Statistics from Altmetric.com
Current guidelines from the American Heart Association/American Stroke Association advocate against the administration of intravenous thrombolysis (IVT) in patients with acute ischaemic stroke (AIS) with a prior ischaemic stroke (IS) within 3 months (Class III: Level of Evidence B).1 This concern is based on a presumed increased risk of symptomatic intracranial haemorrhage (sICH) and mortality2 and therefore led to the respective guideline recommendation. Thus, patients with a history of prior IS have been excluded in the majority of the pivotal randomised controlled clinical trials that have established IVT as a treatment for AIS.2
However, the rationale for excluding patients with a history of IS within the prior 3 months in the aforementioned trials was not based on pathophysiological considerations but instead on information extrapolated from the relevant recommendations for IVT in myocardial infarction.2 In the present systematic review and meta-analysis, we sought to compare the safety and efficacy of IVT treatment between patients with AIS with a history of previous IS within 3 months and patients with AIS but without a history of previous IS within 3 months from the index event.
This meta-analysis is presented according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for systematic reviews and meta-analyses.
From eligible studies, we extracted data on the primary safety outcomes of sICH (according the definition used in each study) and in-hospital or 3-month mortality. We also evaluated the following efficacy outcomes: early neurological improvement (ENI) at 24 hours (according to the definition used in each study), 3-month favourable functional outcome (FFO, modified Rankin Scale (mRS) scores 0–1), 3-month functional independence (FI, mRS scores 0–2). We also performed sensitivity analyses after adjustment for potential confounders in studies with available data. All associations were reported with the use of ORs or adjusted OR (ORsadj), as appropriate. Further information on the methods are available in the online supplementary file.
The complete search algorithm and flow chart of the literature search (online supplementary figure e-1) are available in the online supplementary file. After excluding seven studies that were either case series or provided overlapping data with more recent publications (online supplementary table e-1), we identified six eligible cohorts (52 631 total IVT-treated patients with AIS; 1.7% with history of prior IS within 3 months). Study and baseline patient characteristics are presented in online supplementary tables e2-e3.
In unadjusted analyses (online supplementary figures e2-e6), we documented that IVT administration in patients with AIS with a history of previous IS within 3 months from the index event was not associated with increased likelihood for sICH (OR=1.75, 95% CI: 0.84 to 3.65, p=0.13) or mortality (OR=1.36, 95% CI: 0.99 to 1.87, p=0.06) compared with alteplase use in patients with AIS without prior IS history or with history of previous IS more than 3 months from the index event. Likewise, the odds of ENI (OR=1.12, 95% CI: 0.86 to 1.44, p=0.40), 3-month FFO (OR=0.81, 95% CI: 0.60 to 1.10, p=0.18) and 3-month FI (OR=0.77, 95% CI: 0.58 to 1.00, p=0.05) were comparable between the two groups (table 1). There was no evidence of substantial heterogeneity (I2 <35%; p for Cochran Q>0.20) across the included studies for all unadjusted analyses evaluating safety and efficacy outcomes.
In analyses adjusted for potential confounders (online supplementary figures e7-e10), we detected no difference in the odds of sICH (ORadj=1.75, 95% CI: 0.21 to 14.67, p=0.61) or mortality (ORadj=1.27, 95% CI: 0.59 to 2.77, p=0.54) between patients with AIS with history of previous IS within 3 months and patients with AIS with no history of AIS within 3 months from the index event. The adjusted efficacy outcomes of ENI (ORadj=1.16, 95% CI: 0.86 to 1.58, p=0.33) and 3-month FI (ORadj=0.91, 95% CI: 0.62 to 1.33, p=0.62) were also comparable between the two groups (table 1).
In the present meta-analysis of patients treated with tissue plasminogen activator (t-PA), we found no evidence of increased sICH or mortality risk for the specific condition of these patients having a history of prior IS within 3 months from the index event. We also found comparable early and long-term functional outcomes between t-PA-treated patients with AIS with a history of IS within 3 months and t-PA-treated patients with AIS with no prior IS history or history of IS more than 3 months from the index event.
Although in the everyday clinical practice of high-volume stroke centres, history of stroke in the previous 3 months is already regarded to be an inadequate reason for withholding IVT,3 the reported rates of patients with AIS with a history of previous IS within 3 months receiving IVT treatment is surprisingly low (0.3%–2.0%; online supplementary table e-1). Taking into account that nearly 5% of patients after a first-ever IS will experience recurrence within the first 3 months,4 one can assume that many patients with AIS with a history of previous IS within 3 months are excluded from IVT treatment either due to other co-existing contraindications (eg, anticoagulant intake) or in accordance to the current guidelines.1
Some limitations of the current report also need to be acknowledged. As noted above, data on co-existing relevant contraindications are not available in the two groups. Moreover, the number of patients with IS in the previous 3 months in included studies was very limited, suggesting thus a highly selected population group. Many of the included studies also do not provide data on the baseline characteristics of patients with a history of previous IS within 3 months, while they adjust the reported associations for different confounders (online supplementary tables e-2 & e-3). A potentially higher prevalence of vascular comorbidities in patients with a history of IS within 3 months can be extrapolated from the shift in the reported associations with mortality between unadjusted (OR=1.36, 95% CI: 0.99 to 1.87) and adjusted for potential confounders analyses (ORadj=1.27, 95% CI: 0.59 to 2.77). It should also be noted that a negligible proportion of patients within the included studies received endovascular reperfusion therapies alone (1.3%) or in combination with IVT (4.2%). Additionally, since only one of the studies provided information on the elapsed time from the previous IS with the event of acute cerebral ischaemia treated with IVT (online supplementary table e-3), we were not able to assess the association of the aforementioned parameter with the outcomes of interest. The association of neuroimaging data (previous infarct size, topography and validation on neuroimaging) and/or stroke severity with the outcomes of interest could not be assessed. As with oral anticoagulant administration after an IS5 timing and severity of the previous IS could have a significant impact on the safety and efficacy of IVT treatment in patients with AIS with a history of recent IS.
The findings of the present meta-analysis do no support the safety concerns regarding the use of IVT in patients with AIS with a history of previous IS within the last 3 months. However, the present study cannot compare risk of safety events between early versus later time points from the prior stroke that occurred within 3 months of the index stroke, or assess the relevant risk of other contraindications. Prospective and large observational studies are needed to confirm the findings of the present report and assess further the impact of timing and severity of the previous stroke on the outcomes of patients with AIS with a history of recent IS who are treated with IVT.
Contributors GT and AHK: drafting the manuscript, data extraction, analysis and interpretation of the results. PDS, MK, TS, VC, DS, NA and AVA: revision of the manuscript. LP: revision of the manuscript, data extraction. SS: revision of the manuscript, study supervision.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.