Low Dose Intravenous Cangrelor Versus Glycoprotein IIb/IIIa Inhibitors in Endovascular Treatment of Tandem Lesions

Abstract

Objectives:

Intravenous (IV) periprocedural antiplatelet therapy (APT) for patients undergoing acute carotid stenting during mechanical thrombectomy (MT) is not fully investigated. We aimed to compare the safety profile of IV low dose cangrelor versus IV glycoprotein IIb/IIIa (GP-IIb/IIIa) inhibitors in patients with acute tandem lesions (TLs).

Materials and Methods:

We retrospectively identified all cases of periprocedural administration of IV cangrelor or GP-IIb/IIIa inhibitors during acute TLs intervention from a multicenter collaboration. Patients were divided in two groups according to the IV APT regimen at the time of MT procedure: 1) cangrelor and 2) GP-IIb/IIIa inhibitors (tirofiban and eptifibatide). Safety outcomes included rates of symptomatic intracranial hemorrhage (sICH), parenchymal hematoma type 1 and 2 (PH1-PH2), and hemorrhagic infarction type 1 and 2 (HI1-HI2).

Results:

Sixty-three patients received IV APT during MT, 30 were in the cangrelor group, and 33 were in the GP-IIb/IIIa inhibitors group. There were no significant differences in the rates of sICH (3.3% vs. 12.1%, aOR=0.21, 95%CI 0.02–2.18, p=0.229), PH1-PH2 (17.9% vs. 12.1%, aOR=1.63, 95%CI 0.29–9.83, p=0.577), and HI1-HI2 (21.4% vs 42.4%, aOR=0.21, 95%CI 0.02–2.18, p=0.229) between both treatment groups. However, there was a trend toward reduced hemorrhage rates with cangrelor. Cangrelor was associated with increased odds of complete reperfusion (aOR=5.86; 95%CI 1.57–26.62;p=0.013).

Conclusions:

In this retrospective non-randomized cohort study, our findings suggest that low dose cangrelor has similar safety and increased rate of complete reperfusion compared to IV GP-IIb/IIIa inhibitors. Further prospective studies are warranted to confirm this association.

Introduction

Endovascular treatment of tandem lesions (TLs) in acute ischemic stroke patients frequently entails acute cervical internal carotid artery stent placement with the objective to maintain vessel patency.¹ Consequently, platelet inhibition is used to reduce the risk of stent thrombosis. Nevertheless, there is no consensus on antiplatelet therapy (APT) when acute stenting is required in TLs.² Intravenous (IV) glycoprotein IIb/IIIa (GP IIb/IIIa) inhibitors are commonly used in neurointerventional procedures. However, the risk of intracranial hemorrhage (ICH)³ and the relatively long half-life are disadvantages when a hemorrhagic complication is encountered.⁴ Recently, IV low dose cangrelor has been proposed as a therapeutic option in managing periprocedural platelet inhibition for stenting in acute ischemic stroke due to its fast and reliable onset of action and reversible effect⁵; however, its use in TLs is limited.^4,6–9 Therefore, we aimed to compare the safety of IV low dose cangrelor versus IV GP IIb/IIIa inhibitors in patients with acute TLs during mechanical thrombectomy (MT).

Methods

We used a pooled, multicenter cohort registry for the study. Patient eligibility and methods of the collaboration have been reported previously.¹⁰ Briefly, the study included adult patients with TL treated with MT within 24 hours after stroke onset from 16 stroke centers (15 hospitals in the United States and 1 in Spain). TL was defined as an intracranial large vessel occlusion (petrous, sigmoid, or terminus segment of the internal carotid artery or M1 or proximal M2 segment of the middle cerebral artery) with concomitant extracranial internal carotid artery stenosis ≥50% according to NASCET (North American Symptomatic Carotid Endarterectomy Trial).¹¹ All intracranial occlusions were treated using a stent retriever and/or contact aspiration catheters. The endovascular and medical therapeutic interventions were performed according to the protocol of each institution under conscious sedation or general anesthesia and at the discretion of the neurointerventionalists.

