Yes -No|3/28 (11%) 25/28 (89%)|2/28 (7%) 26/28 (93%)|Yes|

EVT, endovascular thrombectomy; NIHSS, National Institutes of Health Stroke Scale; QoL, quality of life; LVO, large vessel occlusion; MeVO, medium vessel oc-clusion; PRISMS, Potential of rtPA for Ischemic Strokes With Mild Symptoms; ASPECTS, Alberta Stroke Program Early CT Score.

sus that the net benefit was greater than the risks in patients with poor collaterals, orthostatic deterioration, aphasia, and a proximal clot. There was consensus that the risk of EVT surpass-es potential benefits when the clot is poorly accessible(15/28,54%). Detailed results are presented in Supplementary Table 4.

Recommendations guiding treatment decisions in low NIHSS LVO/MeVO patients are summarized in Table 2.

Discussion

Prudence regarding EVT in low NIHSS patients originates from the delicate balance of the risks of EVT that may not be coun-terbalanced by the possible benefits in patients with minimal

impairment. Absence of level 1 evidence leaves the decision on treatment of low NIHSS stroke patients to the discretion of the clinicians.

This Delphi consensus session among international experts on interventional stroke treatment suggests that the decision in favor or against EVT in low NIHSS stroke patients with LVO or MeVO is often not solely based on a strict NIHSS cut-off, but rather an individualized decision based on the degree of disabil-ity and if possible, based on the patient's perspective. Presumed OoL and independence without treatment were key factors sway-ing treatment decisions. The majority of the expert panel relied on discussions with the patient to arrive at a personalized de-cision whenever possible.

JoS

Boshart et al.EVT in LVO Strokes With Low NIHSS

Table 2. Summary of recommendations guiding EVT treatment decisions in low NIHSS LVO/MeVO patients

|No.|Recommendation| |---|---|

1 In stroke patients with LVO and low NIHSS, treatment decision should be guided by anticipated technical difficulty of EVT and the degree of disability defined as inability to return to work or perform basic activities of daily living. It should not be guided by a strict NIHSS cut-off.

2 The level of disability that warrants the risk of EVT should ideally be determined through a discussion with the patient and/or their family. When such a discussion is not possible, disability was considered sufficient to justify EVT if the current symptoms result in a decreased baseline functional status that prevents the patient from performing their main daily activities.

3 For EVT treatment decisions, aphasia and vision loss/field cut should be regarded as highly disabling symptoms that may justify intervention even in patients with low NIHSS scores.

4 The risk of END in low NIHSS LVO/MeVO stroke patients not undergoing EVT was considered moderate.

END,early neurological deterioration;EVT,endovascular thrombectomy;LVO,large vessel occlusion;MeVO,medium vessel occlusion;NIHSS,National Insti-tutes of Health Stroke Scale.

However, the following factors were identified that influence the decision in a certain direction:

Disabling symptoms

While the NIHSS scale is a well-established measure of neuro-logical deficits, the degree of disability within the total score can vary substantially. In the present analysis, aphasia and vision loss were considered the most disabling symptoms captured on the NIHSS scale, particularly compared to sensory deficits that contribute a similar number of points to the score. It has previ-ously been shown that aphasia significantly affects functional outcome, mood, OoL, social participation and the ability to re-turn to work.20-23 Additionally, aphasia is often co-occurring with dysphagia-a symptom that is not captured on the NIHSS but that has a severe impact on QoL, and increases the likelihood of associated complications such as pneumonia.24,25

Vision loss is less straightforward. For instance, the disability caused by vision loss was rated with 0.60 disability adjusted life year(DALY;0=no disability,1=maximal disability similar to death)in the original Global Burden of Disease(GBD) study in 1990,26 and with 0.19 DALY in the 2010 GBD study.27 The reason for the large difference between the two studies likely lies in the ques-tions asked during the assessment: while the original GBD study focused on introduced impairment and need for assistance, the 2010 GBD study asked"how sick" a completely blind person was.28 Although blindness can lead to significant impairment, one would not necessarily consider a visually impaired person as“sick.” 28 Interestingly, there appears to be a difference in disability de-pending on the age somebody turns blind: vision loss in older adults is associated with reduced QoL, increased risk of cogni-tive impairment, and greater difficulties in performing daily ac-tivities.29 Additionally, older adults with visual impairment have an increased risk of falls, and thus, with associated morbidity.30

The expert panel felt that the degree of disability introduced through aphasia and vision loss justifies EVT if their presence as isolated symptoms was resulting in the low NIHSS.

