Background Procedural time in patients with acute ischemic stroke (AIS) undergoing mechanical thrombectomy may affect clinical outcomes. We performed a pooled analysis of the effect of procedural time on clinical outcomes using data from three prospective endovascular treatment trials.
Objective To examine the relationship between endovascular procedural time and clinical outcomes of patients with AIS following endovascular treatment.
Methods We analyzed data from SWIFT, STAR, and SWIFT PRIME studies, including baseline characteristics: National Institutes of Health Stroke Scale (NIHSS) score on admission, intracranial hemorrhage rates, and modified Rankin Scale score at 3 months. The Thrombolysis in Cerebral Infarction (TICI) scale was used to grade postprocedure recanalization. We recorded two procedural time intervals: (1) symptom onset to groin puncture and (2) groin puncture to angiographic recanalization. A multivariate analysis was performed using a logistic regression model to analyze predictors of unfavorable outcome.
Results We analyzed 301 patients who had undergone endovascular treatment and had near-complete or complete recanalization (TICI 2b or 3). At 3 months, 122 patients (40.5%) had unfavorable outcomes. The rate of favorable outcomes was significantly higher when the procedural time was <60 min compared with ≥60 min (62% vs 45%, p=0.020). Predictors of unfavorable outcome at 3 months were age (unit 10 years, OR=0.62, 95% CI 0.46 to 0.82, p<0.001), onset to groin puncture time (unit hour, OR=0.61, 95% CI 0.48 to 0.77, p<0.001), groin puncture to recanalization (unit 10 min, OR=0.89, 95% CI 0.80 to 0.99, p=0.032), baseline NIHSS score (20–28 vs 8–10, OR=0.17, 95% CI 0.05 to 0.62, p=0.018), and collaterals (OR=1.48, 95% CI 1.04 to 2.10, p=0.029).
Conclusion Procedural time in patients with stroke undergoing mechanical thrombectomy may be an important determinant of favorable outcomes in those with recanalization.
Statistics from Altmetric.com
Previous guidelines from the American Stroke Association recommended endovascular treatment for patients with acute ischemic stroke (AIS) presenting within 16 or 24 hours of symptom onset with an initial National Institutes of Health Stroke Scale (NIHSS) score ≥6.1 These recommendations were based on the results of recent clinical trials demonstrating reduced rates of death or disability among patients with large vessel occlusion treated with mechanical thrombectomy (MT) compared with those treated with intravenous recombinant tissue plasminogen activator (IV rtPA) or no treatment. Subsequent studies have focused on identifying various parameters that can be optimized to achieve the maximum benefit of endovascular treatment.2
Analysis of pooled data shows that a shorter time interval between stroke onset and the initiation of treatment leads to more favorable outcomes,3 demonstrating the importance of rapid endovascular treatment. However, considerable variation in procedural time can occur between patients and institutions,4 defined as the time from femoral arterial access to recanalization or procedure completion. The effect of procedural time on outcomes has not been well-characterized. The goal of this study was to examine the relationship between endovascular procedural time and clinical outcomes of patients with AIS following endovascular treatment.
