Risk Factors of Futile Recanalization Following Endovascular Treatment in Patients With Large‐Vessel Occlusion: Systematic Review and Meta‐Analysis

Abstract

Background

Although successful recanalization (modified Thrombolysis in Cerebral Infarction score 2b–3) can be achieved in >80% of patients experiencing stroke attributable to large‐vessel occlusion, up to 50% of patients may develop poor clinical outcomes (modified Rankin scale score at 90 days of 3–6), termed as futile recanalization (FR). The meta‐analysis aims to determine various risk factors associated with FR.

Methods

In February 2021, a comprehensive literature search on risk factors associated with FR was performed with keywords, including “stroke,” “thrombectomy,” “treatment outcome,” and “risk factors.” Their correlations with FR were evaluated using the random effect size meta‐analysis model.

Results

Twenty studies with 3037 patients were included with an FR rate of 51.0%. Our meta‐analyses showed that age (mean difference [MD], 5.6; 95% CI, 4.7–6.6), National Institutes of Health Stroke Scale score (MD, 4.2; 95% CI, 3.2–5.1), Alberta Stroke Program Early Computed Tomography (CT) Score (MD, −0.5; 95% CI, 0.8–0.3), hypertension (odds ratio [OR], 1.5; 95% CI, 1.3–1.9), systolic blood pressure (MD, 6.9; 95% CI, 3.6–8.7), atrial fibrillation (OR, 1.5; 95% CI, 1.2–1.8), IV tPA (tissue‐type plasminogen activator) (OR, 0.7; 95% CI, 0.5–0.8), puncture to recanalization time (MD, 9.6; 95% CI, 5.3–13.8), and time of onset to recanalization (MD, 32.1; 95% CI, 6.5–47.7) were significantly associated with FR (P<0.001).

Conclusions

Age, admission National Institutes of Health Stroke Scale score, Alberta Stroke Program Early CT Score, comorbidities, including hypertension, systolic blood pressure, atrial fibrillation, and use of IV tPA, as well as time frames, including onset to recanalization and onset to arrival, were significant influencing factors for FR after mechanical thrombectomy. Future research on the mechanism underlying FR is warranted.

Nonstandard Abbreviations and Acronyms

EVT

endovascular therapy

FR

futile recanalization

NIHSS

National Institutes of Health Stroke Scale

Clinical Perspective

• This is a novel meta‐analysis to identify risk factors which may predict futile recanalization after endovascular thrombectomy.

• Futile recanalization is a complex and common multifactorial entity with a rate of up to 54% in patients undergoing endovascular therapy.

• Various modifiable and nonmodifiable factors contribute to futile recanalization. Early identification of modifiable risk factors, like duration of onset to recanalization, IV tPA (tissue‐type plasminogen activator) administration, and management of high systolic blood pressure and glucose level, help to decrease likelihood of futile recanalization.

Early recanalization is the main goal in acute ischemic stroke treatment and is strongly associated with an increased probability of a good clinical outcome at 3 months. ¹ , ² Recent randomized clinical trials have consistently demonstrated a clear benefit of endovascular therapy (EVT) in patients with acute ischemic stroke with large‐vessel occlusion. ² , ³ This benefit is driven by a higher rate of early angiographic recanalization with no difference in symptomatic intracranial hemorrhage or 3‐month mortality. Despite the high rate of successful recanalization with EVT, more than half of patients experience a poor clinical outcome. ⁴ The literature suggests that factors other than early recanalization play a significant role in clinical outcome. The early identification of factors involved in poor clinical outcome despite successful recanalization could lead to the development of adjunctive therapeutic strategies to increase the benefit of EVT. Recent articles have described numerous factors for futile revascularization. We performed a meta‐analysis to determine the risk factors of poor clinical outcomes in patients with acute ischemic stroke following successful recanalization.

Methods

Data Sources and Search Strategies

A comprehensive search of several databases from inception to January 2021 was conducted. The databases included Ovid MEDLINE, EMBASE, EBM Reviews, Web of Science, and Scopus. The search strategy was designed and conducted by an experienced librarian with input from the study's principal investigator. We also searched the reference lists of included articles for additional article‐controlled vocabulary supplemented with keywords such as “stroke,” “thrombectomy,” “treatment outcome,” and “risk factors.” Search strategies were formulated using a combination of standardized index terms and keywords. The study is in accordance with Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. The data that support the findings are available from the corresponding author on reasonable request. A complete search strategy is provided in the Appendix.

