Abstract
Background:
Patients with acute ischemic stroke harboring a large vessel occlusion (LVO) admitted to non endovascular-capable centers often require inter-hospital transfer for thrombectomy. We evaluated the incidence and predictors of arterial recanalization during transfer, as well as the relationship between inter-hospital recanalization and clinical outcomes.
Methods:
We analyzed data from two cohorts of patients with an anterior circulation LVO transferred for consideration of thrombectomy to a comprehensive center, with arterial imaging at the referring hospital and upon comprehensive stroke center arrival. Inter-hospital recanalization was determined by comparison of the baseline and post-transfer arterial imaging, and was defined as revised Arterial Occlusive Lesion (rAOL) score 2b-3. Pre-transfer variables independently associated with inter-hospital recanalization were studied using multivariable logistic regression analysis.
Results:
Out of the 520 included patients (Montpellier, France, n=237; Stanford, USA, n=283), 111 (21%) experienced inter-hospital recanalization (partial [rAOL=2b] in 77% and complete [rAOL=3] in 23%). Pre-transfer variables independently associated with recanalization were intravenous thrombolysis (adjusted OR=6.8 [95%CI 4.0–11.6]), more distal occlusions (intracranial carotid occlusion as reference: adjusted OR=2.0 [0.9–4.5] for proximal M1, 5.1 [2.3–11.5] for distal M1 and 5.0 [2.1–11.8] for M2), and smaller clot burden (clot burden score 0–4 as reference: adjusted OR=3.4 [1.5–7.6] for 5–7 and 5.6 [2.4–12.7] for 8–9). Recanalization on arrival at the comprehensive center was associated with less inter-hospital infarct growth (rAOL=0–2a: 11.6mL, rAOL=2b: 2.2mL, rAOL=3: 0.6mL, P for trend<0.001) and greater inter-hospital NIHSS score improvement (0 vs. −5 vs. −6, P for trend<0.001). Inter-hospital recanalization was associated with reduced 3-month disability (adjusted common OR=2.51 [95%CI 1.68–3.77]), with greater benefit from complete than partial recanalization.
Discussion:
Recanalization is frequently observed during inter-hospital transfer for thrombectomy, and is strongly associated with favorable outcomes, even when partial. Broadening thrombolysis indications in primary centers, and developing therapies that increase recanalization during transfer, will likely improve clinical outcomes.
Graphical Abstract
INTRODUCTION
In patients with acute ischemic stroke due to a large vessel occlusion (LVO), the benefit of endovascular therapy (EVT) to improve functional outcome is well established up to 24 hours from symptom onset, owing to >85% rates of arterial recanalization.1 However, despite successful recanalization the efficacy of EVT decreases as time elapses, due to the progression of the irreversibly injured ischemic brain tissue over time.2,3 Comprehensive stroke centers (CSCs) that perform EVT are scarce in most countries, and the majority of LVO-related stroke patients are first evaluated at primary stroke centers (PSCs) before being transferred to a CSC for EVT, leading to delayed recanalization, larger infarcts, and worse functional outcome.4–6
Therapies that aim to achieve early arterial recanalization during inter-hospital transfer − before the EVT procedure − have the potential to substantially improve functional outcome. Currently, intravenous thrombolysis (IVT) is the only recommended therapy, but only leads to 15–30% rates of inter-hospital recanalization in LVO patients.7–9 Moreover, post-IVT recanalization occurring during transfer is often partial, and distal clot migration could reduce the likelihood of achieving complete recanalization during EVT.10–14 This possibility has led to uncertainty whether IVT should be administered before EVT.
Demonstrating the clinical benefits of recanalization – even if partial – occurring during inter-hospital transfer for EVT is of major importance, as it may inform whether IVT is useful before a planned EVT procedure. Moreover, it may reinforce the need for developing new therapies targeting recanalization during transfer, as well as broadening IVT indications. To our knowledge, no study has reported the association between recanalization during transfer and inter-hospital infarct growth or 3-month clinical outcome. The association between pre-EVT recanalization and functional outcome has been reported in a few studies that combined data from patients directly admitted to CSCs and those transferred from PSCs. Some of these studies reported improved outcomes in patients with pre-EVT arterial recanalization,11,15 while others reported no differences.10,12
In this study, we aimed to determine the incidence of recanalization during inter-hospital transfer for EVT, the clinical and imaging predictors of recanalization, and the relationship between recanalization and immediate post-transfer and 3-month clinical outcomes.
METHODS
The data supporting the study findings are available upon reasonable request.
Study design, data sources and inclusion criteria
This study combined data from two prospectively-collected cohort studies of acute stroke patients consecutively admitted to two CSCs (Stanford Hospital, Palo Alto, USA, and Montpellier Hospital, France) for consideration of EVT. Patients were included in the analysis if they fulfilled the following criteria: (1) initial admission at a PSC where a standard-of-care MRI with MR-angiography or CT with CT-angiography was performed showing an occlusion of the intracranial internal carotid artery (ICA) and/or the first (M1) or second (M2) segment of the middle cerebral artery, (2) subsequent transfer to a CSC for consideration of EVT, regardless of whether EVT was eventually attempted, and (3) arterial imaging performed upon arrival in the CSC − either CT-angiography, MR-angiography or digital substraction angiography (DSA). Inclusion dates were November 2019 to January 2023 for the US CSC and January 2015 to January 2017 for the French CSC. In the two cohorts, the arterial imaging evaluation in the PSC and upon arrival at the CSC were both required by protocol. In the US cohort, 6 out of the 18 referring hospitals were in the CSC metropolitan area, and 4/18 were part of a telestroke program. In the French cohort the five referring PSCs were not located in the same metropolitan area than the CSC and were not part of a telestroke program.
Our analysis was reported according to the Strengthening the Reporting of Observational Studies in Epidemiology criteria for observational studies.16 The research was approved by the Stanford review board for the US Cohort and by the Rothschild Foundation Hospital review board for the French cohort. In the US Cohort, each participant signed a written informed consent. In the French cohort, the requirement for written informed consent was waived as this study only implied retrospective analysis of anonymized data collected as part of routine care.
