Treatment and Outcomes in Patients With Basilar Artery Occlusion
基底动脉闭塞患者血管内治疗至治疗时间与结局的相关性
翻译:研究生:张洮铭;科研秘书:刘虹;导师:陈晨
导语:缺血性脑卒中是全球严重的长期残疾和死二的主要原因,也是我院神经内科收治最多的病种,这篇来自1区37.8分的2022年最新文献提示,基底动脉闭塞患者血管内治疗至治疗时间与结局相关,通过多因素分析提示,在接受血管内治疗治疗基底动脉闭塞的患者中,症状发作后更快的治疗与结局改善相关。这些发现支持了对基底动脉闭塞患者进行血管内治疗快速治疗的重要性。
高质量的文献和研究既提示我们学科融合、学科交叉的重要性,也给我们科学研究带来了新的思路。通读本文后,给我们研究者带来继续进行深入研究的思路和冲动。身处心血管病专科医院,上述病例和数据不难获得,要的就是十年如一日的科研坚持和随访。
神经内科3病区目前在研科研项目4项,其中卫健委“四个一批”重大科技攻关项目一项。今年出版专著1部,待出版专著2部,已发表SCI论文3篇,最高为2区6.05分的脑卒中代谢学相关研究。我们有完备的科研团队和研究生团队欢迎兄弟科室常来常往,交流科研经验,共同进行学科融合和多学科交叉科研思路的探讨起申请立项学科交叉的课题。
下面我们就来阅读整篇文献。
BACKGROUND: Basilar artery occlusion (BAO) is a devastating condition without definitive evidence to guide treatment. Whereas the association between faster treatment times with endovascular therapy (EVT) and better outcomes in anterior circulation is well established, whether this relationship exists for patients with BAO is not well delineated.
背景:基底动脉闭塞(BAO)是一种很严重的疾病,目前没有明确的临床证据来指导治疗。虽然血管内治疗(EVT)的治疗时间更快与前循环更好的预后之间的关系已得到公认,但这种关系是否存在于BAO患者中还没有得到很好的界定。
METHODS: We used individual-level patient data from the Get With The Guidelines–Stroke nationwide US registry prospectively collected from January 2015 to December 2019. We identified individuals with BAO treated with EVT within 24 hours of symptom onset. The primary outcomes examined were in-hospital mortality, discharge home, ambulatory at discharge,
discharge,independent at discharge (modified Rankin Scale score 0 to 2), substantial reperfusion (modified Thrombolysis in Cerebral
Infarction score 2b or 3), and symptomatic intracranial hemorrhage. Using logistic regression models, we evaluated the association between time from symptom onset to treatment with EVT and outcomes.
RESULTS: Among 3015 patients with BAO treated with EVT, the mean age was 65.9 years, 38.8% were women, and the median National Institutes of Health Stroke Scale score at presentation was 17 (interquartile range, 8–26). Median onset to EVT time was 406 minutes (interquartile range, 252–688). From 2015 to 2019, there was an overall increase in the
median onset to EVT times (380–411 minutes; P=0.016) but no significant change in the proportion of patients treated within 6 hours of symptom onset (48.4%–44.0%; P=0.17). After risk adjustment for patient and hospital-level factors, there were significantly lower odds of in-hospital mortality (adjusted odds ratio [aOR], 0.55 [95% CI,0.45–0.68]) and symptomatic intracranial hemorrhage (aOR, 0.52 [95% CI, 0.32–0.84]) and significantly higher odds of ambulation at discharge (aOR,1.72 [95% CI, 1.37–2.16]), discharge home (aOR, 2.19 [95% CI, 1.73–2.77]), and independence at discharge (aOR, 2.21 [95% CI, 1.66–2.95]) when onset to EVT time was ≤6 hours compared with >6 hours. The fastest decay in good outcomes per hour occurred within 6 hours of symptom onset.
CONCLUSIONS: Among patients receiving EVT for BAO, faster treatment from symptom onset was associated with improved outcomes. These findings support efforts to achieve rapid treatment with EVT for patients with BAO.
结论:在接受EVT治疗BAO的患者中,症状发作后更快的治疗与结局改善相关。这些发现支持了对BAO患者进行EVT快速治疗的重要性。
Basilar artery occlusion (BAO) is a devastating and frequently fatal condition. BAO represents only 1% to 4% of all ischemic strokes but is
associated with a disproportionately high level of mortality and morbidity.Recanalization rates with intrave-nous (IV) thrombolysis in BAO alone are relatively low.Endovascular therapy (EVT) is highly effective and safe for anterior circulation proximal occlusion4 and fastertreatment is associated with better outcomes.Evi-dence for use of EVT for treatment of basilar occlusion remains inadequate.
