Analysis of 565 thrombectomies for anterior circulation stroke: A Brazilian registry

Abstract

Introduction

The benefits of mechanical thrombectomy in the treatment of patients with acute stroke due to large vessel occlusions (LVOs) have been extensively demonstrated by randomized trials and registries in developed countries. However, data on thrombectomy outside controlled trials are scarce in developing countries. The aim of this study was to assess the safety and efficacy, and to investigate the predictors for good and poor outcomes of thrombectomy for treatment of AIS due to anterior circulation LVOs in Brazil.

Materials and Methods

This was a single center registry of thrombectomy in the treatment of stroke caused by anterior circulation LVOs. Between 2011 and 2019, a total of 565 patients were included.

Results

the mean baseline NIHSS score on admission was 17.2. The average baseline ASPECTS was 8, and 91.0% of patients scored ≥6. Half of the patients received intravenous thrombolysis. The mean time from symptom onset to arterial puncture was 296.4 minutes. The mean procedure time was 61.4 minutes. The rates of the main outcomes were recanalization (TICI 2b-3) 85.6%, symptomatic intracranial hemorrhage (sICH) 8,1%, good clinical outcome (mRS=0-2) 43,5%, and mortality 22.1% at three months.

Conclusions

This study demonstrates the efficacy and safety of mechanical thrombectomy for treatment of patients with AIS of the anterior circulation in real-life conditions under limited facilities and resources. The results of the present study were relatively similar to those of large trials and population registers of developed countries.

Introduction

Since the publication of the first large randomized clinical trials (RCTs), mechanical thrombectomy has been shown to be an effective and safe method for the treatment of acute ischemic stroke (AIS) with large vessel occlusions (LVOs) of the anterior circulation. Consequently, it has been incorporated into several clinical practice guidelines and institutional protocols worldwide.^1–7

Other studies have demonstrated the benefits of thrombectomy in populations beyond the controlled environment of randomized studies.^8–13 However, all data in the literature refer to developed countries, where patients have access to comprehensive and standardized healthcare that covers treatment through to the rehabilitation process.

The recently published RESILIENT trial demonstrated the safety and efficacy of thrombectomy in the Brazilian population, whose majority do not have access to the thrombectomy. ¹⁴ Broadly speaking, patients who experienced AIS due to LVOs and who fail to meet inclusion criteria of RCTs on thrombectomy have poorer outcomes relative to those patients treated in the RCTs.^8–13 Therefore, the outcomes of thrombectomy in a real-world population treated in daily care practice in developing countries remain poorly studied. Whether thrombectomy performed under limited facilities like monoplane angio-suite, single thrombectomy devices, and standard guiding catheters, and no intensive rehabilitation program would result in good clinical outcomes is unknown.

The objectives of this study were to evaluate the results of mechanical thrombectomy, and the independent predictors of the main outcomes of thrombectomy, in a Brazilian registry of patients treated at a single University Hospital in Brazil.

Materials and methods

We analyzed retrospectively collected data between June 2011 and January 2020 from a consecutive series of patients with AIS caused by LVO of the anterior circulation (internal carotid artery, M1 and M2 segments of the middle cerebral artery, or anterior cerebral artery), isolated or in association (tandem), who underwent endovascular treatment with mechanical thrombectomy in a Brazilian University Hospital, which belongs to the Universal Public Health Care System. During this period, 657 patients were treated for thrombectomy, 12 of whom were excluded because they were treated and cross-referenced to other hospitals and clinical follow-up data could not be obtained. A further 80 patients were excluded who had occlusion of vessels in the posterior circulation Ultimately, 565 patients with occlusion of the anterior circulation were included. All data were collected during routine clinical care. This study was approved by the institution’s ethics committee, and the review board dispensed the need to obtain written informed consent from participants.

Clinical and radiological evaluation

Patients were selected for endovascular treatment according to the acute stroke protocol. ¹⁵ A brain computed tomography (CT) scan was performed at admission followed by additional imaging to identify the location of the artery occlusion, which was largely achieved with computed angiotomography (CTA). The resultant images were evaluated by medical teams from general radiology, vascular neurology, and interventional neuroradiology, and all patients were examined by certified vascular neurologists.