Clinical Data
The following variables were collected: age, sex, vascular risk factors, NIHSS scores at the PSC and upon CSC admission, IVT use, systolic blood pressure and serum glucose at PSC, transfer duration, and 3-month modified Rankin Scale (mRS) score. Transfer duration was defined as the time between the PSC imaging and repeat non-invasive imaging or groin puncture (for patients with direct transfer to the angiosuite) upon CSC arrival. NIHSS score change during inter-hospital transfer was defined as NIHSSCSC – NIHSSPSC; a negative value indicated clinical improvement and a positive value indicated clinical deterioration. The NIHSS score at the PSC were evaluated by on-site neurologists in the French cohort, and by on-site or remote (telemedicine) neurologists or emergency department physicians in the US cohort. The NIHSS score upon CSC admission was evaluated by neurologists in both cohorts. The 3-month mRS was assessed by a stroke neurologist or research nurse either during a face-to-face visit or by phone interview.
Radiological Data
Referring hospital
All included patients underwent either MRI or CT on admission at the PSC. Multimodal CT, including non-contrast CT, CT-angiography and CT-perfusion was the routine first-line imaging technique for EVT candidates in all PSCs in the US cohort. MRI was the routine first-line imaging tool in all PSCs in the French cohort, with a standardized protocol at each institution, which systematically included diffusion-weigthed imaging (DWI), T2*, and intracranial MR-angiography; perfusion imaging was rarely performed.
Comprehensive Stroke Center
During the study period, routine non-invasive post-transfer imaging was obtained as standard of care in both centers prior to the EVT procedure. The routine first-line imaging technique was MRI in both cohorts, which included DWI, T2*, and intracranial MR-angiography. In rare instances patients went directly to the cathlab (i.e. bypassing the non-invasive imaging at the CSC), at the discretion of the attending physicians. In these cases the first run of the DSA was used to assess inter-hospital recanalization.
Imaging analysis
One stroke neurologist with 10-years of clinical expertise in stroke imaging (PS) reviewed all imaging, blinded to clinical data. The following variables were collected: (1) Occlusion site on MR- or CT-angiography, divided into intracranial ICA, proximal M1, distal M1 and M2. The M1 segment was defined as the first portion of the middle cerebral artery up to the main bifurcation and dichotomized as proximal or distal based on the middle cerebral artery origin-to-clot interface distance (<10 and ≥10 mm, respectively).9,17 (2) Extent of intracranial thrombus, assessed using the clot burden score (from 0 to 10, a score of 0 indicates complete occlusion of the ipsilateral anterior circulation arteries; a score of 10 implies no occlusion) on CT-angiography18,19 or on T2*/ susceptibility weigthed imaging using the susceptibility vessel sign.20 For patients with MRI, the clot burden score was not evaluable in patients without susceptibility vessel sign. (3) Infarct volume on DWI or CT-perfusion. On DWI, infarct volume was manually outlined based on DWI signal intensity encompassing the entire area of bright DWI signal intensity using Horos (Horos Project, version 3.3.6).21 On CT-perfusion, infarct volume was automatically measured using RAPID software using the relative cerebral blood flow <30% of normal brain. Artifacts were removed manually using an in-house tool. (4) Inter-hospital infarct growth. In patients who had MRI in both centers this was assessed as the volumetric difference between the PSC and the CSC infarct volumes.21 In patients with CT-perfusion in the PSC and MRI upon arrival in the CSC, infarct growth was assessed following co-registration of the two imaging modalitites, since brain coverage was usually incomplete on CT-perfusion. Co-registration was done using SimpleElastix, version 2.0.0. Mutual information similarity metrics were used for cross modality registration. All registration results were inspected visually to ensure good alignment. (5) Leptomeningeal collateral status was assessed at the PSC by the Hypoperfusion Intensity Ratio (HIR), whenever perfusion imaging was available, and defined as the proportion of the Tmax>6s lesion volume that had a Tmax delay of >10s (i.e. Tmax>10s volume / Tmax>6s volume). Low HIR indicates milder hypoperfusion and better collaterals.22 (6) Evidence of inter-hospital hemorrhagic transformation on non-contrast CT or T2* imaging was assessed according to the ECASS (European Cooperative Acute Stroke Study) II classification.
Inter-hospital recanalization evaluation
Inter-hospital recanalization was evaluated by one reader (PS) with head-to-head comparison of the PSC angiography (MR- or CT-angiography) and the post-transfer CSC angiography (MR or CT or first run of the DSA when direct transfer to angiosuite occurred), and rated on the revised Arterial Occlusion Lesion score (rAOL, from 0, no recanalization to 3, complete recanalization).18 The evaluator was blinded from clinical and other brain imaging data. Substantial recanalization was defined as rAOL 2b (partial or complete recanalization of the primary lesion with thrombus/occlusion in minor vascular branch [M2 or beyond] or partial recanalization of the primary lesion with no thrombus in the vascular tree at or beyond the primary occlusive lesion) or 3 (complete recanalization of the primary occlusion with no clot in the vascular tree at or beyond the primary occlusive lesion).18 Examples of rAOL classification are provided in Supplemental Figure 1. To assess reproducibility of inter-hospital recanalization assessment, another neurologist (AW) evaluated recanalization of a random subset of 52 patients (n=24 with rAOL 2b or 3), blinded from the first reader’s ratings.
