Highlights

  • Dual antiplatelet therapy is associated with a significant reduction in recurrent strokes, with no increased risk of vascular events or mortality in comparison with monotherapy

  • Dual antiplatelet therapy is associated with higher rates of bleeding than monotherapy

  • The risk of bleeding is minimized by administering the combination drugs shortly following the ischemic event (≤ 1 week) and for a short period of time (≤ 1 month)

Introduction

The highest risk of recurrent stroke after ischemic stroke and transient ischemic attack (TIA) is immediately following the index event and the risk declines over the ensuing weeks [1]. Multiple randomized clinical trials (RCTs) have confirmed the beneficial effect of aspirin in reducing the risk of recurrent stroke by ~ 20% [2,3,4,5,6]. Compared with aspirin, clopidogrel monotherapy has shown a relative risk reduction of recurrent stroke of 7.3% [7].

Nevertheless, dual antiplatelet therapy (DAPT) may provide more potent synergistic inhibition of different platelet activator pathways. Therefore, DAPT is the cornerstone of management in patients with acute coronary syndrome (ACS) [8]. Nevertheless, the combination of aspirin plus extended-release dipyridamole and/or clopidogrel in the secondary prevention of stroke have shown conflicting efficacy results and increased bleeding risks [9,10,11,12,13,14,15,16,17]. Current American Heart Association (AHA)/American Stroke Association (ASA) guidelines recommend a short duration of DAPT (21 days) with aspirin and clopidogrel only following minor stroke (Class IIa) [18]. However, the generalizability of such intervention in non-Asian populations remains unknown. Therefore, we conducted our meta-analysis to evaluate the efficacy and safety of clopidogrel plus aspirin compared with monotherapy in all patients who presented with ischemic stroke and TIA and to a subset of patients with minor stroke. In addition, we aimed to identify the best duration and timing of DAPT following the index event in this population.

Methods

Data sources

We conducted our meta-analysis based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) Statement 2015 and the PRISMA checklist was followed [19]. We registered our prespecified study protocol with the International Prospective Register of Systematic Reviews (PROSPERO; ID: CRD42018098979). Two authors (BK, AA) independently performed a literature search of PubMed, Embase, and the Cochrane Collaboration Central Register of Controlled Trials from inception to May 2018. Any discrepancies were resolved via a third author (MO). We used Boolean operators for connections of the following headings: “stroke,” “transient ischemic stroke,” “TIA”, “cerebral infarction,” “cerebrovascular infarction,” AND “clopidogrel.”

Selection criteria and data extraction

The pre-specified inclusion criteria of the meta-analysis were as follows: (1) The study design was an RCT, (2) The RCT used clopidogrel-aspirin arm vs single antiplatelet (monotherapy), and (3) The RCT evaluated clinical outcomes. Exclusion criteria were as follows: (1) The RCT included patients undergoing interventions such as stent placement, angioplasty, or thrombectomy, (2) The RCT included DAPT other than clopidogrel or studied triple antiplatelets, (3) The stroke subtype was hemorrhagic or mainly due to cardioembolic phenomenon, and (4) No clinical outcomes were reported. Two authors (TH, SA) extracted the data (baseline trial characteristics, study-level patient demographics, and clinical outcomes) and any discrepancies were resolved via consultation with a third author (BK).

Outcomes and definitions

The pre-specified primary efficacy outcome was any recurrent stroke (ischemic or hemorrhagic). Other efficacy outcomes were recurrent ischemic stroke, TIA, and myocardial infarction (MI). The primary safety outcome was intracranial bleeding, defined as any intracerebral, subdural, epidural, or subarachnoid bleeding. Other safety outcomes were major bleeding (defined per each included RCT as serious or life-threatening bleeding causing hemodynamic compromise that required blood or fluid replacement, inotropic support, hospitalization, or surgical intervention, or resulting in functional sequelae or death), vascular death, and any death. All outcomes were reported at the longest follow-up duration.

Quality assessment

The quality of the included RCTs were assessed independently by two authors (BK and MO) based on the study design and number of sites, blinding to outcomes, treatment assignment generation, and the proportion of follow-up completion.

Statistical analysis

We calculated summary risk ratios (RRs) and 95% confidence intervals (CIs) using the Mantel–Haenszel method for dichotomous data, and we used a random-effects model to account for the between-study heterogeneity. We measured the heterogeneity of the included RCTs using the Cochrane’s Q statistics and I2 test.