However, there is a potential bias here, as proprioceptive and kinesthetic functions, which depend on sensation and may ini-tially present with subtle impairments, can later become more overt during the recovery phase and still lead to significant dis-abling effects.

The discussion among the expert panel acknowledged that even seemingly mild symptoms can be associated with a high degree of disability for certain individuals. For example, facial palsy in a professional trumpet player can lead to full invalidity.McDonough et al.13 have shown that such factors in fact play a role in real-world decision-making on EVT for low-NIHSS pa-tients with MeVOs. However, in our analysis, there was consen-sus that the importance of the profession and hobbies is sur-passed by the QoL and the level of independence. Even though some deficits may be debilitating for someone's profession, they may still not justify risking a patient's independence. It was ar-guided that taking"low-level disability" into account for EVT de-cision-making in low NIHSS patients may underestimate the risk of the intervention. Considering individual patient factors could thus lead interventionalists to be overly aggressive or conserva-tive in their treatment decision. Additionally, the term"individual patient factors" is not trivial, taking into account that essential skillsets of a housewife or a retired person with caregiving re-sponsibilities ought to be equally considered as a profession.

The experts thus coincided that differentiating disability ver-sus no disability was most appropriate approach for selecting patients who qualify for EVT, but the final decision should be guided by the discussion with the patient or their proxy.

A measure of disability is needed that can be efficiently ap-plied in the time-sensitive setting of stroke care. To date, such scores have been implemented in prospective trials, but are not yet common in real-world practice. The expert panel agreed that the PRISMS trial's operational definition of the presence or ab-sence of disabling deficits was currently the most appropriate:"disabling is defined as either inability to return to work or per-form basic activities of daily living based on current deficits."31

Vol.27/ No.3/ September 2025

JoS

In terms of long-term outcomes for trial settings, experts ex-pressed their concern that a single conventional outcome mea-sure of stroke patients, such as the ordinal mRS, may not be granular enough to reliably measure differences in treatment effect in patients who had a low NIHSS to begin with. A com-bination of two established scores was proposed as a pragmatic solution to capture small changes in neurological deficit. How-ever, the results of the survey at the end of each Delphi round reflect that experts would prefer a single, new score tailored to patients with mild to moderate disability(i.e., a low NIHSS score).

Anticipated technical difficulty

The potential benefit of an EVT encompasses the alleviation of current symptoms, the prevention of long-term sequelae(e.g.,pneumonia resulting from dysphagia), and the mitigation of END.Conversely, the risk associated with the intervention reflects a combination of technical complexity, the interventionalist's ex-pertise, and individual patient characteristics.

Factors known to increase the risk of an intervention are, among others, the need for multiple passes with thrombectomy devices,32 a fibrin-rich clot composition,33 and underlying ICAD.34 Those still cannot be reliably predicted based on routine imaging, nor are they modifiable in a given patient with acute stroke. The inter-ventionalist's experience is positively correlated with recanal-zation rates and inversely correlated with the duration of the procedure,35 but this, again, is a non-modifiable factor in the acute setting.

The expert panel agreed that the anticipated technical diffi-culty of EVT was mostly a function of the accessibility of the clot-mainly defined by occlusion location and presence of ICAD-as well as the interventionalist's experience. Individual patient characteristics such as age and diabetes mellitus were thought to have a lower priority when estimating the risk of EVT.

END/risk without treatment

END is a frequent phenomenon of neurological worsening of≥4 NIHSS points within 24 hours after a baseline assessment or treatment.36 Its incidence lies between 9%-10% in strokes due to LVO, and it is a reliable predictor of poor outcome in stroke patients.36.37 Risk factors for END include age, glucose level at admission, symptom onset to treatment time, high systolic blood pressure, history of hypertension,diabetes mellitus, atrial fibril-lation, and large artery atherosclerosis.36 Collateral status is the only imaging-based risk factor for END.

END in low NIHSS patients likely has a larger impact on func-tional outcome compared to high-NIHSS patients. Hence, the identification of risk factors for END in the low NIHSS LVO pop-ulation is of particular interest.