The American Heart Association/American Stroke Association guidelines state that the use of EVT may be reasonable for basilar occlusion within 6 hours but the benefits are uncertain and the evidence is limited.7 Some experts have hypothesized that the treatment window in BAO might be longer because of the greater resilience of brainstem white matter to prolonged ischemia.8,9 Late window treatment, after 6 hours, has been reported in observational case series10 but the frequency of such treatment in clinical practice is uncertain and the outcomes compared with earlier, <6-hour treatment have not been investigated thoroughly in a large sample.
We aimed to evaluate the frequency of early (≤6 hour) and late (>6 hour) EVT for BAO in clinical practice and the association between time to treatment with EVT and outcomes in a large US cohort of patients treated at hospitals participating in the Get With The Guidelines (GWTG)–Stroke Registry. We hypothesized that earlier treatment, particularly ≤6 hours, would be associated with improved outcomes after BAO.
METHODS
方法
For our primary cohort, we selected patients who presented with BAO to the subset of GWTG hospitals that report comprehensive stroke center EVT metrics between January 2, 2015, and December 31, 2019, and underwent EVT. We selected this period to reflect care occurring after publication of positive EVT trials for anterior circulation occlusions.We excluded patients who did not receive EVT as well as those with time from last known well to EVT >24 hours. We also excluded patients with >25% missing information in medical history, missing sex, and with incorrect order of last known well/hospital arrival/treatment time (see Figure S1 for study flowchart).
The primary exposure was time from stroke onset (defined as last known well) to start of EVT for BAO. Time of EVT was defined as the time of arterial puncture; if the time of arterial puncture was not available, then the time of intraarterial tis- sue plasminogen activator (tPA) or revasculariz-ation was used instead, if it was documented (6% of cases). The primary clinical outcomes were in-hospital mortality, independent ambulation status at discharge, and discharge home. Secondary clinical outcomes were in-hospital death or disch-arge to hospice and dichotomized modified Rankin Scale (mRS) scores of 0 to 1 versus 2 to 6, 0 to 2 versus 3 to 6, and 0 to 3 versus 4 to 6.The main technical efficacy outcome was modified Thrombolysis in Cerebral Infarction score of 2b or 3 (50% to 100%) reperfusion.15 The main safety outcome was symptomatic intracranial hemorrhage (ICH), defined as neurologic worsening attributed to ICH verified by computed tomography or magnetic resonance imaging within 36 hours of EVT.
Statistical Analysis
统计分析
Baseline patient characteristics and clinical factors were compared between onset and EVT epochs (≤6 hours and >6 hours). Categorical variables were summarized by frequencies with percentages and compared using Pearson χ2 test or Fisher exact test for sparse data. Continuous variables were summarized by the median and 25th and 75th percentiles and compared using the Wilcoxon rank-sum test. Spearman correl-ation tests were used to assess trends in onset to EVT time from 2015 to 2019 and Cochrane-Armitage trend tests were used to assess the change in proportion of patients treated within 6 hours.
Multivariable logistic regression analysis was performed to assess the association of onset to EVT times with each outcome. Generalized estimating equations were used in all regression models to account for within-hospital clustering.The multivariable models adjusted for age, sex, insurance,race/ethnicity, atrial fibrillation/flutter, previous stroke, previous transient ischemic attack, coronary artery disease/previous myocardial infarction, carotid stenosis, diabetes, peripheral vascular disease, hypertension, dyslipidemia, smoking, heart
failure, renal insufficiency, arrival with emergency medical services, arrival on versus off hours, National Institutes of Health Stroke Scale (NIHSS) score, receipt of IV tPA, region, hospital type, number of beds, annual ischemic stroke volume, annual IV tPA volume, annual EVT volume, and primary/comprehensive stroke center status.
