Interventions for altering blood pressure in people with acute subarachnoid haemorrhage

Abstract

Background

Subarachnoid haemorrhage has an incidence of up to nine per 100,000 person‐years. It carries a mortality of 30% to 45% and leaves 20% dependent in activities of daily living. The major causes of death or disability after the haemorrhage are delayed cerebral ischaemia and rebleeding. Interventions aimed at lowering blood pressure may reduce the risk of rebleeding, while the induction of hypertension may reduce the risk of delayed cerebral ischaemia. Despite the fact that medical alteration of blood pressure has been clinical practice for more than three decades, no previous systematic reviews have assessed the beneficial and harmful effects of altering blood pressure (induced hypertension or lowered blood pressure) in people with acute subarachnoid haemorrhage.

Objectives

To assess the beneficial and harmful effects of altering arterial blood pressure (induced hypertension or lowered blood pressure) in people with acute subarachnoid haemorrhage.

Search methods

We searched the following from inception to 8 September 2020 (Chinese databases to 27 January 2019): Cochrane Stroke Group Trials register; CENTRAL; MEDLINE; Embase; five other databases, and five trial registries. We screened reference lists of review articles and relevant randomised clinical trials.

Selection criteria

Randomised clinical trials assessing the effects of inducing hypertension or lowering blood pressure in people with acute subarachnoid haemorrhage. We included trials irrespective of publication type, status, date, and language.

Data collection and analysis

Two review authors independently extracted data. We assessed the risk of bias of all included trials to control for the risk of systematic errors. We performed trial sequential analysis to control for the risks of random errors. We also applied GRADE. Our primary outcomes were death from all causes and death or dependency. Our secondary outcomes were serious adverse events, quality of life, rebleeding, delayed cerebral ischaemia, and hydrocephalus. We assessed all outcomes closest to three months' follow‐up (primary point of interest) and maximum follow‐up.

Main results

We included three trials: two trials randomising 61 participants to induced hypertension versus no intervention, and one trial randomising 224 participants to lowered blood pressure versus placebo. All trials were at high risk of bias. The certainty of the evidence was very low for all outcomes.

Induced hypertension versus control

Two trials randomised participants to induced hypertension versus no intervention. Meta‐analysis showed no evidence of a difference between induced hypertension versus no intervention on death from all causes (risk ratio (RR) 1.60, 95% confidence interval (CI) 0.57 to 4.42; P = 0.38; I² = 0%; 2 trials, 61 participants; very low‐certainty evidence). Trial sequential analyses showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more.

Meta‐analysis showed no evidence of a difference between induced hypertension versus no intervention on death or dependency (RR 1.29, 95% CI 0.78 to 2.13; P = 0.33; I² = 0%; 2 trials, 61 participants; very low‐certainty evidence). Trial sequential analyses showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more.

Meta‐analysis showed no evidence of a difference between induced hypertension and control on serious adverse events (RR 2.24, 95% CI 1.01 to 4.99; P = 0.05; I² = 0%; 2 trials, 61 participants; very low‐certainty evidence). Trial sequential analysis showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more.

One trial (41 participants) reported quality of life using the Stroke Specific Quality of Life Scale. The induced hypertension group had a median of 47 points (interquartile range 35 to 55) and the no‐intervention group had a median of 49 points (interquartile range 35 to 55). The certainty of evidence was very low.

One trial (41 participants) reported rebleeding. Fisher's exact test (P = 1.0) showed no evidence of a difference between induced hypertension and no intervention on rebleeding. The certainty of evidence was very low. Trial sequential analysis showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more.

One trial (20 participants) reported delayed cerebral ischaemia. Fisher's exact test (P = 1.0) showed no evidence of a difference between induced hypertension and no intervention on delayed cerebral ischaemia. The certainty of the evidence was very low. Trial sequential analysis showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more.

None of the trials randomising participants to induced hypertension versus no intervention reported on hydrocephalus. No subgroup analyses could be conducted for trials randomising participants to induced hypertension versus no intervention.

Lowered blood pressure versus control

One trial randomised 224 participants to lowered blood pressure versus placebo. The trial only reported on death from all causes. Fisher's exact test (P = 0.058) showed no evidence of a difference between lowered blood pressure versus placebo on death from all causes. The certainty of evidence was very low.

Authors' conclusions

Based on the current evidence, there is a lack of information needed to confirm or reject minimally important intervention effects on patient‐important outcomes for both induced hypertension and lowered blood pressure. There is an urgent need for trials assessing the effects of altering blood pressure in people with acute subarachnoid haemorrhage. Such trials should use the SPIRIT statement for their design and the CONSORT statement for their reporting. Moreover, such trials should use methods allowing for blinded altering of blood pressure and report on patient‐important outcomes such as mortality, rebleeding, delayed cerebral ischaemia, quality of life, hydrocephalus, and serious adverse events.

