Comparison of functional and safety outcomes between the extended versus early time window after intravenous thrombolysis and endovascular thrombectomy

. Regression


Introduction
In all the pivotal trials showing the efficacy and safety of endovascular thrombectomy (EVT) in acute ischemic stroke (AIS), intravenous thrombolysis (IVT) was administered to all eligible patients [1][2][3][4][5][6][7].In the early time window, from 0 to 4.5 h from last seen well (LSW), recent trials have not shown benefit from omitting IVT in patients eligible to both IVT and EVT [8][9][10][11][12][13].Yet data concerning the remarkable proportion of patients presenting later than 4.5 h after stroke onset or with wake-up stroke (WUS), is limited [14,15].The European Stroke Organisation (ESO) guidelines recommend IVT for patients with a core and perfusion mismatch, with a known stroke duration of 4.5 to 9 h and in WUS, if the midpoint of sleep is <9 h ago, and for whom EVT is not indicated or planned [16,17].However, the Extending thrombolysis to 4.5-9 h and wake-up stroke using perfusion imaging (EXTEND) trial was commenced before EVT was proven beneficial in the extended time window [18].Concerning EVT-eligible patients presenting from 4.5 h to 9 h since stroke onset, a consensus could not be reached whether IVT should be used before EVT, or not [17].Prior to 2019 according to the guidelines of Helsinki University Hospital (HUS), IVT was administered Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; ESO, European Stroke Organization; EVT, endovascular thrombectomy; EXTEND, Extending thrombolysis to 4.5-9 h and wake-up stroke using perfusion imaging; HSQR, Helsinki Stroke Quality Registry; IVT, intravenous thrombolysis; IQR, interquartile range; LSW, last seen well; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; sICH, symptomatic intracerebral hemorrhage; OR, odds ratio; uaOR, unadjusted odds ratio; WUS, wake-up stroke.. only in the early time window.After the results of the EXTEND trial became available in the beginning of 2019 [19], the HUS guidelines were updated to include patients according to the inclusion criteria of the EXTEND trial [19], and proceed to EVT if indicated.

Aims and hypothesis
We aimed to investigate the role of IVT in combination with EVT in the extended time window by comparing functional and safety outcomes between patients in the extended time window and matched controls in the early time window.

Study design and study population
We included all consecutive patients who received IVT, EVT, or IVT plus EVT in the extended time window at HUS between January 2021 and December 2022 from the Helsinki Stroke Quality Registry (HSQR).IVT was administered according to the EXTEND criteria [19], namely for all patients eligible to IVT presenting 4.5 to 9 h from stroke onset and in WUS, if the mid-point of sleep <9 h ago, with the following criteria: a perfusion lesion volume of >10 ml, infarct core volume ≤ 70 ml, and a mismatch ratio between perfusion and core >1.2.For all the consecutive patients, we identified a matched control treated in the early time window, between January 2016 and December 2020, from the HSQR.Matching was performed based on the acute recanalization treatment (IVT, EVT, or IVT plus EVT), age, stroke severity evaluated on the National Institutes of Health Stroke Scale (NIHSS), occlusion site (right or left anterior, or posterior circulation), and amount of vascular risk factors, including history of hypertension, dyslipidemia, diabetes mellitus, prior ischemic stroke, atrial fibrillation, and coronary artery disease or chronic heart failure.We selected controls who had the same acute recanalization treatment, the same occlusion site, the closest possible NIHSS (±2), age (±3 years), and amount of vascular risk factors (±1).If a matched control was not identified, we expanded the criteria to include also NIHSS ±3 and age ±5 years.Patients for whom a matched control could not be identified were excluded from the final analysis.The need for obtaining informed consent was waived due to retrospective nature of the study.