Statistical analysis
Continuous variables were described as mean±standard deviation or median (interquartile range, IQR), as appropriate, and categorical variables as counts and percentages. Interobserver agreement for inter-hospital recanalization classification was measured using weighted Kappa statistics. Univariable relationships between pre-transfer characteristics and inter-hospital recanalization were assessed using Student’s t test or the Mann-Whitney U test for continuous variables, and the Chi-square test or Fisher’s exact test for categorical variables, as appropriate. To adjust for potential confounders, multivariable binary logistic regression analysis was performed with inter-hospital recanalization as the dependent variable. Variable selection was performed stepwise, whereby variables that were significant at P<0.20 in the univariable analysis were entered into the model and were retained only if they remained associated at P<0.05 with the dependent variable in the multivariable model. Because occlusion site and clot burden score both measure thrombus extent, these two variables were included in separate models. To ensure that the association between clot burden score and recanalization is similar across imaging modalities (MRI vs. CT), an imaging modality*clot burden score interaction was assessed. The univariable associations between inter-hospital recanalization and key immediate post-transfer outcomes (inter-hospital clinical change, infarct growth, hemorrhagic transformation) were assessed. We used Jonckheere-Terpstra test for ordered alternatives to assess hypothesis for gradual differences in NIHSS change and infarct growth during transfer along levels of inter-hospital recanalization (rAOL 0–2a vs. 2b vs. 3). The relationship between inter-hospital recanalization and 3-month mRS was assessed using ordinal logistic regression, adjusted for center, age, pre-transfer occlusion site and NIHSS score, and time from last-seen-well to CSC arterial imaging. The model was not adjusted for subsequent EVT treatment, since this variable is strongly affected by inter-hospital recanalization. Hypothesis for proportional odds assumption was tested and results were reported as a common OR (cOR) and its 95% CI. All statistical tests were 2-tailed, and the threshold for statistical significance was set to P<0.05. Statistical analyses were conducted using SPSS 28.0 (IBM, Armonk, NY) and SAS v.9.4 (SAS Institute).
RESULTS
Study Population
A total of 520 patients were eligible for the study (237 in the French cohort and 283 in the US cohort, see flow chart in Supplemental Figure 2). The pre-transfer characteristics of included patients are detailed in Table 1. Median age was 72 years (IQR, 61–81), 55% of patients were male, median NIHSS score in the PSC was 15 (IQR, 10–20) and last seen well to PSC imaging time was 2.3 hrs (IQR, 1.3–6.4). Baseline intracranial occlusion site was ICA in 21%, M1 in 61% and M2 in 19%. The clot burden score, evaluable in 489/520 (94%) patients, was 0 to 4 (i.e. larger clots) in 21%, 5–7 in 46% and 8–9 (i.e. smaller clots) in 33%. IVT was administered before the transfer in 259/520 (50%) patients, with alteplase 0.9mg/kg in 254 (98%) and tenecteplase 0.25mg/kg in 5 (2%). Reasons for withholding IVT were being outside the 4.5hr time-window in 147/261 (56%) patients, anticoagulants use in 56/261 (21%), recent surgery or bleeding in 15/261 (6%), recent prior ischemic stroke in 13/261 (5%), large core in 7/261 (3%), other reason in 10/261 (4%) and multiple reasons in 13/261 (5%). Median transfer time was 3.0 hrs (IQR, 2.5–3.7). The comparison of main characteristics in the two participating sites is provided in Supplemental Table 1.
Table 1 –
Univariable Comparison Of Pre-transfer Characteristics According To The Inter-Hospital Recanalization Status
| Overall N=520 |
Recanalization N=111 |
No Recanalization N=409 |
P-value | |
|---|---|---|---|---|
| Age | 72 (61–81) | 74 (62–82) | 71 (61–81) | 0.23 |
| Male | 284 (55) | 66 (60) | 218 (53) | 0.25 |
| Hypertension | 349 (67) | 66 (60) | 283 (69) | 0.05 |
| Diabetes | 112 (22) | 16 (14) | 96 (24) | 0.04 |
| Dyslipidemia | 200 (39) | 41 (37) | 159 (39) | 0.71 |
| Pre-stroke antiplatelets | 166 (32) | 39 (35) | 127 (32) | 0.48 |
| Pre-stroke anticoagulant | 81 (16) | 12 (11) | 69 (17) | 0.10 |
| Center | 0.34 | |||
| Stanford | 283 (54) | 56 (50) | 227 (56) | |
| Montpellier | 237 (46) | 55 (50) | 182 (44) | |
| Clinical characteristics | ||||
| NIHSS score | 15 (10–20) | 15 (10–20) | 15 (10–20) | 0.75 |
| Glucose, mg/dL | 119 (104–143) | 114 (101–129) | 120 (106–149) | <0.01 |
| Systolic blood pressure, mmHg | 148 (128–167) | 148 (127–162) | 148 (130–168) | 0.20 |
| Imaging characteristics | ||||
| LSW-to-imaging, hrs | 2.3 (1.3–6.4) | 1.8 (1.1–2.9) | 2.8 (1.4–7.7) | <0.01 |
| Occlusion site | <0.01 | |||
| ICA | 109 (21) | 9 (8) | 100 (24) | |
| M1 proximal | 175 (34) | 29 (26) | 146 (36) | |
| M1 distal | 140 (27) | 46 (41) | 94 (23) | |
| M2 | 96 (19) | 27 (24) | 69 (17) | |
| Associated cervical ICA occlusion | 99 (19) | 17 (15) | 82 (20) | 0.26 |
| Clot burden score (n=489)a | <0.01 | |||
| 0–4 | 103 (21) | 8 (8) | 95 (25) | |
| 5–7 | 224 (46) | 49 (46) | 175 (46) | |
| 8–9 | 162 (33) | 49 (46) | 113 (30) | |
| Infarct volume, mL (n=474)b | 10 (0–25) | 10 (0–21) | 10 (0–25) | 0.82 |
| HIR (n=275)c | 0.43 (0.23–0.54) | 0.42 (0.22–0.54) | 0.43 (0.23–0.54) | 0.98 |
| IV-thrombolysis use | 259 (50) | 91 (82) | 168 (41) | <0.01 |
| Transfer time, hours | 3.0 (2.5–3.7) | 3.1 (2.5–3.7) | 3.0 (2.5–3.7) | 0.32 |
Categorical variables are expressed as numbers (%) and continuous variables as median (interquartile range).