We assessed potential publication bias by visual inspection of the funnel plot. Furthermore, we explained any heterogeneity (≥ 20%) by performing sensitivity analyses and meta-regression analyses. Sensitivity analysis was performed by removing trials that compared clopidogrel-aspirin vs clopidogrel monotherapy. Meta-regression analyses were performed for the primary outcome based on the clopidogrel initial dose (loading or maintenance dose).

We conducted a subset analysis for patients with minor stroke, defined as a score of ≤ 3 on the National Institutes of Health Stroke Scale (NIHSS) (scores range from 0 to 42, with higher scores indicating greater severity). Furthermore, subgroup analyses were conducted for the following study-level variables: follow-up duration (≤ 3 months vs ≥ 6 months), DAPT duration (short duration ≤ 1-month, intermediate duration up to 3 months, and long duration ≥ 1 year), timing of treatment following the index stroke/TIA (acute within 48 h, subacute within 1 week, or chronic after 1 month), and the studied population (East Asian vs Western). All data were analyzed using RevMan v5.3 Windows and Comprehensive Meta-Analysis software v3.

Results

Study selection and trial characteristics

Throughout our electronic databases search, we identified 16 eligible RCTs [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37]. We included the clinical outcomes of the studies that reported both short and long follow-up duration separately, and we also included subgroup studies that reported clinical outcomes for minor stroke/TIA such as CLAIR and CLAIR subset trials [20, 38,39,40,41]. The study selection process is shown in Fig. 1.

Fig. 1
figure 1

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram

Full size image

The total number of patients was 29,032. Most of the RCTs compared clopidogrel-aspirin vs aspirin. However, 2 RCTs compared DAPT against clopidogrel 75 mg/day [22, 25], and in the sensitivity analysis we removed these studies from the pooled estimates to explain any heterogeneity. Although most of the included trials administered clopidogrel at a dose of 75 mg/day, some trials initiated the drug with a loading dose (300 or 600 mg) [20, 22, 26, 27, 29, 31, 36, 39]. However, most of these trials used the maintenance dose for a short duration (≤ 21 days) and only two trials which used loading doses extended the treatment to 3 months [29, 36]. There were 7 RCTs conducted exclusively in Asia [23, 24, 26, 27, 31, 32, 35], 7 trials based solely in Western countries [21, 22, 25, 29, 33, 34, 36], and 2 trials conducted in both Asian and Western countries [28, 30]. The reported NIHSS score ranged between 0 and 22 and the duration of follow-up ranged from 7 days to 5 years. The characteristics of the included RCTs are detailed in Table 1. The study-level baseline demographics were similar between the two groups. The most common risk factors for stroke/TIA were hypertension, diabetes mellitus, and dyslipidemia. Table 2 shows the patients’ demographics.

Table 1 The characteristics of the included randomized clinical trials
Full size table
Table 2 Patients’ demographics
Full size table

Efficacy outcomes

Clopidogrel plus aspirin was associated with a significant reduction of any stroke compared with monotherapy (7.6% vs 9.2%; RR 0.80; 95% CI 0.72–0.89; P < 0.001; I2 = 27%). In addition, there was a significant reduction in recurrent ischemic strokes favoring the clopidogrel-aspirin group (6.7% vs 8.6%; RR 0.75; 95% CI 0.66–0.85; P < 0.001; I2 = 36%). Sensitivity analysis by removing the trials that compared DAPT with clopidogrel monotherapy showed persistent significant reductions of the total and ischemic strokes with considerable improvements in between-study heterogeneity (I2 = 0% and 10%, respectively). There were no significant differences between the groups for TIA (RR 0.88; 95% CI 0.72–1.07; P = 0.19) and MI (RR 1.04; 95% CI 0.84–1.29; P = 0.70) (Fig. 2).

Fig. 2
figure 2

Forest plot of the efficacy outcomes in patients who presented with ischemic stroke and transient ischemic attack. CARESS clopidogrel and aspirin for reduction of emboli in symptomatic carotid stenosis, CHANCE clopidogrel in high-risk patients with acute nondisabling cerebrovascular events, CHARISMA clopidogrel for high atherothrombotic risk and ischaemic stabilisation, management and avoidance, CLAIR clopidogrel plus aspirin for infarction reduction in acute stroke or transient ischaemic attack patients with large artery stenosis and microembolic signals, COMPRESS combination of clopidogrel and aspirin for prevention of recurrence in acute atherothrombotic stroke study, FASTER fast assessment of stroke and TIA to prevent early recurrence, MATCH management of atherothrombosis with clopidogrel in high-risk patients, POINT platelet-oriented inhibition in new TIA and minor ischemic stroke, SPS3 secondary prevention of small subcortical strokes

Full size image

Meta-regression analysis of the primary outcome based on the initial clopidogrel loading dose did not suggest any significant effect modifier (Supplementary Fig. 1).