All continuous variables included in the models were evaluated for nonlinearity with the outcome
using restricted cubic splines and the likelihood ratio test and splines were used for those that violated the linearity assumption. Three approac-hes were used to explore the association of time to EVT with outcomes. First, because the American Heart Association/American Stroke Association guidelines recommend that treatment of BAO within 6 hours is reasonable but do not comment on later treatment windows, we examined the association between time to EVT ≤6 hours compared with 6 to 24 hours and outcomes using logistic regression. In this model,we also tested whether the association of early treatment with outcomes was modified by age, sex, NIHSS <10 compared with ≥10, and IV tPA use. Second, we examined time to EVT as a linear continuous variable, estimating the odds ratio for each1-hour faster treatment with EVT and testing for modification of the association by ≤6 hours treatment compared with >6 hours treatment. Third, we generated time–benefit curves on the basis of calculating outcome-specific predicted probabilities for each value of onset to EVT within the observed range while setting all other variables in the model to their mean value using restricted cubic splines.
Rates of missingness of baseline patient characteristic data were low (<10%) and missing patient medical history or medication before admission was imputed to “no.” Patients with missing NIHSS score were excluded from the models (2.7%).Mortality, discharge destination, and intracranial hemorrhage outcomes data were missing for <0.1%. Data on ambulation at discharge were missing in 3.3% and on discharge mRS in 15.7%; patients with missing data for outcomes were excluded from models of those outcomes.
SAS (version 9.4; SAS Institute Inc) software was used for all statistical analyses. All P values were 2-sided and statistical significance was defined as P<0.05. IQVIA serves as the data collection and coordination center and the Duke Clinical Research Institute serves as the data analysis center.
RESULTS
结果
Among 715 423 patients with ischemic stroke during the study period,5762patients (0.8%) from 368 sites presented with documented BAO and 3015(0.4%) patients from 351 sites received EVT for BAO and were included in our final cohort (Figure S1). Among patients in the final cohort, also had an internal carotid artery occlusion, had a middle cerebral artery occlusion, had another cerebral artery branch occlusion, and 188 had a vertebral artery occlusion. The proportion of patients with basilar occlusion who were treated with EVT was relatively similar from 2015(78.5%) to 2019(82.6%) with no significant linear trend (P=0.08). There was an increase in the median time from stroke onset to EVT from 2015 (380 minutes) until 2019(411 minutes; P=0.016) but no significant change in the proportion of individuals receiving EVT within 6 hours (48.4% vs 44.0%; P=0.17;Figure 1).
Among those treated, median onset to EVT time was 406 minutes (interquartile range [IQR],252–688) overall,235minutes (IQR,179–295) when treatment was ≤6 hours, and 646 minutes (IQR,466–874)when treatment was >6 hours. A total of 1304(43.3%)patients were treated within 6 hours of symptom onset,620(20.6%) at 6 to 9 hours(12.9%) at 9 to 12hours, and 703(23.3%) at 12 to 24 hours.
Patient characteristics stratified by treatment time (≤6 hours and >6 hours) are shown in the Table. The mean age was 65.9 years (SD 14.8),39.3% were women, and median presenting NIHSS score was (IQR,8–27)。 Mode of arrival was by emergency medical services in 96.3% (direct arrival in 42.2% and transfer from another hospital in 54.1%). IV tPA was adminis-tered before EVT in 35.3%. Those who received EVT within 6 hours were slightly older (mean 67.2 vs 64.9 years; P<0.001), had higher median NIHSS score on presentation (19 vs17; P=0.003), and were more likely to receive IV tPA (60.9% vs 15.8%; P<0.001)。 In those who received IV tPA, median time from last known well to adminis-tration was 134 minutes (IQR,92–175) in the ≤6 hours group and 226(IQR,136–269) in the >6 hours group.
Among the 3015 patients treated with EVT,28.1%died in the hospital,36.8%were ambulatory at discharge,22.7%were discharged home,35.1%either died in the hospital or were discharged to hospice, and 3.4% had symptomatic intracranial hemorrhage. At discharge,12.4%had mRS 0 to 1,17.8%had mRS 0 to 2, and 26.0% had mRS 0 to 3. Thrombolysis in Cerebral Infarction grade 2b/3 reperfusion was achieved in 90.9%. There were more favorable rates of all outcomes amongpatients who received EVT ≤6 hours compared with >6 hours, including in-hospital mortality (23.7% vs 31.5%),
ambulatory at discharge (41.1% vs 33.2%), discharge home (27.6% vs 18.9%), and symptomatic ICH (2.6% vs4.0%; Table S1).
