Loading ...

Patients with ischemic stroke on antiplatelet therapy are at a significantly higher risk for symptomatic intracranial hemorrhage (sICH) following thrombolysis, but the absolute increased risk is small, results of a new study showed.

The findings should be reassuring to clinicians who might hesitate to prescribe thrombolysis therapy such as IV tissue plasminogen activator (IV-tPA) to patients on antiplatelet therapy, particularly dual antiplatelet therapy (DAPT), study investigator Richa Sharma, MD, MPH, assistant professor, Department of Neurology, Yale School of Medicine, New Haven, Connecticut, told Medscape Medical News.

"The study showed the rates of hemorrhage from thrombolysis are not excessive and do not warrant withholding the potentially life-saving therapy of thrombolysis when a patient with an ischemic stroke presents to the emergency department," Richa said.

The findings were published online on May 20 in JAMA Neurology.

An Unclear Risk

IV-tPA is recommended for patients presenting with acute ischemic stroke (AIS) within 4.5 hours of symptom onset to reduce stroke-related morbidity. This carries a risk for sICH, but it was unclear to what extent pre-stroke antiplatelet treatment is associated with this risk.

The study included 321,819 patients (mean age, 69 years; 16% Black and 69% White) from the Get With The Guidelines (GWTG)-Stroke registry, an ongoing national reporting system sponsored by the American Heart Association and American Stroke Association. Participants had an AIS between 2013 and 2021 and were treated with IV-tPA within 4.5 hours of symptom onset.

Of the totalstudy population, 56.7% did not use antiplatelet therapy, 36.5% used single antiplatelet therapy (SAPT), and 6.8% used DAPT before IV-tPA.

From 2013 to 2021, the number of patients not using any antiplatelet therapy increased (from 54% to 61%; P = .03), SAPT use decreased (from 40% to 33%; P = .04), and DAPT use increased (from 6% to 7%; P = .009).

The increase in DAPT use was due to more patients taking aspirin and clopidogrel, said the authors. Ticagrelor monotherapy and aspirin-ticagrelor dual therapy increased numerically, although the trends were not significant.

Researchers used a propensity score subclassification algorithm to account for pre-stroke risk factors. Potential confounders included age; sex; race; body mass index (BMI); medical history including hypertension, dyslipidemia, diabetes, and atrial fibrillation; and current use of blood pressure, cholesterol, or glucose-lowering medication.

SAPT exposure was defined as aspirin, clopidogrel, prasugrel, or ticagrelor monotherapy. DAPT exposure was defined as aspirin-clopidogrel, aspirin-ticagrelor, or aspirin-prasugrel combination therapy.

Estimating sICH Risk

The primary outcome was sICH, documented by imaging and a physician's note indicating clinical deterioration due to hemorrhage, within 36 hours of IV-tPA administration.

Symptomatic ICH occurred in 2.9% among those not taking antiplatelet therapy, 3.8% of those taking SAPT, and 4.1% among patients treated with DAPT.

After propensity score subclassification and adjustments, sICH risk was 13% higher with pre-stroke SAPT exposure (odds ratio [OR], 1.13; P < .001) and 28% higher with pre-stroke DAPT exposure (OR, 1.28; P < .001) than with no antiplatelet exposure.

DAPT exposure was independently associated with an increased risk for sICH compared with SAPT exposure (OR, 1.18; P < .001).

Compared with patients taking no pre-stroke antiplatelet therapy, patients in the SAPT and DAPT groups had a 0.9% and 1.2% absolute increased risk for sICH, respectively.

"The absolute rate of symptomatic intracranial hemorrhage was no higher than the rates in all the landmark thrombolysis trials that we use to inform our clinical practice," said Sharma.

There was no significant interaction between exposure groups and sex, BMI, history of hypertension, coronary artery disease and/or myocardial infarction, history of diabetes, serum creatinine level, and endovascular thrombectomy treatment.

Researchers found an increased sICH risk in patients treated with IV-tPA in the late 3- to 4.5-hour window compared with in those treated in the less than 3-hour window, particularly in patients taking DAPT.

