Thrombolysis Limits Stroke Regardless of Age, Severity

Susan Jeffrey

February 13, 2014

SAN DIEGO, California — A new meta-analysis confirms that regardless of patient age or stroke severity, thrombolytic treatment of ischemic stroke is associated with less long-term disability.

Although earlier treatment was associated with better outcomes, patients in every subgroup of age and stroke severity receiving tissue plasminogen activator (tPA) within 4.5 hours of symptom onset benefited from treatment, with more patients left with little or no disability, including patients over 80 years of age.

Symptomatic intracerebral hemorrhage (ICH) was higher in patients treated with tPA and was associated with an early excess in mortality vs nontreated patients, but tPA did not affect other causes of early death.

"Among those treated earlier, there was a suggestion that this early hazard may be followed by later mortality benefits," Jonathan Emberson, PhD, senior statistician and university research lecturer at the University of Oxford's Clinical Trial Service Unit (CTSU), Nuffield Department of Population Health, United Kingdom, concluded.

He presented the findings on behalf of the Stroke Thrombolysis Trialists' Collaboration here at the American Stroke Association (ASA) International Stroke Conference (ISC) 2014.

In a press release, Dr. Emberson was quoted as saying that these findings, "may have implications for treatment guidelines on both sides of the Atlantic." In a press conference, he said as a statistician, he wouldn't try to influence guideline makers, but underlined that "the results of the analyses we have done are quite clear at least up to about 4.5 hours after the initial stroke."

He referred the question to one of his colleagues, Kennedy Lees, MD, coauthor and professor of Cerebrovascular Medicine at the University of Glasgow, United Kingdom.

Clinical guidelines in Europe do currently allow for treatment of elderly patients, as well as treatment within the extended window of 4.5 hours, Dr. Lees said. However, he noted, "the licensing in Europe does not include the elderly population at the moment, so one might wonder whether the licensing authorities might look at the data and see that there's now no justification for that 'ageist' approach."

In the United States, tPA is approved by the US Food and Drug Administration (FDA) to treat ischemic stroke up to 3 hours after symptom onset, but American Heart Association (AHA)/ASA guidelines released in 2013 allow for treatment out to 4.5 hours, based on results of the third European Cooperative Acute Stroke Study (ECASS 3) study. Because ECASS 3 excluded patients older than 80 years of age, the AHA/ASA guideline also excludes patients older than 80.

Changing that recommendation, then, "also seems to be something that my American colleagues are likely to wish to do," Dr. Lees noted. "It seems to be entirely reasonable that now everybody who is eligible for thrombolysis should be treated out to 4.5 hours, as early as possible within that time, and regardless of age."

Asked to comment on the findings, Steven M. Greenberg, MD, PhD, professor of neurology at Harvard Medical School, Boston, Massachusetts, and chair of the program committee for this meeting, said the AHA/ASA guidelines are nearly constantly under revision and the question of how to handle age is an issue that they are dealing with.

"But I think certainly that this evidence, that came mostly from the IST-3 [third International Stroke Trial] and the meta-analysis you saw today really makes a good argument that age should probably not be considered a major barrier to tPA, and the guidelines will have to work their way through that question," he told Medscape Medical News.

Stroke Thrombolysis Trialists' Collaboration

It's well established that tPA is effective in treatment of acute ischemic stroke, Dr. Emberson said, "but there remains considerable debate regarding its use in longer times since stroke, its use in older people, and its use in patients who had the least or most severe strokes."

The Stroke Thrombolysis Trialists' Collaboration, which includes trialists, meta-analysts, and other stroke experts, was set up before the 2012 report of results of IST-3. "Those trialists agreed that once IST-3 was complete, they would share their individual patient data for independent statistical analysis by the CTSU at the University of Oxford, which is not otherwise involved in any of the individual trials," he said.

The results of the current analysis include all 6756 stroke patients who were enrolled in 9 thrombolytic trials: Alteplase ThromboLysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) A and B; ECASS I, II, and III; Echoplanar Imaging Thrombolysis Evaluation Trial (EPITHET); IST-3; and National Institute of Neurologic Diseases and Stroke (NINDS) A and B.

The objectives of the meta-analysis, Dr. Emberson said, were first to assess the extent to which treatment delay modifies the proportional effect of tPA on stroke outcome, to assess the extent to which age or stroke severity might further modify the effects, and to examine the effects of tPA on the risk for symptomatic ICH and mortality.

Patients in IST-3 differed from those in previous trials in that they were an average of 11 years older, with just over half of patients older than 80 years of age. "Average treatment delay, however, wasn't much different between the 2 groups of trials, being about 20 minutes longer in IST-3," while stroke severity was the same, he noted. Baseline characteristics across the 9 trials were well balanced, he added.

