Hypertension (HT) is the most important modifiable risk factor for all types of stroke,[1] although its effect is observed to be stronger in cases of intracerebral hemorrhage
(ICH) compared with that in ischemic stroke.[2] However, the effect of HT on stroke varies based on geographical regions. The INTERSTROKE
study, based on evaluation of 13,000 stroke cases and 13,000 controls in 32 countries,
has shown that the population attributable risk of HT for all types of stroke was
higher in Southeast Asia (59.6%) than that in Western Europe and North America (38.8%)
suggesting that HT is a more important risk factor in Asian countries than in the
West.[2]
Regardless of prior history of HT, approximately 60 to 70% of stroke patients demonstrate
a systolic blood pressure (SBP) > 140 mm Hg on presentation, this proportion is higher
in patients diagnosed with ICH.[3] A population-based study showed that SBP was substantially increased compared with
premorbid levels and fell significantly within 24 hours in patients diagnosed with
ICH. However, patients with ischemic stroke showed only a mild elevation of BP.[4] Among the different subtypes, lacunar stroke was associated with a higher level
of BP than the levels associated with other subtypes. All these findings suggest that
HT significantly affects small perforating vessels leading to lacunar infarcts or
ICH.
It has been observed that an elevated BP may spontaneously decrease over a few days
after presentation.[5]
[6] A persistent elevation of BP is associated with cardiac complications, hemorrhagic
transformation (HTF), hematoma expansion (HE), or enlargement of perihematomal edema—complications
that are associated with poor functional outcomes.[7]
[8]
[9]
[10] Therefore, although a certain degree of BP lowering might be required, the optimal
target level and rapidity of BP reduction has not clearly been identified in patients
with ischemic or hemorrhagic stroke.[11]
[12] We present a review of current literature and summarize the goals of BP management
in patients presenting with acute ischemic stroke and ICH.
Hemodynamic Changes in Acute Stroke
Autoregulation
Cerebral autoregulation (CA) is a mechanism that maintains constant cerebral blood
flow (CBF) regardless of changes in cerebral perfusion pressure (CPP) or mean arterial
pressure (MAP).[13] When CA is intact, a drop in CPP may induce cerebral vasodilation via a decrease
in vascular resistance to maintain constant CBF.[14] Real-time assessment of this dynamic mechanism of CA may require sophisticated tools
for multimodality neuromonitoring.[15] However, in real-world clinical practice, it is difficult to assess the status of
CA in stroke patients without continuous monitoring. Based on studies performed in
patients with severe traumatic brain injury, we extrapolate that CA failure might
contribute to further brain injury in stroke patients and that its severity is directly
proportional to stroke severity.[16] It can be hypothesized that a certain degree of impairment in CA exists in patients
with severe stroke.[17]
When the MAP ranges between 50 and 150 mm Hg, a stable CBF is maintained if CA is
intact.[18]
[19] An increase in MAP > 150 mm Hg may lead to forced dilation of cerebral blood vessels
and cerebral hyperperfusion in the autoregulatory breakthrough zone.[15] On the contrary, a drop in MAP to < 50 mm Hg results in a passive collapse of blood
vessels and leads to ischemia secondary to hypoperfusion. CA is known to be altered/defective
in patients presenting with severe stroke; therefore, an abrupt BP drop in such patients
could lead to a concomitant decrease in CBF, which may predispose them to secondary
cerebral ischemia.
