Aggressive BP Management in Patients with ICH
- Jun 21st, 2014
- Ben Cooper
Incoming Chief Resident
UT Southwestern / Parkland Memorial Hospital
Case Scenario: A 46-year-old male was brought in by EMS from work for syncope. There were no neurologic deficits reported by EMS, and the patient was hypertensive with SBP in 220s without a history of hypertension. Upon my exam, I found him to be altered, and unable to move his right lower extremity. A stat CT scan revealed a large intraparenchymal hemorrhage. The patient’s head of bed was elevated, and a nicardipine drip started.
Clinical Question: What’s the blood pressure goal? And how quickly should the patient reach that goal?
I remember learning in medical school that there is no value in acutely and aggressively lowering a patient’s BP except in one condition: Aortic Dissection. The traditional target for BP lowering in the ED has been a reduction of 20-25%, otherwise we risk “making the patient symptomatic.” What if the patient already is symptomatic?
The field of neurology has advanced significantly in the past several years regarding BP management in patients with intracerebral hemorrhage (ICH). Several trials have prompted some to add another diagnosis (ICH) to the list of conditions in which aggressive BP reduction is advantageous (I’d also add acute pulmonary edema, but that’s another post, and Swami’s smarter than me). In theory, elevated BP may contribute to hematoma expansion, peri-hematoma edema, and re-bleeding leading to adverse outcomes (1). As of 2008, no such evidence existed to corroborate the theory (2).
In May of 2008, Anderson et al published a pilot study in The Lancet known as the INTERACT trial (Intensive blood pressure reduction in acute cerebral hemorrhage) that convinced many that high BP contributes to expansion of intracerebral hematomas (3). The goal of this study was to be a pilot study proving feasibility and safety of early and aggressive blood pressure reduction, which it accomplished. The investigators also determined the rate of hematoma expansion as their primary efficacy endpoint. Prior studies have demonstrated that hematoma expansion is predictive of morbidity and mortality (4-5), so this seems like a reasonable surrogate to use for a study that is under-powered to detect clinical outcomes. 404 patients with spontaneous ICH from Australia, China, and South Korea were enrolled and randomized to receive guideline-based treatment (SBP goal of 180 based on AHA guidelines in 1999) versus early intensive treatment (SBP goal of 140 within 1 hour of randomization, and 6 hours of symptom onset). The investigators found that the rate of hematoma growth from initial CT to 24-hour CT was slower in the intensive group. However, the two groups had different baseline volumes of hematoma, so the results were not statistically significant after adjusting for baseline volume. More importantly, intensive BP lowering did not alter the risk of adverse events…feasibility established. Not practice-changing in and of itself, but it did set the stage for INTERACTII, which was powered to detect clinical outcomes…kind of.
INTERACTII was published in May of 2013 in NEJM (6). 2,839 participants from 144 hospitals and 21 countries were randomized to receive guideline-based treatment versus intensive treatment as in the prior study. The primary outcome was death or major disability defined as a modified Rankin score of 3 to 6 (for further explanation of this score, click here). Of those receiving intensive treatment, 52% had a primary event (modified Rankin of 3 to 6) vs 55.6% of those receiving guideline-based treatment for an odds ratio of 0.87 (95% CI 0.77 to 1.00, p=0.06). Oh snap! So close. So close, in fact, that the investigators couldn’t help themselves…they had to find another way to analyze the data to yield a positive result (IST-3 style!). They used an ‘ordinal analysis,’ which I’ll spare the details of (mostly because I don’t understand the details), but suffice it to say that the study is a positive one if used. Arguably, they should have specified this analysis in the protocol anyway; or as some have pointed out, used a modified Rankin score of 2 to 6 and then there would have been a positive outcome (7). As the eldest grumpy old man in that great movie Grumpier Old Men states, “you can wish in one hand, and crap in the other, and see which gets filled up first.”
Well, despite this negative result, it’s really hard to look at this study and not believe that there must be a statistically significant effect, although probably a modest one based on the INTERACT-2 results. Of note, there was no increase in serious adverse events for those in the intensive group. And, interestingly, the rate of hematoma expansion was not slower in the intensive group (as observed in the first trial). Also, their use of antihypertensives was all over the place – everything from furosemide to nitroglycerin to nicardipine to hydralazine. Some used mannitol, some didn’t…lots of opportunity for confounding. At our shop, we tend to use nicardipine. It sure would be nice if there were a study that only used nicardipine to reduce the noise…enter ATACH!
The Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH) trial is North America’s version of INTERACT, using only a single agent – nicardipine. This trial should be completed in 2016 (7). The first phase demonstrating safety is already complete (8).
In conclusion, early BP reduction in ICH is safe and probably efficacious, although the data doesn’t definitively say so yet. The AHA recommends “modest” reduction if SBP>180 or MAP>130 to a goal of 160/90 or MAP of 110, but the ATACH results may prompt a change in the guidelines…we shall see (1).