Intracranial hemorrhage tends to have a poor prognosis – the cascade of inflammation, vasospasm, and necrosis leaves the vast majority with some residual disability. INTERACT-2 offered some glimpse of hope, at least, from a surrogate standpoint – finding that patients with “intensive” blood pressure control showed reduced hematoma growth. It seems logical that reduced hematoma growth would directly correlate with improved clinical outcomes, and there was a reasonable suggestion this was present, as well.
And, so, ATTACH-2.
This trial randomized patients with ICH to “standard” 140-179 mmHg blood pressure control versus “intensive” 110-139 mmHg blood pressure control, with a primary outcome of modified Rankin scale of 4 to 6. All patients were treated using nicardipine by continuous infusion, and generally achieved their blood pressure targets within 2 hours of randomization. They included 500 in each arm of the trial before stopping prematurely when a pre-specified threshold for futility was crossed.
Since I’ve used the word “futility” twice so far, you’ve probably already discerned the result. At 90 day follow-up, there were 38.7% with mRS 4-6 in the intensive group versus 37.7% in the standard group. No other long-term clinical outcomes seemed to reflect an advantage to one arm or the other. In both adjusted and unadjusted analyses, there were a few interesting tidbits. Hematoma expansion was, indeed, attenuated by intensive control. However, as a counterweight, the intensive treatment group was more likely to suffer neurologic deterioration within 24 hours (11.0% vs. 8.0%) and more likely to suffer a serious adverse event within 3 months (25.6% vs. 20.0%).
It certainly seemed plausible such intensive lowering might be of value – but, instead, we have an excellent example where patient-oriented outcomes trump surrogates, and the adverse effects from treatment seem to counterbalanced any benefit picked up along the way.
“Intensive Blood-Pressure Lowering in Patients with Acute Cerebral Hemorrhage”
http://www.nejm.org/doi/full/10.1056/NEJMoa1603460