Safety outcomes

There was a significant increase in the rate of intracranial bleeding with DAPT compared with monotherapy (1.1% vs 0.7%; RR 1.55; 95% CI 1.20–2.01; P < 0.001; I2 = 0%). Sensitivity analysis by comparing DAPT with aspirin monotherapy showed a higher rate of intracranial bleeding (1.8% vs 1.0%; RR 1.42; 95% CI 1.00–2.01; P = 0.05; I2 = 0%). In addition, there was a significant increased rate of major bleeding in the clopidogrel-aspirin group (2.2% vs 1.1%; RR 1.90; 95% CI 1.33–2.72; P < 0.001; I2 = 47%). Sensitivity analysis showed persistent increased intracranial bleeding but with low heterogeneity (1.8% vs 1.1%; RR 1.63; I2 = 1%). There were no significant differences between the two groups with regard to vascular death (RR 0.99; 95% CI 0.82–1.19; P = 0.91) or any death (RR 1.12; 95% CI 0.88–1.42; P = 0.36) (Fig. 3).

Fig. 3
figure 3

Forest plots of the safety outcomes in patients who presented with ischemic stroke and transient ischemic attack. CARESS clopidogrel and aspirin for reduction of emboli in symptomatic carotid stenosis, CHANCE clopidogrel in high-risk patients with acute nondisabling cerebrovascular events, CHARISMA clopidogrel for high atherothrombotic risk and ischaemic stabilisation, management and avoidance, CLAIR clopidogrel plus aspirin for infarction reduction in acute stroke or transient ischaemic attack patients with large artery stenosis and microembolic signals, COMPRESS combination of clopidogrel and aspirin for prevention of recurrence in acute atherothrombotic stroke study, FASTER fast assessment of stroke and TIA to prevent early recurrence, MATCH management of atherothrombosis with clopidogrel in high-risk patients, PLUTO-Stroke primary hypothesis of the plavix use for treatment of stroke, POINT platelet-oriented inhibition in new TIA and minor ischemic stroke, SPS3 secondary prevention of small subcortical strokes

Full size image

Minor stroke/TIA

Among patients who presented with minor stroke and TIA (NIHSS score 0–3; 5 RCTs), DAPT was associated with significant reductions of any stroke and ischemic stroke compared with monotherapy [(RR 0.75; 95% CI 0.67–0.85; P < 0.001) and (RR 0.74; 95% CI 0.65–0.84; P < 0.001), respectively]. In addition, there were no differences between the two groups in TIA (RR 0.94; 95% CI 0.75–1.17; P = 0.57), MI (RR 1.00; 95% CI 0.44–2.24; P = 0.99), intracranial bleeding (RR 1.50; 95% CI 0.86–2.64; P = 0.16), vascular death (RR 1.13; 95% CI 0.57–2.27; P = 0.72), or any death (RR 1.02; 95% CI 0.54–1.94; P = 0.94). However, the rate of major bleeding was increased with DAPT (0.5% vs 0.3%; RR 1.96; 95% CI 1.03–3.71; P = 0.04) (Fig. 4).

Fig. 4
figure 4

Forest plots of the clinical outcomes in patients who presented with minor ischemic stroke and transient ischemic attack. CHANCE clopidogrel in high-risk patients with acute nondisabling cerebrovascular events, CLAIR clopidogrel plus aspirin for infarction reduction in acute stroke or transient ischaemic attack patients with large artery stenosis and microembolic signals, FASTER fast assessment of stroke and TIA to prevent early recurrence, POINT platelet-oriented inhibition in new TIA and minor ischemic stroke

Full size image

Subgroup analyses

According to follow-up duration, there were significant reductions of any stroke and recurrent ischemic stroke with DAPT in both short- (≤ 3 months) and long-term (> 3 months) follow-up compared with monotherapy (both P < 0.01). Although the risk of intracranial bleeding increased with long-term follow-up in the DAPT group, there were no significant differences between the two groups with short-term follow-up. Both short- and long-term follow-up showed increased rates of major bleeding (Supplementary Fig. 2).