Plain language summary

Assessment of different treatments for altering blood pressure in people with acute subarachnoid haemorrhage

Background
Every year subarachnoid haemorrhage, a type of stroke caused by bleeding into the space surrounding the brain, affects up to nine people per 100,000 population. Mortality (death) of subarachnoid haemorrhage is 30% to 45%, and 20% of survivors are left dependent in activities of daily living. Currently, short‐term treatment may involve changing blood pressure and either inducing hypertension (that is, increasing blood pressure), or lowering blood pressure; however, the beneficial and harmful effects of blood pressure alteration in people with acute subarachnoid haemorrhage are unclear.

Review question
Our aim was to assess the beneficial and harmful effects of altering blood pressure in people with acute subarachnoid haemorrhage.

Search date
We searched scientific databases from their inception to September 2020.

Study characteristics
We identified three randomised clinical trials (studies where people are randomly put into one of two or more treatment groups) assessing blood pressure alterations in people with acute subarachnoid haemorrhage. Two randomised clinical trials enrolling 61 participants compared induced hypertension with placebo (a pretend treatment) or no treatment. One randomised clinical trial enrolling 224 participants compared lowered blood pressure with placebo. We used trial sequential analysis to estimate the number of participants (amount of information) needed to reject or detect important treatment effects. Trial sequential analysis showed that there was a lack of information for all outcomes. All trials were at high risk of bias and we were very unsure about the results.

Key results and conclusion
Our review shows that there is a lack of information about the important treatment effects of induced hypertension or lowered blood pressure in people with acute subarachnoid haemorrhage. Therefore, the beneficial and harmful effects could not be thoroughly assessed. There is an urgent need for randomised trials assessing the effects of altering blood pressure in people with acute subarachnoid haemorrhage, reporting on patient‐important outcomes such as mortality, serious side effects, rebleeding, delayed cerebral ischaemia (reduced blood flow to the brain), quality of life, and hydrocephalus.

Summary of findings

Summary of findings 1. Induced hypertension versus placebo or no intervention for people with acute subarachnoid haemorrhage.

Induced hypertension versus placebo or no intervention for people with acute subarachnoid haemorrhage
Patient or population: people with acute subarachnoid haemorrhage Settings: any setting Intervention: induced hypertension Comparison: no intervention Follow‐up: 3 months
Outcomes	Illustrative comparative risks (95% CI)		Relative effect (95% CI)	No of participants (trials)	Certainty of the evidence (GRADE)	Comments
	Assumed risk (no intervention)	Corresponding risk (induced hypertension)
Death from all causes (primary outcome)	133 per 1000	225 per 1000	RR 1.60 (0.57 to 4.42)	61 (2 trials)	⊕⊝⊝⊝ Very lowa,b	TSA could not be conducted due to lack of information (only 1.17% of the required information size was reached with 61 participants in the analysis). The trials and outcome were at high risk of bias.
Death or dependency (primary outcome)	366 per 1000	483 per 1000	RR 1.29 (0.78 to 2.13)	61 (2 trials)	⊕⊝⊝⊝ Very lowa,b	TSA could not be conducted due to lack of information (only 4.2% of the required information size was reached with 61 participants in the analysis). The trials and outcome were at high risk of bias.
Proportion of participants with ≥ 1 serious adverse events	200 per 1000	451 per 1000	RR 2.24 (1.01 to 4.99)	61 (2 trials)	⊕⊝⊝⊝ Very lowa,b	TSA could not be conducted due to lack of information (only 2.41% of the required information size was reached with 61 participants in the analysis).
Quality of life	—	—	—	20 (1 trial)	⊕⊝⊝⊝ Very lowa,b	Only 1 study reported on quality of life. Therefore, no meta‐analysis was conducted. TSA showed that we had insufficient information to confirm or reject our predefined mean difference of the standard deviation divided by '2'. Approximately 11.8% of the required information size was reached with 20 participants in the analyses.
Rebleeding	48 per 1000	100 per 1000	—	41 (1 trial)	⊕⊝⊝⊝ Very lowa,b	Only 1 study reported rebleeding. Therefore, no meta‐analysis was conducted. TSA could not be conducted due to lack of information (only 0.48% of the required information size was reached with 41 participants in the analyses).
Delayed cerebral ischaemia	100 per 1000	0 per 1000	RR 0.33 (0.02 to 7.32)	20 (1 trials)	⊕⊝⊝⊝ Very lowa,b	TSA could not be conducted due to lack of information (only 0.23% of the required information size was reached with 20 participants in the analysis).
Hydrocephalus	—	—	—	—	—	No studies reported on hydrocephalus.
CI: confidence interval; RR: risk ratio; TSA: trial sequential analysis.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

^aDowngraded one level due to high risk of bias.
^bDowngraded two levels due to imprecision.

Summary of findings 2. Lowered blood pressure versus placebo for people with subarachnoid haemorrhage.

