(Last Updated On: August 17, 2016)

Qureshi MD, Adnan, et al. for the ATACH-2 Trial Investigators and the Neurological Emergency Treatment Trials Network. Intensive Blood-Pressure Lowering in Patients with Acute Cerebral Hemorrhage. NEJM 10.1056/NEJMoa1603460. PMID:27276234


Research and guidelines are limited in terms of specific goal blood pressure targets in patients who suffer intracranial hemorrhage, especially those with acute hypertensive responses.  There is data to suggest that expansion of intracranial hemorrhage and its subsequent complications, may be reduced with quick, aggressive blood pressure control.  A prior study, INTERACT2, found that lower blood pressure goals were correlated with a non-significant decrease in death or disability among patients with spontaneous intracranial hemorrhage that presented with systolic blood pressures of 150-220 within six hours of symptom onset.  This decrease was seen in patients targeted for a goal SBP <140 within 1 hour versus a goal of <180 in the standard group.

Research question:

What is the utility of quickly lowering systolic blood pressure in patients with ICH?


Randomized, multi-center, 2-group, open label study.  100 sites, including US, Japan, China, Taiwan, S. Korea, Germany.

Trial Intervention:

  • goal of therapy was to decrease and maintain hourly systolic BP at 140-179 (“standard group”) versus 110-139 (“intensive group”) throughout 24 hours after randomization
  • after randomization, nicardipine was started and increased to a max of 15mg/hour, if SBP still not at target after 30 minutes of max dose, a second drug was used (IV labetolol).
  • IV diltiazem or urapidil used in countries that do not carry labetolol*
  • the ability to attain goal BP was assessed, with primary treatment failure  defined as not attaining <140 mm in “intensive group”, and <180 mm in “standard group”, within 2 hours of treatment initiation.  secondary treatment failure referred to failure to maintain target blood pressures,  which were measured hourly, for 2 consecutive hours


  • 1000 patients; 500 in standard therapy group, 500 in intensive blood pressure control group
  • age >=18, GCS >= 5 on arrival, ICH measuring <60 cm^3 on 1st CT scan, able to initiate BP treatment within 4.5 hours
  • patients required to have at least one SBP >/= 180 between symptom onset and beginning the IV anti-hypertensive therapy
  • average baseline systolic blood pressure: 200 +/- 27
  • average age: 61.9


  • SBP <140 before randomization occurred


  • CT w/o contrast was performed at 24 hrs after starting therapy, and ICH site, whether blood present in ventricles, and volume of the bleeding was ascertained
  • Serious events reported for up to 3 months
  • Nonserious adverse events reported up to 7 days after randomization or discharge from hospital
  • Post-hospital phone call followup was done at 1 month, during which, adverse event and death information were reported
  • In-person visit at 3 months, during which, the following information was obtained: modified Rankin score (measure of disability or dependence in ADLs), quality of life per the ED-5D scale, serious side effects, and physical and neurologic exams

Primary outcome:

  • Moderately severe or severe disability on modified Rankin (scores of 4-6) or death at 3 months

Secondary outcomes:

  • ED-5D utility index and visual-analogue scale (VAS) scores at 3 months; percentage of patients with expansion of ICH by> /= 33% on 24hr follow-up CT scan
  • Safety outcomes: GCS decrease from baseline of >/= 2 points, or increase in NIH stroke scale by >/= 4 points, lasting at least 8 hours; death within 3 months


  • Trial was stopped secondary to futility before 1280 patients were enrolled
  • Primary outcome (death or disability): 38.7% of intensive treatment group, 37.7% in standard group, RR 1.04, confidence interval 0.85- 1.27
  • No significant difference in EQ-5D scores or expansion in ICH between groups
  • No significant change in rates or death at 3 months or unfavorable change in NIH scale or GCS scores at 24 hours between groups
  • Serious adverse events attributed to treatment in 1.6% of intensive group, v. 1.2% in standard group, within 72 hours
  • Significant difference in number of renal adverse events in the first week after randomization, higher rates in intensive group v. standard group (9% v. 4%, P=0.002.)


  • More patients in intensive therapy group had primary treatment failure; perhaps, there is under-estimation of potential treatment effect as a result
  • Rate of death or disability at 3 months was lower in standard group (37%) than observed in prior studies (quoted at 60%), which could be due to higher baseline GCS scores in this study compared to prior studies, which make the results difficult to interpret
  • Overall, aggressive blood pressure reduction does not appear to improve outcomes in patients who suffer ICH.  It is theorized that the act of intervening itself, simply blunting the acute hypertensive response to any extent, is in itself protective, and may not be dependent on the precise degree of reduction
  • This trial does not apply to those patients who suffer large ICH, those who have raised intracranial pressure, or compromised cerebral perfusion, in whom aggressive blood pressure control may result in harm by way of compromising cerebral blood flow

How will this affect my practice?

I found it helpful to learn that in patients with an acute ICH with favorable GCS scores on arrival (most of the study patients had GCS >12, and greater than 50% in each group had GCS of 15), a systolic blood pressure target of 140-179 mm resulted in similar outcomes compared to intensive blood pressure control.  I think that more studies are needed which can more uniformly achieve the target blood pressure in the intensive therapy group patients, and perhaps then the outcomes will prove better.  Until that time, I may use 140-179 as my target.
Mary Photo  Dr. Naughton is an EM/IM PGY-3.