The expert panel agreed that poor collaterals, a large penum-bra on perfusion imaging, proximal occlusion location, and fluc-tuating symptoms were the most reliable predictors for END in low NIHSS LVO/MeVO patients.That is, the panel relied on im-aging markers for which limited evidence is available, and their value to predict END has yet to be studied.

Interestingly, to date, there is no convincing evidence that EVT actually reduces the risk of END compared to IVT, as there seems to be a similar incidence among patients treated with EVT and IVT only.36,38 Of note, however, rescue EVT in case of END seems to improve outcomes.39 In the discussion among the expert panel it was brought up that, given the known time-sensitive nature of stroke, the potential harm of END in low NIHSS LVO/MeVO patients is higher in institutions that cannot guarantee close neurological monitoring. This could thus be a valuable consider-ation for triage and transport decisions.

Of note, despite its correlation with poor outcome in stroke patients, END is just an intermediary endpoint, and as such, a surrogate for the final 90-day outcome. As not all cases of END result in long-term disability, further research is needed to iden-tify patients who benefit from END treatment or prevention.

Limitations

Delphi consensus sessions are conducted on topics with little or no evidence, and thus represent opinions of the experts surveyed.The selective group of participants may not represent the glob-al perspective on EVT in low NIHSS patients. Recommendations may not apply to all healthcare systems due to variations in re-sources, expertise, and patient demographics.The study is quali-tative and does not provide definitive recommendations. The conclusions drawn are interim, reflect the opinions of the Delphi panel members only, and may change upon completion of EN-DOLOW,MOSTE, and other ongoing trials.For example, the 9-level mRS, selected as the most suitable outcome measure for trials involving low NIHSS LVO/MeVO stroke patients, has not yet been published or validated. As a result, its suitability will need to be evaluated once evidence is available.

Conclusions

This Delphi consensus panel suggests that a single, universal NIHSS cut-off may not be appropriate to guide EVT decisions in low NIHSS LVO and MeVO stroke. Instead, factors such as the degree of disability, anticipated technical difficulty, and the risk of END should be taken into account. The panel emphasized the importance of patient-centered discussions and individualized treatment approaches, and highlighted critical gaps in evidence,

Bosshart et al.EVT in LVO Strokes With Low NIHSS

including the need for more robust predictors of END in low NI-HSS populations. These findings provide an interim framework for clinicians while awaiting results from ongoing randomized trials.

Supplementary materials

Supplementary materials related to this article can be found online at https://doi.org/10.5853/jos.2025.00304.

Funding statement

None

Conflicts of interest

The authors have no financial conflicts of interest.

Author contribution

Conceptualization: SLB, AS, JMO, MDH. Study design: SLB, AS,JMO. Methodology: SLB, JMO, MDH. Data collection: SLB, AS.Investigation: SLB, MK,AS, JMO. Statistical analysis: SLB, JMO.Writing-original draft: SLB, MK, AS, JMO. Writing-review&editing: all authors. Approval of final manuscript: all authors.

References

1. Berkhemer OA,Fransen PS, Beumer D, van den Berg LA, Lings-ma HF, Yoo AJ,et al.A randomized trial of intraarterial treat-ment for acute ischemic stroke.N Engl J Med 2015;372:11-20.

2. Campbell BC, Mitchell PJ, Kleinigin TJ, Dewey HM, Churilov L,Yassi N, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015;372:1009-1018.

3. Ospel JM,Kim B,Heo JH,Yoshimura S,Kashani N,Menon B,et al. Endovascular treatment decision-making in acute isch-emic stroke patients with large vessel occlusion and low Na-tional Institutes of Health Stroke Scale: insights from UNMASK EVT, an international multidisciplinary survey. Neuroradiology 2020;62:715-721.

4. Duloquin G,Crespy V,Jakubina P,Giroud M,Vergely C,Béjot Y.Large vessel occlusion in patients with minor ischemic stroke in a population-based study.The Dionj Stroke Registry. Front Neurol/2022;12:796046.

5. Schwarz G, Cascio Rizzo A, Matusevicius M, Giussani G, In-vernizzi P, Melis F,et al. Reperfusion treatments in disabling versus nondisabling mild stroke due to anterior circulation vessel occlusion. Stroke 2023;54:743-750.

6. Khatri P, Kleindorfer DO, Devlin T, Sawyer RN Jr, Starr M, Mejil-la J,et al.Effect of alteplase vs aspirin on functional outcome for patients with acute ischemic stroke and minor nondis-abling neurologic deficits: the PRISMS randomized clinical trial. JAMA 2018;320:156-166.