Lowered blood pressure versus placebo for people with subarachnoid haemorrhage
Patient or population: people with acute subarachnoid haemorrhage Settings: any setting Intervention: lowered blood pressure Comparison: placebo
Outcomes	Illustrative comparative risks (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Assumed risk (placebo)	Corresponding risk (lowered blood pressure)
Death from all causes (primary outcome)	226 per 1000	117 per 1000	RR 0.52 (0.28 to 0.98)	204 (1 trial)	⊕⊝⊝⊝ Very lowa,b	TSA showed that we had insufficient information to confirm or reject our predefined relative risk reduction of ≥ 20% (only 7.3% of the required information size was reached with 204 participants in the analysis).
Death or dependency (primary outcome)	—	—	—	—	—	—
Proportion of participants with ≥ 1 serious adverse events	—	—	—	—	—	—
Quality of life	—	—	—	—	—	—
Rebleeding	—	—	—	—	—	—
Delayed cerebral ischaemia	—	—	—	—	—	—
Hydrocephalus	—	—	—	—	—	—
CI: confidence interval; RR: risk ratio; TSA: trial sequential analysis.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

^aDowngraded one level due to high risk of bias.
^bDowngraded two levels due to imprecision.

Background

Description of the condition

Subarachnoid haemorrhage is an acute condition in which blood extravasates to the subarachnoid space. Subarachnoid haemorrhage is most commonly caused by head trauma, but may also be non‐traumatic (spontaneous) (Marder 2014). Spontaneous subarachnoid haemorrhage may occur due to intracranial aneurysm rupture (85%); non‐aneurysmal perimesencephalic subarachnoid haemorrhage (10%); or, less commonly, arteriovenous malformation, arterial dissection, cerebral venous sinus thrombosis, and vasculitis (Marder 2014; van Gijn 2007).

The estimated worldwide incidence of aneurysmal subarachnoid haemorrhage is six to nine per 100,000 person‐years, but there is regional variability, with markedly higher incidences in Finland and Japan (de Rooij 2007; Ingall 2000; van Gijn 2007). The primary symptom of a spontaneous subarachnoid haemorrhage is a sudden, severe headache; associated symptoms may be nausea, vomiting, photophobia, impaired consciousness, and neck stiffness (Edlow 2000; Suarez 2015). Between 15% and 40% of people with aneurysmal subarachnoid haemorrhage have a thunderclap‐like 'warning headache' that precedes the major haemorrhagic event by hours to days (Dupont 2010). The most common locations of aneurysmal rupture, in order of incidence, are the anterior communicating artery, the posterior communicating artery, the basilar artery, and the middle cerebral artery (Zacharia 2010). Initial diagnostic work‐up consists of a computed tomography (CT) scan of the brain; further investigations may include lumbar puncture and angiography, in order to visualise and characterise possible vascular malformations and aneurysms (Suarez 2015). Risk factors for developing spontaneous subarachnoid haemorrhage are smoking, hypertension, and excessive alcohol consumption (Feigin 2005).

Aneurysmal subarachnoid haemorrhage is associated with high mortality and morbidity rates (Steiner 2013). The case fatality rate is approximately 30% to 45%, and 20% of people are left dependent in activities of daily living (Ingall 2000; Nieuwkamp 2009). Rebleeding and delayed cerebral ischaemia are major causes of death and dependency after the initial haemorrhage (Koenig 2012). Other important complications are hydrocephalus and hyponatraemia (Diringer 2011).

Non‐aneurysmal perimesencephalic subarachnoid haemorrhage, in which there is a spontaneous bleeding in the perimesencephalic and prepontine cisterns without an angiographically confirmed source of bleeding, has an incidence of 0.5 per 100,000 person‐years, and occurs at a mean age of 50 to 55 years (Flaherty 2005). Compared with aneurysmal subarachnoid haemorrhage, the prognosis is excellent, with almost no risk of rebleeding, and only a mild risk of delayed cerebral ischaemia (Steiner 2013).

Traumatic subarachnoid haemorrhage frequently accompanies traumatic brain injury (Armin 2006; Wu 2010). Severity ranges from mild cases with bleeding in one or more sulci of the cerebral convexities to rare, but fatal, non‐aneurysmal basal artery ruptures (Bunai 2000; Takahara 1993). Delayed cerebral ischaemia may also occur in relation to traumatic subarachnoid haemorrhage, although it rarely occurs, has a milder course, and occurs earlier than spontaneous subarachnoid haemorrhage (Armin 2006; Kramer 2013).

Delayed cerebral ischaemia occurs in approximately 30% to 40% of people with subarachnoid haemorrhage, primarily in those with an aneurysmal aetiology (Budohoski 2014; Koenig 2012). Delayed cerebral ischaemia usually develops three to 14 days after the aneurysmal subarachnoid haemorrhage, despite preventive treatment with the calcium‐channel antagonist nimodipine, and avoidance of hypovolaemia (Connolly 2012; Vergouwen 2010). Typical clinical features of delayed cerebral ischaemia are focal or global neurological deficits, or decreases in level of consciousness unrelated to the initial subarachnoid haemorrhage. Delayed cerebral ischaemia may either reverse or progress to cerebral infarction (Vergouwen 2010). Rebleeding affects 4% to 17% of people with an aneurysmal subarachnoid haemorrhage and up to 87% of rebleedings occur within the first six hours of the initial haemorrhage (Connolly 2012; Fujii 1996; Oheda 2015; Ohkuma 2001; Starke 2011; Tang 2014). Systolic blood pressure greater than 160 mmHg is associated with an increased risk of rebleeding (Ohkuma 2001; Tang 2014). Early surgical intervention with neurosurgical clipping or endovascular coiling, aimed at securing the source of bleeding, is the mainstay in the prevention of rebleeding (Starke 2011; Steiner 2013). The use of antifibrinolytic treatment (e.g. tranexamic acid) for aneurysmal subarachnoid haemorrhage appears to decrease the risk of rebleeding, but might increase the risk of delayed cerebral ischaemia. However, antifibrinolytic therapy does not appear to affect clinical outcomes (Baharoglu 2013).