Clot burden score was evaluated on CT-angiography or T2*/SWI imaging (susceptibility vessel sign). Data was missing in 31/520 (6%) patients without susceptibility vessel sign on T2*/SWI imaging (5 [5%] in the recanalization group and 26 [6%] in the non recanalization group).
Infarct volume was measured on diffusion-weighted imaging (manual delineation) or CT-perfusion (relative cerebral blood flow <30%). Data was missing in 46/520 (9%) patients without CT-perfusion or diffusion-weighted imaging (8 [7%] in the recanalization group and 38 [9%] in the non recanalization group).
HIR was defined as the proportion of Tmax>6s volume with Tmax>10s (i.e. Tmax>10s volume / Tmax>6s volume), low HIR indicating milder hypoperfusion and better collaterals. Data was missing in 245/520 (47%) patients without perfusion imaging (52 [47%] in the recanalization group and 193 [47%] in the non recanalization group).
Incidence Of Inter-Hospital Recanalization
Upon arrival in the CSC, the arterial imaging was MR-angiography in 74% of cases, CT-angiography in 17% and first run of DSA in 9%. Arterial recanalization was observed in 111/520 (21%) patients, and was rAOL 2b in 86/111 (77%) patients and rAOL 3 in 25/111 (23%). Among patients with rAOL 2b, the remaining occlusion site was proximal M1 in 1/86 (1%) patient (partially occlusive M1 clot, with clear improvement as compared to the PSC arterial imaging), proximal M2 in 37/86 (43%), distal M2 in 24/86 (28%), and distal (beyond M2) middle cerebral artery occlusion in 24/86 (28%). There was an excellent agreement between the two observers for recanalization classification (weighted Kappa: 0.95; 95%CI 0.90–1.00).
Pre-Transfer Characteristics Associated With Arterial Recanalization
Patients who recanalized during transfer were less likely to have diabetes, had lower blood glucose, shorter last seen well-to-PSC imaging time, more distal occlusions, higher clot burden scores (i.e. lower thrombus burden) and were more frequently treated with IVT than patients without recanalization (Table 1). Recanalization rate was similar across the two participating centers (P=0.34). Recanalization occurred in 91/259 (35%) IVT-treated patients compared with only 20/261 (8%) patients not treated with IVT. In IVT-treated patients, there was no association between the time from symptom onset to IVT start and inter-hospital recanalization occurrence (median time in no-recanalizers: 2.4hrs, IQR 1.8–3.4; recanalizers: 2.4hrs, IQR 1.8–3.4; P=0.97).
In multivariable analysis, a more distal occlusion site and IVT therapy were independently associated with inter-hospital recanalization (Table 2). In an alternative multivariable model with clot burden score instead of occlusion site, clot burden score (i.e. smaller thrombi) and IVT therapy were independently associated with arterial recanalization (Table 2). There was no clot burden score*imaging modality interaction (P for interaction = 0.10), indicating that there was no difference in association between clot burden score and recanalization across imaging modalities. The rates of inter-hospital recanalization according to occlusion site, clot burden score and IVT use are provided in Figure 1. They ranged from 2% in patients with ICA occlusion and no IVT use, to 50% in IVT-treated patients with high clot burden score (i.e. smaller thrombi) or distal M1/ M2 occlusion sites.
Table 2 –
Logistic Regression Models Showing Pre-Transfer Variables Independently Associated With Inter-hospital Arterial Recanalization
| Model with occlusion site | |||
|---|---|---|---|
| Unadjusted OR (95%CI) | Adjusted OR (95% CI)a | P value | |
| Intravenous thrombolysis use | 6.5 (3.9–11.0) | 6.8 (4.0–11.6) | <0.001 |
| Occlusion site | <0.001 | ||
| ICA | Reference | Reference | |
| M1 proximal | 2.2 (1.0–4.9) | 2.0 (0.9–4.5) | |
| M1 distal | 5.4 (2.5–11.7) | 5.1 (2.3–11.5) | |
| M2 | 4.3 (1.9–9.8) | 5.0 (2.1–11.8) | |
| Model with clot burden score | |||
| Unadjusted OR (95%CI) | Adjusted OR (95% CI)a | P value | |
| Intravenous thrombolysis use | 6.5 (3.9–11.0) | 6.6 (3.8–11.3) | <0.001 |
| Clot burden score | <0.001 | ||
| 0–4 | Referenc | Reference | |
| 5–7 | 3.3 (1.5–7.3) | 3.4 (1.5–7.6) | |
| 8–9 | 5.1 (2.3–11.4) | 5.6 (2.4–12.7) | |
Variables not retained in the model: hypertension, diabetes, pre-stroke anticoagulants, glucose level, last-seen-well to imaging time.
Figure 1 –
Arterial Recanalization During Inter-hospital Transfer According To Intravenous Thrombolysis Use, Occlusion Site And Clot Burden Score
Immediate Post-Transfer Outcomes Associated With Inter-hospital Recanalization
The immediate post-transfer clinical and radiological outcomes are provided in Table 3. Upon arrival in the CSC, median NIHSS score change was 0 (IQR, −3 ; 2), with greater NIHSS score improvement observed in patients with inter-hospital recanalization. Median inter-hospital change in NIHSS was 0 (IQR, −2 ; 2) in patients without inter-hospital recanalization, −5 (IQR, −9 ; −1) in patients with rAOL 2b and −6 (IQR, −10 ; −3) in those with rAOL 3 (P for trend test <0.001).