Clinical data collected in this study include gender, age, history of arterial hypertension, hyperlipidemia, coronary disease, atrial fibrillation, diabetes, chronic kidney disease, National Institutes of Health Stroke Scale (NIHSS) score, modified Rankin Scale (mRS; range: 0 [no symptoms] to 6 [death]), Alberta Stroke Program Early CT Score (ASPECTS), intravenous recombinant tissue plasminogen activator (rtPA), site of anterior circulation occlusion, time of symptom onset, number of patients admitted with an unknown time window, time from symptom onset to arterial puncture, puncture performed 6 h after symptom onset, and procedure time (interval between arterial puncture until maximum revascularization of the affected vascular territory). Recanalization was defined as an extended Treatment in Cerebral Ischemia (eTICI) score of 2b or 3.

Patient selection

In general, endovascular treatment by mechanical thrombectomy was indicated for patients with a NIHSS score ≥6, ASPECTS ≥6, occlusions of large vessels of the anterior circulation isolated or in association (tandem), and <6 h since the onset of symptoms or the last time the patient has been seen asymptomatic. Cases that presented within 6 h and 24 h of symptom onset, or cases that presented at an unknown time of symptom onset, were treated by thrombectomy if they had ASPECTS ≥6 or if they had favorable results after evaluation of perfusion by CT, when available, following the DAWN or the DEFUSE trials criteria.^3,7 There were no exclusion criteria for the indication of mechanical thrombectomy based exclusively on the patient’s age, previous clinical conditions, or specific mRS score.

Patients were assessed for eligibility for intravenous rtPA thrombolysis by the stroke team, and treatment was indicated for patients admitted within 4.5 h of symptom onset, based on current criteria and protocols. ¹⁵

Endovascular procedure

All mechanical thrombectomies were performed by one of the members of a team of five interventional neuroradiologists, with 24/7 coverage for emergency care. Anesthesiology team support was requested for all patients. The indication between conscious sedation or general anesthesia was defined on a case-by-case basis. The target mean blood pressure values ​​were ≥100 mmHg if there was no rTPA infusion, and <180 × 105 mmHg if rTPA was performed. Intravenous heparin was infused (3000 IU bolus) for cases that did not receive rTPA. No patients required intra-arterial thrombolysis with rTPA.

Arterial accesses, devices, and thrombectomy techniques were defined by the interventionist neuroradiology team. Arterial accesses were obtained with 8 or 9F femoral sheaths. The access catheters used were conventional guides 7 or 8F, long sheaths Destination 7F (Terumo), NeuronMax 088 (Penumbra, Alameda, CA), balloon guide catheter (BGC) Cello 8 or 9F (Medtronic; Irvine, CA), or Merci 8F (Stryker; Kalamazoo, MI). The catheters were continuously perfused with a 10 mg solution of Milrinone diluted in 1000 mL of saline 0.9%. The thrombectomy devices used were the stentriever Solitaire AB, Solitaire FR, Solitaire Platinum (Medtronic; Irvine, CA), Trevo XP Stentriever (Stryker; Kalamazoo, MI), and ACE aspiration system (Penumbra, Oakland, CA). Usually, at least three passes were made with the first-choice device before performing device association (stentriever and aspiration catheter). Both aspiration catheters and BGCs were only used during the late period of the study, after being made available at our institution. A total of ≥6 passes were performed before the procedure was deemed a failure in recanalization. For cases with occlusion in the proximal segment of the carotid artery, angioplasty with X-Act (Abbott Vascular) or Wallstent (Boston Scientific Target, Fremont, CA) was performed. The procedure time was defined as the duration of arterial puncture until the maximum eTICI obtained.

Outcome measures

Control CT scans of the brain were performed within 12–48 h of treatment, or at other times when necessary. Symptomatic intracranial hemorrhage (sICH) was defined based on the criteria of the European Cooperative Acute Stroke Study II (ECASS II); that is, when parenchymal hemorrhage in the infarcted region resulted in a ≥4-point increase on the NIHSS scale in the period of 48 h.

