“Drip-and-ship” intravenous thrombolysis and outcomes for large vessel occlusion thrombectomy candidates in a hub-and-spoke model

Abstract

Background:

Randomized trials have not demonstrated benefit from intravenous alteplase among patients undergoing endovascular thrombectomy (EVT). However, these trials included primarily patients presenting directly to an EVT-capable hub center. We sought to study outcomes for EVT candidates who presented to spoke hospitals and were subsequently transferred for EVT consideration, comparing those administered alteplase at spokes (i.e. “drip-and-ship” model) versus those not.

Methods:

Consecutive EVT candidates presenting to 25 spokes from 2018 to 2020 with pre-transfer CTA-defined emergent large vessel occlusion (ELVO) and ASPECTS ≥6 were identified from a prospectively maintained Telestroke database. Outcomes of interest included adequate reperfusion (TICI 2b-3), intracerebral hemorrhage (ICH), discharge functional independence (modified Rankin Scale, mRS ≤2), and 90-day mRS ≤2.

Results:

Among 258 patients, the median age was 70 years (IQR 60-81), median NIHSS was 13 (6-19), and 50% were female. Ninety-eight (38%) were treated with alteplase at spokes, and 113 (44%) underwent EVT at the hub. Spoke alteplase use was independently associated with discharge mRS ≤2 (aOR=2.43, 95%CI=1.08,5.46, p=0.03) and 90-day mRS ≤2 (aOR=3.45, 95%CI=1.65,7.22, p=0.001), even when controlling for last known well, NIHSS, and EVT; it was not associated with increased risk of ICH (OR=1.04, 95%CI=0.39,2.78, p=0.94), and there was a trend toward association with TICI 2b-3 (OR=3.59, 95%CI=0.94,13.70, p=0.06).

Conclusions:

Intravenous alteplase at spoke hospitals may improve discharge and 90-day mRS and should not be withheld from EVT-eligible patients who first present at alteplase-capable spoke hospitals that do not perform EVT. Additional studies are warranted to confirm and further explore these benefits.

Introduction

With the advent of highly-effective endovascular thrombectomy (EVT) for emergent large vessel occlusion (ELVO) strokes, the use of bridging intravenous thrombolysis (IVT, primarily with alteplase) has come under higher scrutiny.[1-3] Advantages to bridging IVT include the possibility of reperfusion prior to EVT, enhanced reperfusion success when followed by EVT, distal microemboli dissolution, treatment in absence or failure of EVT, and theoretical enhancement of EVT.[2,4-9] However, with ELVO recanalization rates as low as 4-20% with IVT alone, the potential for logistical delay in EVT, added costs of IVT, and side effects associated with bleeding risk and clot disruption, the use of IVT as standard-of-care before EVT has reasonably been questioned.[1-6,10-12] Recent randomized trials of IVT-EVT bridging vs direct-to-EVT have independently shown non-inferiority of the direct-to-EVT approach among patients presenting initially to so-called “mothership” EVT centers.[1-3] Specifically, these trials have demonstrated similar functional outcomes as well as safety profiles. However, in the alternative “drip-and-ship” model, there remains a population where both potential benefits and risks may be accentuated given prolonged time-to-EVT associated with transit from “spoke” hospital to “hub” hospital capable of EVT.[8,9,13] This transit time may allow for higher dose completion rates and longer dwell time for true chemical thrombolysis from IVT.[14,15] As such, we aimed to compare the safety and efficacy of bridging alteplase IVT for ELVO stroke patients who first presented to a spoke hospital and were transferred for EVT consideration in a large Telestroke hub-and-spoke network.

Methods

This study was compliant with the Health Insurance Portability and Accountability Act and was approved by the local institutional review board. Informed consent was waived based on minimal patient risk and practical inability to perform the study without the waiver. The data that support the findings of this study will be made available from the corresponding author upon reasonable request and pending approval of local institutional review board.

ELVO stroke patients who presented to 25 spoke hospitals from January 1, 2018 to June 30, 2020 were retrospectively identified from a prospectively maintained database (Figure 1). This database includes demographics, medical history, clinical presentations, treatments, and functional outcomes for consecutive patients.[16] Inclusion criteria were pre-transfer spoke CTA-defined ELVO and Alberta Stroke Program CT score (ASPECTS) ≥6, and hub transfer for EVT consideration.

