Rapid Blood Pressure Lowering in Patients with Acute Intracerebral Hemorrhage (INTERACT-2)

When a patient rolls in unresponsive and hypertensive intracranial hemorrhage is at the top of our differential. When the CT scan shows a large hemorrhage, we are quick to reverse coagulopathy if present, intubate if indicated, discuss the case with neurosurgery and pray. However, the management of blood pressure tends to vary based on the attending physician and their underlying training and experience. The dogma of avoiding blood pressure control because of concern over neurologic deterioration in acute intracerebral hemorrhage has been debated intensely.


Preliminary data had suggested that it reduced hematoma growth, a marker for poor outcomes in patients with ICH, but patient-oriented outcomes were lacking. A recent Study based on the original pilot study INTERACT1; INTERACT2 has just been published on May 29th in the NEJM.

1.  Why is this topic important?

The current blood pressure management of patients with ICH is unknown. Guideline recommendations suggest maintaining a systolic blood pressure of <180 mmHg. However, many physicians still practice on old dogma that permissive hypertension be allowed to a systolic of < 180 and to avoid relative hypotension in order to maintain perfusion to the ischemic penumbra surrounding the hemorrhage.

2.  What does this study attempt to show?

This study attempts to demonstrate that in patients with spontaneous intracranial hemorrhage with:

1.   Hypertension, systolic BP > 180 mmHg

2.   Not early surgical candidates (who these people are no one knows)

3.   Had an admission GCS >5

4.    Did not have a large hematoma with a poor prognosis based on attending physician discretion

Were randomized to intensive blood pressure lowering to a systolic BP <140 within one hour of randomization and continuing with a goal blood pressure of <140 for the next 7 days or guideline therapy maintaining a systolic blood pressure below/near 180 mmHg.  The type of anti-hypertensive agent was not specified, and 7 main anti-hypertensive agents were used.

  The study attempts to demonstrate that in patients who receive early aggressive anti-hypertensive treatment, maintained for 7 days, there is improvement in death and disability, defined as a mRS >2.


3.  What are the essential findings?

A total of 2,839 participants were enrolled in the study, with > 60% of participants in each arm coming from China. 1,403 patients enrolled in the early intensive treatment arm, and 1,436 assigned to receive guideline-recommended treatment.

  Baseline characteristics were similar (NIHSS was slightly lower in the intensive treatment arm 10 vs. 11) Table 1.

  Also the hematoma volume (<15ml) and superficial location of the hematoma favored the treatment arm, (both known to be better prognostic markers) Figure 1.

   At 90 days 719 participants (52%) in the intensive-treatment group, as compared with 785 (55.6%) in the standard-treatment group, had a poor outcome defined as a mRS >2 (OR 0.87; 95% CI, 0.75-1.01; p = 0.06). Table 3.

   On ORDINAL analysis (different prespecified vantage point) there was a shift in the distribution of scores on the mRS favoring the intensive blood pressure-lowering treatment (OR 0.87; 95% CI, 0.77 to 1.00; P =0.04). Table 3.

   There was no difference in the pre-specified safety outcomes of ACS, hypotension, and neurologic deterioration in the first 24 hours, Table 3.

   An interesting finding was that there was no statistically significant decrease in size in hematoma expansion in the intensive lowering of blood pressure group at 24 hours, the presumed mechanism by which blood pressure management was thought to be protective.


4.  How is patient care impacted?

Given that there were few adverse events and that there is a small signal of potential benefit I can see the guidelines changing to recommend a more strict control of blood pressure after ICH. See below.


5.  Is this an area of controversy?

Acute management of blood pressure has been a critical question in all neurologic emergencies; stroke, subarachnoid hemorrhage and ICH. It is believed that auto-regulation of blood pressure by the cerebral vasculature is lost during neurologic insults such as ICH. The authors of the INTERACT2 trial believed the mechanism of benefit to be a reduced hematoma size, based on their pilot study INTERACT1. The trial failed to demonstrate this, but it must be mentioned that finding statistical significance among means may not be appropriate. There is potential that some patients had marked reduction in their hematoma size and they accounted for most of the benefit, but were washed out by the overwhelming number of patients who did not see such a reduction.

   Also it is possible that small reductions in hematoma volume (on the order of 2mL could have profound impact on neurologic sequelae, but this question remains unanswered.)

   It is also interesting that they picked a primary outcome of a mRS >2, because had they of picked a mRS of >1, as is typical in most acute stroke trials, the evidence of benefit would have been statistically significant with a lower end point (OR 0.83; 95% CI 0.70 to 0.98; p =0.03).

   This study although suggestive is certainly not definitive, and will be aided by the ATATCH2 study to be published in 2016. Proponents of blood-pressure control will argue that there is no downside to aggressive blood pressure control, and potential a benefit in an otherwise group of patients with a high morbidity and mortality rate. Others will argue that the baseline characteristics of the treatment arm were heavily favored, based on NIHSS, location of hemorrhage, and % of patients with initial hematoma volume < 15mL.


6.  Major Limitations of the study?

The study has several strengths. First it is a large multi-center randomized controlled trial. Their outcomes are patient-centric and they pre-specified their secondary outcomes and sub-group analysis.

   However, multiple flaws in the study design are present. First, the choice of anti-hypertensive agents was markedly different than is used primarily in the United States. Urapidil, a potent alpha antagonist was used in a third of patients in the intensive blood pressure control arm, and nicardipine and labetalol only accounted for another third. Other less titratable medications such as nitroprusside, furosemide, hydralazine and nitroglycerine accounted for the other third. This was almost certainly because 2/3’s of patients recruited in the study were from China. Whether or not this limits the generalizability of the results is questionable.

   In addition it is important to note that the results of ATATCH II (antihypertensive treatment of acute cerebral hemorrhage) is almost completed with recruitment, and the results of the study will be much anticipated in 2016. The study design is in North America, with similar inclusion criteria to the INTERACT2 study, except that nicardipine was used as the sole agent, and patients had to receive intensive blood pressure control in < 4 hours, whereas in the INTERACT2 trial 50% of patients were randomized after 4 hours.


Despite the limitations several an important concept was addressed:

 Intensive blood pressure control may have a neurologic benefit in patients with intracranial hemorrhage and at worst does not seem to result in any increased morbidity. The mechanism by which this occurs is not entirely clear.


Until next time!



Anderson et al. Rapid Blood-Pressure Lowering in patients with Acute Intracerebral Hemorrhage. NEJM. 2013; 1-11.