Clinical assessments were performed at admission, during hospitalization and at the three-month follow-up. Patients who were not present for their 3-month follow-up appointment or who were unable to make phone calls had their mRS score established at the time of hospital discharge or most recent patient evaluation. A good clinical outcome was considered functional independence for daily activities (mRS ≤2).

Statistical analysis

Categorical variables were presented in numerals and percentages. The mean and standard deviation (SD), median, and interquartile range (IQR) were calculated for the description of statistical data for the numerical variables, when appropriate. The Student’s t-test, chi-square test (χ²) and Fisher’s exact test were used, as appropriate.

Univariate and multivariate analyses were performed to find the predictors for eTICI of 2b–3, good neurological outcomes at 3 months post-procedure (mRS = 0–2), sICH and mortality (Tables 3 and 4). The Stata program v.16.1 (StataCorp, College Station TX) was used for statistical analysis, and p values <0.05 were considered significant.

Table 3.

Logistic regression analysis: Predictors of recanalization and good neurologic outcome at three months.

	Univariate (unadjusted)Odds ratio (95% CI)	p value	Multivariate (adjusted) Odds ratio (95% CI)	p value
Predictors of recanalization (eTICI=2b-3)
Baseline NIHSS	0.95 (0.92–0.99)	0.029	0.99 (0.94–1.04)	0.742
Procedure time	0.97 (0.97–0.98)	<0.001	0.99 (0.98–1.00)	0.095
Number of passes	0.50 (0.43–0.58)	<0.001	0.53 (0.45–0.63)	<0.001
Aspiration	1.50 (1.04–2.16)	0.027	1.19 (0.78–1.82)	0.399
Carotid occlusion	0.59 (0.37–0.94)	0.026	1.26 (0.67–2.35)	0.467
M2 occlusion	7.10 (1.70–29.53)	0.007	14.2 (1.74–115.7)	0.013
Predictors of good outcome (mRS=0–2) at 3 months
Age	0.96 (0.95–0.97)	<0.001	0.95 (0.93–0.96)	<0.001
Diabetes	0.51 (0.32–0.81)	0.005	0.49 (0.28–0.86)	0.013
Baseline NIHSS	0.90 (0.88–0.93)	<0.001	0.93 (0.89–0.96)	<0.001
Baseline ASPECTS	1.35 (1.22–1.50)	<0.001	1.32 (1.15–1.52)	<0.001
Carotid occlusion	0.62 (0.44–0.88)	0.008	0.72 (0.44–1.17)	0.193
M2 occlusion	2.25 (1.35–3.74)	0.002	1.33 (0.65–2.71)	0.425
Number of passes	0.70 (0.62–0.78)	<0.001	0.84 (0.65–1.08)	0.181
Procedure time	0.98 (0.97–0.98)	<0.001	0.99 (0.98–1.00)	0.098
eTICI=2b–3	9.5 (4.49–20.11)	<0.001	5.9 (1.31–26.6)	0.021
Balloon guiding catheter	2.45 (1.24–4.83)	0.009	2.04 (0.93–4.46)	0.073
sICH	0.05 (0.01–0.21)	<0.001	0.06 (0.01–0.29)	<0.001

Note: (mRS) modified Rankin Scale; (NIHSS) National Institutes Health Stroke Score; (ASPECTS) Alberta Stroke Program Early Computed Tomography Score; (eTICI) extended Treatment In Cerebral Infarction scale; (sICH) symptomatic Intra Cranial Hemorrhage.

Table 4.

Logistic regression analysis: predictors of symptomatic intracranial hemorrhage and mortality at three months.