The primary efficacy outcome was a modified Rankin scale (mRS) score of 0 or 1 at 3 to 6 months after stroke, indicating little or no residual disability. Predefined safety outcomes included 90-day mortality and symptomatic ICH using 2 definitions: (1) parenchymal hemorrhage 2 (PH2) within 7 days or the Safe Implementation of Thrombolysis in Stroke Monitoring Study (SITS-MOST) definition of a PH2-type hemorrhage within 36 hours and (2) fatal ICH within the first 7 days.

On the primary efficacy outcome, Dr. Emberson reported that the odds of achieving an mRS score of 0 or 1 was significantly increased by tPA, "with significantly bigger proportional benefits arising from earlier treatment. The further tests of interaction found no evidence that the shape of this slope was significantly altered, either by age or by stroke severity."

Table. Stroke Thrombolysis Trialists' Collaboration: Primary Efficacy Outcome by Treatment Delay

Time From Symptom Onset Odds Ratio (95% Confidence Interval)
≤3 h (average: 2 h, 2 min) 1.75 (1.35 - 2.27)
3 - 4.5 h 1.26 (1.05 - 1.51)
>4.5 h 1.15 (0.95 - 1.40)


"In particular," he added, "there was clear evidence of benefit in the subgroup of patients aged over 80 years, who, on average, received treatment after 3 hours and 40 minutes."

In terms of safety, the researchers found that without tPA, the average risk for ICH by any definition was low but was increased significantly by tPA. The relative risk for fatal ICH within the first 7 days was 7.1 (95% confidence interval, 3.98 - 12.8), Dr. Emberson said, "reflecting 91 vs 13 deaths, or 2.7% vs 0.4%, an average absolute excess risk of about 2%."

By 90 days, however, there was a nonsignificant 11% excess of death from any cause. Within the first week, the excess mortality was all related to fatal ICH, without any evidence of increase in other causes of death (191 vs 191).

"Among patients who survived that first week, there were no significant differences in subsequent mortality up to day 90, and perhaps even a suggestion from both this and other analyses of a later protective effect on mortality with increased follow-up," he said.

Therefore, he noted, the overall relative risk for mortality at 90 days (17.9% with tPA vs 16.5%) of 1.11 (0.99 - 1.25), "doesn't really mean that much by itself because it mixes the real early excess risk from ICH with no significant effect on other early causes of death, or on later causes of death."

The proportional increase in the early risk for ICH was similar irrespective of treatment delay, age, or stroke severity, and was present even for those treated within 3 hours. "While the proportional increases in risk were similar across these categories, however, the absolute excess risk increased with increasing stroke severity," he said.

The data also suggested an increasing risk for overall 90-day mortality with increasing treatment delay.

He applied their findings to a hypothetical group of 100 stroke patients. Without treatment, about 25 patients would have a good outcome, and an average of 6 would die in the first 7 days. With tPA, about 10 additional patients would have a good outcome, "which is a big effect if you think of that as a number needed to treat of 10," Dr. Emberson noted, "but 2 extra might die of ICH within that first week."

However, many of the other patients, if treated, might do better, even if they didn't achieve an mRS score of 0 or 2, he added, particularly if given early.

Coauthor Peter Sandercock, DM, professor of neurology at the University of Edinburgh, United Kingdom, principal investigator of the IST-3 trial, the most recent addition to the meta-analysis, said it may be a difficult conversation to have in the emergency setting, but many patients say they would rather not survive than live with significant disability. "They would take the risk of an immediate bleed to have the prospect of being able to live independently," he said in an interview.

"In Europe, the vast majority of the stroke clinicians, as opposed to emergency clinicians who are not dealing with stroke patients, believe in treating regardless of age and out to 4 and a half hours, with knowledge of and acceptance of the small risk, that they will communicate to the patients, but with an absolute conviction that, on balance, they are doing a huge amount of good," Dr. Lees added. "The feeling when you treat a patient who looks as though they have a moderate stroke and could be badly disabled, and they get up and walk out within 24, 48 hours is absolutely compelling."

Dr. Sandercock added that "the real problem, on both sides of the Atlantic," is that older people are badly undertreated in general.

The study was funded by the UK Medical Research Council, the British Heart Foundation, and the University of Glasgow. Dr. Emberson has disclosed no relevant financial relationships. Dr. Lees reports he is chair of the Data Monitoring Committee for trials conducted by Boehringer Ingelheim, Grifols, and Lundbeck.

International Stroke Conference (ISC) 2013. Abstract #LB2. Presented February 12, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.