Analysis of Recent Papers in Hypertension Treatment of Hypertension in the Setting of Acute Intracerebral Hemorrhage: Still No Clear Answer on the Best BP Level to Intervene or What BP Goal to Achieve
- Publications reviewed: Anderson CS, Heeley E, Huang Y, et al. Rapid blood pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med. 2013;368:2355–2365.
Whether or not and how aggressively to treat elevated blood pressure (BP) in the face of acute intracerebral hemorrhage (ICH) have been questions stroke neurologists and hypertension experts have been debating for decades. Even in patients without a known history of hypertension, BP tends to elevate after ICH. This effect is thought to be primarily an adaptive response to increased intracerebral pressure (ICP), although other mechanisms may also be involved. Whether elevated BP after acute ICH is simply a marker of increased ICP and a poor prognosis or is a potential target for intervention is unknown. Certainly, increased BP might lead to a greater risk of hematoma expansion and poor outcome, but lowering BP with antihypertensive medication in the face of increased ICP might also be associated with decreased cerebral perfusion and a greater likelihood of ischemia.
Current guidelines from the American Stroke Association (ASA) published in 2010 suggest aggressively treating BP in patients with acute ICH with intravenous antihypertensive agents if:
- The systolic BP (SBP) is >200 mmHg (or mean arterial pressure [MAP] is >150 mm Hg) or
- The SBP is >180 mmHg (or MAP >130 mm Hg) and there is suspicion of increased ICP.
They also suggest consideration of modest reduction in BP in patients with SBP >180 mm Hg (or MAP >130 mm Hg) if there is no evidence of increased ICP.
Importantly, these guidelines recommend careful and near-continuous monitoring (every 5–15 minutes) in patients with acute ICH on antihypertensive therapy to assess for worsening of neurological status or drop in cerebral perfusion pressure. Each of these recommendations has a level of evidence C, indicating opinion, as the area lacks well-designed clinical trials.
Given the relatively poor prognosis of patients presenting with acute ICH and the lack of good quality evidence as to how aggressively to treat BP, clinicians caring for patients with acute ICH are often faced with a “damned if you do and damned if you don't” scenario. While there are certainly many barriers to enrolling patients with acute ICH in clinical trials, such clinical uncertainties require prospective, randomized, controlled clinical trials to provide the best evidence to develop clear, well-supported clinical practice guidelines.
Summary of Paper and Comment
On the basis of results of a pilot study (the Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial [INTERACT 1]), investigators developed and executed an international clinical trial, known as INTERACT 2, to determine the safety and efficacy of early, intensive lowering of BP in patients with acute ICH. In INTERACT 2, 2839 patients (mean age 63.5 years) with acute ICH from 21 countries were assigned to receive intensive treatment to lower their SBP to a target level of <140 mm Hg within 1 hour and be maintained for at least 7 days or to a less intensive SBP target of <180 mm Hg, closer to the current standard of care in the United States. There were no US sites included, and the majority (68%) of patients were enrolled at Chinese hospitals. In both groups, the choice of medication was at the discretion of the local health care delivery team. While there were many different agents utilized, the most commonly used intravenous medication in both groups was an α-adrenergic–blocking agent, used in 32.5% of patients in the intensively treated group and 13.4% of patients in the less intensive group.
After 90 days of follow-up, intensive treatment of BP led to no significant improvement in the primary outcome, death or disability. There was also no significant difference on follow-up imaging in hematoma volume and, importantly, no increase in treatment-related adverse events in the intensively managed group. While there was no demonstrated advantage in the primary endpoint, the intensively treated group did show significantly lower Rankin scores, a prespecified secondary endpoint, indicative of better functional status. Subgroup analysis did not demonstrate enhanced clinical benefit or harm in any particular group.