	Univariate (unadjusted) Odds ratio (95% CI)	p value	Multivariate (adjusted) Odds ratio (95% CI)	p value
Predictors of sICH
Baseline NIHSS	1.07 (1.02–1.13)	0.004	1.06 (1.01–1.12)	0.013
Baseline ASPECTS	0.81 (0.69–0.93)	0.005	0.83 (0.72–0.97)	0.023
Predictors of mortality at 3 months
Age	1.03 (1.01–1.04)	<0.001	1.03 (1.01–1.05)	0.001
High blood pressure	1.65 (1.01–2.68)	0.042	0.75 (0.40–1.40)	0.376
Diabetes	2.13 (1.33–3.40)	0.001	2.60 (1.48–4.55)	0.001
Chronic kidney disease	2.55 (1.29–5.03)	0.007	1.14 (0.45–2.89)	0.773
Baseline NIHSS	1.13 (1.08–1.17)	<0.001	1.11 (1.06–1.17)	<0.001
Baseline ASPECTS	0.83 (0.75–0.92)	0.001	0.92 (0.80–1.05)	0.259
General anesthesia	1.72 (0.99–2.98)	0.05	1.49 (0.79–2.81)	0.213
Number of passes	1.26 (1.14–1.39)	<0.001	1.13 (0.98–1.30)	0.077
eTICI 0-2a	3.57 (2.16–5.89)	<0.001	2.48 (1.21–5.11)	0.013
sICH	5.48 (2.94–10.23)	<0.001	4.50 (2.16–9.36)	<0.001

Note: (mRS) modified Rankin Scale; (NIHSS) National Institutes Health Stroke Score; (ASPECTS) Alberta Stroke Program Early Computed Tomography Score; (eTICI) extended Treatment In Cerebral Infarction scale; (sICH) symptomatic Intra Cranial Hemorrhage.

Results

The baseline characteristics and diagnostic imaging data for the 565 patients included in this study are shown in Table 1. The data indicate a slight predominance of males (53%), a mean age of 66.9 years (9–97; SD ± 15.3), and a mean NIHSS base score on admission of 17.2 (0–35; SD ± 6.27). The average ASPECTS was 8 (2–10; IQR = 3), and 91.0% of patients scored ≥6. Half of the patients (n = 283; 50.1%) received intravenous rTPA, and a significant prevalence was found of occlusions located in the M1 segment of the middle cerebral artery (n = 261; 46.2%). The mean period between symptom onset and puncture was 296.4 min (3–2880; SD ± 189.1), with 95 patients (16.9%) having an unknown period. For 72 patients (12.7%), the puncture was performed 6 h after symptom onset.

Table 1.

Patient’s baseline clinical and imaging data.

	N= 565
Male (n, %)	297 (53)
Female (n, %)	268 (47)
Age (mean, range, SD)	66.9 (9–97, SD±15.3)
High blood pressure (n, %)	411 (72.8)
Hyperlipidemia (n, %)	184 (32.6)
Coronary disease (n, %)	109 (19.3)
Atrial fibrillation (n, %)	189 (33.5)
Diabetes (n, %)	110 (19.4)
Chronic kidney disease (n, %)	36 (6.4)
Baseline NIHSS (mean, range, SD)	17.2 (0–35, SD±6.27)
Baseline mRS (median, range, SD)	0 (0–5, IQ=0)
Baseline ASPECTS (median, range, SD)	8 (2–10, IQ=3)
6–10 (n, %)	514 (91.0)
3–5 (n, %)	48 (8.5)
0–2 (n, %)	3 (0.53)
Intravenous rTPA (n, %)	283 (50.1)
Vessel occluded
Internal carotid artery (n, %)	131 (23.2)
Carotid tandem (n, %)	91 (16.1)
M1 (n, %)	261 (46.2)
M2 (n, %)	71 (12.6)
M3 (n, %)	1 (0.2)
Anterior cerebral artery	2 (0.4)
Bilateral occlusions	8 (1.4)
Unknown time window (n, %)	95 (16.9)
Time from symptoms onset to puncture (min, mean, range, SD)	296.4 (3–2880, SD±189.1)
Puncture > 6 hours from symptoms onset	72 (12.7)
Procedure time (min, mean, range, SD)	61.4 (5–255, SD±36.8)

Note: (mRS) modified Rankin Scale; (NIHSS) National Institutes Health Stroke Score; (ASPECTS) Alberta Stroke Program Early Computed Tomography Score; (rTPA) recombinant Tissue Plasminogen Ativactor.

