The most common forms of hypertensive crisis in developed countries are pulmonary oedema/heart failure, acute coronary syndrome, acute intracerebral haemorrhage and acute ischaemic stroke.  Hypertension can be both cause and consequence of stroke and is associated with poor functional outcome and increased mortality.  It does not necessarily follow that lowering of blood pressure acutely will confer benefit, so the question that arises is when and when not to treat?  Current recommendations are to consider patients with acute spontaneous intracerebral haemorrhage with a SBP of 150-220 mmHg for urgent treatment within 6 hours of symptom onset, using a locally agreed protocol for BP lowering, aiming to achieve a systolic BP between 130-139 mmHg within one hour and sustained for at least 7 days.   By contrast the only certainty for BP lowering after acute ischaemic stroke is a BP target <185/110mmHg for thrombolysis.  The many randomised trials in acute ischaemic stroke are inconclusive about stopping or continuing antihypertensive drugs, or specific BP targets, in almost all other settings. 

The management of hypertensive crisis has been the subject of many recent reviews [1-3].  These have tended to draw on recommendations of the AHA [4], ESC [5-6] and NICE [7] and are based on expert opinion rather than randomised treatment trials, given the lack of the latter in this particular field of medicine.  In this review, we will examine in more detail the evidence relating to the management of one particular hypertensive crisis, namely that of hypertension in acute stroke.  To our knowledge, acute stroke is the only hypertensive crisis that has been subjected to randomised trials. Our review will concentrate on three large randomised trials in stroke due to intracerebral haemorrhage (ICH) [8-10], and one recently published large trial in acute ischaemic stroke [11].   For additional analyses of these and several smaller trials, see the reviews by [12-13].

Intracerebral haemorrhage

INTERACT-2 (Investigators in the second Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial) was a randomised controlled trial in 2,839 patients with ICH and elevated systolic BP (SBP) 150–220 mm Hg who were allocated to receive intensive (target SBP <140 P=0.06). P=0.04).  P=0.002)>

INTERACT-3 was a randomized controlled trial in 7036 patients, with ICH and SBP 140mmHg or greater, who were allocated to early intensive BP reduction as part of a bundle of hyperacute care incorporating protocols for hyperglycaemia, pyrexia and abnormal anticoagulation, or to usual care [10].  Target systolic blood pressure in the care bundle group was less than 140 mm Hg within one hour of the initiation of treatment, with a systolic blood pressure of 130 mm Hg being the threshold for the cessation of treatment.  An intravenous agent, commonly urapidil, sodium nitroprusside, labetalol or nicardipine, was prescribed in 78.9% of those allocated to intensive treatment and to 70.9% of controls.  The primary outcome was functional recovery measured by the modified Rankin scale at 6 months. The likelihood of a poor functional outcome was lower in the care bundle group (odds ratio 0·86; 95% CI 0·76 to 0·97; p=0·015) [10].   There was no significant difference in the combined end point of death or disability (modified Rankin scores 3–6) at 6 months (odds ratio 0·89; 95% CI 0·78–1·02; p=0·10).

ATACH-2 (Antihypertensive Treatment of Acute Cerebral Haemorrhage) randomised 1000 patients with acute ICH and hypertension to the same two systolic BP targets as INTERACT-2 (ie, <140>5 ml/hr) who received treatment within 2 hours after symptom onset, intensive BP lowering was associated with improved functional independence (odds ratio 1.98, 95% CI 1.06–3.69; p = 0.031) [11].