Study Selection Process

Inclusion criteria were the following: (1) English language, (2) study reporting futile rate or poor clinical outcomes and its predictive factors despite good recanalization (Thrombolysis in Cerebral Infarction [TICI] score 2b or 3) in patients with acute anterior and posterior ischemic stroke with large‐vessel occlusion, (3) age group of >17 years, and (4) case series of ≥4 patients reporting clinical outcomes. Exclusion criteria included review articles, comments, guidelines, technical notes, anatomical studies, and editorials.

Titles and abstracts were screened for inclusion by 2 authors under supervision of senior author. Full‐text articles were retrieved for the included abstracts and screened by the 2 authors.

Outcomes and Data Extraction

For each study, we extracted the following baseline information: futile rate, age, race, sex, comorbidities (hypertension, diabetes, atrial fibrillation [AF], coronary artery disease [CAD], smoking, systolic and diastolic blood pressure, and blood glucose levels), history of stroke or transient ischemic attack, antiplatelet and anticoagulant intake, use of IV tPA (tissue‐type plasminogen activator), time from symptom onset to arrival, time from groin puncture to recanalization, source of thrombus (large‐artery atherosclerosis affecting internal carotid or middle cerebral artery M1/M2 or basilar artery, cardiac embolic, or others), general anesthesia, National Institutes of Health Stroke Scale (NIHSS) score, and Alberta Stroke Program Early Computed Tomography (CT) Score on admission, procedural complications, and symptomatic hemorrhage.

Definition

For this study, futile recanalization (FR) is defined by the occurrence of poor functional outcome (modified Rankin scale score of >2 at 3 months) despite successful angiographic recanalization (defined as a modified TICI score of 2b–3).

Risk of Bias

Risk of bias of individual studies was assessed in accordance with Newcastle‐Ottawa Quality Assessment Scale. We modified the Newcastle‐Ottawa Quality Assessment Scale for case control and cohort. We assessed study risk of bias based on selected items from the tool, focusing on the following questions: (1) Did the study include all patients or consecutive patients versus a selected sample? (2) Was the study retrospective or prospective? (3) Was the futile revascularization rate defined and mentioned at 3 months?, and (4) Were the predicting factors clearly mentioned for futile and nonfutile groups?

Statistical Analysis

The mean±SD was used for all continuous outcomes. When these were not available, statistical conversions were made using the presented data to approximate these values. Specifically, the median and interquartile ranges were used to approximate the mean and SD (Wan et al). ⁵ We calculated the difference in means and the 95% CI of age, NIHSS score, Alberta Stroke Program Early CT Score, onset to puncture time, puncture to recanalization, onset to emergency department arrival, and blood pressure between patients with and without FR. For dichotomous values, including hypertension, diabetes, dyslipidemia, AF, CAD, sex, smoking, and IV tPA, antiplatelet, and anticoagulant use, we calculated the odds ratio (OR) with 95% CI. A random‐effects model was used for meta‐analysis. A forest plot was used to display the results of each study and the overall result. We evaluated the heterogeneity of studies using Cochrane Q and I² statistics. We explored the impact of publication bias by constructing funnel plots and checking for symmetry. The Egger regression test was also used to evaluate publication bias.

Results

Study Selection and Characteristics

The initial literature search yielded 1834 studies. On review of the abstracts and titles, 1292 studies were excluded. Sixty studies were selected for full‐text screening. In total, 20 studies including patients with futile rate and their risk factors were included. The Preferred Reporting Items for Systematic Reviews and Meta‐Analyses flowchart for study selection is provided in Figure 1. Of these, 3707 patients underwent successful mechanical thrombectomy with TICI score 2b/3, and 1891 patients had poor clinical outcomes with an overall rate of 51.0% (95% CI, 45.8%–54.7%) (Figure 2). Successful recanalization was observed in both the groups.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) flowchart for meta‐analysis.

Figure 2.

Forest plot for futile recanalization rate for various studies.

In our study, comprised of a total sample of 3707 patients, there were 2097 women (56.5%), and mean age was 68.4 years. Of the 20 studies, 13 had a moderate risk of bias and 7 had an elevated risk of bias. Sixteen studies of the 20 articles were retrospective studies and 4 were clinical trials. The characteristics of the included studies are shown in Supplementary Table S1.

Overall Outcomes

Univariate analysis of various risk factors between both groups was compared as follows and is shown in Supplementary Figures S1–S17.