Acute Ischemic Stroke
Mechanisms of Blood Pressure Elevation
Elevated BP is commonly observed in patients presenting with acute ischemic stroke,
especially in a setting of a prior history of HT.[3]
[20]
[21] An exact pathomechanism to explain the BP elevation remains unclear. However, proposed
pathomechanisms are a stress response secondary to sympathetic hyperactivity, disturbed
parasympathetic activity, elevated levels of circulating catecholamines or brain natriuretic
peptide, tissue ischemia, or a combination of the above factors.[3]
[22]
[23]
[24] In patients demonstrating large vessel occlusion, an elevated BP may return to baseline
level following complete recanalization, suggesting that BP elevation is closely linked
to brain tissue ischemia.[25] Therefore, BP elevation by itself might be beneficial in augmenting CBF in the penumbra
in patients with ischemic stroke. Clinical data combined with perfusion imaging have
shown that a relative filling delay decreases as the BP increases, suggesting the
augmentation of cerebral perfusion via leptomeningeal collaterals.[26] To summarize, the BP surge observed in cases of acute ischemic infarction may be
a natural response to tissue ischemia.
American Stroke Society Guideline Suggestion
Guidelines suggest that a level of 220/120 mm Hg, which is the upper limit of BP corresponding
to adequate function of CA, is usually considered an acceptable level of BP elevation.
BP elevation beyond this level might be associated with hyperperfusion syndrome or
HTF in the ischemic tissue. As mentioned earlier, the reference range of intact CA
is a MAP between 50 and 150 mm Hg. The recommended BP level of 220/120 mm Hg can based
off of a MAP of 150 mm Hg.[27]
Points to Consider with Use of Intravenous Thrombolysis
Symptomatic ICH must be recognized as the most serious complication when considering
infusion of recombinant tissue-type plasminogen activator (tPA) ([Fig. 1]). When administration of tPA is considered, BP should be controlled to a level < 185/100 mm
Hg and should be maintained at a level < 180/105 mm Hg even after completion of infusion.[28]
[29] In a pilot dose-finding study performed in patients with ischemic stroke, based
on a previous study examining administration of tPA in patients with myocardial infarction,
the original exclusion criterion of BP elevation was a level of 200/120 mm Hg. However,
three patients demonstrated symptomatic ICH in the high-dose tier (0.95–1.08 mg/kg),[30] and all patients showed BP level > 160/90 mm Hg, while 38% (27 of 71) patients without
ICH showed BP > 160/90 mm Hg. Therefore, the protocol was amended to exclude patients
with MAP > 133 mm Hg, which could be interpreted as a BP reading of 185/110 mm Hg.[30] Subsequent analysis showed that the dose of infused tPA and the level of DBP were
significantly associated with ICH after intravenous (IV) thrombolysis.[31] Following these findings, strict BP control was recommended in patients who were
administered tPA.
Fig. 1 Brain CT in a patient who experienced a significant symptomatic intracerebral hemorrhage
after recombinant tPA. A 68-year-old patient was presented with left side hemiparesis.
NIHSS was 11, and brain CT showed subtle hypoattenuation in the right frontal lobe
(Panel A). tPA was administered 3.5 hours after symptom onset. Blood pressure was
strictly maintained below 180/105 mm Hg. Five hours after tPA infusion, this patient
suddenly became somnolent, and follow-up brain CT showed a symptomatic parenchymal
hematoma with midline shifting (Panel B). CT, computed tomography; NIHSS, National
Institutes of Health Stroke Scale; tPA, tissue-type plasminogen activator.
Some patients may need antihypertensive drugs to maintain BP < 185/110 mm Hg for safe
administration of tPA infusion. The aforementioned study showed that BP reduction
to initiate tPA treatment does not affect adverse outcomes in patients with ischemic
stroke. A substudy from tPA National Institute of Neurological Disorders and Stroke
(NINDS) trial showed that antihypertensive therapy administered prior to tPA was not
associated with differences in early or late outcomes.[32] In addition, another study performed by independent investigators showed that antihypertensive
treatment prior to tPA infusion was not independently associated with poor neurological
outcomes.[33] Among the 427 patients studied, 89 received aggressive BP control prior to thrombolysis,
65 received standard BP control, and 273 required no BP control. Although patients
requiring BP control scored higher on the National Institutes of Health Stroke Scale
(NIHSS), BP reduction was not associated with adverse events after adjusting for baseline
characteristics. Therefore, careful attention to and gentle management of BP are warranted
in patients presenting with stroke who are considered candidates suitable for tPA
treatment.
