Article Text
Abstract
Background The first pass effect has been recently reported as a predictor of good clinical outcome after stroke thrombectomy. We evaluate the first pass effect on outcome and the influence of revascularization in these and other patients.
Methods We performed a retrospective analysis of a prospectively maintained database on anterior cerebral circulation stroke thrombectomy cases from April 2012 to April 2018. Data compiled included patient demographics, presenting National Institutes of Health Stroke Scale score, vessel occlusion site, thrombectomy procedural details, and 90 day modified Rankin Scale scores.
Results 205 patients were included. The numbers of patients who underwent one, two, three, four, five, and six passes were 69, 70, 55, 9, 1, and 1, respectively. Successful revascularization was achieved in 87%, 83%, and 64% of patients in the one, two, and 3 or more passes groups, respectively (p=0.002). Good functional outcome was inversely correlated with number of passes when comparing the one, two, and three or more passes groups (54%, 43%, 29%; P=0.014). In patients with full revascularization, there was no significant difference in good functional outcomes between the one, two, and three or more passes groups (64%, 65%, 50%; P=0.432). Number of passes was not an independent negative predictor of good clinical outcome (OR 1.66, 95% CI 0.82 to 3.39; P=0.165).
Conclusions First pass thrombectomy patients have better functional outcomes compared with beyond-first pass patients. This effect is related at least in part to a higher rate of revascularization in one pass patients. Revascularization beyond the first pass should continue to be the goal of stroke thrombectomy.
- thrombectomy
- technique
- stroke
- stent
Statistics from Altmetric.com
Introduction
Several landmark randomized clinical trials have established mechanical thrombectomy as the standard of care in selected patients presenting with large vessel occlusion related acute ischemic stroke.1 2 Major technical advances in mechanical thrombectomy have resulted in significantly improved rates of revascularization in these patients. Achieving complete or near complete reperfusion has been associated with improved clinical outcomes and reduced adverse effects.1
Recently, there has been investigation into how many attempts, or passes, are required during mechanical thrombectomy and whether the number of passes influences patient outcome.3 4 According to a recent report, complete or near complete reperfusion is obtained from the first pass in approximately 25% of cases, and these patients have significantly better clinical outcomes in comparison with other patients.3 However, the influence of revascularization when comparing these groups of patients has not been reported.
In this article, we again evaluate this recently described metric, the first pass, and its influence on outcomes after stroke thrombectomy. We compare a one pass cohort with other cohorts of patients: >1 pass, 2 passes, and ≥3 passes. Furthermore, we investigate the rate of revascularization in each of these groups and seek to assess if lower rates of revascularization may be a factor in the worse clinical outcomes seen in higher pass patients. Here the frequency of patients undergoing one pass, two passes, and three or more passes is evaluated, as well as the likelihood of revascularization in these groups and clinical outcomes.
Methods
Local institutional review board approval was obtained for data collection. The data that support the findings of this study are available from the corresponding author on request. A retrospective review of a prospectively maintained database was performed. Data collected included demographics, clinical presentation, procedural details, angiographic data, and clinical outcome data on consecutive anterior cerebral circulation acute ischemic stroke patients treated at a single university hospital’s comprehensive stroke center from April 2012 to April 2018. Angiographic revascularization grade was reviewed by two independent neurointerventional surgeons with 8 and 3 years of experience, respectively. Discrepancies were resolved by consensus. Reviewers were blinded to one another’s scores, patient outcomes, and original angiographic reports. Successful revascularization was defined as TICI 2b, 2c, or 3. Full revascularization was defined as TICI 2c or 3. Clinical and angiographic outcomes were compared among groups of patients who underwent one thrombectomy pass and more than one pass. Clinical and angiographic outcomes were additionally compared among groups of patients who underwent one pass, two passes, and three or more passes. A pass was defined as either a stent retrieval thrombectomy or aspiration thrombectomy without a stent retriever. The primary clinical outcome was the modified Rankin Scale (mRS) score at 90 days, as assessed by a neurologist or neurology nurse who did not perform the thrombectomy procedure; mRS scores of 0–2 were considered good clinical outcomes.
Procedures were performed using local protocols and standard of care. Patient selection was performed by utilizing a non-contrast CT scan, CT angiography, and/or MRI/MR angiography. Institutional exclusion criteria for mechanical thrombectomy during the study period were: (1) mild stroke symptoms, defined as an admission National Institutes of Health Stroke Scale (NIHSS) score <6; (2) the presence of a large completed territorial infarction by non-contrast CT, defined as an Alberta Stroke Program Early CT score <5, or by MRI, defined as an infarction volume of >70 mL of brain; (3) functional dependence prior to stroke onset, defined as a pre-stroke mRS score of >3; or (4) target territory intracranial hemorrhage. Procedures were performed on a biplane neuroangiography suite (Artis Zee; Siemens, Munich, Germany) by one of four attending neurointerventional surgeons. Procedures were performed, in most cases, under moderate conscious sedation using intravenous fentanyl and midazolam or, in a minority of cases, under general anesthesia.