Demographics and Comorbidities

Mean age was found to be higher in the futile group than in the nonfutile group, with mean difference of 5.6 (95% CI, 4.7–6.6) with P<0.001. The proportion of women in the futile group was higher than in the nonfutile group, with an OR of 1.33 (95% CI, 1.1–1.6) with P<0.01. Mean admission NIHSS score was found to be significantly higher in the futile group compared with nonfutile revascularization, with mean difference of 4.2 (95% CI, 3.2–5.1) with P<0.001 (Table 1).

Table 1.

Various Continuous Outcomes in Futile and Nonfutile Groups

Variables	Futile group*	Nonfutile group*	Mean difference (95% CI)	P value	I2
Age, y	71.3 (69.4 to 73.1)	65.6 (64.3 to 67.8)	5.6 (4.7 to 6.6)	<0.001	18.5
NIHSS score	19.0 (18.2 to 19.7)	14.8 (12.3 to 16.7)	4.2 (3.2 to 5.1)	<0.001	76.0
ASPECTS	7.7 (7.2 to 8.1)	8.2 (7.8 to 8.9)	−0.5 (−0.8 to −0.3)	<0.001	78.1
Onset to puncture time, min	308.4 (279.1 to 337.7)	284.1 (254.1 to 300.9)	24.3 (9.9 to 38.7)	<0.01	53.1
Puncture to recanalization, min	66.9 (53.5 to 80.5)	57.4 (48.2 to 69.2)	9.58 (5.3 to 13.8)	<0.001	42.0
Onset to recanalization, min	336.6 (302.5 to 370.8)	333.5 (252.4 to 331.2)	32.1 (6.5 to 47.7)	<0.001	58.3
Onset to emergency department arrival, min	214.3 (120.2 to 308.3)	194.1 (85.3 to 233.7)	20.1 (4.4 to 35.8)	<0.01	12.8
SBP, mm Hg	149.0 (147.0 to 151.9)	142.1 (140.2 to 146.1)	6.9 (3.6 to 8.7)	<0.001	0
DBP, mm Hg	79.4 (76.7 to 82.2)	78.6 (76.3 to 80.9)	1.31 (−1.0 to 3.6)	0.26	0

ASPECTS indicates Alberta Stroke Program Early CT [Computed Tomography] Score; DBP, diastolic blood pressure; NIHSS, National Institutes of Health Stroke Scale; and SBP, systolic blood pressure.

^*Data are given as median (interquartile range).

Those in the futile group had higher odds of hypertension, with an OR of 1.5 (95% CI, 1.3–1.9) with P<0.001. Similarly, significantly greater odds for diabetes, systolic blood pressure, CAD, AF, use of IV tPA, previous stroke, and symptomatic intracranial hemorrhage were seen in the futile group. Smoking was common in the nonfutile group, with an OR of 0.6 (95% CI, 0.5–0.7) with P<0.01 (Table 2).

Table 2.

Various Categorical Outcomes in Futile and Nonfutile Groups

Variables	Futile group*	Nonfutile group*	Odds ratio (95% CI)	P value	I2
Female sex	50.1 (802/1600)	43.7 (684/1565)	1.3 (1.1–1.6)	<0.01	28.7
Hypertension	69.3 (1411/2035)	57.3 (1006/1755)	1.5 (1.3–1.9)	<0.001	44.5
Diabetes	24.4 (497/2035)	16.3 (286/1755)	1.5 (1.1–2.1)	0.01	68.9
Smoking	19.8 (337/1699)	29.5 (397/1345)	0.6 (0.5–0.7)	<0.01	0
CAD	28.1 (307/1094)	19.6 (173/881)	1.4 (1.1–1.8)	<0.01	20.5
AF	39.0 (694/1779)	31.3 (481/1538)	1.5 (1.2–1.8)	<0.001	30.3
Hyperlipidemia	38.0 (662/1741)	32.1 (465/1447)	1.1 (0.9–1.3)	0.20	6.1
Previous stroke	17.9 (127/711)	12.4 (60/485)	1.4 (1.03–2.04)	0.03	0
Intravenous tPA	41.7 (796/1910)	53.0 (823/1553)	0.7 (0.5–0.8)	<0.001	35.2
Antiplatelet	25.2 (130/514)	53.9 (117/489)	1.1 (0.8–1.4)	0.58	0
Anticoagulant	17.8 (45/253)	23.0 (50/217)	0.5 (0.1–1.6)	0.23	94.0
GA	45.1 (367/813)	42.4 (341/804)	1.2 (0.78–2.01)	0.34	74.5
sICH	10.1 (69/681)	1.4 (11/797)	5.7 (2.8–11.65)	<0.01	8.8
Procedure complications	5.7 (29/507)	6.7 (17/254)	0.8 (0.4–1.8)	0.61	18.8

AF indicates atrial fibrillation; CAD, coronary artery disease; GA, general anesthesia; sICH, symptomatic intracranial hemorrhage; and tPA, tissue‐type plasminogen activator.