The Safe Implementation of Thrombolysis in Stroke-International Stroke Thrombolysis
Register (SITS-ISTR) is a prospective multinational register containing data of patients
treated with thrombolysis following an acute ischemic stroke. Secondary analysis of
data from this monitoring registry has shown that persistently high SBP up to 24 hours
after thrombolysis was associated with worse outcomes.[34] Moreover, the relationship between SBP and outcomes showed a U-shaped association
suggesting that a higher proportion of patients with SBP 141 to 150 mm Hg demonstrated
favorable outcomes defined as a modified Rankin scale (mRS) of 0 to 2 at 3 months.[34] Considering that the SITS-ISTR registry is not a randomized trial, these results
are inconclusive, and we can only hypothesize regarding the findings.
The Enhanced Control of Hypertension and Thrombolysis Stroke Study was originally
designed to compare two doses of tPA (0.6 vs. 0.9 mg/kg). In this trial, patients
with elevated SBP (150–220 mm Hg) were randomly assigned to an intensive BP-lowering
group (SBP < 140 mm Hg within 1 hour) and a conventional guideline group (SBP < 180 mm
Hg).[35] Choice of antihypertensive used was at the discretion of the treating physicians.
The primary outcome regarding tPA dose has already been published; however, data from
different BP targets continue and are scheduled to be completed in 2018.[36] Results of this trial may provide an answer regarding the optimal target BP that
needs to be achieved after IV thrombolysis.
Endovascular Thrombectomy
No strong data exist regarding target BP after endovascular thrombectomy (EVT). A
study involving 674 patients who underwent IV or intra-arterial thrombolysis showed
that a J-shaped relationship was identified in terms of the association between BP
over the first 24 hours and functional outcomes in patients who failed recanalization,
whereas a linear correlation was identified in terms of the association between BP
and functional outcome in patients who demonstrated complete recanalization. These
findings suggest that recanalization status should be considered an important modifier
of neurological outcomes.[37]
Intraprocedural BP drop might be detrimental in patients who undergo EVT. Earlier
studies focusing on the type of anesthesia show that general anesthesia (GA) was inferior
to conscious sedation (CS) in terms of neurological outcomes in patients presenting
with acute large vessel occlusion.[38]
[39] However, it is unclear whether anesthesia itself or the ensuing drop in BP was the
cause of this finding; most patients who received GA showed a lower BP during EVT.[40] Low BP during an EVT procedure is considered to be an important contributor to poor
functional outcomes.[40]
[41] A subsequent study comparing GA and CS has since shown that the two anesthesia methods
are equivalent in terms of functional outcomes.[42] That study utilized strict protocols to maintain target SBP between 140 and 150 mm
Hg with variability in BP maintained at < 10 mm Hg.[42] Based on these results, it can be deduced that a very low BP in nonrecanalized patients
might be harmful when undergoing EVT. However, the optimal intraprocedural target
BP level has not yet been established. Expert opinions vary with some authors recommending
a level of approximately 140 to 160 mm Hg and others recommending a level of 120 to
140 mm Hg.[43]
[44]
Moreover, the optimal target BP after complete recanalization has not yet been conclusively
established/defined. Modest BP control might be helpful in reducing reperfusion injury
or HTF. In a retrospective study that included 228 patients who showed complete recanalization
after EVT, patients were categorized into three groups (those without HFT, those with
asymptomatic HTF, and those with symptomatic HTF), and mean SBP was compared between
these groups. It was observed that patients without HFT showed lower SBP (mean 159 mm
Hg) compared with those with asymptomatic HTF (mean SBP 169 mm Hg) and symptomatic
HTF (mean SBP 170 mm Hg) suggesting that lower BP might be associated with a lower
risk for HTF.[45] Another study has shown that the BP level after EVT was one of the factors related
to functional outcomes at 3 months. Patients with moderate BP control (SBP < 160 mm
Hg) showed lower odds for mortality at 3 months than those with permissive HT (SBP < 220
or < 180 mm Hg in cases warranting IV administration of tPA).[46] To summarize, it can be stated that modest BP control might be needed in patients
with complete recanalization following EVT, although the optimal target SBP that needs
to be achieved warrants further investigation for a definitive answer.