There were differences in the clinical outcomes between the two groups based on the duration of DAPT. Patients treated with short (≤ 3 months) DAPT showed greater reductions of the total/ischemic strokes without increased risk of intracranial bleeding compared with monotherapy. Shorter DAPT (≤ 1 month) was even safer in terms of major bleeding risk than intermediate or long duration therapy. Interestingly, patients treated with long duration DAPT showed similar rates of any stroke incidence compared with monotherapy. These patients also exhibited increased rates of intracranial and major bleeding (Supplementary Fig. 3).

Among patients who presented shortly after the index stroke/TIA (acute [48 h] or subacute [7 days]), DAPT was associated with lower rates of any/ischemic stroke without significant increased intracranial bleeding compared with monotherapy. In contrast, patients who presented later (chronic [≥ 1 month]) and were treated with DAPT experienced no significant reduction of any stroke but an increase in both intracranial and major bleeding compared with monotherapy (Supplementary Fig. 4). In a subset of patients with minor stroke and TIA, compared with monotherapy, early DAPT in the acute ischemic phase (within 48 h) was associated with a lower rate of any/ischemic stroke without increased risk of intracranial bleeding than in those who presented later (Supplementary Fig. 5).

Based on the population studied, there were no differences between Asian and Western studies on multiple outcomes (any/ischemic strokes, TIA, MI, vascular death, or intracranial bleeding). However, Western studies showed greater increased risk of major bleeding and death than Asian studies with DAPT compared with monotherapy (Supplementary Fig. 6).

Discussion

In the present meta-analysis of 16 RCTs (n = 29,032) comparing DAPT with monotherapy, we made several potentially useful observations. First, in patients who presented with ischemic stroke and TIA, DAPT significantly reduced the risk of any/ischemic stroke with no increased risk of TIA, MI, any death, or vascular death. Second, the risk of intracranial and major bleeding was increased with DAPT compared with monotherapy. However, the increased risk of intracranial bleeding was nonsignificant with DAPT compared with aspirin monotherapy. Third, the reduction of strokes with DAPT was observed during all follow-up durations (short- and long-term). However, the increased risk of intracranial bleeding was observed only with longer follow-up (> 3 months). Fourth, unlike prolonged DAPT (≥ 1 year), shorter duration dual therapy (up to 3 months) resulted in a reduction of any/ischemic strokes without increased risk of intracranial bleeding. Furthermore, the 1-month DAPT regimen did not significantly increase the risk of either intracranial or major bleeding. Fifth, DAPT was more effective in reducing any/ischemic strokes without increased risk of intracranial bleeding in patients who presented early during the acute ischemic phase (within 7 days) than those presenting late (≥ 1 month). Sixth, in a subset of patients with minor stroke/TIA and treated with DAPT, the risk of any/ischemic stroke was significantly reduced without increased risk of TIA, MI, intracranial bleeding, or mortality. However, an increased risk of major bleeding was observed with DAPT.

Previous studies of DAPT with aspirin and clopidogrel in patients with ischemic stroke and TIA have shown conflicting results [20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. Nevertheless, studies of microembolic signals, as an independent marker of future stroke risk, have shown a greater reduction of recurrent stroke with the clopidogrel-aspirin combination [20, 22, 27]. In our study, we found a greater benefit with clopidogrel plus aspirin in the secondary prevention of stroke. DAPT was associated with reductions of both any stroke (relative risk reduction [RRR] 18%; absolute risk reduction [ARR] 1.6%) and ischemic stroke (RRR 22%; ARR 1.9%). In addition, we found a greater benefit of DAPT immediately following the index event (within 7 days), including in patients treated within 48 h after minor stroke/TIA. Interestingly, longer durations (≥ 1 year) of DAPT showed no benefit in reducing the total stroke rate compared with monotherapy (7.6% vs 8.2%; P = 0.16) and were associated with increased bleeding events. This observation was based on pooling the estimates from 3 RCTs [22, 28, 34].

Despite the increased risk of intracranial bleeding and major bleeding with DAPT, the frequency of such events was relatively low (number needed to harm (NNH) = 258 and 113, respectively). Compared with aspirin monotherapy, DAPT was associated with a higher rate of intracranial bleeding. Nevertheless, our results showed that the initiation of DAPT in the acute ischemic phase was not associated with a significantly increased risk of intracranial bleeding (vs DAPT initiation in post-ischemic stroke phase > 1 month).