Figure 1. Map of spoke hospitals that transferred patients for consideration of thrombectomy at the hub from 2018-2020.

Red “H” corresponds to hub hospital; Blue crosses correspond to spoke hospitals. Scale shows 30 miles.

Presenting National Institutes of Health Stroke Scale (NIHSS) score was determined by the hub neurologist as described, with higher numbers reflecting increased clinical stroke severity.[17] All patients underwent CT and CTA at the spoke hospital prior to transfer.[18,19] ASPECTS and presence of ELVO on CTA were determined by a vascular neurologist and confirmed by a neuroradiologist. ELVO was defined as occlusion of the internal carotid artery (ICA) terminus, first (M1) and proximal second (M2) segments of the middle cerebral artery or the basilar artery. Cervical ICA disease was defined as severe stenosis (>70%) or occlusion related to atherosclerosis or dissection.[20] IVT treatment decisions at spokes were guideline-based at the discretion of a vascular neurologist through telemedicine.[21] Thrombectomy treatment decisions at the hub were at the discretion of a vascular neurologist and neurointerventionalist.

Thrombolysis in Cerebral Infarction (TICI) scores were determined by a neurointerventionalist using the modified scale: 2a partial filling <50%, 2b partial filling ≥50%, 3 complete perfusion (available for all undergoing EVT).[22] Adequate reperfusion was considered TICI 2b-3.[23] Intracerebral hemorrhage (ICH) was defined as intraparenchymal, intraventricular, or subarachnoid hemorrhage during hospitalization.[24] Discharge modified Rankin Scale (mRS) was determined by a vascular neurologist at the time of hospital discharge (available for 99%).[25] 90-day mRS score was obtained by telephone call using a validated approach (available for 85%).[26] Functional independence was defined as 90-day mRS ≤2.[27,28]

Median values with interquartile range (IQR) were reported for continuous variables. Percent and count were reported for categorical variables. Differences were assessed using nonparametric Wilcoxon rank-sum for continuous variables and Fisher’s Exact tests for categorical variables. Logistic regression analyses were performed to assess associations with dichotomous outcomes; odds ratios were reported as unadjusted and adjusted for a priori determined last known well (LKW), NIHSS, and EVT. Two-tailed p-values <0.05 were interpreted as statistically significant. Analyses were performed with SPSS version 23.0 (IBM Corp).

Results

Through the hub-and-spoke Telestroke network, 337 consecutive patients were transferred to the hub after presenting to a spoke hospital with stroke symptoms. Two hundred fifty-eight had CTA-defined ELVO, ASPECTS ≥6, and were clinically determined to be potential EVT candidates at the time of spoke presentation.

For these 258 patients meeting inclusion criteria, the median age was 70 years (IQR 60-81), 50% were female, 9% were non-white race/ethnicity, and the median NIHSS was 13 (6-19). 98 patients (38%) were treated with IVT at spoke hospitals. The reasons for not administering IVT were LKW >4.5 hours (68%; of which 59% were wake-up strokes), bleeding diathesis or anticoagulation (11%), intracranial surgery or severe head trauma or prior stroke in the preceding 3 months (2%), other major surgery or serious trauma in the preceding 2 weeks (2%), stroke severity too mild (2%), gastrointestinal or urinary hemorrhage in the preceding 3 weeks (1%), concern for active internal bleeding (1%), elevated blood pressure despite treatment (1%), patient/family refusal (1%), and unknown (12%). Baseline demographics, medical history, and stroke severity were similar comparing those treated with alteplase and those not treated (Table 1). LKW-to-Telestroke time (median 1.6 vs 6.7 hours, p<0.001) and Telestroke-to-hub arrival time (median 2.2 vs 2.6 hours, p=0.01) were different between groups.

Table 1. Demographics, medical history, presentation, treatments, and outcomes.

IQR interquartile range; TIA transient ischemic attack; NIHSS NIH stroke scale; LKW las known well; EVT endovascular thrombectomy; TICI thrombolysis in cerebral infarction; ICH intracerebral hemorrhage; mRS modified Rankin scale.