Table 3 –
Clinical and Imaging Outcomes Upon Arrival In the Comprehensive Center According To The Inter-Hospital Arterial Recanalization Status
| Overall N=520 |
Recanalization N=111 |
No Recanalization N=409 |
P-value | |
|---|---|---|---|---|
| NIHSS change during transfer (n=502)a | 0 (−3 ; 2) | −6 (−9 ; −1) | 0 (−2 ; 2) | <0.01 |
| Infarct growth, mL (n=313)b | 8.7 (1.4–23.9) | 1.5 (0–11.7) | 11.6 (2.1–25.9) | <0.01 |
| MRI-to-MRI (n=136) | 3.8 (0.8–18.8) | 0 (0–2.9) | 7.2 (1.4–19.7) | <0.01 |
| CTP-to-MRI (n=177) | 12.2 (2.5–26.6) | 3.2 (0.5–20.0 | 13.8 (4.8–28.6) | <0.01 |
| Any hemorrhagic transformation (n=479)c | 23 (5) | 8 (8) | 15 (4) | 0.14 |
| Endovascular therapy performed | 324 (62) | 23 (21) | 301 (74) | <0.01 |
Categorical variables are expressed as numbers (%) and continuous variables as median (interquartile range).
Defined as NIHSS at the comprehensive center – NIHSS at the referring hospital. A negative value indicates clinical improvement and a positive value clinical deterioration. Data missing in 18/520 (3%) patients (2 [2%] in the recanalization group and 16 [4%] in the non recanalization group).
Infarct growth was assessed in patients with MRI in both hospitals, or with CT-perfusion in the referring hospital and MRI upon arrival in the comprehensive center (see Methods for details). Data was missing in 207/520 (40%) patients (46 [41%] in the recanalization group and 161 [39%] in the non recanalization group).
Inter-hospital hemorrhagic transformation was not available in 41/520 (8%) patients with direct transfer to angiosuite (i.e. lack of non-invasive control imaging upon arrival in the comprehensive center); 4 [4%] in the recanalization group and 37 [9%] in the non recanalization group).
Adequate quality imaging to determine inter-hospital infarct growth was available in 313/520 (60%) patients (MRI-to-MRI in 136/313 and CT-perfusion-to-MRI in 177/313), with a median infarct growth of 9 mL (IQR, 1–24). Patients with infarct growth measurement had similar characteristices than those without (Supplemental Table 2). As compared to patients without recanalization during transfer, those who recanalized had less inter-hospital infarct growth, with similar results across the imaging modality used (Table 3). Less infarct growth was observed in patients with more complete recanalization: median inter-hospital infarct growth was 11.6mL (IQR, 2.1–25.9) in patients without inter-hospital recanalization, 2.2mL (IQR, 0–14.0) in patients with rAOL 2b and 0.6mL (IQR, 0–3.3) in those with rAOL 3 (P for trend test <0.001).
Inter-hospital hemorrhagic transformation occurred in 23/479 (5%) patients with post-transfer parenchymal imaging, without between group differences (Table 3). EVT was performed in 324/520 (62%) patients, less frequently in patients with than without inter-hospital recanalization (21% vs 74%, P<0.01). Among the patients who received EVT, an expanded TICI 2c or 3 reperfusion was reached in 178/299 (60%) patients without inter-hospital recanalization, and 11/23 (48%) patients with inter-hospital recanalization (P=0.26).
Relationship Between Inter-Hospital Recanalization And 3-month Functional Outcome
The 3-month mRS score was recorded for 478/520 (92%) patients. Inter-hospital recanalization was independently associated with a shift toward better 3-month outcome as compared to those without (unadjusted cOR=2.18; 95% CI 1.48–3.21; adjusted cOR=2.51; 95%CI 1.68–3.77; P<0.001; Figure 2A). There was an increase in the odds of improved outcome with increasing recanalization grade: with no recanalization as the reference, adjusted cOR=2.24 (95%CI 1.44–3.49) for rAOL 2b (unadjusted cOR=1.99 [95%CI 1.30–3.05]), and adjusted cOR=3.70 (95%CI 1.73–7.92) for rAOL 3 (unadjusted cOR=2.94 [95%CI 1.42–6.10]), P<0.001; Figure 2B.
Figure 2 – Relationship Between Inter-hospital Arterial Recanalization And 3-month Functional Outcome.
Panel A: inter-hospital arterial recanalization was associated with better 3-month functional outcomes (adjusted common OR=2.51; 95%CI 1.68–3.77; P<0.001). Panel B: There was an increase in the odds of improved outcome with increasing recanalization grade (no recanalization as the reference, adjusted common OR=2.24 [95%CI 1.44–3.49] for rAOL 2b and adjusted common OR=3.70 [95%CI 1.73–7.92] for rAOL 3, P<0.001. The 3-month mRS was missing in 42/520 (8%) patients (7/111 [6%] with recanalization and 35/409 [9%] without). mRS indicates modified Rankin Scale score. The models were adjusted for center, age, pre-transfer occlusion site and NIHSS score, and time from last-seen-well to comprehensive stroke center arterial imaging.
DISCUSSION
This study of 520 acute stroke patients with an LVO transferred from a PSC to a CSC for consideration of EVT has 3 key findings. First, the rate of inter-hospital recanalization is substantial, occurring in 1 out of 5 transferred patients. Second, IVT use and thrombus location or burden are strong independent predictors of recanalization. Third, inter-hospital recanalization − even when partial − is strongly associated with inter-hospital clinical improvement, less infarct growth, and better 3-month functional outcome.
The rate of inter-hospital recanalization observed in our study is similar to a large prospective multicenter study using the same scale and definition for recanalization, and with similar time between the baseline and early follow-up arterial imaging.18 In the prior study, recanalization occurred in 27% vs 21% in our study; in IVT-treated patients: 30% vs 35%; without IVT use: 13% vs 7%.18 Lower, 8–10%, inter-hospital recanalization rates were observed in two recent studies, which used a more restrictive definition of recanalization.7,8 In one study, recanalization was defined as a thrombus migration, which precluded EVT.7 Therefore, patients with substantial recanalization but with persistent M2 occlusion who underwent EVT were not considered as recanalized. Similarly, in the other study, patients with M2 occlusions upon CSC arrival were not considered recanalized, thereby underestimating the actual recanalization rate.8 In our study, recanalization during transfer was mostly partial with 56% of recanalizers still having an M2 occlusion. However, as further described below, improved clinical outcomes were also observed in the subgroup of patients with partial inter-hospital recanalization, suggesting this degree of recanalization is clinically relevant.