As with ACS patients, DAPT with clopidogrel plus aspirin could provide an attractive option in high thrombotic risk patients immediately following acute stroke/TIA and with short therapy duration. Our results showed no significant increase in intracranial/major bleeding events as well as persistent reduction of ischemic events, even with a shorter DAPT duration (i.e., 1 month). In patients at higher bleeding risk, the tradeoff between ischemic and bleeding risks should be weighed individually.

Some RCTs (n = 6) used a loading dose of clopidogrel (300 and 600 mg) [22, 26, 27, 29, 31, 36]; these trials either used DAPT for a short period (≤ 21 days) or assessed short-term outcomes (3 months), and the optimal loading dose in this setting remains uncertain. In a previous RCT, a loading dose of clopidogrel was not effective in reducing the risk of recurrent stroke compared with a maintenance dose only [42]. Furthermore, our meta-regression analysis for the effect of initial clopidogrel dose on recurrent stroke did not suggest any significant effect modification.

Current AHA/ASA guidelines recommend 21 days of clopidogrel plus aspirin only in patients presenting with minor stroke (Class IIa; level of evidence: moderate) [18]. The recommendation is mainly based on the Clopidogrel in High-Risk Patients with Acute Nondisabling Cerebrovascular Events (CHANCE) trial, which recruited 5170 Chinese patients and found a greater benefit of DAPT in preventing early recurrent stroke without a higher rate of bleeding [26, 39]. Nevertheless, in a subset analysis of both Asian and non-Asian populations from 5 RCTs with NIHSS scores of 3 or less or TIA, we found significantly lower rates of any stroke (RRR 25% and ARR 2.5%) and ischemic stroke (RRR 26% and ARR 2.6%) without a significantly increased rate of intracranial bleeding in the DAPT group compared with monotherapy. Although the rates of major bleeding were significantly increased with DAPT, the NNH was high (rates 0.5% vs 0.3%; NNH 375).

In patients with stroke or TIA, antiplatelet “resistance” might represent a factor for the occurrence of stroke due to several pharmacogenetic and pharmacodynamic properties [43, 44]. Genetic polymorphisms, notably CYP2C19, are associated with increased clopidogrel response variability [45]. Among patients with stroke and TIA, carriers of CYP2C19 loss-of-function (LOF) alleles had an associated increased risk of stroke compared with noncarriers [46]. However, the frequency of CYP2C19 LOF varies substantially between white and East Asian populations (up to 30% and 60%, respectively) [45]. In our subgroup analysis of Asian vs Western studies, the rates of recurrent any or ischemic strokes in the DAPT group were almost a third more in Asians vs Westerners (9.3% and 8.2% vs 6.1% and 5.3%, respectively), possibly reflecting the higher prevalence of poor/intermediate clopidogrel metabolizers among Asian populations. Nevertheless, compared with monotherapy, we found that the combination of clopidogrel plus aspirin was associated with significant reductions of any/ischemic stroke in East Asians. In Western studies, there was an increased risk of any death, potentially driven by significantly increased rates of major bleeding. However, the average duration of DAPT in the Western studies was 211 days (vs 29 days in Asian studies), which inherently carries a higher bleeding risk and thus potentially mortality. This also supports the use of DAPT for short durations in these patients.

There are limitations to the present study. First, there were differences between the included studies with regard to the baseline stroke severity (NIHSS score), follow-up duration, drug dosage, and duration of treatment, as well as differences in the timing of randomization after stroke and TIA. However, we have tried to provide consistent results through sensitivity, subgroup, and meta-regression analyses. Second, we could not investigate the clinical outcomes based on the stroke subtypes as we lacked individual patient-level data. Third, we were not able to perform subgroup analysis based on the index stroke severity, as most trials did not provide an initial stroke assessment score, such as the NIHSS score. Fourth, we were not able to assess any functional outcomes, such as the modified Rankin score, as most trials did not provide such outcomes.

Conclusions

Among patients who presented with ischemic stroke and TIA, the combination of clopidogrel plus aspirin compared with antiplatelet monotherapy was associated with a significant reduction of recurrent stroke with no increased risk of vascular events (TIA, MI, or vascular death) or mortality. Although there was increased risk of bleeding events, mainly with long therapy duration and late stroke/TIA presentation, short course clopidogrel and aspirin was associated with less recurrent stroke and equivalent bleeding complications compared with monotherapy. Similar findings were also observed in patients who presented with minor stroke and TIA.