	Total		Spoke Alteplase		No Spoke Alteplase
	Median/ Count	IQR/ %	Median/ Count	IQR/ %	Median/ Count	IQR/ %	p
Age, Years	70	(60,81)	67	(58,81)	71	(61,81)	0.13
Female	129	50%	48	49%	81	51%	0.90
Black	9	4%	5	7%	4	3%	0.29
Hispanic	9	4%	1	1%	8	6%	0.16
Asian	6	3%	2	3%	4	3%	1.00
White	184	91%	66	90%	118	91%	1.00
Hypertension	162	63%	60	61%	102	64%	0.69
Diabetes	61	24%	21	21%	40	25%	0.55
Atrial Fibrillation	78	30%	23	24%	55	34%	0.07
Coronary Disease	44	17%	18	18%	26	16%	0.73
Prior Stroke/TIA	42	16%	14	14%	28	18%	0.60
Dyslipidemia	120	47%	45	46%	75	47%	0.90
Obesity/Overweight	89	35%	33	34%	56	35%	0.89
Renal Insufficiency	21	8%	9	9%	12	8%	0.65
Smoking	47	18%	23	24%	24	15%	0.10
NIHSS	13	(6,19)	12	(6,19)	14	(6,18)	0.98
LKW-Telestroke, Hours	3.6	(1.5,9.6)	1.6	(1.1,2.6)	6.7	(3.9,11.7)	<0.0001
LKW-Hub Arrival, Hours	6.4	(4.1,12.0)	4	(3.3,5.1)	9.6	(6.4,14.6)	<0.0001
Telestroke-Hub Arrival, Hours	2.5	(1.8,3.3)	2.2	(1.8,2.9)	2.6	(1.9,3.5)	0.01
Alteplase-Hub Arrival, Hours	-	-	1.6	(1.2,2.3)	-	-	-
Hub EVT	113	44%	51	52%	62	39%	0.04
TICI 2b-3, N=107	93	87%	46	94%	47	81%	0.08
ICH	18	7%	7	7%	11	7%	1.00
Discharge mRS≤2, N=255	50	20%	28	29%	22	14%	0.005
90-Day mRS≤2, N=219	84	38%	45	52%	39	30%	0.001

Among the entire cohort, 44% underwent EVT (87% achieved adequate reperfusion), 7% experienced ICH (all were symptomatic), and 38% achieved 90-day functional independence (Table 1). In univariable analyses, spoke alteplase use significantly increased the odds of discharge mRS ≤2 (OR=2.51, 95%CI=1.34,4.71, p=0.004) and 90-day mRS ≤2 (OR=2.56, 95%CI=1.46,4.49, p=0.001) and was not associated with ICH (OR=1.04, 95%CI=0.39,2.78, p=0.94). We observed a trend toward association with greater TICI 2b-3 reperfusion (OR=3.59, 95%CI=0.94,13.70, p=0.06) (Table 2).

Table 2. Associations of spoke-administered alteplase with outcomes after transfer for EVT consideration.

Spoke alteplase was not associated with intracerebral hemorrhage (ICH) but increased the odds of discharge and 90-day modified Rankin Scale (mRS) ≤2. Univariable associations are shown. OR odds ratio; CI confidence interval; TICI thrombolysis in cerebral infarction.

	OR (95%CI)	p
TICI 2b-3	3.59 (0.94,13.70)	0.06
ICH	1.04 (0.39,2.78)	0.94
Discharge mRS≤2	2.51 (1.34,4.71)	0.004
90-Day mRS≤2	2.56 (1.46,4.49)	0.001

In a multivariable model for determinants of discharge mRS ≤2, spoke alteplase independently increased the odds of discharge mRS ≤2 (aOR=2.43, 95%CI=1.08,5.46, p=0.03) when controlling for other relevant variables. Higher NIHSS independently reduced the odds of discharge mRS ≤2 (aOR=0.84, 95%CI=0.79,0.90, p<0.0001), while LKW-to-Telestroke consult time (aOR=0.98, 95%CI=0.92,1.05, p=0.55), Telestroke-to-hub arrival time (aOR=0.85, 95%CI=0.63,1.15, p=0.29), and hub EVT (aOR=1.46, 95%CI=1.65,3.26, p=0.36) were not associated (Table 3).