We retrospectively analyzed previously published data from three prospective studies of MT in patients with AIS—Solitaire FR with the Intention for Thrombectomy (SWIFT), Solitaire FR Thrombectomy for Acute Revascularization (STAR), and Solitaire FR With the Intention For Thrombectomy as Primary Endovascular Treatment for Acute Ischemic Stroke (SWIFT PRIME)5–7—to determine the effect of procedural time on angiographic and clinical outcomes. These studies evaluated MT with a stent retriever (Solitaire revascularization device) in patients with large vessel occlusion. The design and methodologies of these studies have been previously published.5–7
Data and design
We recorded two procedural time intervals: (1) symptom onset to femoral arterial access (onset to groin puncture) and (2) femoral arterial access to angiographic recanalization at the site of occlusion (groin puncture to recanalization). We documented baseline demographic and clinical characteristics, symptomatic intracranial hemorrhage; defined as non-contrast CT documented intracranial hemorrhage related to ≥4 point deterioration on the NIHSS) within 24 hours postprocedure, NIHSS score on admission and at 3 months postprocedure, intracranial hemorrhage (ICH) rates at 24 hours' follow-up, and modified Rankin Scale (mRS) score at 3 months postprocedure. The mRS score was ascertained by an independent investigator for the SWIFT PRIME trial. All relevant angiographic images were submitted for central analysis performed by an independent reviewer. The Thrombolysis in Cerebral Infarction (TICI) scale was used to grade postoperative angiographic recanalization. Successful recanalization was defined by a modified TICI score of 2b–3 (near-complete to complete). The primary endpoint was functional dependence at 3 months postprocedure, as defined by a mRS score of 3–6. Patients with a TICI score <2b were excluded from this study, as we aimed to determine the effect of procedure time on outcome in patients with near-complete or complete recanalization.
Demographics, baseline, and clinical characteristics were compared between subjects with unfavorable (mRS score 3–6) and favorable (mRS score 0–2) outcome at 90 days, and were also compared between subjects with procedure times <60 min and ≥60 min. For continuous measures, means and SD are presented and p values calculated with a paired t-test. For categorical measures, frequencies and percentages are presented and p values calculated with a Fisher’s exact test.
A multivariate analysis was performed using a logistic regression model with unfavorable outcome at 3 months as the dependent variable. The variables included in the multivariate analysis as predictor variables included age, onset to groin puncture, groin puncture to recanalization, baseline NIHSS score, and collaterals. Odds ratios and 95% confidence intervals (CIs) were calculated using SAS version 9.4. The rates of favorable outcome at 90 days in subjects with different procedure times of <60 min vs ≥60 min were compared in subjects using Fisher’s exact test. A similar multivariate analysis was performed in patients with procedure time ≥60 min in order to determine predictors of unfavorable outcome in this subgroup.
We analyzed 301 patients (mean age: 70±11 years in the group with poor outcome, 65±13 years in the group with good outcome; 57% female) who underwent MT with successful recanalization. Of the 301 patients, 122 (40.5%) had an unfavorable/poor outcome and 179 (59.5%) had a favorable outcome at 3 months after treatment (table 1).
Patients with favorable outcomes had lower procedure times than those with unfavorable outcomes: onset to groin puncture time averaged 228±81 min in patients with good clinical outcome versus 283±92 in patients with poor clinical outcomes (p<0.001). Patients with favorable outcomes also had a lower groin puncture to recanalization time (44±25 min) versus patients with poor clinical outcomes (51±33 min; p<0.040). Similarly, time from onset to recanalization time was significantly lower in patients with favorable outcome (273±86 min) than in those with poor clinical outcome (336±96 min; p<0.001).
In the univariate analysis, statistically significant differences in baseline characteristics between patients with poor outcome and those with good outcome were found for age (70±11 vs 65±13, p<0.001), baseline NIHSS score (18.2±4.4 vs 15.9±4.6, p<0.001), baseline Alberta Stroke Program Early CT Score (ASPECTS; 8.0±1.8 vs 8.6±1.5, p=0.004), collaterals (1.9±0.9 vs 2.3±0.9, p=0.005), and glucose levels (138.8±61.8 vs 122.4±50.1, p=0.013), as shown in table 1.
Angiographic results showed that final revascularization was significantly different between patients with poor versus good outcomes (p=0.043), with the unfavorable outcome group having a higher percentage of TICI 2b revascularization (47/122 [39%] vs 48/179 [27%]) and a lower percentage of TICI 3 (75/122 [61%] vs 131/179 [73%]), as shown in table 1. There were also a higher percentage of patients who had an asymptomatic ICH in patients with unfavorable outcomes (16/96 [17%] vs 9/132 [7%], p=0.03).