Blood Pressure Variability
In addition to absolute BP levels, variability of BP levels may be an independent
predictor for poor clinical outcomes. Marked BP variability measured as significant
differences in successive variation was shown to be associated with poor functional
outcomes (mRS of 3–6), higher rate of mortality, and symptomatic ICH in stroke patients
treated with IV tPA.[47]
[48]
Blood Pressure Control in a Nonthrombolysis Setting
Several observational studies have examined the relationship between BP and outcomes
in nonthrombolysis settings. A post hoc analysis of the International Stroke Trial
reported that SBP measured immediately after randomization was associated with poor
functional outcomes and mortality in 17,398 patients with acute ischemic stroke. The
association showed a U-shaped relationship, and SBP of 140 to 179 mm Hg (a nadir at
∼150 mm Hg) was associated with the best outcome.[49] A U-shaped relationship has also been demonstrated in another independent study.[5] However, these studies did not focus on the effect of treatment with antihypertensive
agents.
Several clinical trials have focused on modulation of BP within several days after
onset of an ischemic stroke.[50]
[51]
[52]
[53]
[54] The Acute Candesartan Cilexetil Therapy in Stroke Survivors (ACCESS) trial was designed
to assess the outcomes of modest BP reduction following administration of candesartan
in 342 patients with elevated BP (defined as BP > 220/110 mm Hg within 24 hours after
admission or > 180/105 mm Hg between 24 and 36 hours after admission). Candesartan
was administered for 7 days after which the BP profile was reviewed. When mean BP
was > 135/85 mm Hg, candesartan was continued.[50] Results showed that early use of candesartan after onset of an ischemic stroke significantly
improved cardiovascular mortality and morbidity. A subsequent study involving a larger
number of patients was performed to validate these results.[51] The Scandinavian Candesartan Acute Stroke Trial was a phase III clinical trial that
included 2,029 patients using study protocols that were very similar to the ACCESS
trial. However, functional outcomes were not different in patients with or without
candesartan treatment. Moreover, patients who were treated with candesartan showed
less favorable outcomes at 6 months compared with those who received a placebo. Interestingly,
15% of the included patients were those diagnosed with a hemorrhagic stroke. However,
subgroup analysis did not show any benefit even in patients with a hemorrhagic stroke.
Moreover, acute BP management using valsartan showed a negative effect on functional
outcomes. The Valsartan Efficacy oN modesT blood pressUre REduction in acute ischemic
stroke trial included 393 patients diagnosed with acute ischemic stroke within 48
hours and showed that early BP lowering was significantly associated with a higher
risk for early neurological deterioration.[55] Based on these findings, it can be stated that if BP reduction is a risk factor
for poor neurological outcomes, discontinuing the use of antihypertensive agents might
be considered in patients with acute ischemic stroke who are currently administered
antihypertensives.