^*Data are given as percentage (number/total).

Time Period

The mean durations of onset to recanalization and puncture to recanalization were significantly higher in the futile group than in the nonfutile group, with values of 32.1 (95% CI, 6.5–47.7) and 9.6 (95% CI, 5.3–13.8) minutes with P<0.001, respectively. Moreover, mean durations of onset to puncture and onset to emergency department arrival were also significantly higher in the futile group with P<0.01 (Table 1).

Radiological Variable

Compared with those with nonfutile revascularization, those with futile revascularization had a significantly lower Alberta Stroke Program Early CT Score (difference in means, −0.5; 95% CI, −0.8 to −0.3 HEME) with P<0.001. The data are summarized in Table 1.

Discussion

Early and complete recanalization is the main goal in stroke management to prevent ischemic morbidity and mortality. Despite the recent approval of thrombectomy as the standard of care in patients with acute stroke, the benefits of thrombectomy remain uncertain even after successful or complete recanalization. Therefore, we conducted a systematic review and meta‐analysis comprised of 20 studies, including 3707 patients, and examined several factors potentially associated with poor clinical outcomes. We identified several modifiable and nonmodifiable variables associated with FR. For modifiable variables, longer procedure duration was significantly associated with poor clinical outcomes. For nonmodifiable variables, cardiovascular risk factors, baseline NIHSS score, and lower Alberta Stroke Program Early CT Score were significantly associated with FR. These findings are important as they provide important information that can aid in risk stratification. Furthermore, identifying clinical variables associated with FR could lead to the development of adjunctive therapeutic strategies to increase the benefit of EVT in these patient populations.

In the Highly Effective Reperfusion Using Multiple Endovascular Devices (HERMES) collaboration, Goyal et al found that the rate of FR can be up to 54% in patients undergoing EVT. ⁴ Results from 2 more recent trials, Clinical Mismatch in the Triage of Wake Up and Late Presenting Strokes Undergoing Neurointervention with Trevo Thrombectomy Procedure (DAWN) ⁶ and Endovascular Therapy Following Imaging Evaluation for Ischaemic Stroke (DEFUSE) 3, ⁷ found an FR rate of 50%. These were analogous to our findings, where the rate of FR was found to be 51%. Other retrospective studies demonstrated that FR rates were in the range of 29% to 77%. This relentless soaring proportion of FR demands further research. We discovered that FR during mechanical thrombectomy after large‐vessel occlusive stroke is associated with many clinical variables.

Demographic Variables

As per current literature, neurointerventionalists are able to achieve similar recanalization rates in elderly patients (median age, 68 years) as in nonelderly patients. Despite this achievement, older patients still exhibit a lower rate of 90‐day good clinical outcome. ⁴ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ This can be explained by various phenomena that increase with aging, like impaired cerebral autoregulation and collateral circulation, ¹⁹ and increasing brain atrophy. ²⁰ In addition, white matter hyperintensity or leukoaraiosis, which is common in the aging population, is an independent risk factor for FR, according to some studies. ²¹ , ²²

The effect of sex on EVT is arguable. ²³ Prior reports have also suggested that female patients have worse outcomes than male patients after EVT, even after controlling for demographic differences, stroke severity, comorbidities, vessel recanalization, and EVT complications. ²⁴ , ²⁵ , ²⁶ In our analysis, women were also more likely than men to have poor outcomes after recanalization. One possible explanation for this finding is that there may be differences in malignancy infarction thresholds between sexes for patients with large cores. Other measureless factors, like medical comorbidities, arthritis, osteoporosis, frailty, and some mental health disorders, and less social support, are potentially more common in women, which could contribute to poor outcomes. ²⁴ These findings should be examined in future registries.

Clinical Variables

Our meta‐analysis pointed out that a higher baseline NIHSS score was significantly associated with FR. A similar finding was demonstrated by various other studies in the literature. ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ²⁰ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ Interestingly, a recent meta‐analysis of all major positive endovascular trials showed that a score of ≥20 was associated with a more significant relative benefit of EVT compared with medical treatment, including IV tPA. ³⁶ A possible explanation of this observation is that patients with a high NIHSS score will invariably end up with an unfavorable functional outcome if untreated, leading to a greater relative benefit of EVT compared with those who have a lower NIHSS score.