Clinical variables
We recorded data on age, sex, stroke severity as measured with NIHSS, the above-mentioned vascular risk factors, acute recanalization treatment (IVT, EVT, or IVT plus EVT) and time from symptom onset to initiation of IVT (hours) from the HSQR.Regarding patients with EVT, we recorded time from symptom onset to groin puncture.Concerning WUS, symptom-onset was calculated as the time from the mid-point of LSW and the time symptoms were noticed [19].Reason for the possible ineligibility to IVT was recorded.EVT was not conducted, if there was no target for EVT, namely an appropriate vessel occlusion based on the clinical judgement aligned with the current treatment guidelines [20,21].Modified Rankin Scale (mRS) was collected at baseline and at 90 days follow-up.We recorded rates of symptomatic intracerebral hemorrhage (sICH), based on the European Cooperative Acute Stroke Study -II criteria [22], namely as blood in the brain on the follow-up imaging, documentation of clinical deterioration, or adverse events indicating clinical worsening or causing an increase in NIHSS of 4 or more points.
The primary outcome was a shift towards better functional outcome on the mRS scale at 90 days in the extended versus early time window, whereas the safety outcome was the occurrence of sICH.

Statistical analysis
Statistical analyses were performed with IBM SPSS Statistics for Windows, version 29.0 (IBM Corp., Armonk, N.Y., USA).Categorical variables were presented as counts and percentages, and continuous variables were described by mean and standard deviation or by median and interquartile range (IQR).Comparisons of patients' characteristics regarding categorical variables were performed with Pearson Chi-Square test whilst comparisons of continuous, normally distributed variables with unpaired t-test and continuous, non-normally distributed variables with Mann-Whitney U test.Normality of continuous data was inspected and tested with Kolmogorov-Smirnov's test.The univariate and multivariable logistic regression models were conducted respectively among all the patients and among patients with IVT, EVT or, IVT plus EVT to compare the occurrence of sICH and the outcomes at 90 days follow-up for a 1-step shift on the mRS score.Based on comparable literature, we adjusted for potential confounding factors including age, sex, baseline mRS and NIHSS, occlusion site and amount of vascular risk factors.For statistical comparisons and regression analyses, patients with missing data for the respective parameter were excluded.Results of the regression analyses are presented as odds ratio (OR), and 95% confidence interval (CI).A p-value <0.05 was considered as statistically significant.

Study population and patient characteristics
We identified 156 patients treated in the extended time window between January 2021 and December 2022 (Fig. 1).A matched control was identified for all the patients with IVT alone.For 6 patients who received IVT plus EVT, and for 6 patients who received only EVT, the closest matched control was found with the previously mentioned expanded criteria.A matched control could not be identified for altogether 22 patients, who were thus excluded from the final analysis.Compared to other patients in the extended time window, they were younger (66 years versus 71 years), had more vascular risk factors (1.5 versus 1.0), higher NIHSS (11 versus 9), and more often posterior circulation occlusion (27% versus 13%) and only 3 of them received IVT.
The final study population consists of 134 patients treated in the extended time window and 134 matched controls treated in the early time window.In both time windows, altogether 51 patients (38%) received IVT alone, 37 patients (28%) EVT alone, whereas 46 patients (34%) received both IVT and EVT.There were no significant differences concerning age, sex, baseline NIHSS, or pre-stroke mRS between the two time windows (Table 1).The controls had more often history of hypertension and atrial fibrillation, but the total amount of vascular risk factors did not significantly differ.Time from symptom onset to initiation of IVT significantly differed between the extended versus early time window (5.7 [4.9-6.8]hours) versus 1.8 [1.4-2.8]hours, respectively).
No significant shift towards better outcome on mRS was detected among patients treated in the extended time window versus early time window, unadjusted odds ratio (uaOR) 1.08; 95% CI 0.71-1.65 (Table 2).For patient receiving IVT alone, uaOR was 1.60; 95% CI 0.80-3.22,for patients with EVT alone, uaOR was 0.90; 95% CI 0.40-2.03,whereas for patients with IVT plus EVT, uaOR was 0.95; 95% CI 0.46-1.98.The shift on mRS towards better outcome remained unsignificant after multivariate adjustments.The adjusted ORs (aOR) for a favorable outcome shift in the extended versus early time window was 1.4; 95% CI 0.89-2.12.For patients receiving IVT alone, aOR was 1.91, 95% CI 0.93-3.90, for patients with EVT alone, aOR was 0.90, 95% CI 0.38-2.11,and for patients with IVT plus EVT, aOR was 1.15; 95% CI 0.54-2.43(Table 2).Patients with no matching control identified had a higher mRS score at 90 days (3 [2][3][4][5] versus 2 [1][2][3][4]), but they also had more severe symptoms on NIHSS at admission.The trend towards a greater proportion of patients with functional independence (mRS 0-2) and less worse outcomes (3)(4)(5)(6) in the extended versus early time window is presented in the Fig. 2.This trend was observed in the extended time window among all the patients, and patients with IVT, and IVT plus EVT (Fig. 2).However, among patients with IVT alone, there was a trend to a greater proportion of excellent outcome (mRS 0-1) in the early time window, as well as a trend for a greater proportion of patients with good outcome (mRS 0-2) among patients with EVT alone in the early time window (Fig. 2).