The Continue Or Stop post-Stroke Antihypertensives Collaborative Study (COSSACS) trial
compared the effect of continuing or stopping antihypertensive drugs for 2 weeks on
functional outcomes. The study included 763 patients and continuation of antihypertensives
lowered SBP and DBP by 13 and 8 mm Hg, respectively. However, this trial could not
conclusively demonstrate the differences in outcomes between the groups probably because
of the small number of patients included.[54] The most recent trial that studied the effect of BP lowering was the China Antihypertensive
Trial in Acute Ischemic Stroke (CATIS) trial. This study enrolled 4,071 patients with
acute ischemic stroke within 48 hours and in whom SBP was measured between 140 and
220 mm Hg. Compared with the placebo group, the active BP lowering group (SBP lowering
by 10–25% within a day) did not show benefit in terms of neurological disability at
14 days.[52] Based on these clinical trials, a meta-analysis was performed and concluded that
early management of BP in acute ischemic stroke was not associated with better clinical
outcomes.[56]
Brain Edema and Blood Pressure
Brain edema is one of the major complications known to occur after an acute ischemic
stroke. Significant brain edema leads to brain tissue herniation secondary to an elevation
in intracranial pressure and subsequent poor neurological outcomes. Cytotoxic edema
develops due to energy failure within the first few hours after permanent ischemia
and is followed by the development of ionic edema owing to disruption of the osmotic
pressure gradient. Vasogenic edema begins to develop a few days later due to the breakdown
of the blood–brain barrier (BBB), which is linked to HTF.[57] In brain tissues with BBB breakdown, hyperperfusion or HT is mechanistically associated
with brain edema and HTF.[58] A study has shown that higher SBP was associated with brain edema in patients with
acute ischemic and hemorrhagic stroke, which is in agreement with the afore-mentioned
statement.[59] Moreover, patients with brain edema showed persistently elevated SBP compared with
those without brain edema and usually had a spontaneous decline in SBP over time.[59]
A study has investigated the association between BP variability and brain edema. Among
the included patients, 75.8% had presented with ischemic strokes. Development of brain
edema was significantly associated with stroke severity and BP variability measured
as “time rate” implying the rapidity of SBP change, while the absolute SBP failed
to show a statistically significant relationship with brain edema.[60] A time rate of BP variation of 0.1 mm Hg/min was significantly associated with a
13.9% increased probability of developing brain swelling.
Hemorrhagic Transformation
HTF develops in approximately 10 to 15% of cases of ischemic stroke and is associated
with poor functional outcomes.[61] When HTF produces a mass effect, it leads to symptoms in the patient and is related
with neurological deterioration. A high BP has been recognized as a risk factor for
HTF.[62] Moreover, wide fluctuations in BP plays a role in development of HTF.[63] SBP variability and successive variation of SBP were associated with HTF in patients
who underwent IV thrombolysis.[63]
Hemodynamic Stroke with Low Perfusion
In animals, therapeutic approaches targeting enhancement of collateral circulation
have shown better functional outcomes without significant complications.[64] Those approaches include induced HT using phenylephrine (PE) infusion, intravascular
volume expansion, cerebral arteriolar vasodilation, or CBF version using a head-down
tilt. A study has investigated the effect of norepinephrine (NE) administration in
patients with acute ischemic stroke within 24 hours from symptom onset. These patients
presented with an NIHSS score ≥ 5 on admission, SBP ≤ 140 mm Hg before initiation
of infusion, and without clinical signs of hypovolemia or other treatable causes for
low blood pressure. Continuous infusion of NE was performed in 34 patients, targeting
an SBP > 10% from baseline. After 12 hours of infusion, NE infusion was tapered, and
BP returned to the baseline level.[65] Based on this study, short-term NE infusion was observed to be a feasible treatment
option without a significantly increased risk of complications including ICH or cardiac
arrhythmia. In another study, PE administration was used to increase SBP by 20%. Functional
outcome assessment was not possible because the number of patients included in the
study was very small (n = 13).[66] Another study has investigated the effect of NE-infusion-induced HT wherein surrogate
markers such as mean flow velocity were calculated using transcranial Doppler ultrasound.
It was observed that NE-infusion-induced HT augmented mean Doppler flow velocities
with concomitant elevation of intracranial pressure.[67] However, to date, no high-quality randomized clinical trial has recommended the
utility of catecholamine-induced HT in hemodynamic stroke.