A total of 67/288 patients (23%) had procedure times ≥60 min (table 2). In these patients, both groin puncture to recanalization and onset to recanalization times were significantly higher than for patients with procedure time <60 min (87±32 vs 35±13 [p<0.001] and 351±100 vs 284±89 [p<0.001], respectively). Onset to groin puncture time was not significantly different between the groups, and there were no differences in baseline characteristics or site of occlusion.
A higher percentage of patients with procedure time <60 min received IV tPA (155/208 [75%] vs 37/61 [61%]), although the groups were not significantly different (p=0.052). There was a significant difference in the number of device passes (p<0.001), where 48% (31/65) patients in the ≥60 min group had three or more device passes vs 8% (15/198) in the <60 min group, and just one pass was performed in 23% (15/65) of the ≥60 min group vs 70% (139/198) in the <60 min group. There was also a significant difference in angiographic results (p<0.001), where the ≥60 min group had a higher percentage of TICI 2b revascularization (35/67 [52%] vs 58/221 [26%]) and a lower percentage of TICI 3 (32/67 [48%] vs 163/221 [74%]).
The multivariate analysis (table 3) showed that overall predictors of unfavorable outcome at 3 months after treatment were age (unit 10 years, OR=0.62, 95% CI 0.46 to 0.82, p<0.001), onset to groin puncture time (unit hour, OR=0.61, 95% CI 0.48 to 0.77, p<0.001), groin puncture to recanalization time (unit 10 min, OR=0.89, 95% CI 0.80 to 0.99, p=0.032), baseline NIHSS score (20–28 vs. 8–10, OR=0.17, 95% CI 0.05 to 0.62, p=0.018), and collaterals (OR=1.48, 95% CI 1.04 to 2.10, p=0.029). The rates of favorable outcomes were significantly higher when the procedural time was <60 min compared with ≥60 min (62% vs 45%, p=0.020) (figure 1B). A similar multivariate analysis among the subgroup of patients with procedure time ≥60 min showed no significant predictors of unfavorable outcome (table 3). See online supplementary table 1 for baseline characteristics in this subgroup separated by outcome).
Supplementary file 1
This study highlights the importance of MT procedure time in determining clinical outcome in patients with AIS, showing that patients with procedure times <60 min have significantly better neurological outcomes. Our multivariate analysis revealed that longer onset to groin puncture time and groin puncture to recanalization time are significant predictors of unfavorable clinical outcome. With increased use of MT for patients with AIS, procedure time becomes an important determinant of favorable outcomes in those who achieve complete recanalization.
Our findings corroborate that of a previous study, in which we reported a higher rate of favorable outcome for patients with procedure times <30 min.4 In that study, we also found that rates of favorable outcome for patients with procedure times >60 min were lower than for patients treated with placebo, and unfavorable outcomes were significantly associated with age, admission NIHSS score, and time from symptom onset. A study by Spiotta et al also identified longer procedure times as being associated with a lower likelihood of achieving recanalization, higher rates of intraprocedural complications, and significantly higher device costs.8
The impact of MT procedural time has not been well studied, thus it has been difficult to define an optimal time window. The brain attack coalition recommends that door-to-administration of IV rtPA should be <60 min; however, outcomes vary for MT, especially when IV rtPA is contraindicated.9 10 In our analysis of patients with procedure times <60 min compared with ≥60 min, onset to groin puncture time did not differ significantly, while both groin puncture to recanalization and onset to recanalization times were significantly longer in the ≥60 min group. Because there was no significant difference in IV tPA pretreatment rates, the difference in procedure times is probably explained by the higher percentage of patients in the ≥60 min group who had two or more device passes compared with the <60 min group. A range of additional factors may influence procedure times, including vascular tortuosity, clot consistency, and limitations of available technology for mechanical clot retrieval.11
Increased time to reperfusion may largely depend on the degree of medical decision-making for each procedure, according to the difficulty and level of procedural complications. A recent study used the Stroke Triage Education, Procedure Standardization, and Technology (STEPS-T) program, an integrated approach developed to decrease the treatment time for patients with stroke while maintaining quality of care and improving clinical outcomes.12 The program uses multiple departments and has enabled iterative evidence-based process improvements, thereby sustaining significant reductions in intervention and recanalization times, and also increasing operator efficiency. The program significantly reduced total intervention and recanalization times, improved the mRS score, and reduced the time required in the angiography suite. Although the process of recanalization is difficult, standardized protocols will help the neurointerventionalist estimate procedure time based on admission characteristics and whether the outcome is favorable.