Moreover, patients with hypertension, systolic blood pressure, diabetes, AF, CAD, and previous strokes have higher odds of FR. However, systolic blood pressure, hyperlipidemia, use of antiplatelets and anticoagulant drugs, and general anesthesia were not found to be significantly related to FR. Our findings are supported by a multicentric study by Hussein et al performed in 130 patients undergoing EVT, which found that FR was higher in women, patients with diabetes, patients with CAD, and patients with higher NIHSS scores. ²⁹ Concordant findings of several comorbidities were also seen by Shi et al. ¹⁷

Another important variable that is identified as a conceivable predictor of FR is the amount of collateral circulation in acute ischemic stroke. Although not studied in the current meta‐analysis, it emerged as a significant radiological biomarker for good prognosis. ³⁷ , ³⁸ , ³⁹ Robust collaterals suggest better prognosis following recanalization and warrant consideration for the EVT. Also, vice versa, better collateral results were found with lower NIHSS score on admission and small lesion at 24‐hour CT scan, resulting in slower progression and good functional outcomes. Hassan et al also evaluated the importance of collateral circulation and its impact on better clinical and angiographical outcomes in patients receiving EVT. ⁴⁰ Liebeskind et al demonstrated the importance of collateral circulation in a selection of patients after 6 hours of ischemic stroke for mechanical thrombectomy. ³⁸

Procedural Variables

In addition to other variables, various time frames, including long durations for onset to puncture, puncture to recanalization, onset to recanalization, and onset to arrival, were all significantly associated with poor outcome at 90 days (modified Rankin scale score, >2). This finding was supported by Hassan et al, who analyzed the data from SOLITAIRE™ FR With the Intention for Thrombectomy (SWIFT), ⁴¹ Solitaire FR Thrombectomy for Acute Revascularisation (STAR), ⁴² and Solitaire With the Intention for Thrombectomy as Primary Endovascular Treatment (SWIFT PRIME) ⁴³ trials to examine the relationship between vascular procedure time and clinical outcomes. The phenomenon could be explained by a delay in achieving reperfusion, which prolongs ischemic time and causes irreversible damage to brain tissue. Our findings and previous reports demonstrate that the prognosis of patients with ischemic stroke is improved by timely restoration of blood flow to salvageable ischemic brain tissue. ⁴⁴ , ⁴⁵ Shortening the time between arrival in the emergency department and onset to recanalization also reduces neurological impairment. ⁴⁶ , ⁴⁷ , ⁴⁸

The effect of the number of passes during EVT, although not studied for the meta‐analysis, theoretically reduces the chances of recovering with further additional passes. Additional passes suggest more occlusion time and more embolism attributable to clot fragmentation and damage to the vessel wall. First‐pass rate and complete recanalization are independent factors for better outcomes. ⁴⁹ Flottmann demonstrated that 3 attempts performed during persistent occlusion result in a good outcome. ⁵⁰ But continuing thrombectomy beyond TICI score 2b should be weighed against the risk involved in additional passes. Therefore, first‐pass TICI score 2b is superior to TICI score 2b after ≥2 retrievals, and comparable to TICI score 3 in ≥3 passes. ⁵¹

Limitations

This study has limitations. First, our meta‐analysis primarily included retrospective studies, and the inevitable bias in retrospective studies might have an unexpected impact on our overall conclusions. Other factors, like infarct volume, blood‐brain barrier disruption, and biological factors (neutrophil counts were also significantly related to futile outcomes) were not studied. Moderate risk of biases could be explained by the heterogeneity of the studies involved. Moreover, potential differences may exist between anterior and posterior circulation stroke, and separate analysis with further studies may be more valuable.

Conclusions

In summary, FR is common during EVT. Identification of various modifiable risk factors, like reducing the duration for onset to recanalization, IV tPA administration, blood pressure, and glucose level, would help in reducing futile outcomes. Also, understanding various nonmodifiable risk factors, like older age, female sex, as well as various comorbidities, like CAD, AF, previous stroke, and smoking, helps in prognostication and proper selection of patients before undergoing intervention and can also identify targets for future study.

Sources of Funding

None.

Disclosure

None.

Supporting information

Table 1

Figures 1–17