General anesthesia was performed in 78.9% of the procedures. The average duration of the procedure was 61.4 min (5–255; SD ± 36.8). Regarding recanalization rates, an eTICI score ranging from 2b to 3 was achieved in 85.5% of patients, and an eTICI score ranging from 2c to 3 was achieved in 58.4% of patients. A maximum recanalization was obtained after the first pass in 30.1% of patients. As for the type of device, the stent retriever alone was adopted in 66.1% of patients, and the thrombus aspiration system alone in 19.5% of patients. Three months after the procedure, a favorable neurological result (mRS ≤2) was obtained in 43.5% of patients, 20.5% of whom obtained an excellent result (mRS 0–1). The rate of sICH was 8.1%, and the overall mortality rate was 22.1% (Table 2).

Table 2.

Thrombectomy data.

	N=565
General anesthesia (n, %)	446 (78.9)
Thrombectomy technique
Stent retrieval (n, %)	373 (66.0)
Aspiration (n, %)	110 (19.5)
Stent retrieval and aspiration (n, %)	61 (10.8)
Carotid angioplasty stenting only (n, %)	10 (1.8)
Microcatheter only (n, %)	8 (1.4)
Intracranial angioplasty stenting (n, %)	3 (0.5)
Balloon guiding-catheter (n, %)	39 (6.9)
Recanalization (eTICI) (n, %)
eTICI=0 or 1	46 (8.1)
eTICI=2a	36 (6.4)
eTICI=2b	153 (27.1)
eTICI=2c	89 (15.8)
eTICI=3	241 (42.7)
eTICI=2b-3	483 (85.5)
eTICI=2c or 3	330 (58.4)
First-pass recanalization (n, %)	170 (30.1)
Number of passes (median, range, IQ)	2 (0–10, IQ=3)
NIHSS at 24h (mean, range, SD)	14.2 (0–35, SD±10.1)
Symptomatic intracranial hemorrhage (n, %)	46 (8.1)
mRS at 3 months (median, range, IQ)	0–4 (IQ=0)
mRS=0 (n, %)	61 (10.8)
mRS=1 (n, %)	57 (10.1)
mRS=2 (n, %)	128 (22.7)
mRS=3 (n, %)	47 (8.3)
mRS=4 (n, %)	90 (15.9)
mRS=5 (n, %)	57 (10.1)
mRS=0–1 (n, %)	116 (20.5)
mRS=0–2 (n, %)	246 (43.5)
Mortality (n, %)	125 (22.1)

Note: (mRS) modified Rankin Scale; (NIHSS) National Institutes Health Stroke Score; (eTICI) extended Treatment In Cerebral Infarction scale; (sICH) symptomatic Intra Cranial Hemorrhage.

Comparisons were made of the percentage values of successful recanalizations (eTICI = 2b–3), functional independence at 3 months post-procedure (mRS ≤2), sICH and mortality rates at 3 months post-procedure; between data from the present study and the main clinical trials previously published (Table 5).^{1–10,12–14,16}

Table 5.

The outcomes of the present study compared to the largest trials and registries on thrombectomy for acute ischemic stroke due to anterior circulation large vessel occlusion.