Compared with time from symptom onset to EVT>6 hours, time from symptom onset to EVT ≤6 hours was associated with lower odds of in-hospital mortality(adjusted odds ratio [aOR], 0.55 [95% CI, 0.45–0.68]) and higher odds of ambulat-ion at discharge (aOR, 1.72 [95% CI, 1.37–2.16]) and discharge home (aOR, 2.19 [95% CI, 1.73–2.77]). There were also lower odds of symptomatic ICH, greater odds of reperfusion, and lower odds of disability on the mRS (Figure 2 and Table S1).
The odds of each outcome for each 1 hour of faster time to EVT are shown in Figure 3. Odds of improvement per hour were greater within 6 hours of onset compared with >6 hours. For example, there was a 16% decrease in the odds of in-hospital mortality per hour of faster treatment with EVT within 6 hours compared with 4% after 6 hours. Similarly, the odds of ambulation at discharg increased by 21% per hour of faster treatment within 6 hours from symptom onset and 4% per hour afterward. This pattern of results was consistent across outcomes.
Predicted probability plots are shown for all outcomes (Figure 4). With the exception of symptomatic ICH, these plots demonstrate a nonlinear decline in benefit with longer onset to EVT times: degree of benefit declines in a steep linear manner from 1 to 4.5 hours, then declines in a gradually less steep manner from 4.5 to 6 hours, and then declines in a shallow manner from 6 to 12 hours.
图 4 显示了所有结果的预测概率图。除无症状 ICH 外,这些图显示,随着起病至 EVT 时间的延长,获益程度呈非线性下降。的获益程度呈非线性下降。从 1 小时到 4.5 小时,获益程度呈陡峭的线性下降,然后从 4.5 小时开始,获益程度逐渐下降。从 4.5 小时到 6 小时的下降幅度逐渐减小,然后从 6 小时到 1 小时的下降幅度逐渐减小。然后在 6 至 12 小时内缓慢下降。
There was a significant interaction between age (<70 and ≥70 years) and onset to treatment time for the outcome of in-hospital mortality (interaction P=0.03). The benefit of more rapid treatment for in-hospital mortalitywas stronger in patients <70 years of age (aOR, 0.43 [95% CI, 0.32–0.58] for EVT ≤6 hours) compared with those ≥70 years of age (aOR, 0.71 [95% CI, 0.53–0.95];Table S2 and Figure S2). There was no interaction with age for the remainder of the outcomes, or with sex, NIHSS score ≥10, or IV tPA use with any outcome.
DISCUSSION
讨论
In this large retrospective analysis of a nationwide stroke registry, EVT for BAO was performed in 0.4% of all acute ischemic stroke admissions. Consistent with past literature, patients with BAO in this study had severe stroke and were at high risk for poor outcomes, with 1 in 3 dying in hospital or discharged to hospice, and only 1 in 5 being discharged to home despite receiving EVT. Among >3000 patients analyzed with EVT for BAO, earlier compared with late treatment was associated with improved outcomes, including reduced in-hospital mortality, increased ambulation at discharge, increased functional independence at discharge, and decreased symptomatic ICH. The greatest associated benefit of earlier treatment with EVT occurred within 6 hours of onset.
Our results are consistent with previous analyses of EVT for anterior circulation occlusion, demonstrating a similar relationship between faster onset to treatment times and better outcomes.Previous studies have found an association between time to treatment with EVT in BAO and outcomes, although were prior to the modern EVT era or sample sizes were small. The inflection point for time to treatment association with outcomes in anterior circulation occurred at ≈4.5 hours and in our study of basilar occlusion was ≈6 hours. The potentially longer time window for enhanced associated benefit may be attributable to differing physiologyof posterior circulation ischemia, specifically retrograde perfusion of posterior circulation through the circle of Willis and pial collaterals.
However, time from symptom onset to EVT (median 406 minutes in our study)was substantially longer compared with EVT in anterior circulation occlusion (median 230 minutes),possibly because of difficulty in recognition and diagnosis of BAO, lack of certainty with regards to efficacy of EVT,goals of care discussions given the potential for survival with severe disability,and more willingness to provide treatment in late windows for selected patientsowing to the poor natural history of basilar occlusion in the absence of reperfusion.