Table 3. Adjusted associations of spoke-administered alteplase with discharge and 90-day disability among patients transferred for EVT consideration.

Spoke alteplase independently increased the odds of discharge and 90-day modified Rankin Scale (mRS) ≤2, even when controlling for other determinants of outcome. Two multivariable models are shown. OR unadjusted odds ratio; aOR adjusted odds ratio; CI confidence interval; LKW last known well; NIHSS NIH stroke scale; EVT endovascular thrombectomy.

Discharge mRS≤2 Model	OR (95%CI)	p	aOR (95%CI)	p
Spoke Alteplase	2.51 (1.34,4.71)	0.004	2.43 (1.08,5.46)	0.03
LKW-Telestroke Consult	0.95 (0.90,1.02)	0.13	0.98 (0.92,1.05)	0.55
Telestroke-Hub Arrival	0.96 (0.76,1.22)	0.74	0.85 (0.63,1.15)	0.29
NIHSS	0.87 (0.83,0.92)	<0.0001	0.84 (0.79,0.90)	<0.0001
Hub EVT	0.91 (0.49,1.70)	0.76	1.46 (0.65,3.26)	0.36
90-Day mRS≤2 Model
Spoke Alteplase	2.56 (1.46,4.49)	0.001	3.45 (1.65,7.22)	0.001
LKW-Telestroke Consult	0.97 (0.93,1.02)	0.24	1.01 (0.96,1.06)	0.75
Telestroke-Hub Arrival	1.03 (0.84,1.26)	0.77	1.11 (0.86,1.43)	0.41
NIHSS	0.89 (0.86,0.93)	<0.0001	0.86 (0.81,0.90)	<0.0001
Hub EVT	1.97 (1.13,3.41)	0.016	4.34 (2.01,9.39)	<0.0001

In a multivariable model for determinants of 90-day mRS ≤2, spoke alteplase independently increased the odds of 90-day mRS ≤2 (aOR=3.45, 95%CI=1.65,7.22, p=0.001) when controlling for other relevant variables. Hub EVT also independently increased the odds of 90-day mRS ≤2 (aOR=4.34, 95%CI=2.01,9.39, p<0.0001) while higher NIHSS independently reduced them (aOR=0.86, 95%CI=0.81,0.90, p<0.0001). Neither LKW-to-Telestroke consult time (aOR=1.01, 95%CI=0.96,1.06, p=0.75) nor Telestroke-to-hub arrival time (aOR=1.11, 95%CI=0.86,1.43, p=0.41) were associated (Table 3).

Discussion

In this retrospective cohort of 258 patients who first presented to spoke hospitals as potential EVT candidates with CTA-defined ELVO and ASPECTS ≥6, treatment with intravenous alteplase at spokes before transfer significantly increased the odds of discharge and 90-day functional independence without increasing the odds of intracerebral hemorrhage.

Our findings suggest that within hospital systems utilizing the “drip and ship” model, IVT should still be considered for patients requiring transfer for EVT. This aligns with analyses of the MR CLEAN registry concerning IVT before EVT, as well as with current guidelines that support bridging IVT.[7,21] As stated in the introduction, recent evidence has called this dogma into question. Perhaps since patients initially presenting to a “mothership” have rapid access to highly effective EVT, bridging IVT did not add significant benefit in the recently published ANGEL-ACT[4] registry or in the DEVT,[2] SKIP,[3] and DIRECT-MT[1] randomized trials. While patients initially presenting to a “mothership” have rapid access to EVT (often minutes), there can be significant delay to EVT associated with transport in the “drip-and-ship” model (sometimes hours). Indeed, most patients in our cohort completed their IVT infusion prior to arrival to the hub, as evidenced by a median alteplase administration-to-hub arrival time of 1.6 hours.

The time delay from IVT to EVT during transfer may better allow the possibility of chemical thrombolysis.[6,7,13,29-32] While recent studies focused on patients that could undergo EVT within 4.5 hours,[1-3] our model included patients who underwent EVT beyond this window but were restricted to IVT within it. Indeed, the incidence of partial or total recanalization from IVT alone over this timeframe has been described,[29,31,33,34] and likely benefitted our patients in a way not permitted outside the “drip-and-ship” model. With the clear utility of EVT well beyond 6 hours from symptom onset,[21] our results suggest bridging IVT in the “drip and ship” model has particular clinical relevance given the number of patients who can be treated with IVT at spoke hospitals within 4.5 hours but are unable to undergo EVT until hours later. Furthermore, there is mounting evidence IVT may be beneficial for select patients beyond 4.5 hours.[21] This model also allows spoke hospital integration of benefit from future advances in “pre-treatment” before EVT, such as with tenecteplase[35] or other treatments targeting metabolic[36] or inflammatory[37] pathologies.

