Introduction
Acute large vessel occlusion (LVO) is a major public health concern worldwide because of its high rates of disability and mortality [1]. Management of acute LVO within 4.5 hours of symptom onset relies on the guideline-recommended strategy of intravenous thrombolysis (IVT) prior to mechanical thrombectomy (MT) [2]. Pre-MT thrombolysis promotes thrombus resolution in occluded intracranial arteries, thereby contributing to improved functional outcomes in patients with LVO [3,4]. Moreover, it may increase the incidence of thrombus migration [5].
Thrombus migration is defined as the movement of a thrombus from the initial occlusion site to distal vessels during thrombolysis or thrombectomy [6]. These thrombus dynamics may influence clinical outcomes through two mechanisms. On the one hand, thrombus migration may reflect partial recanalization induced by thrombolytic agents, thereby potentially improving perfusion in the ischemic penumbra [7]. On the other hand, thrombus fragmentation into more distal arteries may increase the difficulty of thrombectomy and reduce the rate of complete recanalization [8,9]. However, the relationship between thrombus migration and 3-month functional outcomes remains uncertain [10-12].
The traditional thrombolytic agent alteplase is widely used but has limitations because of its short half-life (5 min), which necessitates a 60-minute infusion [13]. Previous studies have shown that thrombus migration occurs in 10%-20% of patients treated with alteplase and is associated with small thrombus size, distal thrombus composition, and longer intervals from thrombolysis to puncture [9,14,15]. Tenecteplase is a genetically modified variant of tissue plasminogen activator with a longer half-life (20-24 min), enabling single-bolus administration [16]. Although recent trials have supported the noninferiority of tenecteplase compared with alteplase [17-20], its superiority in patients with LVO undergoing bridging therapy remains controversial [21,22]. Moreover, systematic data on thrombus migration following tenecteplase administration—including its incidence, predictors, and clinical implications—are lacking [23].
In this study, we conducted a comparative analysis of thrombus dynamics and functional outcomes in patients with LVO treated with either tenecteplase or alteplase before MT. We additionally assessed the impact of predictors of thrombus dynamics— including IVT-to-puncture time, clot burden score, and occlusion site—and further evaluated whether thrombus migration and resolution were associated with functional outcomes.
Methods
Study population
We performed a retrospective analysis of a prospectively collected, multicenter cohort to compare bridging therapy with tenecteplase versus alteplase followed by MT in patients with LVO treated at five comprehensive stroke centers in China between 2022 and 2024 (the list of centers is provided in Supplementary Table 1). Administration of IVT and eligibility for endovascular therapy followed contemporary guidelines. Following the 2023 Chinese guideline update, which granted tenecteplase and alteplase equal (Class IA) recommendations, participating centers increasingly incorporated tenecteplase into their institutional protocols during the study period. In addition, all decisions regarding thrombectomy and procedure termination were made by experienced neurointerventionalists based on individualized anatomical and clinical risk-benefit assessments, in accordance with contemporary guidelines. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Ethics Committee of Xuanwu Hospital, Capital Medical University (Approval No. [2021]215; approved October 20, 2021), and informed consent was obtained from all participants.
The inclusion criteria were as follows: (1) age ≥18 years; (2) management under the mothership model with a one-stop computed tomography (CT)/CT angiography (CTA) imaging protocol, with baseline CTA demonstrating LVO, defined as occlusion of the internal carotid artery (ICA), the M1 or proximal M2 segment of the middle cerebral artery (MCA), or the basilar artery (BA); (3) presentation within ≤4.5 hours of symptom onset or the presence of a hypoperfusion-ischemic core mismatch on CT perfusion when the exact onset time was unknown; and (4) receipt of IVT with either 0.25 mg/kg tenecteplase (rhTNK-tPA; China Shijiazhuang Pharmaceutical Company Recomgen Pharmaceutical, Guangzhou, China) or 0.9 mg/kg alteplase, followed by endovascular treatment.
The exclusion criteria were as follows: (1) ischemic stroke not attributable to LVO, (2) receipt of other thrombolytic therapies, (3) inadequate image quality or substantial missing clinical data; and (4) transfer to another hospital without available follow-up imaging data.
Data collection
The study data were primarily extracted from the registry database, and additional information was obtained through medical record review when necessary. Data collected included demographic characteristics (age and sex), relevant medical history (hypertension, hyperlipidemia, diabetes, coronary heart disease, atrial fibrillation, prior stroke, antiplatelet use, and smoking status), baseline National Institutes of Health Stroke Scale (NIHSS) score, Alberta Stroke Program Early Computed Tomography Score (ASPECTS), prestroke modified Rankin Scale (mRS) score, and critical time points (including symptom onset, IVT administration, groin puncture, and recanalization).