In summary, three large randomised trials of antihypertensive therapy in patients with ICH have shown no reduction in a combined end-point of death or disability.  Two reported small but significant improvements in disability and quality of life scores, while the third didn’t.   Although target systolic BP was the same, there were differences in trial design and execution.  INTERACT-2 and ATACH-2 were trials purely of early BP reduction whereas INTERACT-3 tested a package of hyperacute care of which early BP reduction was but one component.  BP lowering had to start within 6 hours of presentation in both INTERACT trials and within 4.5 hours in ATACH-2.  The choice of drug and the route of administration in INTERACT-2 and INTERACT-3 were at the discretion of the trial physician while all patients in ATACH-2 were to receive intravenous nicardipine as first line therapy.   BP reduction was faster and more pronounced in ATACH-2 (average SBP over the first 24 hours was 120-130 mmHg in ATACH-2 compared to 135-145 mmHg in INTERACT-2 and INTERACT-3).  It is conceivable therefore that this degree of reduction may have been excessive in ATACH-2 [15], though it has recently been suggested that a subgroup of ‘fast bleeding’ patients who receive intravenous antihypertensive therapy within two hours may benefit [11].   Importantly, all three trials excluded patients who had an underlying structural cause eg tumour, arteriovenous malformation or aneurysm; or Glasgow Coma Scale below 6; or early neurosurgery to evacuate the haematoma; or a massive haematoma with a poor expected prognosis [8-10].

Acute ischaemic stroke

CATIS-2 (China Antihypertensive Trial in acute Ischaemic Stroke) was a multicentre, randomised open label trial of early versus delayed antihypertensive treatment in 4810 patients with acute ischaemic stroke and elevated SBP 140-220mmHg [11].   Patients were allocated to receive antihypertensive treatment within 24-48 hours of stroke onset (target 10-20% SBP reduction within 24 hours and SBP <140 P=0.08). >

Intracerebral haemorrhage

We have summarised current recommendations for control of hypertension in acute ICH in Table 1 [7, 16-18].  There is a general consensus that patients in whom treatment can be started within 6 hours of symptom onset and whose SBP lies between 150 and 220mmHg, will benefit from blood pressure lowering.  There are slight differences in recommended blood pressure targets (Table 1).

Table 1: Recommendations for BP reduction in acute ICH

Acute ischaemic stroke

CATIS-2 is the latest study to have concluded that early treatment of hypertension in ischaemic stroke does not confer benefit, and may possibly be harmful (CATIS-2). The only certainty for BP lowering after acute ischaemic stroke is a BP target <185>

Schematic cerebral auto regulatory curve showing the relation between cerebral blood flow and mean arterial pressure.   This is shifted to the right in chronic hypertension.  Cerebral blood flow becomes pressure dependent for a time after acute stroke.[21]

There are a few exceptions to the general rule that it is best not to lower BP too quickly in acute ischaemic stroke (apart from the recommendation to lower BP for thrombolysis above).   If BP is >220/120mmHg in the acute phase then it is considered reasonable to lower MAP by no more than 15% over 24 hours to maintain perfusion of the penumbra [18-19], Early treatment of hypertension is also indicated in acute ischaemic stroke if this is associated with one or more hypertensive emergencies, namely hypertensive encephalopathy, hypertensive nephropathy, hypertensive heart failure, aortic dissection or pre-eclampsia/eclampsia [7,19]

Choice of Medication 

Although the AHA [4] and the ESC [5] provide dosing schedules for no fewer than 12 intravenous antihypertensives, we feel it is unlikely that clinicians will have experience of more than a few of these, given that hypertensive emergencies are uncommon and the indications for their intravenous use infrequent [3].  Labetalol, nitroglycerin and nicardipine have all been used intravenously for control of hypertension acutely in intracerebral haemorrhage (Table 2).  Of note, intravenous nicardipine is not stocked in all UK hospitals [3].   There remains uncertainty about the best time to start antihypertensive therapy in ischaemic stroke.  The National Clinical Guideline for Stroke states that antihypertensive medication should generally be initiated prior to the transfer of care out of hospital or at 2 weeks, whichever is the soonest, or at the first clinic visit for people not admitted [7], by which time oral calcium channel blocker, ACE inhibitor or angiotensin receptor blocker would all be acceptable choices.

Current recommendations for the management of hypertension in acute stroke reflect the need for a tailored approach. The evidence in favour of treating hypertension acutely in intracerebral haemorrhage is not as strong as we would wish, but is probably strong enough to consider those with a SBP of 150-220 mmHg for urgent treatment within 6 hours of symptom onset, using a locally agreed protocol for BP lowering, aiming to achieve a systolic BP between 130 139 mmHg within one hour and sustained for at least 7 days [7] (Figure 2). By contrast the only certainty for BP lowering after acute ischaemic stroke is a BP target <185>