Statistical analysis
We assessed the distribution, frequency, and mean values of all clinical and demographic variables. Covariates were compared between patients grouped according to the number of thrombectomy passes using the χ2 test, Student t test, and ANOVA. Procedural details and clinical outcomes were also compared among patients grouped according to the number of thrombectomy passes. Logistic regression analysis was used to assess which variables were independently associated with a good clinical outcome at 90 days. All clinical and procedural variables with P values <0.15 in univariate analysis were included in a final multivariate logistic regression model. We assessed for potential interactions among all independent variables prior to inclusion in the final model. A receiver operating characteristic analysis was performed to measure the predictive power of the final regression model. All analyses were conducted with STATA/SE (V.15 for Windows, StataCorp LLC). A P value <0.05 was considered statistically significant.
Results
Frequency and characteristics of patients grouped by number of passes
A total of 212 patients underwent anterior circulation intracranial thrombectomy for acute ischemic stroke during the study period. Five patients were lost to follow-up and two patients had missing data. Hence 205 patients were included in our analysis. The number of patients who underwent a total of one, two, three, four, five, and six passes were 69, 70, 55, 9, 1, and 1, respectively. A total of 421 thrombectomy passes were performed on 205 patients (mean 2.05 passes/patient). Aspiration thrombectomy alone without a stent retriever was used a total of 37 times in 31 cases in the last 2 years of the study period. All remaining 384 thrombectomy passes were performed using a stent retriever.
No difference was noted in gender, age, comorbidities, site of occlusion, laterality, stroke severity, or use of intravenous thrombolysis between the one, two, and three or more passes groups (table 1). There was a trend towards more internal carotid artery (ICA) occlusions and higher stroke severity in the three or more passes group. Median time of symptom onset to revascularization was not statistically different in the one, two, and three or more passes cohorts (6.11, 5.63, and 6.31 hours, respectively; P=0.297) (table 2). Median time from groin arteriotomy to revascularization was significantly correlated with the number of passes in the one, two, and three or more passes groups (38, 60, and 90 min; P=0.001) (table 2).
Angiographic and clinical outcomes in the one, two and three or more pass cohorts
A total of 160 of 205 patients (78%) achieved successful revascularization (TICI 2b/2c/3) and 106 patients (52%) achieved full revascularization (TICI 2c/3). There was a significant difference in the rate of successful revascularization in the one, two, and three or more passes groups (87%, 83%, and 64%, respectively; P=0.002) (table 2). Good functional outcomes were more likely in a dichotomized first pass group versus non-first pass group comparison (54% vs 36%; P=0.016) (table 2). Good functional outcome was also inversely correlated with number of passes when comparing the one, two, and three or more passes groups (54%, 43%, and 29%; P=0.014) (table 2). In patients with successful revascularization, there was no significant difference in good functional outcomes between the one, two, and three or more passes groups (57%, 50%, and 40%; P=0.274). In patients with full revascularization, there was no significant difference in good functional outcomes between the one, two, and three or more passes groups (64%, 65%, and 50%; P=0.432).
Mortality was less likely in the first pass group versus the non-first pass group (12% vs 28%; P=0.008) (table 2). Mortality was also correlated with the number of passes when comparing the one, two, and three or more passes groups (12%, 26%, and 30%; P=0.024) (table 2). In patients with revascularization or full revascularization, however, there was no significant difference in mortality between the one, two, and three or more passes groups: 13%, 24%, and 26%, respectively, in revascularization patients (P=0.204) and 9%, 8%, and 21%, respectively, in full revascularization patients (P=0.246).
On univariate analysis of 205 patients, age, revascularization grade, history of atrial fibrillation, NIHSS score, and thrombectomy attempts beyond the first pass were significantly associated with functional outcome (P<0.05) (table 3). On multivariate logistic regression, only older age (OR 1.04, 95% CI 1.01 to 1.07; P=0.01) and higher NIHSS (OR 1.11, 95% CI 1.04 to 1.19; P=0.002) were independent predictors of poor functional outcome, while only TICI 2c (OR 0.04, 95% CI 0.01 to 0.24; P<0.001) and TICI 3 revascularization grades (OR 0.09, 95% CI 0.02 to 0.46; P=0.004) were associated with a reduced odds of poor functional outcome. The number of thrombectomy attempts was not a predictor of functional outcome (OR 1.66, 95% CI 0.82 to 3.39; P=0.165).