We identified all cases of periprocedural administration of IV cangrelor or GP IIb/IIIa inhibitors with no additional oral APT during acute TLs intervention. Patients were divided into two groups: 1) Low dose cangrelor at half the dose of the myocardial infarction protocol, and 2) GP IIb/IIIa inhibitors (tirofiban and eptifibatide). The endovascular and medical therapeutic interventions were performed according to each institution’s protocol and at the discretion of the interventionalists. Details about the APT regimens are available in Supplementary Table 1. Institutional Review Board approval was obtained from all sites. Original data are available upon reasonable request.

The primary study outcomes were symptomatic intracranial hemorrhage (sICH), a composite of sICH and parenchymal hematoma type 2 (PH2), and petechial hemorrhage (hemorrhagic infarction type 1 [H1] and type 2 [H2]) on computed tomography at 24 ± 12 hours follow-up according to the Heidelberg Bleeding Classification.¹² sICH was defined according to the European Collaborative Acute Stroke Study (ECASS-3) criteria as any type of intracranial hemorrhage observed on follow-up imaging between 22 and 36 hours and 7 days after stroke onset, and an increase of ≥4 points on the National Institutes of Health Stroke Scale from baseline or the lowest value within 7 days, or resulting in mortality.¹³ We assessed for hemorrhagic transformation rate and categorized it as no hemorrhage, H1-H2-PH1, and sICH-PH2. Additional exploratory outcomes included successful reperfusion (mTICI 2b-3), complete reperfusion (mTICI 3), ordinal mTICI (0–2a, 2b and 2c-3), first-pass effect, 90-day 0–2 modified Rankin scale, and in-hospital and 90-day mortality.

We performed univariate and multivariable logistic regression adjusting for critical confounders to compare outcomes between the groups. All the statistical analyses were conducted using R version 4.1.3.

Results

Of the 691 patients from the registry, 628 were excluded [Supplementary Figure 1]. Sixty-three patients were included, 30 in the cangrelor group (used at 4 sites) and 33 in the GP IIb/IIIa inhibitors group (used at 10 sites). Demographics and baseline characteristics of the two groups are presented in Supplementary Table 2.

Compared to GP IIb/IIIa inhibitors, cangrelor showed a non-significantly lower rate of sICH (3.3% vs. 12.1%, p=0.197). After adjusting for ASPECTS (Alberta Stroke Program Early Computed Tomography Score) and procedural heparin use, the risk of sICH between the two groups remained comparable (aOR=0.21, 95% CI 0.02–2.18, p=0.229). Similarly, we observed similar rates of PH1-PH2 (17.9% vs. 12.1%; aOR=1.63, 95% CI 0.29–9.83, p=0.577) and HI1-HI2 (21.4% vs 42.4%; aOR=0.21, 95% CI 0.02–2.18, p=0.229) between both treatment groups, both before and after adjusting for ASPECTS and procedural heparin use [Figure 1A]. The overall distribution of hemorrhagic conversion was comparable across both groups (aOR=1.16; 95% CI 0.35–3.80; p=0.811) [Figure 1B].

Figure 1.

(A) Bar chart of sICH, PH1-PH2, and HI1-HI2 for patients treated with cangrelor and glycoprotein IIb/IIIa inhibitors; and (B) shift analysis of hemorrhagic transformation according to the severity. *Adjusted for: ASPECTS, procedural heparin.

The rate of complete reperfusion was significantly higher with cangrelor (56.7% vs. 24.2%; aOR=5.86, 95% CI 1.57–26.62, p=0.013). Additional exploratory outcomes were similar between groups [Table 1].

Table 1.

Additional outcomes of patients with endovascularly treated tandem lesions according to cangrelor versus glycoprotein IIb/IIIa inhibitors.

