aspirin right after tia/stroke

A meta-analysis challenged the published statistic of the protective effect of aspirin after a TIA or ischemic stroke, showing that the effect of aspirin is much greater than believed, especially if given within days of the event (see stroke TIA early ASA rx lancet2016 in dropbox, or doi.org/10.1016/S0140-6736(16)30468-8 ). details:

--background:

    --risk of recurrent stroke is 10% in the week after a TIA or minor stroke

    --but many patients delay seeking medical attention for days-to-weeks after these events

    --in the UK, 1/2 of recurrent strokes happen prior to seeking medical attention

    --pre-hospital use of self-administered aspirin is often discouraged for fear of exacerbating intracerebral bleed

    --BUT, hemorrhage is a rare cause of TIAs and is in <5% of minor strokes

    ​--there are minimal data from RCTs on the effect of aspirin after TIA or minor stroke, with only observational data on the effect of aspirin on early benefits after TIA or minor stroke

--12 trials with 15,778 participants comparing aspirin vs placebo for secondary prevention of ischemic stroke, all with data on recurrent vascular events within 12 weeks of randomization. the data from these trials were reanalyzed, looking at individual patient outcomes

--results:

    --aspirin vs placebo reduced the 6 week risk of:

        -- ischemic stroke by 58% [(HR 0.42 (0.32-0.55), p<0.0001)]; 84 of 8452 on aspirin, vs 175 of 7326

        --disabling or fatal ischemic stroke by 71% [(HR 0.9 (0.2-0.42), p<0.0001)]; 36 of 8452 on aspirin, vs 110 of 7326

        --and, greatest benefit was in those put on aspirin within 2 weeks à for disabling or fatal ischemic stroke the risk was decreased by 93%:  [(HR 0.07 (0.02-0.31), p=0.004)]; 2 of 6691 on aspirin, vs 23 of 5726

    --the effect of aspirin on recurrent strokes was partly by reduction in severity (by modified Rankin Score, 58% decrease), and was independent of dose, patient characteristics, or etiology of TIA/stroke

    --there was still some benefit at 6-12 weeks after the initial event, but none after 12 weeks

    --in the 3 trials (40,531 people) in people with major acute stroke, recurrent ischemic stroke at 14 days was mostly in patients with less severe baseline deficits and was 63% less by the 2nd day after starting treatment [(HR 0.37 (0.25-0.57), p<0.0001)]

    ​--in terms of the dreaded complication of intracerebral hemorrhage, there was no difference between those on aspirin vs control at 12 weeks (3 of 4125 on aspirin vs 5 of 4137 in control group), though there was a trend to more events in those on high-dose aspirin

    --the effect of aspirin on preventing recurrent ischemic strokes was independent of the dose (low dose = <100mg/d, high-dose =  >300 mg).

    --also no difference if patient diabetic, hypertensive, or if current smoker 

    ​--also no difference in patients with atrial fibrillation at baseline (HR 0.28 (0.087-1.00), p=0.0508), or in those with lacunar strokes [I could not find the data on this in the supplementary materials, so cannot comment further]

    --dipyridamole plus aspirin vs aspirin alone (7 trials, with 6602 participants)

        --no difference in risk of ischemic stroke or their severity in first 12 weeks (OR=0.90)

        --but after 12 weeks, there was a 24% decreased risk [(OR 76 (0.63-0.92), p=0.005)], particularly for disabling or fatal ischemic strokes [(OR 0.64 (0.49-0.84), p=0.001)]

Conclusion:  it seems that aspirin worked much better when given as soon as possible (in the above analysis, in the 0-2 weeks after a TIA or minor stroke), and also pretty dramatically reduced the severity of a further ischemic stroke when that happens (in the 70% range), also with the greatest reduction when given within the first 2 weeks. Dipyridamole did nothing early on, but the combo with aspirin was superior to aspirin in the long-term.

so, this poses the difficult clinical conundrum: how do we reach out to patients to make sure they get aspirin therapy as soon as they have one of these minor events: a resolved TIA or a minor stroke, for which many a denier type may just write-off and not seek care. this brings up several issues:

--it is likely much safer/preferable to reinforce/have public health initiatives to educate patients to seeking care as soon as possible after any stroke-like symptoms.  and to reduce obstacles to care: this would likely happen more often if there were universal coverage for health care, and if there were not the rather daunting co-pays that people may be confronted with if they go to an ER.

--the UK did a major public education campaign (Oxford Vascular Study), showing that there was there was a major improvement in patients having a major stroke seeing emergency attention within 3 hours and simultaneous decrease in people just making future medical appointments; but for TIA and minor stroke, there was not much difference in either of these, suggesting that reaching those with TIA/minor stroke and convincing them to seek urgent medical attention is not so easy. (graphs in the supplemental material of this study)

--given the low likelihood of there being a cerebral hemorrhage causing the TIA/minor stroke (and this study found no increase in the 12-week risk of intracerebral hemorrhage on low dose aspirin vs control), it seems that there would be pretty clear benefit to the public health imperative: self-administer aspirin as soon as possible after a patient has symptoms suggestive of TIA/minor stroke (though, again, being assessed in the ER is the preferable action). And it should be emphasized that the most dramatic effect of aspirin in those with TIA/mild stroke, a 93% reduction, was in preventing of disabling or fatal subsequent ischemic strokes.

--and, I do try to use the combo of aspirin/dipyridamole over just aspirin if the patient can tolerate it (it is twice a day and has more adverse effects) but is pretty expensive (though, both aspirin and dipyridamole are really, really old meds….), since the data do suggest some more efficacy over plain low-dose aspirin. But with this data, I think it is likely to be more useful to start aspirin right away (cheaper and better tolerated), just to make sure the patient has the aspirin in their system as soon as possible, and then transition to the combo aspirin/dipyridamole over the next couple of months (making sure the patient reverts to aspirin if the combo causes any problems).