Intracerebral Hemorrhage
Hematoma Expansion in Intracerebral Hemorrhage
HT plays a more significant role in ICH compared with its effect in ischemic stroke.[2] In addition to initial hemorrhage volume, HE is closely related with poor functional
outcomes.[15]
[68] HE is defined as an increased ICH volume by 33% from the baseline and is reported
in approximately up to 38% of patients within the first 24 hours of symptom onset.[69] Considering that the driving force of hemorrhage is hydrostatic pressure, BP reduction
might be mechanistically helpful in reducing the risk of HE. Several nonrandomized
observational studies have shown a possible benefit of BP reduction in patients with
ICH.[70]
[71] The concern associated with disproportionate BP lowering is precipitation of ischemia
in the perihematomal area. However, several neuroimaging studies using positron emission
tomography have shown that global or perihematomal CBF does not change significantly
even with a 15% reduction in MAP, suggesting that CA is relatively intact in patients
with small ICH.[72]
Another study used CT perfusion to assess possible CBF changes in perihematomal tissue
in patients with ICH. The ICH Acutely Decreasing Arterial Pressure Trial included
75 patients and randomized them into two groups (SBP < 150 mm Hg and SBP < 180 mm
Hg). CT perfusion was performed 2 hours after randomization and showed that perihematomal
CBF was lower compared with the contralateral homologous region and that rapid reduction
in BP did not decrease perihematomal CBF, suggesting that there is no penumbra in
the perihematomal area.[73] Several clinical trials have been performed to verify the safety and efficacy of
BP reduction in patients with ICH.[74]
[75]
[76]
[77]
Clinical Trials
The Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH) study was a feasibility
trial using IV nicardipine in 60 patients with acute ICH. The results showed that
acute BP reduction to 110 to 140 mm Hg was relatively safe.[76] Although the neurological deterioration rate was higher in the intensive therapy
group (target SBP of 110–140 mm Hg) than in the group with a target SBP of 140–170 mm
Hg (tier 2) and the group with a target SBP of 170 to 200 mm Hg (tier 1), there was
no temporal relationship observed between the infusion of IV nicardipine and the time
of onset of neurological deterioration.
The Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial (INTERACT)
compared the effect of BP lowering on HE.[75] The 404 patients included in the trial were randomized to an intensive BP lowering
group (target SBP < 140 mm Hg within 6 hours) and a guideline-based BP lowering group
(SBP < 180 mm Hg). Results showed that intensive BP lowering produced a less proportional
increase in hematoma volume. However, this trial was underpowered to show clinical
benefit. The subsequent INTERACT2 trial involving 2,839 patients showed that intensive
lowering of BP did not lead to significant reduction in poor functional outcomes or
mortality, although a secondary outcome parameter using ordinal mRS analysis showed
that intensive treatment was associated with better functional outcomes.[74]
The ATACH-2 trial compared the effect of intensive BP reduction (SBP of 110–139 mm
Hg) and modest BP reduction (SBP of 140–179 mm Hg) on functional outcomes in 1,000
patients with acute ICH.[77] The majority of patients included demonstrated ICH in the basal ganglia and thalamus.
Rapid BP reduction was attempted within 4.5 hours from stroke onset and maintained
over the initial 24 hours of presentation using IV nicardipine infusion. The mean
minimum SBP in the first 2 hours was observed to be 128.9 mm Hg (intensive group)
and 141.1 mm Hg (standard treatment group), respectively. Results of the ATACH-2 trial
show that intensive BP reduction to < 140 mm Hg (median 120 mm Hg) was not superior
to maintaining an SBP of > 140 mm Hg (median 140 mm Hg) in terms of the proportion
of mRS of 4 to 6 noted at 3 months and the rate of HE noted at 24 hours. In addition,
the intensive BP reduction group showed a 2.3-fold higher incidence of adverse renal
events than the standard group. Therefore, SBP reduction to a level < 140 mm Hg is
not recommended because while aggressive BP reduction was not observed to be effective
in improving functional outcomes, it did increase the risk of renal complications.