Intravenous thrombolysis administration was independently associated with a 7-fold higher likelihood of inter-hospital recanalization, consistent with previous studies.7,8,18 Occlusion site or clot burden was also independently associated with recanalization, in line with previous reports.9,18,23 Importantly, in patients with ICA occlusion or very low clot burden score (i.e. large thrombi) who received IVT, the inter-hospital recanalization rate was low but not negligible (~15%). Transfer time was not associated with recanalization in our study, similar to other studies in this population.7,8 Longer time from IVT to recanalization assessment has a well-known association with higher recanalization rates,9,15,18 but this relationship is not linear, with a steep increase in the first 90 minutes due to the short half-life of alteplase.9 In our study, median transfer time was longer than 90min (median time of 3hrs), which likely explains the lack of a relationship between transfer time and recanalization.
Inter-hospital recanalization was strongly associated with less inter-hospital infarct growth, greater inter-hospital clinical improvement, and reduced 3-month disability. Importantly, we observed greater benefit from complete (rAOL 3) than partial (rAOL 2b) recanalization on these outcomes. However, even rAOL 2b recanalization was associated with improved outcomes, suggesting that partial recanalization before EVT is benefical. These findings suggest that strategies that can increase recanalization rates and grade during inter-hospital transfer – before a planned EVT procedure – may reduce long-term disability.
Several new strategies look promising. First, using more efficient fibrinolytic therapies than alteplase. The randomized controlled trials comparing alteplase 0.9mg/kg to tenecteplase 0.25mg/kg in LVO-related acute stroke patients mostly focused on patients directly admitted in CSCs, and randomized data are lacking for patients transferred from a PSC.24,25 Two studies provide observational data in this setting, suggesting similar inter-hospital recanalization rates across the 2 thrombolytics,15,26 although there was a trend towards higher recanalization following tenecteplase in one of them.15 Further studies are warranted.
Second, current thrombolytic therapies only target the fibrin component of the clot. Accumulating evidence indicates that drugs aimed at disrupting the non-fibrin components of intracranial thrombi such as extracellular DNA, von Willebrand factor multimers, or platelet aggregates could increase the efficacy of thrombolytic therapies.27 Several trials are ongoing.27 Besides pharmalogical therapies, trans-cranial doppler (“sonothrombolysis”) plus IVT doubled the rate of early arterial recanalization as compared to IVT alone in a recent meta-analysis of pilot clinical trials.28 However, these results did not translate in improved outcome in a phase-3 trial using an operator-independent device, yet LVO identification before enrollment was not required, as it was in the pilot trials.29 One randomized trial testing sonothrombolysis in transferred patients is currently ongoing (NCT 03519737).
Broadening the indications of current thrombolytic therapies has the potential to substantially improve inter-hospital recanalization and 3-month outcome. Patients presenting outside the 4.5hr time-window was the main reason for withholding IVT (56% in our cohort) and may therefore represent the primary subgroup of interest for future trials. Several randomized trials have demonstrated the benefit of IVT in patients with unwitnessed stroke onset or in the 4.5–9h time-window selected with advanced imaging – either diffusion-FLAIR mismatch or core-perfusion mismatch –, but these trials excluded patients planned for EVT.30 Therefore, the benefit of IVT administered in the situation of planned EVT is uncertain. In the recently presented TIMELESS trial, tenecteplase administered within the 4.5–24h time-window in LVO-related acute stroke patients with a core-perfusion mismatch – 75% of whom being planned for EVT – did not improve functional outcome vs placebo, although it increased 24-hr recanalization rate and had no safety concerns.31 However, 96% of patients were enrolled in CSC, and further studies testing this approach in PSCs is warranted.
This study has limitations. First, clot burden score evaluation is different on CT-angiography and MRI (T2*), which may have impacted our results. However, the relationship between clot burden score and recanalization was not modified by imaging modality. Second, thrombus perviousness evaluated on CT-angiography, which has been reported as an independent predictor of recanalization,18 was not collected in our study because half of the population had MRI. Third, inter-hospital infarct growth could be measured in only 60% of patients, with an MR-to-MR comparison in 45% and a CT-perfusion-to-MR comparison in 55%. However, patients with and without infarct growth measurement had similar characteristics, and similar results were observed in patients with MRI-to-MRI vs CT-perfusion-to-MRI infarct growth measurement. Last, our transfer times were longer than in some prior studies. However, we did not find an association between transfer time and recanalization rates.
CONCLUSION
This study demonstrated that 1 out of 5 patients who are transferred from a PSC to a CSC for consideration of EVT experience inter-hospital recanalization, which can be predicted by IVT use and thrombus site or clot burden on baseline imaging. Inter-hospital recanalization − even when partial − is strongly associated with improved 3-month functional outcome. Therapies targeting arterial recanalization during inter-hospital transfer have the potential to substantially improve functional outcome.
Supplementary Material
Source of funding:
Pierre Seners received grants from Edmond de Rothschild Fellowship program, Bettencourt-Schueller Foundation, Servier Institute and Philippe Foundation.
Anke Wouters received a grant from Fonds Wetenschappelijk Onderzoek Vlaanderen (FWO) (V415223N)
The CRISP-2 study was funded by the National Institutes of Health.
Disclosures:
Dr Wouters reports grants from Remmert Adriaan-Laan-Fonds.
Dr Arquizan reports compensation from Amgen for other services and compensation from Medtronic Vascular, Inc. for other services.
Dr Christensen reports stock holdings in Ischemaview.