An important safety concern of bridging IVT is ICH given both EVT and IVT hold some risk independently. In the original pivotal EVT trials, the combination of EVT and IVT did not confer higher hemorrhage risk; this was also demonstrated in the more recent direct EVT trials.[1-3,7] However, as stated the previously studied population has included primarily patients who initially present to an EVT-capable “mothership”. Our cohort of exclusively “drip-and-ship” patients provides important further evidence that the hemorrhage risk of bridging IVT, even in a population with prolonged interval to EVT, is not significantly higher.

Our analyses suggest possible efficacy and safety of bridging IVT for ELVO patients within the “drip and ship” model. However, all geographic models are unique and further research is needed to make definitive conclusions.[38] Differences in transfer distances and times can vary widely according to region. Aside from purposefully selecting patients with initial presentations to spoke hospitals prior to transfer and the associated delay to EVT, our cohort likely has good generalizability given similar baseline characteristics to previously published studies. Age, sex, comorbidities, and median NIHSS aligned with the previously published population of interest.[1-4] While many of the recent studies included primarily Asian populations, our study included predominantly white patients. This may raise speculative considerations about the prevalence of intracranial atherosclerosis as ELVO etiology, though whether this would impact the utility of IVT before EVT is not clear.[20,39,40]

Given the mounting evidence supporting direct EVT non-inferiority for ELVO, understanding the role of early bridging IVT within “drip-and-ship” models is crucial to optimizing triage in current systems of care. Similar to the evolution of cardiac intervention,[41-43] there remains an ongoing need to balance early catheter-based intervention with the more universal availability of immediate IVT. Our study provides some insight into the benefit of IVT when EVT is not immediately available. Considering present data and since acute therapies have historically had the greatest impact on outcomes after stroke,[44,45] IVT treatment in this patient population may be preferred if there are no contraindications. However, future studies are needed to confirm the benefit of IVT and better define the minimum transfer distance and transfer time for which benefit exists. Additional non-randomized studies will be limited by selection bias, however there are unique ethical considerations for randomized trials including this ELVO population. There may be some situations in which clinical equipoise is lacking, such as the case of withholding IVT from patients who will have appreciable transfer delay to EVT.

Our study has several limitations. First, despite multivariable analyses to control for potential confounders, the limitations of retrospective studies apply. In particular, time from LKW was the most frequent reason IVT was not administered in the non-bridged cohort. Furthermore, other variables not included in this dataset have been associated with outcomes after stroke, such as social support networks, socioeconomic status, collateral status, and perfusion mismatch, that require further study.[46-48] The heavy reliance on Telestroke, represents a relatively understudied variable in virtually triaging EVT candidacy. We sought to evaluate bridging IVT in this population specifically as they are not well reflected in recent international EVT RCTs. Third, some patients were recommended for transfer to our hub but did not arrive or did not undergo EVT. Future studies will investigate reasons these patients do not undergo EVT. In addition, there was a small minority of patients (15%) who were lost to follow-up and could not be reached for 90-day outcomes. Finally, similar to the limitations of the recent Asian population-based trials (DIRECT-MT, SKIP, and DEVT), our study’s generalizability may be limited by being heavily weighted to non-Hispanic white patients, as opposed to a more diverse patient cohort.[49]

In conclusion, intravenous alteplase at spoke hospitals may improve discharge and 90-day mRS for EVT candidates with ELVO. Careful consideration is warranted before withholding intravenous alteplase from ELVO patients who require transfer for EVT. Additional studies are essential to confirm benefits of spoke-administered bridging IVT in this transfer population and to better understand the minimum transfer distance and time where benefit is likely.

Data Availability:

The data that support the findings of this study will be made available from the corresponding author upon reasonable request and pending approval of local institutional review board.