While the results of INTERACT 2 do not support treating patients with acute ICH to a more aggressive BP goal, the lack of harm in the intensively treated group and the improvement in the secondary endpoint of functional outcome suggest enough clinical equipose to study this question further. This is especially true given the fact that the population studied (mostly from China) and the agents used (mostly α-adrenergic–blocking agents) in INTERACT 2 differ from the US population and our current standard of care.
Hopefully, results of the ongoing Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATTACH 2) trial, which employs escalating doses of intravenous nicardipine in the intensively treated group, will further educate us about how best to treat elevated BP in patients with acute ICH. There is certainly nothing in the results from INTERACT 2 which would suggest that ATTACH 2 should not continue. While we eagerly await results of this and future trials in this area, our suspicion is that management of BP, intensive or not, will actually make little difference in the overall poor prognosis of most patients with ICH. Avoiding extremes of BP (ie, >180 mm Hg or <100 mm Hg) seems prudent, but defining an exact optimal target for BP across the entire population of ICH will likely be difficult, especially since BP in these patients often fluctuates markedly over short periods of time owing to the loss of cerebral autoregulation associated with ICH. Determining which clinical characteristics are associated with benefit from antihypertensive therapy will be just as important as determining the optimal BP goal across the spectrum of patients with ICH.
Detailed Review of Methods
INTERACT 2 was an international, multicenter, prospective randomized, open-treatment, blinded end-point (PROBE) trial. The primary inclusion criteria were a diagnosis of acute spontaneous ICH with the onset of symptoms within 6 hours confirmed by neuroimaging and a baseline SBP of 150 mm Hg to 220 mm Hg. Patients were excluded if they had a structural cause for ICH (ie, aneurysm), deep coma at presentation (defined as a score of 3–5 on the Glasgow Coma Scale), massive hematoma with poor prognosis, or if early surgery to evacuate the hematoma was planned. Written informed consent was obtained from each patient or legal surrogate in accordance with each nation's regulations, but in many cases informed consent apparently was not obtained until after randomization, which would have been difficult to accomplish in a US trial.
The randomization schema was designed to ensure balance between the groups with respect to country, hospital, and time between onset of symptoms and randomization (>4 vs <4 hours). In the patients randomized to intensive treatment of BP, the goal was to achieve an SBP <140 mm Hg within 1 hour of randomization and to maintain that level for the next 7 days. Treatment could include both intravenous and oral medications and was initiated according to protocols based on the local availability of antihypertensive agents. In the patients randomized to the less intensive or standard treated group, antihypertensive therapy (either intravenous or oral) was to be administered only if SBP was >180 mm Hg. After 7 days or at discharge, patients in both groups were given an SBP goal of <140 mm Hg for long-term follow-up.
Follow-up was accomplished in person or by telephone at 28 and 90 days by trained staff blinded to treatment allocation. The prespecified primary outcome, calculated according to the intention-to-treat principle, was the proportion of participants with death or major disability (defined as a score of 3–5 on the modified Rankin scale) at 90 days after randomization. The modified Rankin scale is a well-validated measurement of functional ability in neurological conditions with scores ranging from 0 to 6. A score of 0 indicates no symptoms, a score of 5 indicates severe disability such as confinement to bed or incontinence, and a score of 6 indicates death. During the course of the trial, but before results were analyzed, the investigators prespecified a secondary endpoint of functional status. Other secondary outcomes included all-cause mortality, cause-specific mortality, a plethora of quality-of-life scales, duration of hospitalization, residence in a care facility, and hematoma volume on follow-up imaging (when available). The principal safety outcomes of interest included early neurologic deterioration and episodes of severe hypotension requiring intervention.
Detailed Review of Results
From October 2008 through August 2012, a total of 2839 participants were enrolled at 144 hospitals in 21 countries. The mean age of the study population was 63.5 years, 63% were males, and 72% of participants had a known history of hypertension with 44% taking anti-hypertensive medications upon randomization. Sixty-eight percent of participants were recruited from China. Mean baseline BP was 179/101 mm Hg. Baseline characteristics were well balanced between the two groups.