Description of the intervention

Management of arterial blood pressure in subarachnoid haemorrhage may be aimed at either deliberately lowering blood pressure below the patient's initial blood pressure or inducing hypertension (that is, deliberately increasing blood pressure above the patient's initial blood pressure), depending on the situation.

Some clinicians suggest that interventions aimed at lowering blood pressure may reduce the risk of rebleeding in people with hypertension with aneurysmal subarachnoid haemorrhage, before clipping or coiling (Connolly 2012; Steiner 2013). Calcium‐channel antagonists (e.g. nimodipine, nicardipine) and beta‐adrenergic receptor antagonists (e.g. labetalol) are medications commonly used to induce hypotension (Connolly 2012; Steiner 2013).

For decades, interventions aimed at inducing hypertension have been used alone, or in combination with hypervolaemia and haemodilution to treat delayed cerebral ischaemia, following coiling or clipping (Diringer 2011; Kosnik 1976). Triple‐H therapy is a combination of induced hypertension, hypervolaemia, and haemodilution, previously used in subarachnoid haemorrhage to reduce the risk of delayed cerebral ischaemia. More recently, there has been a shift in management towards induced hypertension and euvolaemia, but there is sparse evidence supporting either management strategy (Findlay 2016; Francoeur 2016; Sen 2003; Veldeman 2016). The most commonly used treatments for induced hypertension are sympathomimetic drugs, such as noradrenaline and phenylephrine (Koenig 2012; Raya 2014).

How the intervention might work

High systolic blood pressure is associated with an increased risk of rebleeding in unsecured aneurysmal subarachnoid haemorrhage (Alfotih 2014; Tang 2014). It has been suggested that antihypertensive treatment may decrease the risk of rebleeding, possibly by decreasing the pressure against the weakened wall of the aneurysm (Wijdicks 1990). People with other types of subarachnoid haemorrhage might also benefit from lowering blood pressure by an overall decrease in arterial pressure.

The pathophysiology of delayed cerebral ischaemia following subarachnoid haemorrhage is thought to be a combination of loss of autoregulation, formation of microthrombi, microcirculatory constriction, and the breakdown of extravasated blood, with the release of reactive oxygen species, leading to inflammation and vasospasm, which results in reduced cerebral blood flow, leading to ischaemia (Brathwaite 2014; de Rooij 2013; Macdonald 2014). Since autoregulation is frequently disturbed in subarachnoid haemorrhage, inducing hypertension with the aim of increasing cerebral blood flow may ensure sufficient cerebral perfusion, which in turn, might minimise cerebral parenchymal ischaemia and thereby lower the risk of progression to infarction (Koenig 2012).

Alteration of blood pressure in people with subarachnoid haemorrhage may be associated with adverse effects. Lowering blood pressure in order to prevent rebleeding could potentially increase the risk of delayed cerebral ischaemia (Wijdicks 1990), while induced hypertension for delayed cerebral ischaemia may carry an increased risk of cardiopulmonary complications (e.g. congestive heart failure and pulmonary oedema) and cerebral complications (e.g. haemorrhagic transformation of infarctions, hypertensive encephalopathy, and cerebral oedema) (Amin‐Hanjani 1999; Diringer 2011; Otsubo 1990; Steiner 2013).

Why it is important to do this review

The current evidence on the effects of altering arterial blood pressure in people with subarachnoid haemorrhage is unclear. Our preliminary search identified few randomised clinical trials on this topic, despite the fact that medical alteration of blood pressure has been clinical practice for more than three decades (Connolly 2012; Diringer 2011; Kosnik 1976; Steiner 2013). International guidelines from the American Heart Association/American Stroke Association (AHA/ASA), the European Stroke Organization (ESO), and the Neurocritical Care Society (NSC) recommend inducing hypotension when systolic blood pressure exceeds 160 mmHg (Connolly 2012; Diringer 2011), or 180 mmHg (Steiner 2013), before occluding the aneurysm. These guidelines either recommend inducing hypertension in people with delayed cerebral ischaemia (Connolly 2012; Diringer 2011), or refrain from making such a recommendation due to lack of evidence (Steiner 2013). These guidelines are based on observational studies, case reports, and expert opinion.

One systematic review found no beneficial effect of induced hypertension in preventing or treating delayed cerebral ischaemia following aneurysmal subarachnoid haemorrhage (Veldeman 2016). However, the review did not search all relevant databases, and did not take the risks of systematic errors and random errors into account (Higgins 2021; Jakobsen 2014). One previous Cochrane Review found no evidence supporting the use of volume expansion (Rinkel 2004). In one systematic review, two of four uncontrolled studies found a significant increase in cerebral blood flow following induced hypertension (Dankbaar 2010). However, one small randomised clinical trial found no significant effect of induced hypertension on overall cerebral blood flow in people with subarachnoid haemorrhage when compared to controls with normal blood pressure (Gathier 2018).