Although good outcomes correlate with shorter procedures, each procedure should be examined on its own merits, incorporating the type of stroke, location, risk factors, imaging modality, organizational aspects, and interventionalist skill level/operator efficiency. Our findings reinforce the notion that prolonged procedures can have a negative effect on a patient’s outcomes. By convention, studies assessing treatment success after acute ischemic stroke have focused on time from symptom onset to reperfusion, and this study provides evidence that shorter procedural times are a significant predictor of a higher odds of favorable outcome. Endovascular treatment that rapidly restores blood flow should be a central focus of new treatment paradigms, whether this intervention is mechanical alone or carried out in combination with thrombolytic agents.
Limitations of our study include its retrospective design and reliance on the pooled analysis of SWIFT, STAR, and SWIFT PRIME studies, which each had their own limitations.5–7 Most patients had anterior rather than posterior circulation occlusions. An additional limitation is that the measure of procedural time does not describe the time spent on decision-making for each procedure—for example, procedure time variability increases when tPA is contraindicated. Individual patient anatomy, such as more tortuous anatomy or more refractory occluded thrombus, can also influence procedure times and outcomes. As a result, further research is necessary to fully confirm procedural time as an independent predictor of patient outcomes in the setting of other confounding factors such as operator experience and procedural complexity, together with patient and family medical decisions. Finally, this analysis included small numbers of outcome events, limiting the ability to adjust for potential confounding variables. Inclusion of confounding variables may still be unable to deal with all sources of unrecorded systematic bias between the groups.
Procedural time in patients undergoing MT for AIS is an important determinant of favorable outcomes in patients with recanalization. Our results illustrate that endovascular procedural time is an important factor affecting patient outcome, in addition to the more thoroughly studied onset to revascularization time. Time guidelines and standardized treatment protocols should aim to reduce procedural time from symptom onset to decrease delays in endovascular treatment and improve outcomes among patients with AIS.
We acknowledge Medtronic for editing assistance.
Contributors All authors made substantial contributions to conception and design, analysis, and interpretation of data; drafted or critically revised the article; and gave final approval of the version to be published.
Funding This work was supported by Medtronic, Inc.
Competing interests AEH serves as a Consultant, Proctor and Advisor and has received honoraria from Medtronic, Stryker, Penumbra, GE Healthcare, Genentech, and Microvention. JLS is an employee of the University of California, which holds a patent on retriever devices for stroke. DL has received an NIH grant and serves as a scientific consultant to Stryker and Medtronic. MG serves as consultant for Medtronic Neurovascular and as co-principal investigator for ESCAPE and SWIFT PRIME trials. The Regents of the University of California receives funding for RJ’s services as a scientific consultant regarding trial design and conduct to Medtronic/Covidien, and RJ is an employee of the University of California, which holds a patent on retriever devices for stroke. AQ has received grant support from Boston Scientific, Cordis Therapeutics, Cor Therapeutics, Centocor therapeutics, C R Bard, and San Francisco Science/PICS. He has also received consultant fees from EndoTex Interventional Systems and Boston Scientific and an honorarium from Boehringer Ingelheim.
Provenance and peer review Not commissioned; externally peer reviewed.
Patient consent for publication Not required.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.