7. Seners P, Ben Hassen W, Lapergue B, Arquizan C, Heldner MR,Henon H,et al. Prediction of early neurological deterioration in individuals with minor stroke and large vessel occlusion in-tended for intravenous thrombolysis alone. JAMA Neurol 2021;78:321-328.

8. Lin CH, Saver JL, Ovbiagele B, Tang SC, Lee M, Liebeskind DS.Effects of endovascular therapy for mild stroke due to proxi-mal or M2 occlusions: meta-analysis.J Neurointerv Surg 2023;15:350-354.

9. Schwarz G,Bonato S, Lanfranconi S, Matusevicius M,Ghione I,Valcamonica G,et al. Intravenous thrombolysis+ endovas-cular thrombectomy versus thrombolysis alone in large ves-sel occlusion mild stroke: a propensity score matched analy-sis.Eur JNeurol 2023;30:1312-1319.

10. Wang P,Chen W,Chen C,Bivard A,Yu G,Parsons MW,et al.Association of perfusion lesion variables with functional out-come in patients with mild stroke and large vessel occlusion managed medically. Neurology 2023;100:e627-e638.

11. Xu Y, Liu X, Li H. A comparison of endovascular therapy and medical management in patients with large vessel occlusion mild stroke treated between 2015 and 2023: a systematic re-view and meta-analysis. J Stroke Cerebrovasc Dis 2024;33:107721.

12. Ospel JM, Nguyen TN, Jadhav AP, Psychogios MN, Clarencon F,Yan B,et al.Endovascular treatment of medium vessel oc-clusion stroke. Stroke 2024;55:769-778.

13. McDonough R, Cimflova P, Kashani N,Ospel JM,Kappelhof M,Singh N,et al. Patient-relevant deficits dictate endovascular thrombectomy decision-making in patients with low NIHSS scores with medium-vessel occlusion stroke. AJNR Am J Neu-roradiol 2021;42:1834-1838.

14. Dayé C. How to train your oracle: the Delphi method and its turbulent youth in operations research and the policy scienc-es. Soc Stud Sci 2018;8:48:846-868.

15. Steurer J. The Delphi method: an efficient procedure to gen-erate knowledge. Skeletal Radiol/2011;40:959-961.

16. Goyal M, Yoshimura S, Milot G, Fiehler J, Jayaraman M, Dorn F,et al. Considerations for antiplatelet management of ca-rotid stenting in the setting of mechanical thrombectomy: a Delphi consensus statement. AJNR Am J Neuroradiol 2020;41:2274-2279.

17. Ospel JM, Brouwer P, Dorn F, Arthur A, Jensen ME, Nogueira R,et al. Antiplatelet management for stent-assisted coiling

Vol.27/ No.3/ September 2025

JoS

and flow diversion of ruptured intracranial aneurysms: a DEL-PHI consensus statement. AJNR Am J Neuroradiol 2020;41:1856-1862.

18. Ospel JM, Mirza M, Clarengon F, Siddiqui A, Doyle K, Consoli A, et al. What is a challenging clot?: a DELPHI consensus statement from the CLOTS 7.0 summit. Clin Neuroradiol 2023;33:1007-1016.

19. Whelehan DF, Conlon KC, Ridgway PF. Medicine and heuris-tics: cognitive biases and medical decision-making. Ir J Med Sci 2020;189:1477-1484.

20. Ferro JM, Madureira S. Aphasia type, age and cerebral infarct localisation.J Neurol/1997;244:505-509.

21. Mazaux JM, Lagadec T, de Seze MP, Zongo D, Asselineau J,Douce E,et al. Communication activity in stroke patients with aphasia. J.Rehabil Med 2013;45:341-346.

22. Carod-Artal FJ, Egido JA. Quality of life after stroke: the im-portance of a good recovery. Cerebrovasc Dis 2009;27(Suppl 1):204-214.

23. Koleck M,Gana K, Lucot C,Darrigrand B, Mazaux JM,Glize B.Quality of life in aphasic patients 1 year after a first stroke.Qual Life Res 2017;26:45-54.