All enrolled patients underwent baseline noncontrast CT and CTA, followed by digital subtraction angiography (DSA) for further evaluation. We assessed the occlusion site on baseline CTA and initial DSA, the clot burden score on CTA, and the modified Thrombolysis in Cerebral Infarction (mTICI) grade on the initial and final DSA runs. The clot burden score was evaluated using a 10-point scoring system (range, 0-10), in which a score of 10 indicates the absence of thrombus with normal vasculature, one point is deducted for each predefined arterial segment not visualized with contrast, and a score of 0 indicates complete occlusion of the major ipsilateral circulation vessels [24].
Definition of outcome variables
The primary outcome was thrombus dynamics, classified as resolution, migration, or stability. Thrombus resolution was defined as complete disappearance of the target occlusion on initial DSA, corresponding to an mTICI grade of 3. Thrombus migration was defined as distal displacement of the occlusion site on initial DSA relative to the initial thrombus position identified on CTA. The location of the migrating thrombus was classified according to the involved cerebral circulation segment: the ICA; the proximal or distal M1 segment; the M2, M3, or M4 segments; the A1 or A2 segments of the anterior cerebral artery; the BA; or the P1 or P2 segments of the posterior cerebral artery. Proximal and distal M1 occlusions were defined relative to the midpoint of the contralateral M1 segment on coronal images. Two experienced neuroradiologists, blinded to all clinical data—including treatment allocation (tenecteplase or alteplase) and patient outcomes—independently assessed the occlusion site on baseline CTA and on initial DSA. In cases of disagreement, a third senior neuroradiologist with more than 15 years of experience served as an arbiter, and the adjudicated decision was considered final.
The primary clinical outcome was the mRS score at 3 months, ascertained by neurologists through telephone interview or clinic visit. Assessors who collected clinical outcome data were blinded to care pathway information. Functional independence was defined as an mRS score of 0-2.
Imaging outcomes were assessed using the mTICI scale on the initial and final DSA runs. Early reperfusion before MT was defined as achievement of mTICI 2b-3 on the initial DSA run. Successful reperfusion after MT was defined as mTICI 2b-3 on the final DSA run. Complete reperfusion after MT was defined as mTICI 3 on the final DSA run.
Safety outcomes included parenchymal hematoma type 2 (PH-2) according to the European Cooperative Acute Stroke Study II criteria [25], symptomatic intracranial hemorrhage (sICH) according to the Safe Implementation of Thrombolysis in Stroke-Monitoring Study criteria [26], and all-cause mortality (defined as an mRS score of 6).
Statistical analysis
Data are presented as median (interquartile range) for continuous variables and number (percentage) for categorical variables. Group comparisons were performed using the χ2 test, Fisher’s exact test, or Mann-Whitney U test, as appropriate. Interrater reliability was assessed using Cohen’s kappa. To account for potential confounding, we performed 1:1 optimal propensity score matching without replacement, with a caliper width of 0.2 times the standard deviation of the logit of the propensity score. The propensity score model included the following baseline variables: age, sex, baseline NIHSS score, occlusion site, clot burden score, history of prior stroke, onset-to-IVT time, onset-to-puncture time, and year of treatment.
For all primary and secondary outcomes, mixed-effects logistic regression models were fitted, with center included as a random effect. Fixed-effect adjustments comprised prestroke mRS score, ASPECTS, baseline NIHSS score, onset-to-puncture time, occlusion site, clot burden score, history of antiplatelet use, diabetes, and year of treatment. Interaction effects between treatment and prespecified subgroups (IVT-to-puncture time, occlusion site, and clot burden score) were assessed by including multiplicative interaction terms in multivariable logistic regression models. The P-value for interaction was derived from the likelihood ratio test. The potentially nonlinear association between IVT-to-puncture time and thrombus migration was examined using restricted cubic splines. A threshold-effect analysis was subsequently conducted using a two-piecewise linear regression model, adjusting for the same covariates as in the primary analysis, to identify the specific time cutoff at which the probability of thrombus migration exhibited an inflection point. All P-values were two-sided. A two-sided P-value <0.05 was considered statistically significant. All analyses were performed using IBM SPSS Statistics (version 27; IBM Corp., Armonk, NY, USA) and R software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria).