	Total (N=63)	Cangrelor (N=30)	GP IIb/IIIa inhibitors (N=33)	Unadjusted analysis		Adjusted analysis*
				OR (95% CI)	p value	aOR (95% CI)	p value
Successful reperfusion*, N° (%)	62 (98.4)	30 (100)	32 (97)	2.81 (0.11 – 71.78)	0.336	-	-
Complete reperfusion*, N° (%)	25 (39.7)	17 (56.7)	8 (24.2)	4.08 (1.39 – 11.97)	0.009	5.86 (1.57 – 26.62)	0.013
mTICI*, N° (%)				5.39 (1.66 – 17.51)	0.005	4.83 (1.17 – 20.0)	0.030
0–2a	1 (1.6)	0 (0)	1 (3)
2b	21 (33.3)	5 (16.7)	16 (48.5)
2c-3	41 (65.1)	25 (83.3)	16 (48.5)
First-pass effect*, N° (%)	42 (68.9)	16 (57.1)	26 (78.8)	0.36 (0.11 – 1.08)	0.073	0.73 (0.18 – 3.02)	0.656
90-day 0–2 mRS*, N° (%)	33 (56.9)	15 (55.6)	18 (58.1)	1.22 (0.36 – 4.25)	0.751	2.68 (0.61 – 13.93)	0.209
In-hospital mortality*, N° (%)	3 (4.8)	2 (6.7)	1 (3)	2.29 (0.21 – 50.8)	0.509	1.77 (0.11 – 53.1)	0.699
90-day mortality*, N° (%)	6 (10.3)	3 (11.10	3 (9.7)	1.17 (0.20 – 6.82)	0.858	0.85 (0.11 – 7.04)	0.874

GP IIb/IIIa: Glycoprotein IIb/IIIa; mTICI: modified Thrombolysis in Cerebral Infarction; mRS: modified Rankin scale.

^*Adjusted for: ASPECTS, procedural heparin use.

Discussion

We found that, in exclusively acute TLs patients, low dose cangrelor was associated with a non-significant trend toward lower rates of sICH, PH1-PH2, and HI1-HI2 when compared to GP IIb/IIIa inhibitors. Cangrelor was also associated with a higher rate of complete reperfusion.

A review study of IV cangrelor and GP IIb/IIIa inhibitors vis-a-vis perioperative bridging after percutaneous coronary intervention suggested that differences in the pharmacokinetic/pharmacodynamic properties, the safety and efficacy profile, and cost considerations must be considered for selection of optimum APT.¹⁴ Cangrelor has the advantage of rapid onset with a half-life of 3–6 min, does not require transformation to active metabolites, and provides immediate and consistent platelet inhibition.^15–17 Its short half-life makes cangrelor ideally suited for acute MT of TLs, where cangrelor can be replaced by an oral APT after stent placement.⁶ In the case of an hemorrhagic complication, unlike IV GP IIb/IIIa inhibitors, cangrelor activity subsides less than an hour after the end of the infusion, eliminating the need for platelet transfusion.¹⁸ However, the current experience using cangrelor for TLs is very limited and mainly comes from populations not exclusively focused in TLs^{4,6–9,19,20} [Supplementary Table 3]. Jumaa et al. reported hemorrhagic transformation (HI1) in 2 out of 9 TL patients and no sICH who received IV cangrelor.⁹ In both patients, a 15 mcg/kg bolus was followed by a 2 mcg/kg/min infusion.⁹

Observational studies have reported that IV tirofiban, a GP IIb/IIIa inhibitor, was associated with higher reperfusion rates among patients with large artery atherosclerosis stroke.^21–23 This may be due to the efficacy of tirofiban on microcirculation thrombi in comparison to the more extensive thrombus in proximal lesions.²¹ Patients in our study treated with cangrelor achieved successful reperfusion, and more than half achieved complete reperfusion, which agrees with previous reports.^4,6–9,20 Similar to tirofiban^21–23, cangrelor’s higher reperfusion rates may be due to its easy access to the more distal sub-occlusive residual thrombi and effective APT activity within the microcirculation.²¹

There are several limitations to our study. First, we included a small sample size in both groups; however, our study has the larger population of TL patients treated with cangrelor to date. Second, because of the retrospective nature of the data a potential selection bias might affected our findings. Similarly, the decision to use cangrelor or GP IIb/IIIa inhibitors was dependent on the institutional protocol of each center, which means a substantial risk of selection bias. Third, the outcomes were self-adjudicated by independent investigators at each center without external control or core imaging laboratory adjudication. Finally, relevant safety outcomes were not available such as stent patency at follow-up and, therefore, we could not assess the effect of cangrelor in chronic stent patency.

Our findings suggest that low dose cangrelor has similar safety and increased rate of complete reperfusion compared to IV GP-IIb/IIIa inhibitors. The preliminary data presented indicate that, in patients with acute TLs treated with stenting, IV low dose cangrelor may represent a potential alternative to the only available fast acting IV agents, GP IIb/IIIa inhibitors. Larger sample prospective studies are required to better define cangrelor safety in TLs.