24. Flowers HL,Silver FL,Fang J,Rochon E, Martino R.The inci-dence, co-occurrence, and predictors of dysphagia, dysarthria,and aphasia after first-ever acute ischemic stroke. J Commun Disord 2013;46:238-248.

25. Banda KJ,Chu H,Kang XL,Liu D,Pien LC,JH J,et al.Preva-lence of dysphagia and risk of pneumonia and mortality in acute stroke patients: a meta-analysis. BMC Geriatr 2022;22:420.

26. Murray CJ. Quantifying the burden of disease: the technical basis for disability-adjusted life years. Bull World Health Or-gan 1994;72:429-445.

27. Salomon JA, Vos T, Hogan DR, Gagnon M, Naghavi M, Mok-dad A, et al. Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010. Lancet 2012;380:2129-2143.

28. Braithwaite T,Taylor H, Bourne R, Keeffe J, Pesudovs K. Does blindness count? Disability weights for vision loss. Clin Exp Ophthalmol 2017;45:217-220.

29. Li S,Ye E,Huang J,Wang J,Zhao Y,Niu D,et al.Global,re-gional, and national years lived with disability due to blindness and vision loss from 1990 to 2019: findings from the Global Burden of Disease Study 2019. Front Public Health 2022;10:

1033495.

30. Singh RR, Maurya P. Visual impairment and falls among older adults and elderly: evidence from longitudinal study of age-ing in India. BMC Public Health 2022;22:2324.

31. Yeatts SD, Broderick JP,Chatterjee A, Jauch EC, Levine SR,Romano JG,et al. Alteplase for the treatment of acute isch-emic stroke in patients with low National Institutes of Health Stroke Scale and not clearly disabling deficits(potential of rtPA for ischemic strokes with mild symptoms PRISMS): ra-tionale and design. Int J Stroke 2018;13:654-661.

32. Ringheanu VM, Tekle WG, Preston L, Sarraj A, Hassan AE.Higher number of stent-retriever thrombectomy passes sig-nificantly increases risk of mass effect, poor functional out-come, and mortality. Interv Neuroradiol 2023;29:674-682.

33. Jolugbo P, Ariens RAS. Thrombus composition and efficacy of thrombolysis and thrombectomy in acute ischemic stroke.Stroke 2021;52:1131-1142.

34. Kim YW, Hong JM, Park DG, Choi JW, Kang DH, Kim YS, et al.Effect of intracranial atherosclerotic disease on endovascular treatment for patients with acute vertebrobasilar occlusion.AJNR Am J Neurorodiol 2016;37:2072-2078.

35. Dimitrios X,Ghozy S,Christina C,Kolovoy A,Ramanathan K,Kallmes DF.The effect of operator's experience on mechani-cal thrombectomy outcomes: a systematic review. Interv Neuroradiol 2025;31:121-127.

36. Shi HX,Li C,Zhang YQ,Li X,Liu AF,Liu YE,et al. Predictors of early neurological deterioration occurring within 24 h in acute ischemic stroke following reperfusion therapy: a systematic review and meta-analysis. J Integr Neurosci2023;22:52.

37. Seners P, Turc G, Oppenheim C, Baron JC. Incidence, causes and predictors of neurological deterioration occurring within 24 h following acute ischaemic stroke: a systematic review with pathophysiological implications. J Neurol Neurosurg Psy-chiatry 2015;86:87-94.

38. Bourcier R, Goyal M, Muir KW, Desal H, Dippel DWJ, Majoie CBLM,et al. Risk factors of unexplained early neurological deterioration after treatment for ischemic stroke due to large vessel occlusion: a post hoc analysis of the HERMES study.J Neurointerv Surg 2023;15:221-226.

39. Broccolini A, Brunetti V, Colo F, Alexandre AM, Valente I, Fal-cou A,et al.Early neurological deterioration in patients with minor stroke due to isolated M2 occlusion undergoing medi-cal management: a retrospective multicenter study. J Neuro-interv Surg 2023;16:38-44.