One trial has indicated that nimodipine in aneurysmal subarachnoid haemorrhage appears to decrease the risk of poor outcome, rebleeding, and delayed cerebral ischaemia, but not death (Mees 2008). Small pilot trials on preventing delayed cerebral ischaemia found significantly worse neuropsychological scores upon follow‐up of participants allocated to blood pressure augmentation (Togashi 2015), and no difference between participants randomised to dobutamine or noradrenaline (Rondeau 2012).

Although a different type of intracranial haemorrhage, and, therefore, not directly comparable, inducing hypotension in people with acute spontaneous intracerebral haemorrhage was found to be safe in randomised and non‐randomised trials (Anderson 2008; Anderson 2013; ATACH 2010; Koch 2008). One previous Cochrane Review concluded that there is a lack of evidence in favour of lowering blood pressure in people with acute ischaemic or haemorrhagic stroke (Bath 2014). The large randomised clinical trial, Anderson 2013, found no difference in the primary composite outcome of death or disability, between participants with intracerebral haemorrhage randomised to a systolic blood pressure target of less than 140 mmHg within one hour, or a conventional target of less than 180 mmHg. However, the secondary outcomes of disability and quality of life were significantly improved in participants who received intensive blood pressure‐lowering therapy (Anderson 2013). In the ATACH‐2 (Antihypertensive Treatment of Acute Cerebral Hemorrhage II) trial, there were no significant differences in rates of death or disability between participants randomised to a systolic blood pressure target of 110 mmHg to 139 mmHg within 4.5 hours using intravenous nicardipine versus a target of 140 mmHg to 179 mmHg (Qureshi 2016 (ATACH‐2)).

Considering the variations in the use, results from trials and reviews, and the recommendations to deliberately alter blood pressure in the management of subarachnoid haemorrhage, it is of great importance that any potential beneficial and harmful effects are well established. Thus, we consider it relevant to conduct this Cochrane Review, taking into account both risks of random errors ('play of chance') and risk of systematic errors ('bias') (Higgins 2021; Jakobsen 2014).

Objectives

To assess the beneficial and harmful effects of altering arterial blood pressure (induced hypertension or lowered blood pressure) in people with acute subarachnoid haemorrhage.

Methods

Criteria for considering studies for this review

Types of studies

We included all parallel‐arm, cross‐over, and cluster randomised clinical trials, regardless of publication status, type, setting, or language. We excluded quasi‐randomised studies and observational studies. However, during the literature search, if we identified observational studies reporting on long‐term adverse events or rare adverse events, we planned to provide a narrative report of these results.

Types of participants

We included participants of any age and sex with subarachnoid haemorrhage (defined by trialists).

Types of interventions

We included any experimental intervention, aimed at inducing hypertension or lowering blood pressure, by either pharmacological or non‐pharmacological means. For pharmaceutical interventions, we accepted any dose, frequency, and formulation, used as either monotherapy or in combination. For non‐pharmaceutical interventions, we were not aware of any such interventions a priori, but we accepted any type and frequency of intervention. As control interventions, we accepted placebo or no intervention. We allowed co‐interventions, if they were planned to be delivered equally to both the experimental and control group.

Types of outcome measures

We assessed two primary outcomes and five secondary outcomes. We primarily assessed the outcomes at three months after subarachnoid haemorrhage (we used the trials' assessment time point closest to three months after subarachnoid haemorrhage) and at maximum follow‐up.

Primary outcomes

• Death from all causes.

• Death (from all causes) or dependency (defined by trialists; measured on the modified Rankin Scale (mRS), or the extended Glasgow Outcome Scale (eGOS): we planned to perform a subgroup analysis of the various methods for measuring dependency as defined by trialists.

Secondary outcomes

• Proportion of participants with one or more serious adverse events. We defined serious adverse events as any untoward medical occurrence that resulted in death, was life‐threatening, was persistent, required hospitalisation, led to significant disability, or jeopardised the patient (ICH‐GCP 1997).

• Quality of life (any valid continuous scale used by trialists).

• Rebleeding (defined by trialists).

• Delayed cerebral ischaemia (defined by trialists).

• Hydrocephalus (defined by trialists).

Search methods for identification of studies

We searched for trials in all languages and, if necessary, arranged for the translation of relevant articles.

Electronic searches

We searched the following electronic databases and trial registry databases:

• Cochrane Stroke Group Register (from inception to 8 September 2020);

• Cochrane Central Register of Controlled Trials (CENTRAL; 2020, Issue 9) in the Cochrane Library from inception to 8 September 2020; Appendix 1);

• MEDLINE Ovid (from 1966 to 8 September 2020; Appendix 2);

• Embase Ovid (from 1980 to 8 September 2020; Appendix 3);

• Science Citation Index Expanded (ISI Web of Knowledge; from 1945 to 8 September 2020; Appendix 4);

• Latin‐American and Caribbean Health Sciences Literature (LILACS; from 1982 to 8 September 2020; Appendix 5);

• Chinese Biomedical Literature Database (CBM; from 1978 to 27 January 2019;  Appendix 6);

• China National Knowledge Infrastructure (CNKI; from 1979 to 27 January 2019; Appendix 7);

• Chinese Science Journal Database (VIP; from 1989 to 27 January 2019; Appendix 8).