Results
A total of 906 consecutive patients with acute ischemic stroke who received IVT (either 0.25 mg/kg tenecteplase or 0.9 mg/kg alteplase) followed by MT between January 2022 and December 2024 were included. After application of the exclusion criteria (unavailable external images, n=26; missing DSA data, n=32; missing CTA data, n=24; no LVO, n=18), 806 patients constituted the final study cohort, including 431 in the alteplase group and 375 in the tenecteplase group. In addition, 3-month mRS outcomes were unavailable for 40 patients in the alteplase group and 28 in the tenecteplase group (Supplementary Figure 1). After propensity score matching, 746 patients (373 in each group) were included in the primary analysis. All baseline characteristics were well balanced between the groups (Table 1).
Among the 746 propensity score-matched patients, thrombus stability was observed in 608 patients (81.5%), thrombus migration in 114 (15.3%), and thrombus resolution in 24 (3.2%). No significant difference was observed in the incidence of thrombus resolution between the alteplase and tenecteplase groups (3.8% vs. 2.7%; P=0.327). However, a significantly higher rate of thrombus migration was observed in patients receiving tenecteplase (19.3%) than in those receiving alteplase (11.3%) (P=0.002) (Supplementary Table 2). Table 2 provides a detailed overview of thrombus migration patterns in both groups. Thrombus migration from M1 to M2, as well as to M3 or M4, was most frequently observed in both groups. The interrater reliability for the assessment of thrombus dynamics was excellent (Cohen’s kappa=0.851).
The impact of IVT-to-puncture time on thrombus migration differed significantly between alteplase and tenecteplase (Pinteraction=0.043). When the interval was <60 minutes, thrombus migration occurred significantly more frequently with tenecteplase than with alteplase (18.6% vs. 6.2%; odds ratio [OR]: 3.77; 95% confidence interval [CI] 1.72-8.29). In contrast, when the interval ≥60 minutes, no significant difference in migration rates was observed between the two agents (19.7% vs. 15.0%; OR: 1.39; 95% CI 0.84-2.30) (Table 3). Restricted cubic spline analysis demonstrated a nonlinear association between IVT-to-puncture time and the likelihood of thrombus migration with alteplase (Pnonlinear=0.004) (Figure 1A). Threshold analysis identified a significant association when the interval was <62 minutes (OR per minute: 1.08; 95% CI 1.01-1.15; P=0.034), but not when it was ≥62 minutes (OR: 0.99; 95% CI 0.99-1.00; P=0.235). For tenecteplase, no nonlinear association with time was observed (Pnonlinear=0.940) (Figure 1B). As shown in Figure 1C, the thrombus migration rate was consistently higher with tenecteplase within the 0-60-minute window (detailed ORs and P-values are provided in Supplementary Tables 3 and 4), whereas the rates between the two agents were comparable beyond 60 minutes.
The treatment effect also varied according to occlusion site. For M1 occlusions, tenecteplase was associated with a significantly higher thrombus migration rate than alteplase (30.7% vs. 15.7%; OR: 2.46; 95% CI 1.35-4.49). No significant differences were observed for occlusions of the ICA, M2, or BA (Pinteraction=0.696). In the subgroup analysis by clot burden, tenecteplase showed a numerically higher thrombus migration rate than alteplase in patients with low clot burden (scores 8-10; 24.3% vs. 13.2%; OR: 2.24; 95% CI 1.09-4.61), whereas no such difference was observed in those with higher clot burden (Pinteraction=0.642) (Table 3).
As shown in Table 4, logistic regression models were used to evaluate the associations between thrombus status and 3-month functional outcomes. Compared with patients with stable thrombus, those with thrombus migration had significantly better functional outcomes (OR: 1.62; 95% CI 1.04-2.53). Similarly, patients with thrombus resolution had improved functional outcomes (OR: 2.93; 95% CI 1.12-7.67). The rate of functional independence (mRS 0-2) did not differ significantly between the thrombus resolution and migration groups (74.9% vs. 61.9%; OR: 0.54; 95% CI 0.18-1.63). The distribution of mRS scores at 3 months according to thrombus dynamics is presented in Supplementary Figure 2.