Discussion
The first pass effect (FPE) has been shown to be a predictor of good functional outcome in a recent report3; our data are similar to the results from this prior report in multiple respects. In the prior report, FPE was defined as single pass/use of the device without rescue therapy and complete or near complete revascularization of the large vessel occlusion and its downstream territory (TICI 3 or TICI 2c). If defined in this way, FPE was present in 25% of cases in the prior report and in 22% of cases in our report. FPE patients had good functional outcomes at 90 days in 61% of cases in the prior report and in 64% of cases in our report ().
Our report addresses two additional important considerations which have not been previously evaluated: the rate of successful revascularization and its influence on outcomes in the first pass cohort versus other, higher pass cohorts. Our data show a significant difference in the rate of successful revascularization among the one, two, and three or more passes groups (87%, 83%, and 64%, respectively). Thrombectomy attempt beyond the first pass was associated with worse outcomes in our univariate analysis. However, in multivariate analysis, when accounting for lower recanalization rates in the higher pass cohorts, number of thrombectomy attempts was not an independent negative predictor of functional outcome. Moreover, when excluding patients without revascularization and evaluating only those patients who achieved successful revascularization (TICI 2b/2c/3) and full revascularization (TICI 2c/3), the first pass favorable effect on clinical outcome was not statistically significant. In multivariate analysis, only age, presenting NIHSS, and revascularization grade were significant predictors of functional outcome; this is consistent with the current stroke literature.1 5
The recent prior study demonstrated an association of FPE with both the use of balloon guide catheters and middle cerebral artery occlusions.3 Our data similarly demonstrate a higher percentage of ICA occlusions in the three or more passes group (29%) compared with the one and two passes groups (16% and 17%, respectively), although this did not reach statistical significance (P=0.097). While we did not use balloon guide catheters, both the rate of FPE and likelihood of FPE resulting in good outcomes in our study were very similar to the previous study. Firstline thrombectomy technique has also been reported as a factor influencing FPE. One recent study demonstrated FPE in 26%, 17%, and 14% of cases when comparing a primary combined approach, aspiration alone, and stent retriever alone, respectively.6 Longer stent retriever length has also been correlated with a higher rate of FPE.7 While we included initial stent retriever length in our data, our study was not designed to make conclusions on these associations.
It should be noted that while not statistically significant, there was numerically a lower percentage of patients with good outcomes and higher percentage mortality in our higher pass cohort even when assessing only those with revascularization and full revascularization. This discrepancy may be accounted for in part by the relatively higher NIHSS scores, higher rates of ICA terminus occlusions, and/or higher rates of TICI 2b scores (as opposed to TICI 2c/3) in the three or more passes revascularization group. These are known predictors of worse outcome compared with lower NIHSS, smaller/middle cerebral artery occlusions, and full TICI 2c/3 revascularization.1 8 9 This discrepancy, however, may also be related to the relatively small sample size and power of our study. Future larger studies addressing these high pass number thrombectomy patients would be helpful in elucidating potentially multiple factors involved in their clinical outcomes. As such, we believe great care must be taken with every thrombectomy attempt and that longer and/or more challenging procedures may be associated with a higher risk. Two additional studies suggesting that higher pass number is associated with worse clinical outcomes have included patients with up to 8–10 passes.4 10 It has been suggested that greater number of stent retriever passes may incite vascular injury which could be associated with worse outcome.3 While our data do not support such a specific hypothesis, we agree on pursuing devices and techniques that minimize patient manipulation.
Our study has limitations. It is a single center study with a retrospective design. A single center series, however, offers homogeneity with regards to procedural and periprocedural care, and several components of our data are maintained prospectively. Dichotomization of passes into first pass and beyond first pass cohorts for our univariate and multivariate analyses ignores potential shifts in outcomes across a spectrum of thrombectomy attempts. Our triage and techniques have evolved over the study period, although this is unlikely to significantly impact the study results aimed at evaluating number of passes on 90 day clinical outcomes. Moreover, our patient cohorts were matched across numerous variables. Our thrombectomy passes were somewhat heterogeneous; however, the great majority (91%) utilized stent retrievers while a minority (9%) used aspiration alone. We do not have data on clot histology to characterize the relationship of clot type with thrombectomy attempts.
In conclusion, one pass stroke thrombectomy patients have better functional outcomes in comparison with patients requiring additional passes. This favorable first pass effect, however, appears to be related at least in part to a higher rate of successful revascularization in this group of patients compared with other patients. Revascularization beyond the first pass should continue to be the goal of stroke thrombectomy. Future larger studies would be helpful to further assess the impact of multiple variables on outcomes in high pass number patients.
Footnotes
Contributors All authors contributed to data collection, manuscript preparation, and/or review. GJ is the guarantor of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests GJ has research grant funding from Stryker Neurovascular and Microvention not directly related to the work presented here.
Ethics approval Local institutional review board approval was obtained for data collection and analysis.
Provenance and peer review Not commissioned; externally peer reviewed.
Correction notice Since this paper was first published online, minor changes have been made to tables 2 and 3.
Patient consent for publication Not required.