Dr COSTALAT reports compensation from Penumbra, Inc. for consultant services; compensation from MicroVention, Inc. for consultant services; compensation from Balt USA, LLC for consultant services; compensation from Stryker Corporation for consultant services; compensation from Medtronic USA, Inc. for consultant services; and compensation from Johnson & Johnson Health Care Systems Inc. for consultant services.
Dr Heit reports consulting fees from Medtronic and MicroVention, and he is a member of the medical and scientific advisory board for iSchemaView
Dr Albers reports stock holdings in iSchemaView; compensation from Biogen, iSchemaView and Genentech for consultant services.
Dr Lansberg reports no disclosures relevant to the manuscript.
Other authors have nothing to disclose.
NON-STANDARD ABBREVIATIONS AND ACRONYMS
- cOR
common odds ratio
- CSC
comprehensive stroke center
- DWI
diffusion-weighted imaging
- DSA
digital substraction angiography
- EVT
endovascular therapy
- HIR
hypoperfusion intensity ratio
- ICA
intracranial internal carotid artery
- IVT
intravenous thrombolysis
- LVO
large vessel occlusion
- M1
first segment of the middle cerebral artery
- M2
second segment of the middle cerebral artery
- mRS
modified Rankin scale score
- PSC
primary stroke center
- rAOL
revised arterial occlusive lesion scale
Footnotes
REFERENCES
- 1.Lapergue B, Blanc R, Gory B, Labreuche J, Duhamel A, Marnat G, Saleme S, Costalat V, Bracard S, Desal H, et al. Effect of endovascular contact aspiration vs stent retriever on revascularization in patients with acute ischemic stroke and large vessel occlusion: The ASTER randomized clinical trial. JAMA. 2017;318:443–452 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Saver JL, Goyal M, van der Lugt A, Menon BK, Majoie CB, Dippel DW, Campbell BC, Nogueira RG, Demchuk AM, Tomasello A, et al. Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: A meta-analysis. JAMA. 2016;316:1279–1288 [DOI] [PubMed] [Google Scholar]
- 3.Mulder M, Jansen IGH, Goldhoorn RB, Venema E, Chalos V, Compagne KCJ, Roozenbeek B, Lingsma HF, Schonewille WJ, van den Wijngaard IR, et al. Time to endovascular treatment and outcome in acute ischemic stroke: MRCLEAN registry results. Circulation. 2018;138:232–240 [DOI] [PubMed] [Google Scholar]
- 4.Gerschenfeld G, Muresan IP, Blanc R, Obadia M, Abrivard M, Piotin M, Alamowitch S. Two paradigms for endovascular thrombectomy after intravenous thrombolysis for acute ischemic stroke. JAMA Neurol. 2017;74:549–556 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Froehler MT, Saver JL, Zaidat OO, Jahan R, Aziz-Sultan MA, Klucznik RP, Haussen DC, Hellinger FR Jr., Yavagal DR, Yao TL, et al. Interhospital transfer before thrombectomy is associated with delayed treatment and worse outcome in the STRATIS registry (systematic evaluation of patients treated with neurothrombectomy devices for acute ischemic stroke). Circulation. 2017;136:2311–2321 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Romoli M, Paciaroni M, Tsivgoulis G, Agostoni EC, Vidale S. Mothership versus drip-and-ship model for mechanical thrombectomy in acute stroke: A systematic review and meta-analysis for clinical and radiological outcomes. J Stroke. 2020;22:317–323 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Flores A, Ustrell X, Seró L, Pellisé A, Rodriguez P, Viñas J, Ribó M, Krupinski J, Más N, Garcia S, et al. Vascular occlusion evolution in endovascular reperfusion candidates transferred from primary to comprehensive stroke centers. Cerebrovasc Dis. 2020;49:550–555 [DOI] [PubMed] [Google Scholar]
- 8.Kraft AW, Regenhardt RW, Awad A, Rosenthal JA, Dmytriw AA, Vranic JE, Bonkhoff AK, Bretzner M, Hirsch JA, Rabinov JD, et al. Spoke-administered thrombolysis improves large vessel occlusion early recanalization: The real-world experience of a large academic hub-and-spoke telestroke network. Stroke Vasc Interv Neurol. 2023;3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Seners P, Turc G, Naggara O, Henon H, Piotin M, Arquizan C, Cho T. Post-thrombolysis recanalization in stroke referrals for thrombectomy: Incidence, predictors, and prediction scores. Stroke 2018;49:2975–2982 [DOI] [PubMed] [Google Scholar]
- 10.Riegler C, Siebert E, Kleine JF, Nolte CH. Thrombus migration in ischemic stroke due to large vessel occlusion: A question of time. J Neurointerv Surg. 2023;15:e216–e222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Alves HC, Treurniet KM, Jansen IGH, Yoo AJ, Dutra BG, Zhang G, Yo L, van Es A, Emmer BJ, van den Berg R, et al. Thrombus migration paradox in patients with acute ischemic stroke. Stroke. 2019;50:3156–3163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kaesmacher J, Giarrusso M, Zibold F, Mosimann PJ, Dobrocky T, Piechowiak E, Bellwald S, Arnold M, Jung S, El-Koussy M, et al. Rates and quality of preinterventional reperfusion in patients with direct access to endovascular treatment. Stroke. 2018;49(8):1924–1932. [DOI] [PubMed] [Google Scholar]