Compared with patients with thrombus stability, those with thrombus migration had a significantly higher rate of early reperfusion (mTICI 2b-3) before MT (39.5% vs. 8.7%; P<0.001). Although the rate of successful reperfusion (mTICI 2b-3) after MT was similar between the two groups (86.8% vs. 86.0%; P=0.971), complete reperfusion (mTICI 3) was achieved less frequently in the thrombus migration group (40.4% vs. 68.8%; P< 0.001) (Table 5). Specifically, among the 114 patients with thrombus migration, 68 did not achieve mTICI 3 reperfusion. Of these, 39 cases were attributed to technical failure, whereas the remaining 29 underwent intentional procedure termination after satisfactory antegrade flow had been restored in the proximal territory. Patients with early reperfusion before MT had a higher rate of functional independence (mRS 0-2) at 3 months compared with those without early reperfusion (72.7% vs. 48.1%; OR: 3.02; 95% CI 1.90-4.81) (Supplementary Figure 3). In contrast, the rate of functional independence did not differ significantly between patients who achieved mTICI 2b and those who achieved mTICI 3 after MT (49.0% vs. 54.1%; OR: 1.22; 95% CI 0.83-1.79) (Supplementary Figure 4). The incidence rates of PH-2 and sICH were similar between the two groups. Comparative analysis of 3-month functional outcomes showed that thrombus migration was associated with improved functional outcome rates. Patients with thrombus migration had significantly higher rates of excellent functional outcome (mRS 0-1) (30.8% vs. 42.9%; OR: 1.67; 95% CI 1.08-2.59), functional independence (mRS 0-2) (49.4% vs. 61.9%; OR: 1.62; 95% CI 1.04-2.53), and good functional outcome (mRS 0-3) (63.4% vs. 74.3%; OR: 1.66; 95% CI 1.02-2.71). Mortality rates (mRS 6) were comparable between the thrombus stability and thrombus migration groups (11.5% vs. 7.6%; OR: 0.66; 95% CI 0.30-1.47) (Table 5).
After adjustment for relevant covariates, the 3-month functional outcomes were analyzed using the mRS score to compare alteplase and tenecteplase. Tenecteplase was associated with a higher rate of 3-month functional independence than alteplase (mRS 0-2: 48.1% vs. 56.2%, OR: 1.43; 95% CI 1.04-1.95). Similar efficacy was observed between the two groups, with comparable proportions of patients achieving mRS 0-1 (31.6% vs. 34.5%, OR: 1.15; 95% CI 0.83-1.59) and mRS 0-3 (64.0% vs. 67.8%, OR: 1.19; 95% CI 0.86-1.66) (Figure 2).
Discussion
Our study found that tenecteplase was superior to alteplase in promoting thrombus migration and was associated with higher rates of 3-month functional independence (mRS 0-2). The thrombus migration effect of tenecteplase was time dependent, and its advantage was most pronounced when the IVT-to-puncture interval was within 60 minutes. Moreover, tenecteplase-induced thrombus migration occurred more frequently in patients with M1 occlusions and a low clot burden. Although this migration was associated with a lower rate of complete reperfusion (mTICI 3) after thrombectomy, it facilitated early reperfusion before MT. Importantly, thrombus migration was independently associated with improved functional outcomes at 3 months.
Analysis of thrombus migration rates revealed a rate of 19.3% in the tenecteplase cohort, consistent with other reports [23,27]. In contrast, the rate observed in the alteplase group was 11.3%. Prior investigations have documented substantial heterogeneity in thrombus migration rates following alteplase administration, with reported figures ranging from 9% to 54% [11,23,28]. This variability across studies was likely driven by differences in the distribution of occlusion sites and IVT-to-puncture intervals. In our cohort, MCA occlusions accounted for less than half of the cases, and the median IVT-to-puncture duration was 66 minutes and 68 minutes, which is relatively short compared with certain prior studies [8,9]. The composition of our cohort may explain the lower rate of thrombus migration observed with alteplase compared with previous reports.
Our results showed that IVT-to-puncture time differentially affected thrombus migration for the two agents. Whereas alteplase-induced migration tended to increase with longer intervals, consistent with prior reports [9,11], tenecteplase-induced migration remained stable over time, resulting in superior efficacy within the first hour. This time-limited advantage is consistent with our previous work, which demonstrated that tenecteplase-induced early recanalization is also time dependent [29]. This pronounced temporal dependence is attributable to tenecteplase’s unique pharmacokinetic profile, which determines the early rate of thrombolysis. Administered as a single full bolus, tenecteplase achieves an immediate high peak plasma concentration, whereas standard alteplase is delivered as a small initial bolus followed by prolonged infusion, resulting in a slower and less intense pharmacodynamic profile without an early lytic peak. With a half-life of approximately 20 minutes, tenecteplase is largely cleared within 1 hour, creating a brief critical window of intense fibrinolytic activity [30,31]. Beyond this window, its incremental reperfusion advantage diminishes, and thrombus migration rates converge between the two agents. This pattern suggests that pharmacokinetics-driven early lysis speed accounts for the time-dependent difference [32]. A thrombus that persists intact through this peak activity period may have a more resistant fibrin architecture or stronger adhesion to the vessel wall, rendering it relatively resistant to further pharmacological lysis [33]. The rapid and potent action of tenecteplase within this critical early window provides a plausible mechanistic explana-tion for its superior capacity to induce early reperfusion and improve functional outcomes.