The median time from the onset of ICH symptoms to the initiation of antihypertensive treatment was shorter in the intensive compared with the standard treatment group (4.0 vs 4.5 hour). Antihypertensive medication was initiated within a mean of 6 min of randomization in the intensively treated group and within 19 min in the standard treatment group. In the intensivley treated group, 90% of patients received at least one intravenous antihypertensive medication as compared with 42.9% in the standard group. In the intensively treated group, 26.6% received more than one intravenous antihypertensive. The most common intravenous antihypertensive used (32.5% of the intensive group and 13.4% of the standard group) was an α-adrenergic antagonist such as urapidil, a class of intravenous medications not commonly used in the United States in any setting. Intravenous agents more commonly used in the United States, such as calcium channel blockers (eg, nicardipine, nimodipine), combined α-β–blockers (labetolol), nitrates, and nitroprusside were each used in no more than 16% of the patients in the intensively treated group. At 1 hour after randomization the mean SBP was 150 mm Hg in the intensively treated group as compared to 164 mmHg in the standard treatment group; differences that persisted throughout the 7-day treatment period.
Primary outcome measures were available for more than 98% of patients in both groups. At 90 days, 719 of the participants (52.0%) in the intensive-treatment group compared with 785 (55.6%) in the standard-treatment group met the criteria for the primary outcome of death or major disability (odds ratio [OR] with intensive treatment, 0.87; 95% confidence interval [CI], 0.75 to 1.01; P=.06). Subgroup analysis, while showing a consistency of effect, was nonsignificant in all major subgroups, including those based on age, region, time to randomization, baseline SBP, history of hypertension, baseline stroke severity score, and baseline hematoma volume and location.
In terms of secondary endpoints, patients in the more intensively treated group had significantly more favorable Rankin scores at 90 days, indicative of better functional ability, but the differences were modest in magnitude. Aggressively treated participants also had modestly better quality of life as measured by a number of well validated instruments. There was no significant difference, however, in all cause mortality (11.9% vs 12.0%), percentage of deaths due to ICH (61.4% vs 65.3%), or mean change in hematoma volume in patients randomized to intensive as opposed to standard therapy. Importantly, there was also no difference in neurological deterioration within the first 24 h (14.5% vs 15.1%), non-fatal serious adverse events (23.3% vs 23.6%), severe hypotension (0.5% vs 0.6%), or any other safety endpoints in the intensive as compared to the standard therapy group.
The barriers to enrolling and randomizing hypertensive patients with ICH into a clinical trial are formidable, particularly in the highly regulated environment of clinical trials in the US. Timing is critical, informed consent is difficult to obtain in these situations, and preconceived notions on the part of treating physicians may make them reluctant to allow critically ill patients' treatment allocation to be left to chance. The INTERACT investigators should be congratulated for performing a scientifically rigorous investigation of a very difficult to study clinical condition.
As the study did not meet its primary endpoint, many would see this as a ‘negative trial,’ but in fact its results are critically important. They tell us that while we should not as a general rule be intensively treating BP in patients with acute ICH, if a clinician elects to do so based on individual patient characteristics, there is no reason to suspect an increased risk of harm, and there may in fact be a modest benefit, mainly in functional status. As the ASA guidelines from 2010 state, “until ongoing clinical trials of BP intervention for ICH are completed, physicians must manage BP on the basis of present incomplete evidence (Class IIb; Level of Evidence C),” but in patients presenting with a “systolic BP of 150–220 mm Hg, acute lowering of BP to 140 mm Hg is probably safe (Class IIa; Level of Evidence B).” The results of INTERACT 2 are consistent with that recommendation but do not significantly change the level of evidence. We eagerly await the results of the ATTACH II trial, which hopefully will allow for stronger recommendations with a higher level of evidence in this critically important clinical area of uncertainty.