Discussion
We found no significant differences in the functional or safety outcomes between the extended versus early time window among patients who received either IVT, EVT, or IVT plus EVT.A trend towards fewer worse outcomes was observed in the extended time window among patients receiving IVT, or IVT plus EVT.In this matched control cohort this might be explained by the requirement to have a perfusion mismatch at baseline imaging in the extended time window, whereas perfusion imaging was not mandatory in the early time window.
In the early time window, the Chinese trials [9,10] found that EVT alone was non-inferior to IVT plus EVT, whereas Japanese [11] and European [12] trials did not show non-inferiority.However, there was between-study heterogeneity regarding the selected non-inferiority margins, patient population, stroke etiology, and workflow organization [8].In a study-level meta-analysis, EVT alone was non-inferior to IVT at non-inferiority margins proposed in the literature, but not the most conservative ones [23].However, the latest randomized controlled trials [8,13] in the early time window concluded that EVT alone could not be shown non-inferior compared to IVT plus EVT, and EVT alone did not show any safety advantages.IVT-associated reperfusion before EVT occurs infrequently [24], as in the present study.However, the final reperfusion rates were higher in patients with IVT plus EVT, a significant difference not reported previously, and a likely a reason for the favorable outcome shifts observed in patients treated with the combination therapy [8].By enhancing the fibrinolytic process, IVT might increase the speed and likelihood of successful reperfusion with EVT, dissolve thrombosis beyond reach of EVT [25,26], and reduce the required number of passes with a stent retriever [27], and hence improve the prognosis among patients with the combination of IVT plus EVT.In the present study, IVT did not significantly increase the risk of sICH, although the statistical power to detect a difference was limited by the small number of sICH.The occurrence of sICH was comparable with the frequency of sICH in the extended time window based on the literature [16].To conclude, our results align with the previous literature.
Our study has limitations.Firstly, due to the observative nature of this register-based study, the treatment decisions have been made by the clinical judgement of the treating physician, based on treatment guidelines [17,20].Secondly, the patients and matched controls were treated in two different time periods, 2021-2022 and 2016-2020, respectively.The possible differences in the workflow organization, technical issues (newer devices, combined methods, and accumulated experience of the interventionists) might have favored the more recent  time period.However, although number of passes was diminished and a trend towards reduced duration of EVT compared to the earlier time period was observed, recanalization rates were smaller in the more recent time period.Nonetheless, we aimed to select the matched controls based on several factors to achieve as similar cohorts as possible.Consequently, baseline characteristics, excluding the different treatment time window, did not significantly differ between the two cohorts.Thirdly, a matched control could not be identified for all the patients.Still, only few of the excluded patients received IVT, which was the focus of interest in the present study.

Conclusions
In the extended time window, there remains paucity of data on the role of IVT prior to EVT.However, the present observative study did not raise any safety concerns on IVT followed by EVT in eligible patients.We did not find any signals indicating, that IVT or EVT should be avoided in patients who fulfill the criteria for IVT in the extended time window and have a significant vessel occlusion which can be treated by means of EVT.

Fig. 2 .
Fig. 2. Distribution of the modified Rankin Scale scores among patients with IVT, EVT, or IVT plus EVT in the extended versus early time window.EVT, endovascular thrombectomy; IVT, intravenous thrombolysis.

Table 1
Comparison of baseline characteristics between the patients and matched controls.
indicates National Institutes of Health Stroke Scale; NCCT, non-contrast computed tomography; SAH, subarachnoid hemorrhage.*Duration of EVT was calculated from the groin puncture to the recanalization, or in case of no recanalization, to the decision to end the procedure.

Table 2
Odds ratios for a 1-step shift towards a better functional outcome on modified Rankin scale between patients in the extended versus early time window.

Table 3
Rates of symptomatic intracerebral hemorrhage and adjusted odds ratios for the occurence of sICH in the extended versus early time window.