In our cohort, the increase in thrombus migration associated with tenecteplase was most pronounced in M1 occlusions, with a similar trend observed in ICA occlusions, whereas no significant difference was observed for M2 or BA occlusions. This sitespecific pattern suggests that the pharmacologic advantage of tenecteplase may be more prominent in the setting of larger, proximal thrombi. Recent evidence indicates that tenecteplase is associated with a higher likelihood of early recanalization in thrombi longer than 10 mm and with improved functional outcomes in patients with large ischemic cores—clinical scenarios typically characterized by proximal LVOs [34,35]. Consequently, the enhanced efficacy of tenecteplase appears most relevant for mobilizing larger, more resistant proximal thrombi, whereas for smaller distal clots, the choice between thrombolytic agents may be less consequential for facilitating migration. Moreover, collateral status has been widely recognized in recent literature as an important determinant of thrombus migration, exerting its effects through hemodynamic forces and distal perfusion pressure [8,36]. Future prospective studies with standardized collateral grading will be critical to further validate and extend these findings.
Our study observed that thrombus migration was associated with lower rates of complete reperfusion (mTICI 3) after MT yet with improved 90-day functional independence (mRS 0-2), which appears paradoxical. A multifactorial explanation may reconcile this finding. First, migration itself may serve as an in vivo marker of favorable thrombus biology. Clots prone to migration are likely more erythrocyte-rich and permeable—characteristics associated with greater lytic responsiveness and intrinsically better outcomes [37]. Second, and importantly, thrombus migration was associated with early reperfusion before thrombectomy. Distal displacement of the occlusion (e.g., from M1 to M2, M3, or M4) may promptly restore antegrade flow to downstream salvageable tissue,7 thereby preserving a substantial penumbral volume and contributing to a smaller final infarct core [38]. Furthermore, by relieving the proximal outflow obstruction, distal migration may augment collateral circulation (e.g., via leptomeningeal pathways), which in turn contributes to improved perfusion [39]. We acknowledge the technical challenge that distal migration can complicate device access and reduce the rate of complete reperfusion (mTICI 3), as reported in prior studies [5,6,8,9,15,23]. However, the overall effect of thrombus migration in our cohort appeared favorable with respect to functional independence.
Our study found that tenecteplase was associated with improved 3-month functional independence compared with alteplase. In patients with LVO, Intravenous Tenecteplase Before Thrombectomy in Stroke established the superiority of tenecteplase bridging therapy over thrombectomy alone, with a median IVT-topuncture time of 16 minutes [40]. The Tenecteplase versus Alteplase before Thrombectomy for Ischemic Stroke study demonstrated that tenecteplase improved pre-thrombectomy recanalization and 3-month functional independence compared with alteplase, with a median IVT-to-puncture time of 43 minutes [21]. However, a prespecified secondary analysis of the intravenous tenecteplase compared with alteplase for acute ischaemic stroke in Canada study showed that pre-thrombectomy recanalization and 90-day functional independence rates were similar for both agents, with a median IVT-to-reperfusion assessment time of 53 minutes [22]. The time-dependent benefit of tenecteplase—primarily confined to a one-hour window—provides a plausible explanation for these discrepant results across trials. Further large-scale randomized controlled trials are warranted to evaluate the efficacy of tenecteplase for rapid bridging to MT in patients with LVO (ClinicalTrials.gov number: NCT06658197).
Our study has several limitations. First, as a retrospective analysis of prospectively collected multicenter registry data, it is subject to potential selection bias and unmeasured confounding, despite adjustment for key covariates in the statistical models. Second, we did not evaluate the histological composition or size of the thrombus, nor did we assess collateral status—factors that are crucial determinants of thrombus migration. Third, the openlabel administration of thrombolytic therapy may have introduced performance bias, although outcome assessors were blinded to treatment allocation.
Conclusions
Tenecteplase demonstrated superior efficacy over alteplase in inducing thrombus migration, particularly when administered within a critical 60-minute window after IVT. Thrombus migration was associated with higher rates of functional independence. These findings suggest that tenecteplase may offer advantages over alteplase for rapid bridging to MT in patients with LVO, although further confirmation in randomized controlled trials is warranted.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
Print