Vol.27/ No.3/ September 2025

$$\begin{array}{l}\\ \\ \\ \\ \\ \\ \end{array}$$

Supplementary Table 1. List of the publications selected for presentation during the meeting

|No.|Publications| |---|---| |1|Seners P, Perrin C, Lapergue B, Henon H, Debiais S, Sablot D, et al; MINOR-STROKE Collaborators. Bridging therapy or IV thrombolysis in minor stroke with large vessel occlusion. Ann Neurol 2020 Jul;88(1):160-169. doi: 10.1002/ana.25756. Epub 2020 May 14. PMID: 32350929.| |2|Lin CH,Saver JL,Ojbagiote B, Tang SC, Lee M, Liebeshind DS. Effects of endovascular therapy for mild stroke due to proximal or M2 occlusions: meta-analysis. J Neurointerv Surg 2023 Apr;15(4):350-354. doi: 10.1136/neurinturg-2022-018662.Epub 2022 Mar 15.PMID:35292565; PMCID: PMC10086511.| |3|Schwarz G,Bonato S, Lanfranconi S, Maturusevici M,Ghione I, Valcamonica G,et al.Intravenous thrombolysis+endovascular thrombectomy versus thrombolysis alone in large vessel occlusion mild stroke: a propensity score matched analysis. Eur J Neurol 2023 May;30(5):1312-1319. doi: 10.1111/ene.15722.Epub 2023 Feb 24. PMID:36746650.| |4|Wang P,Chen W,Chen C,Bivard A,Yu G,Parsons MW,et al;INSPIRE.Association of perfusion lesion variables with functional outcome in patients with mild stroke and large vessel occlusion managed medically. Neurology 2023 Feb 7;100(6):e627-e638. doi: 10.1212/WNL000000000201498.Epub 2022 Oct.28.PMID:36307224;PMID:PMDC:PMDC9946183.| |5|Dunne T, Patel K, Siegles JE. How low do we go? NIHSS in anterior circulation thrombectomy decision-making. Stroke Vasc Interv Neurol 2024;4(5):e001212. doi: 10.1161/SVIN.123.01212.| |6|Xu Y, Liu X, Li H, A comparison of endovascular therapy and medical management in patients with large vessel occlusion mild stroke. 2024/4(5):e001212. doi: 10.1161/SVIN.123.01212.| |6|Xu Y, Liu X, Li H, A comparison of endovascular therapy and medical management in patients with large vessel occlusion mild stroke treated between 2015 and 2023: a systematic review and meta-analysis. J Stroke Cerebrovasc Dis 2024 Jul;33(7):107721. doi: 10.1016/j.jstrokecerebrovasdis.2024.107721. Epub 2024 Apr 13. PMID:38616013.| |7|Seners P,Ben Hassen W, Lapergue B, Arquizan C,Heldner MR, Henon H,et al; MINOR-STROKE Collaborators. Prediction of early neurological deterioration in individuals with minor stroke and large treatment mild stroke: In intravenous thrombolysis alone. JAMA Neurol 2021 Mar 1,78(3):321-328. doi: 10.1001/jamaneurol.2020.4557. PMID:33427887; PMCID: PMC7820007.| |8|Heitkamp C,Winkelmeier L, Flottmann Fschell M, Knieh P, Brocks G,et al; German Stroke Registry-Endovascular Treatment(GSR-ET).Thrombectomy patients with minor stroke: factors of early neurological deterioration. J Neurointerv Surg 2025 Jul 14;17(8):796-802. doi: 10.1136/jnis-2024-021930. PMID:38991731;PMCID: PMC1232244. 9| |9|Romano JG,Gardener H,Campo-Bustillo I, Khan Y,Tai S,Riley N,et al;MaRISS Investigators.Predictors of outcomes in patients with mild ischemic stroke symptoms: MaRISS. Stroke 2021 Jun;52(6):1995-2004. doi: 10.1161/STROKEAHA.120.032809.Epub 2021 May 5.PMID:33947209;PMCID: PMC9254134.|

Supplementary Table 2. Number of panelist votes for ranking outcome measures to assess IV/EVT benefit in patients with baseline NIHSS 3-5

||Rank 1|Rank 2|Rank 3|Rank 4|Rank 5|Rank 6|Rank 7|Rank 8|Rank 9| |---|---|---|---|---|---|---|---|---|---| |mRS 0-1|5|6|3|5|5|0|3|0|2| |mRS 0|6|5|3|4|5|4|0|2|0| |Ordinal mRS alone|1|0|5|4|3|4|7|2|3| |Ordinal mRS in combination with another score.|2|4|3|5|5|4|0|4|2| |9-level mRS with 1a, 1b, 2a, and 2b levels(under development in STEP trial).|8|5|3|4|3|4|1|0|1| |QoL scale|5|2|6|3|3|5|2|2|1| |NIHSS|0|2|3|1|1|3|8|6|5| |Delta-NIHSS(baseline NIHSS minus 30 d NIHSS)|1|4|2|3|2|2|3|10|2| |Delta mRS|1|1|1|0|2|3|5|3|13|

Panelists were asked to assign a rank/priority to the outcome measures to assess IVT/EVT benefit in patients with baseline NIHSS 3-5 in a trial setting.The number of votes per rank for the outcome measures listed in the first column is shown. Rank 1=highest priority; Rank 9=lowest priority.