We agreed not to update the search of the Chinese databases because we did not identify any relevant literature during the first search.

The Cochrane Stroke Group Information Specialist developed the search strategies (except for LILACS).

We used Google Scholar to forward search included articles to identify other relevant literature.

In order to identify ongoing or unpublished trials, we searched the following electronic trial registry databases:

• US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov/; updated 8 September 2020; Appendix 9);

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch; updated 8 September 2020; Appendix 10);

• EU Clinical Trials Register (www.clinicaltrialsregister.eu/; updated 8 September 2020; included in the WHO ICTRP);

• ISRCTN Registry (www.isrctn.com/; updated 8 September 2020; included in the WHO ICTRP);

• the Stroke Trials Registry (www.strokecenter.org/trials; updated 8 September 2018; no longer online).

Searching other resources

We handsearched reference lists of studies and reviews on the topic of therapeutic alteration of blood pressure in subarachnoid haemorrhage.

Data collection and analysis

We conducted data collection and analysis according to the recommendations in the Cochrane Handbook for Systematic Reviews of Interventions Chapters 5, 6, 8, 9, 10, 14, and 15 (Higgins 2021), and as described by Jakobsen and colleagues (Jakobsen 2014).

Selection of studies

Two review authors (MM and WKK) independently screened the title and abstract of all records identified by the search and excluded records that were obviously irrelevant. We obtained full‐text articles for all potentially eligible studies, and independently selected eligible studies according to the inclusion and exclusion criteria. We reported the reasons for trial exclusion in the Characteristics of excluded studies table. We resolved all disagreements by discussion; if required, we consulted a third review author (JCJ). We collated multiple reports of the same study so that each study, not each reference, was the unit of interest in the review. We recorded the selection process and presented a PRISMA flow chart to visually display the selection of studies (Moher 2009).

Data extraction and management

Two review authors (MM and WKK) independently extracted data from included trials using a data collection form. For each included trial, we extracted: year of publication, first author, country, publication status, number of participants in each group, number of participants lost to follow‐up in each treatment group, relevant baseline characteristics (e.g. age range and mean, sex ratio, comorbidities), any relevant scoring (e.g. Glasgow Coma Score (GCS) at admission, World Federation of Neurosurgical Societies (WFNS) score, or Hunt and Hess scale), method of diagnosing subarachnoid haemorrhage, type of subarachnoid haemorrhage (aneurysmal, non‐aneurysmal, traumatic), type and timing of aneurysm treatment, active intervention (generic drug name, timing, route of administration, dose), comparator and co‐interventions (e.g. haemostatic treatment), duration of intervention, follow‐up duration, outcomes, time to rebleeding or delayed cerebral ischaemia, definition of delayed cerebral ischaemia, presence of intraventricular haematoma, presence of parenchymatous haemorrhage (intracerebral haemorrhage), type of non‐pharmaceutical intervention, frequency of non‐pharmaceutical intervention, length of treatment with non‐pharmaceutical intervention, other medical complications, and source of funding. We cross‐checked the completed data collection forms for each individual trial. We resolved all disagreements by discussion; if required, we consulted a third review author (JCJ). We entered data into Review Manager 5 (Review Manager 2014).

Assessment of risk of bias in included studies

We assessed the risk of bias in included studies as described in the Cochrane Handbook for Systematic Reviews of Interventions using the Risk of Bias 1 tool (Higgins 2011a; Higgins 2011b). Two review authors (MM and WKK) independently assessed the included trials; they resolved all disagreements by discussion, or if required, by consulting a third review author (JCJ). We evaluated the risk of bias in random sequence generation, allocation sequence concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other bias (for‐profit bias for each trial) (see Appendix 11).

We classified each risk of bias item as low, unclear, or high. A trial was at overall low risk of bias if all risk of bias items were at low risk of bias. A trial was at overall high risk of bias if one or more risk of bias items were at either unclear or high. We based our primary conclusions on the results of our primary outcomes at low risk of bias. We presented the results of predefined outcomes in the summary of findings tables. We reported bias assessments for each trial in risk of bias tables. We assessed the items 'blinding of outcome assessment', 'incomplete outcome data', and 'selective outcome reporting' for each outcome result. Thus, we were able to assess the bias risk for each outcome result, in addition to each trial.

Measures of treatment effect

We presented dichotomous outcomes as risk ratios (RRs) with 95% confidence intervals (CIs) and corresponding P values. We reported the number needed to treat for an additional beneficial outcome (NNTB) and the number needed to treat for an additional harmful outcome (NNTH).

We presented continuous outcomes as the mean differences (MDs), with 95% CIs, and corresponding P values. If studies reported both end scores and change from baseline scores, we used end scores. If studies reported only change values, we analysed the results together with end scores (Higgins 2021). We also considered pooling similar continuous outcome scales, if we identified such results, using standard mean differences (SMDs), with 95% CIs, and corresponding P values.