Trial (Year)	Recanalization (mTICI=2b/3) N (%)	mRS 0, 1 or 2 at 3 monthsN (%)	sICH N (%)	Mortality at 3 monthsN (%)
MR-CLEAN(2014)	115/196 (58.7)p<0.001	76 (32.6)p=0.005	14 (6.0)p=0.006	49 (21.0)p=0.796
ESCAPE(2015)	113/156 (72.4)p<0.001	87 (52.7)p=0.048	6 (3.6)p=0.070	17/164 (10.4)p=0.001
SWIFT-PRIME(2015)	73/83 (88.0)p=0.575	59 (60.2)p=0.003	0 (0)p=0.006	9 (9.2)p=0.004
EXTEND-IA(2015)	25/29 (86.2)p=0.851	25 (71.4)p=0.002	0 (0)p=0.152	3 (8.6)p=0.090
REVASCATE(2015)	67 (65.0)p<0.001	45 (43.7)p=0.989	2 (1.9)p=0.041	19 (18.4)p=0.476
DAWN trial(2017)	90 (84.1)p=0.945	52 (48.6)p=0.398	6 (5.6)p=0.479	20 (18;7)p=0.501
DEFUSE 3 trial(2017)	69 (75.0)p=0.025	41 (44.5)p=0.954	6 (6.5)p=0.741	13 (14.1)p=0.106
SONIIA registry(2015)	327 (79.7)p=0.043	204 (49.7)p=0.066	10 (2.4)p<0.001	59 (14.4)p=0.002
TRACK registry(2017)	505 (80.3)p=0.042	277 (47.9)p=0.929	44 (7.1)p=0.516	106 (19.8)p=0.024
STRATIS registry(2017)	724/824 (87.9)p=0.133	512/906 (56.5)p<0.001	12/841 (1.4)p<0.001	142 (14,4)p<0.001
German registry(2019)	1857/2236 (83.0) p=0.156	732/1997 (36.7)p=0.003	349/2637 (13.2)p=0.001	570/1997 (28.5)p=0.003
RESILIENT(2020)	91 (82.0)p=0.522	39 (35.1)p=0.121	8 (7.2)p=0.884	27 (24.3)p=0.708
Present Study (Fornazari et al., this study)	483 (85.5)	246 (43.5)	46 (8.1)	125 (22.1)

After univariate logistic regression analyzes, the main predictors for mRS ≤2 were eTICI 2b–3, sICH, use of a BGC, M2 occlusion, diabetes, and initial ASPECTS. The main predictors found for successful recanalization (eTICI 2b–3) were M2 occlusion, number of passages, aspiration, and carotid occlusion. However, after multivariate analysis, the main variables that remained as independent predictors for mRS ≤2 were eTICI 2b–3, sICH, diabetes, and initial ASPECTS. For eTICI 2b–3, the variables that remained as independent predictors were M2 segment occlusion and number of passages (Table 3). The values ​​of the baseline ASPECTS and baseline NIHSS scores were the only independent predictors for the occurrence of sICH, in both analyses.

Many variables were found to be predictors of 90-day mortality in the univariate analysis. However, only age, diabetes, initial NIHSS score, eTICI 2b–3, and sICH were confirmed as independent predictors, the latter being the most significant (OR 4.50; 95% CI 2.16–9.36; Table 4).

Discussion

Thrombectomy is the gold standard for treatment of AIS caused by large vessel occlusion of the anterior circulation.^1–5 The impressive results of thrombectomy in the treatment of stroke motivated health care systems around the world to allocate resources and efforts to implement this indispensable tool in stroke care. In our service, thrombectomy has been performed since 2011; however, the number of procedures has increased progressively over the years, with approximately 80% of the cases reported in our sample being treated after 2016.

Although RCTs differed in some of their selection criteria, they predominantly include relatively homogeneous populations, establishing an environment conducive to a high therapeutic benefit.^1–7,14,16 In addition, other studies have also demonstrated the efficacy and safety of thrombectomy in a daily clinical practice, following more pragmatic selection criteria.^8–13 The results presented in these studies were based on data from national registries of thrombectomy,^9–11 or to comprehensive population registries,^8,12,13 all of which refer to countries with significant financial resources, holders of health services that are equipped with modern technological facilities and that offer comprehensive and standardized care. The clinical results of thrombectomy in developing countries—such as Brazil—remained scarce until the publication of the RESILIENT study.^14,17

Many factors can decisively interfere in the clinical outcome of stroke victims, from pre-hospital care to a complete rehabilitation process. Thus, doubts remain as to the generalizability of the results of thrombectomy obtained in comprehensive stroke center of developed countries.^18,19 Whether the favorable results can be replicated in scenarios with scarce resources remain unknown. In the present study some limitations that could reduce the benefits of thrombectomy were the absence of an intensive rehabilitation program after discharge,^19,20 the use of a monoplane angio-suite, ²¹ the use of a single thrombectomy device, ²² and the use of standard guiding catheters in the majority of the cases. ⁹