- 13.Kaesmacher J, Maegerlein C, Kaesmacher M, Zimmer C, Poppert H, Friedrich B, Boeckh-Behrens T, Kleine JF. Thrombus migration in the middle cerebral artery: Incidence, imaging signs, and impact on success of endovascular thrombectomy. J Am Heart Assoc. 2017;6(2):e005149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ohara T, Menon BK, Al-Ajlan FS, Horn M, Najm M, Al-Sultan A, Puig J, Dowlatshahi D, Calleja Sanz AI, Sohn SI, et al. Thrombus migration and fragmentation after intravenous alteplase treatment: The interrsect study. Stroke. 2021;52:203–212 [DOI] [PubMed] [Google Scholar]
- 15.Yogendrakumar V, Beharry J, Churilov L, Alidin K, Ugalde M, Pesavento L, Weir L, Mitchell PJ, Kleinig TJ, Yassi N, et al. Tenecteplase improves reperfusion across time in large vessel stroke. Ann Neurol. 2023;93:489–499 [DOI] [PubMed] [Google Scholar]
- 16.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Int J Surg. 2014;12:1495–1499 [DOI] [PubMed] [Google Scholar]
- 17.Mishra SM, Dykeman J, Sajobi TT, Trivedi A, Almekhlafi M, Sohn SI, Bal S, Qazi E, Calleja A, Eesa M, et al. Early reperfusion rates with iv tpa are determined by cta clot characteristics. AJNR Am J Neuroradiol. 2014;35:2265–2272 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Menon BK, Al-Ajlan FS, Najm M, Puig J, Castellanos M, Dowlatshahi D, Calleja A, Sohn SI, Ahn SH, Poppe A, et al. Association of clinical, imaging, and thrombus characteristics with recanalization of visible intracranial occlusion in patients with acute ischemic stroke. JAMA. 2018;320:1017–1026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Puetz V, Dzialowski I, Hill MD, Subramaniam S, Sylaja PN, Krol A, O’Reilly C, Hudon ME, Hu WY, Coutts SB, et al. Intracranial thrombus extent predicts clinical outcome, final infarct size and hemorrhagic transformation in ischemic stroke: The clot burden score. Int J Stroke. 2008;3:230–236 [DOI] [PubMed] [Google Scholar]
- 20.Legrand L, Naggara O, Turc G, Mellerio C, Roca P, Calvet D, Labeyrie MA, Baron JC, Mas JL, Meder JF, et al. Clot burden score on admission T2*-MRI predicts recanalization in acute stroke. Stroke. 2013;44:1878–1884 [DOI] [PubMed] [Google Scholar]
- 21.Seners P, Scheldeman L, Christensen S, Mlynash M, Ter Schiphorst A, Arquizan C, Costalat V, Henon H, Bretzner M, Heit JJ, et al. Determinants of infarct core growth during inter-hospital transfer for thrombectomy. Ann Neurol. 2023;93:1117–1129 [DOI] [PubMed] [Google Scholar]
- 22.Guenego A, Fahed R, Albers GW, Kuraitis G, Sussman ES, Martin BW, Marcellus DG, Olivot JM, Marks MP, Lansberg MG, et al. Hypoperfusion intensity ratio correlates with angiographic collaterals in acute ischaemic stroke with m1 occlusion. Eur J Neurol. 2020;27:864–870 [DOI] [PubMed] [Google Scholar]
- 23.Seners P, Turc G, Maier B, Mas JL, Oppenheim C, Baron JC. Incidence and predictors of early recanalization after intravenous thrombolysis: A systematic review and meta-analysis. Stroke. 2016;47:2409–2412 [DOI] [PubMed] [Google Scholar]
- 24.Campbell BC, Mitchell PJ, Churilov L, Yassi N, Kleinig TJ, Yan B, Dowling RJ, Bush SJ, Dewey HM, Thijs V, et al. Tenecteplase versus alteplase before endovascular thrombectomy (EXTEND-IA TNK): A multicenter, randomized, controlled study. Int J Stroke. 2018;13:328–334 [DOI] [PubMed] [Google Scholar]
- 25.Bala F, Singh N, Buck B, Ademola A, Coutts SB, Deschaintre Y, Khosravani H, Appireddy R, Moreau F, Phillips S, et al. Safety and efficacy of tenecteplase compared with alteplase in patients with large vessel occlusion stroke: A prespecified secondary analysis of the ACT randomized clinical trial. JAMA Neurol. 2023;80:824–832 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Seners P, Caroff J, Chausson N, Turc G, Denier C, Piotin M, Aghasaryan M, Alecu C, Chassin O, Lapergue B, et al. Recanalization before thrombectomy in tenecteplase vs. Alteplase-treated drip-and-ship patients. J Stroke. 2019;21:105–107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Desilles JP, Di Meglio L, Delvoye F, Maïer B, Piotin M, Ho-Tin-Noé B, Mazighi M. Composition and organization of acute ischemic stroke thrombus: A wealth of information for future thrombolytic strategies. Front Neurol. 2022;13:870331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tsivgoulis G, Katsanos AH, Eggers J, Larrue V, Thomassen L, Grotta JC, Seitidis G, Schellinger PD, Mavridis D, Demchuk A, et al. Sonothrombolysis in patients with acute ischemic stroke with large vessel occlusion: An individual patient data meta-analysis. Stroke. 2021;52:3786–3795 [DOI] [PubMed] [Google Scholar]
- 29.Alexandrov AV, Köhrmann M, Soinne L, Tsivgoulis G, Barreto AD, Demchuk AM, Sharma VK, Mikulik R, Muir KW, Brandt G, et al. Safety and efficacy of sonothrombolysis for acute ischaemic stroke: A multicentre, double-blind, phase 3, randomised controlled trial. Lancet Neurol. 2019;18:338–347 [DOI] [PubMed] [Google Scholar]
- 30.Campbell BCV, Ma H, Ringleb PA, Parsons MW, Churilov L, Bendszus M, Levi CR, Hsu C, Kleinig TJ, Fatar M, et al. Extending thrombolysis to 4.5–9 h and wake-up stroke using perfusion imaging: A systematic review and meta-analysis of individual patient data. Lancet. 2019;394:139–147 [DOI] [PubMed] [Google Scholar]
- 31.Albers GW, Jumaa M, Purdon B, Zaidi SF, Streib C, Shuaib A, Sangha N, Kim M, Froehler MT, Schwartz NE, et al. Tenecteplase for stroke at 4.5 to 24 hours with perfusion-imaging selection. N Engl J Med. 2024;390:701–711 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.