Unit of analysis issues

For trials using a crossover design, we only planned to include data from the first period. For trials using multiple comparisons, we planned to split the control group between the comparisons to avoid double‐counting of participants.  We did not anticipate identifying any trials using a cluster‐randomised design. However, we intended to include the data after correcting the actual sample size to the effective sample size using an intracluster correlation coefficient. 

Dealing with missing data

We contacted trial investigators if data were missing or unclear. We did not use intention‐to‐treat data if the original report did not contain such data. We did not impute missing values for any outcomes in our primary analysis. In our sensitivity analysis for dichotomous and continuous outcomes, we imputed data (see Sensitivity analysis). If standard deviations (SD) were not reported in a trial, we calculated the SDs using data from the trial, if possible.

Assessment of heterogeneity

We mainly used I² statistics to quantify heterogeneity. Interpretation of the I² statistics depends on the value (Higgins 2021):

• 0% to 40% might not be important;

• 30% to 60% may indicate moderate heterogeneity;

• 50% to 90% may indicate substantial heterogeneity;

• 75% to 100% indicates considerable heterogeneity.

In case of I² values that fall in the overlaps (e.g. 30% to 40%, 50% to 60%, or 75% to 90%), we classified the heterogeneity as, for example, 'might not be important to moderate heterogeneity', 'moderate to substantial heterogeneity', or 'substantial to considerable heterogeneity'.

Statistical tests (e.g. I² statistic) for heterogeneity have two problems: 1. there might not be enough power in the analysis to show even considerable heterogeneity, and 2. there might be 'too much data', so that small and clinically insignificant heterogeneity result in a large I² statistic estimate. This is especially true when analysing continuous outcomes. Therefore, in addition to calculating I² statistics, we planned to visually inspected forest plots to look for signs of study heterogeneity of the included trials if at least 10 trials had been included in an analysis (Jakobsen 2014). Quote: "a forest plot is an essential tool to summarise information on individual studies, give a visual suggestion of the amount of study heterogeneity, and show the estimated common effect, all in one figure" (Fagerland 2015).

We also investigated possible heterogeneity through subgroup analyses (see Subgroup analysis and investigation of heterogeneity). Ultimately, we may have decided that a meta‐analysis (e.g. pooling all types of subarachnoid haemorrhage in one analysis) should be avoided, for instance in the event of statistical heterogeneity due to the trials delivering the intervention in unexpected different ways (Higgins 2021; Jakobsen 2014).

Assessment of reporting biases

We assessed publication bias and other reporting biases by visual inspection of funnel plots for primary outcomes, if at least 10 trials were included for an outcome (Higgins 2021). We assessed the risk of bias using the asymmetry of the funnel plot.

For dichotomous outcomes, we tested asymmetry with the Harbord test, if Tau² was less than 0.1 (Harbord 2006), and with the Rücker test, if Tau² was more than 0.1 (Rücker 2008).

For continuous outcomes, we used the regression asymmetry test (Egger 1997).

Data synthesis

We separately analysed trials that assess interventions aimed at: lowering blood pressure; and increasing blood pressure.

For trials concerning delayed cerebral ischaemia, we separately analysed trials concerning: prevention of delayed cerebral ischaemia; and treatment of delayed cerebral ischaemia.

For statistical analysis, we used Review Manager 5 (Review Manager 2014), trial sequential analysis (TSA) software, provided by the Copenhagen Trial Unit (CTU; TSA 2011), and if necessary, STATA 14 (Stata).

Assessment of statistical and clinical significance

We analysed all primary and secondary outcomes using both a fixed‐effect (DeMets 1987), and random‐effects model meta‐analysis (DerSimonian 1986). It is often likely that a given intervention will have different effects across the included trials depending on different forms of the interventions, different definitions of the outcomes, different types of included participants, etc. The random‐effects model assumption will, therefore, often be more realistic than the fixed‐effect model assumption. If there is absence of statistical heterogeneity, the between‐trial variance of the estimated intervention effects is close to zero, then the fixed‐effect and the random‐effects models will show identical results. If there is substantial statistical heterogeneity, the fixed‐effect meta‐analysis will, in some circumstances, show erroneous results because the between‐trial variance is not appropriately accounted for. In such a case, the random‐effects meta‐analysis result should be regarded as the main result. In contrast, if one or two trials account for approximately 80% or more of the total weight in a fixed‐effect meta‐analysis, then the random‐effects meta‐analysis might show erroneous results because the larger trials with the greatest precision are inappropriately down‐weighted. In such a case, the fixed‐effect meta‐analysis result should be regarded as the main result. We recommend reporting results from both fixed‐effect and random‐effects meta‐analyses. If the fixed‐effect and the random‐effects meta‐analyses show different results, then the most conservative result (the analysis with the highest P value) should be chosen as the main result. Choosing the most conservative result will take account of the mentioned pitfalls of the two analyses (Jakobsen 2014).