Biplane angio-suites have been widely recommended for neuro-endovascular procedures because, compared to the monoplane angio-suites, it allows for a better visualization of the vessels that would reduce the risks of vessel perforations. Therefore, the use of a monoplane angio-suite in this study could be a potential source of complications during thrombectomy. However similar outcomes have been recently reported regarding thrombectomy performed with both mono- or biplane angio-suites. ¹⁹ Another limitation of our center was the use of a single thrombectomy device as a first line strategy. The association of stentrievers and aspiration catheters that results in higher recanalization rates, ²² was indicated in this study for only cases that a single device failed to recanalize the vessel (10.8% of the cases). In the multivariate analysis the use of two devices (stentriever and aspiration catheters) was not a predictor of recanalization. This might be explained because the use of two devices was indicated for only refractory cases, and not as a first-line strategy.

Another limitation of our center was the use of standard guiding catheters to 93% of the cases. Balloon-guiding catheters, that have been associated with better outcomes compared to standard guiding catheters, ⁹ were used to only 7% of the cases in this study. The use of BGC was not a predictor for recanalization in the multivariate analysis. However, BGC had a trend toward good neurologic outcome (Table 3). It is possible that a large sample of patients treated using combined devices as a first line strategy, or using BGC, would improve the rates of recanalization and good neurologic outcomes. On the other hand, this study showed that even under all the limitations described above, thrombectomy for AIS resulted in relatively similar outcomes compared to previous trials and registries of developed countries.

Regarding primary outcomes, the results obtained in our study broadly corroborated those found in the literature (Table 5). Compared to our initial experience with thrombectomy, the present study resulted in a significant higher recanalization rate, which may be explained by a learning curve effect over the time. ¹⁷ Our recanalization rate of 85.5% (eTICI = 2b–3) was similar to the rates reported in recent studies, namely 75.0–87.9%.^6–14,16 That a favorable neurological outcome (mRS ≤2) was obtained in 43.5% of the patients in our study was statistically inferior to the data of some of the trials (ESCAPE, SWIFT-PRIME, EXTEND-IA)^2–4 and to those found in the STRATIS registry, ¹⁰ which obtained mRS ≤2 rates ranging from 52.7% to 71.4%. This outcome was statistically superior to those obtained in the MR-CLEAN, DIRECT-MT, and German Registry studies (mRS ≤2 rates between 32.6% and 36.7%).^1,12,16

The rate of sICH of 8.1% found in our study was slightly higher than the rates reported in some of the RCTs (MR-CLEAN, SWIFT-PRIME, DIRECT-MT)^1,3,16 and in the SONIIA and STRATIS,^8,10 which ranged from 0–6%; however, our rate remained lower than that reported in the German Registry (13.2%). ¹² Furthermore, the overall mortality rate of 22.1% in our study was statistically lower than that reported in the German Registry (28.5%) ¹² and higher than that of the ESCAPE and SWIFT-PRIME trials (9.2% and 10.4%, respectively),^2,3 as well as the SONIIA, TRACK records, and STRATIS (14.4–19.8%).^8–10

As previously reported, the use of a BGC was an independent predictor of good neurological outcome in 3 months. ⁹ Our study, the use of a BGC was not found to be a predictor of recanalization nor of good outcome (mRS ≤2), which likely resulted from the low proportion of procedures in which the BGC was used (6.9%), since it was only recently made available in our service.

This study had limitations, one of which was its lack of control group and of independent data monitoring and analysis. However, the overwhelming majority of evaluations were performed by more than one vascular neurologist. Another limitation was the study’s retrospective design, which suffers from intrinsic biases. The limitations described above limit the generalizability of the results.

Conclusion

This study demonstrates the efficacy and safety of mechanical thrombectomy for treatment of patients with AIS of the anterior circulation in real-life conditions under limited facilities and resources. The results of the present study were relatively similar to those of large trials and population registers of developed countries.