Evidence is limited for the treatment of BAO. The recent BEST trial (Acute Basilar Artery Occlusion: Endovascular Interventions vs Standard Medical Treatment)and BASICS (Basilar Artery International Cooperation Study) were inconclusive for benefit of EVT and further trials may be limited because of logistical or ethical issues of randomizing patients with BAO who havea devastating prognosis to no EVT. The most recent American Heart Association/American Stroke Association stroke guidelines state that the benefits of EVT are unclear but the use of EVT with stent retrievers may be reasonable for selected patients with occlusion of the
basilar artery within 6 hours.7 Our study supports these guidelines and emphasizes the importance of minimizing time to treatment with EVT for basilar occlusion, with a minimum target of <6 hours but faster treatment is favored. Achieving rapid treatment may involve improvements in recognition of posterior circulation symptoms,lower thresholds for vascular imaging to evaluate for basilar occlusion, and enhanced workflows to achieve rapid reperfusion including interfacility transfer of patients. Our study also highlights the persistently poor outcomes of
patients with BAO receiving EVT in a contemporary realworld setting.
In the BASICS registry, 68% of patients had a poor outcome, defined as mRS score 4 or 5 or death,1 although the registry was evaluated before the
modern endovascular era. In a large prospective registry in China, mortality at 90 days occurred in 71.4% of those not receiving EVT and 46.2% of those receiving EVT.26 The 90-day mortality in the control versus EVT groups was 38.5% versus 33.3% in the BEST trial and 43.2% versus 38.3% in BASICS.In our study of patients with BAO receiving EVT, 28% died in hospital, 36% were ambulatory at discharge, and only 22% were discharged
home. However, in those treated within 6 hours, the odds of mortality were approximately halved, and the odds of discharge home, ambulatory at discharge, and independence approximately doubled. Whereas ethical considerations and goals of care determination remain criticalgiven the high rate of severe disability compared with that with anterior circulation occlusion, our study provides evidence that faster treatment with EVT may alter the poor natural history of BAO.
There were some limitations to our study. First, despite adjustment for a large number of patient- and hospital level factors, residual measured and unmeasured confounding may influence our study findings. Second, the results reported relied on the accuracy of data abstraction from the medical record. Multiple methods are used to optimize data quality in the GWTG-Stroke program, including detailed training of site chart abstractors, standardized case definitions and coding instructions, predefined logic and range checks on data fields at data entry, audit trails, and regular data quality reports for all sites. Third, data missingness was present for some outcomes, particularly for the mRS at discharge. However, we retained a large cohort for the analysis of each outcome.Fourth, we did not have information on 90-day outcomes.Therefore, estimates of outcomes from this study cannot be directly compared with some other studies of EVT in BAO, which measured outcomes at 3 months.
However, previous studies have demonstrated that early functional outcome is correlated with 90-day outcomes.Fifth, the GWTG-Stroke database did not collect information regarding the infarct burden on baseline imaging (ieposterior circulation Acute Stroke Prognosis Early Computed Tomography Score29) or which patients underwent advanced physiologic imaging and magnetic resonance imaging. Computed tomography scanning is limited in its evaluation of the posterior circulation.
Physiologic imaging, if performed, is of uncertain usefulness in pos- terior circulation strokes. However, some treatment may have been delayed or avoided because of information obtained from magnetic resonance imaging to assess infarction in the posterior circulation and treatment mayhave differed on the basis of basilar thrombus characteristics and location. Sixth, outcomes were analyzed in an exploratory manner without adjustment for multiplicity and external validation is required to confirm the results. Last, because the GWTG registry is voluntary and data
were collected at GWTG hospitals that report EVT metrics, the data may not be generalizable to all hospitals in the United States. However, Medicare beneficiaries with acute ischemic stroke entered in the GWTG-Stroke program are similar to other Medicare beneficiaries.In addition, we had a large sample from routine clinical practice and our findings regarding the relationship of time to treatment with outcomes are plausible given similar findings in the anterior circulation.
In patients with BAO, shorter time from onset to treatment with EVT was associated with substantial improvement in reperfusion, mortality, and functional outcomes.The results from our analysis encourage efforts to improve workflow and speed of treatment with EVT33 to achieve better outcomes after BAO, analogous to those for patients with anterior circulation occlusion.
在 BAO 患者中,从发病到接受 EVT 治疗的较短时间与再灌注、死亡率和功能预后的大幅改善有关。我们的分析结果鼓励人们努力改进 EVT 的工作流程和治疗速度,以便在 BAO 后获得更好的预后,这与前循环闭塞患者的情况类似。
结束语
神经内科3病区全体,预祝大家万事顺利、心想事成!