Functional Outcome

A secondary outcome was modified Rankin Scale (mRS) score, which measures functional outcome at discharge from 0 (no symptoms) to 6 (death). Researchers defined functional independence as an mRS score of 0-2.

After propensity score subclassification and adjusting for covariates, functional independence at discharge was less likely with SAPT (OR, 0.92; P < .001) or DAPT (OR, 0.94; P = .01) compared to no antiplatelet exposure. However, absolute differences between groups were small.

There was a slight increase in in-hospital mortality in patients with DAPT compared with in those with SAPT (OR, 1.10; P = .007) and no antiplatelet therapy (OR, 1.10; P = .045) but no significant associations between SAPT and no antiplatelet therapy groups.

A separate exploratory analysis found only SAPT was associated with less favorable outcomes at 90 days, but only marginally.

Sharma stressed these results should be taken "with a grain of salt," as long-term outcomes in the registry are incomplete.

Despite the perceived increased risk linked to ticagrelor, there were no differences in rates of sICH, mortality, or functional outcomes between aspirin-ticagrelor and other DAPT treatments. However, the study was not powered to exclude a clinically important difference.

The aspirin-prasugrel group was not evaluated due to its low frequency in the sample.

Know the Antiplatelet History

These new results suggest being on DAPT shouldn't preclude treatment thrombolysis if otherwise indicated, Sharma said.

"Thrombolysis should still be the standard of care in patients on antiplatelet therapy" who suffer a stroke, she added.

Knowing the antiplatelet status of patients receiving thrombolysis could assist clinicians in terms of triage.

"Once the patient has been treated with thrombolytic therapy, this might help in terms of where they might need to go in the hospital and what degree of monitoring, they might need," Sharma said.

As this was a retrospective, observational study, it was subject to selection bias, and the findings may not generalize to populations outside the United States. Other study limitations included the homogeneity of antiplatelet therapies and lack of adequate matching of patient characteristics, which may have introduced an indication bias.

Timely, Important

In an accompanying editorial, Shyam Prabhakaran, MD, Department of Neurology, The University of Chicago,Chicago, and Jose G. Romano, MD, University of Miami, Coral Gables, Florida, said the study "is timely and important" and offers several implications for clinical practice.

"First, though rates of sICH are lower than the rates observed in pivotal trials of alteplase, the authors' finding of an interaction by treatment time window raises the possibility that risks may exceed benefits, especially in the 3- to 4.5-hour window and in patients with mildly disabling deficits," they wrote.

The findings also underscore the value of earlier treatment with alteplase and suggest collecting a more comprehensive patient medication history is important.

Prabhakaran and Romano noted the sICH rates in the study (up to 4.1% in patients taking DAPT) "are still much lower than the 6.4% sICH rate observed in the National Institute of Neurological Disorders and Stroke [NINDS] alteplase trials using the same definition for sICH."

It's unlikely the "modest" sICH rates seen in the study "would mitigate the proven benefits of alteplase," they wrote.

However, they stressed the results should be approached with some caution and in context. For example, there was no central adjudication of sICH in the GWTG-Stroke registry, so misclassification and underreporting of sICH is possible.

Missing pre-stroke disability data is another study limitation, they added. Because 56.8% of patients in the DAPT group had prior stroke or transient ischemic attack compared with only 14.4% in the group not taking antiplatelet therapy, "one might speculate that baseline differences in pre-stroke disability may be a significant residual confounder in these analyses."

The editorialists also noted creatinine and platelet count, prespecified covariates in adjusted models, were missing in many subjects. The study also lacked information on antiplatelet dosage and time of last ingestion.

Sharma reported receiving grants from the National Institutes of Health/NINDS outside the submitted work and having a patent for "Methods of Training an Algorithm to Predict Ischemic Stroke Etiology" pending. Prabhakaran reported grants from the National Institutes of Health and the Agency for Healthcare Research and Quality and personal fees from UpToDate.