EVT,endovascular thrombectomy; IVT,intravenous thrombolysis; mRS,modified Rankin Scale;STEP trial, StrokeNet Thrombectomy Endovascular Platform trial;NIHSS,National Institutes of Health Stroke Scale;QoL,quality of life.

JoS

Boshart et al.EVT in LVO Strokes With Low NIHSS

Supplementary Table 3. Round 3 ratings regarding usefulness of different indicators for END risk

||Not useful at all|Slightly useful|Extremely useful| |---|---|---|---| |How useful are the following factors to identify low NIHSS LVO/MeVO patients with high risk of END without EVT?|||| |Orthostatic challenge|1|20|8| |Any low NIHSS patient with LVO is at risk. Secondary deterioration often leads to disability|8|16|5| |→all low NIHSS individuals with LVO should be treated.|||| |Any LVO not determined to be from ICAD.|9|12|8| |Older age|13|14|2| |Incomplete circle of Willis|8|19|2| |Multiple occlusions|3|15|11| |Long clot length|5|15|9| |Large penumbras|2|6|21| |Proximal occlusion of major intra-cranial vessel|1|5|23| |Potential embolic origin|12|13|3| |Low collateral flow reserve (=poor collaterals)|3|9|17| |High blood pressure|6|17|6| |Diabetes|16|12|0| |Fluctuating symptoms|1|9|19|

END,early neurological deterioration;EVT,endovascular thrombectomy;ICAD,intracranial atherosclerotic disease;LVO,large vessel occlusion;MeVO,medium vessel occlusion;NIHSS,National Institutes of Health Stroke Scale.

Supplementary Table 4. Perceived risk-benefit balance of EVT in low NIHSS LVO patients(Round 3 responses, stratified by patient and treatment character-istics; number of panelist votes)

||Risk is higher than benefit|No benefit, no risk|Small net benefit|Great net benefit| |---|---|---|---|---| |How do you rate the benefit of EVT in low NIHSS patients with...||||| |Ipsilateral ICAD|8|4|13|3| |Poor collaterals|2|3|6|17| |Orthostatic deterioration|0|2|12|14| |Aphasia|0|2|7|19| |Distal clot(MeVO)|6|4|14|3| |Proximal clot(LVO)|0|0|6|22| |Poor accessibility of clot|15|4|9|0| |After TNK application|0|8|14|6| |If EVT is performed by an experienced interventionalist|0|1|9|17| |If EVT is performed by an un-experienced interventionalist|15|5|8|0| |Disabling symptoms|0|0|9|19| |Young patient|0|3|8|17|

Round 3 answers are shown to how participants perceive the risk-benefit balance of EVT in low NIHSS LVO patients stratified for different patient or treat-ment characteristics. The number of votes from panelists are shown per patient or treatment characteristic.

EVT, endovascular thrombectomy; ICAD, intracranial atherosclerotic disease; LVO, large vessel occlusion; MeVO, medium vessel occlusion; NIHSS, National In-stitutes of Health Stroke Scale;TNK, tenecteplase.

相关指南

• 重大突破！中国最新指南，率先延长至 24 小时 · 综合与多学科
• 缺血性卒中脑细胞保护临床实践中国专家共识 · 神经内科
• 中国急性缺血性卒中诊治指南2023 · 神经内科 2023
• 脑机接口康复评估临床应用专家共识 · 神经内科
• 急性缺血性卒中再灌注治疗医疗质量评价与改进专家建议（2026） · 神经内科 2026
• 中国缺血性卒中和短暂性脑缺血发作二级预防指南2022 · 神经内科 2022
• 中国脑卒中防治指导规范(2021年版) · 神经内科 2021
• 2026 AHA／ASA指南：急性缺血性脑卒中患者的早期管理 · 神经内科 2026