With each additional outcome assessed in a meta‐analysis, the risk of falsely rejecting the null hypothesis increases. The Bonferroni adjustment accounts for the problems associated with multiplicity, but may often be too conservative, since the assumption of no correlation between outcomes may not be suitable (Jakobsen 2014; Jakobsen 2016). We calculated a multiplicity‐adjusted significance threshold by taking a pragmatic midway between the Bonferroni adjustment and unadjusted P‐value (commonly 0.05). We calculated the multiplicity‐adjusted significance threshold as the prespecified P‐value threshold divided by the value halfway between 1 (no adjustment) and the number of outcome comparisons (Bonferroni adjustment) (Jakobsen 2014; Jakobsen 2016). By these calculations, we considered a P value of 0.0333 or less as statistically significant for primary outcomes, and a P value of 0.0167 or less for secondary outcomes.

Trial sequential analysis

TSA is a method to minimise risk of random errors resulting from sparse data and repetitive testing for significance in cumulative meta‐analyses (Brok 2008; Brok 2009; Jakobsen 2014; Thorlund 2009; Wetterslev 2008). We used TSA in addition to the methodology currently recommended by Cochrane (Higgins 2021; Jakobsen 2014).

TSA estimates the a priori required number of participants for a meta‐analysis to detect or reject a given intervention effect (the required information size); in addition, trial sequential monitoring boundaries are used to show if the information size adjusted boundaries for significance are crossed. The diversity‐adjusted required information size depends on the anticipated intervention effect, variance of the anticipated difference in intervention effect (e.g. proportion of control participants with an event for dichotomous outcomes and a SD for continuous outcomes), variance of the intervention estimate between trials, and the predefined acceptable risk of statistical error (type I and II errors; Jakobsen 2014). Trial sequential monitoring boundaries can be constructed, depending on the required information size. If the trial sequential monitoring boundaries are crossed by the cumulative Z‐curve before the required information size has been reached, this strongly indicates either a beneficial or harmful intervention effect, in which case, additional trials may be redundant. Conversely, if the monitoring boundaries are not crossed, this indicates that more trials are needed in order to confirm or reject an intervention effect. Monitoring boundaries for futility may be also be constructed which, if crossed by the cumulative Z‐curve, will indicate the absence of the anticipated intervention effect (Thorlund 2011).

If one trial or more has been published, we analysed all primary and secondary outcomes by TSA. We added trial data to the analysis depending on year of publication. If more than one trial had been published in the same year, we added them alphabetically, by the surname of the first author (Wetterslev 2008). We performed the TSA using both a fixed‐effect and a random‐effects model.

We estimated the diversity‐adjusted required information size, based on the anticipated intervention effect (former studies, trials, and meta‐analyses), the proportion of participants with an outcome in the control group for dichotomous outcomes or SD for continuous outcomes, and the variance of the intervention effect between trials, as suggested by the trials in the meta‐analysis. For outcomes with no relevant studies, trials, or meta‐analyses, the anticipated intervention effect would be the observed SD and an anticipated MD of SDs divided by two for continuous outcomes, and the proportion of outcomes in the control group and a relative risk reduction of 20% for dichotomous outcomes (Appendix 12). For primary outcomes, we set alpha to 3.33% and beta to 20%. For secondary outcomes, we set alpha to 1.67% and beta to 20% (Jakobsen 2014).

Subgroup analysis and investigation of heterogeneity

We planned to conduct the following subgroup analyses for the assessment of the primary outcomes.

• Type of subarachnoid haemorrhage:

  • aneurysmal;

  • non‐aneurysmal aetiologies (e.g. traumatic, perimesencephalic, etc.).

• Type of intervention.

• Haemostatic treatment (e.g. tranexamic acid) as co‐intervention.

• Haemostatic treatment (e.g. tranexamic acid) as co‐intervention, with or without delayed cerebral ischaemia‐preventive treatment.

• Dose of any pharmaceutical intervention.

For trials concerning aneurysmal subarachnoid haemorrhage, we also planned to perform the following subgroup analyses.

• Severity of subarachnoidal haemorrhage according to clinical scales (e.g. GCS at admission, WFNS score, and Hunt and Hess scale).

• Type of definitive treatment of the aneurysm:

  • surgical clipping;

  • endovascular treatment;

  • no treatment.

For trials concerning traumatic subarachnoid haemorrhage, we also planned to perform the following subgroup analyses.

• Severity of traumatic brain injury upon admission according to GCS.

• Severity of traumatic brain injury upon admission according to neuroradiological scales (e.g. Marshall classification of traumatic brain injury and Rotterdam CT score of traumatic brain injury).

For trials concerning delayed cerebral ischaemia, we also planned to perform the following subgroup analyses.

• Risk of developing delayed cerebral ischaemia according to neuroradiological scales (e.g. modified Fisher scale).

Sensitivity analysis

To assess the potential impact of bias, we planned to perform a sensitivity analysis where we excluded trials with an overall high risk of bias.

To assess the potential impact of missing data for dichotomous outcomes, we planned to perform the following two analyses on primary outcomes.

• Best‐case, worst‐case scenario: we assumed that all participants lost to follow‐up in the intervention group survived, and had no serious adverse events, and all those with missing outcomes in the control group, did not survive and had a serious adverse event.

• Worst‐case, best‐case scenario: we assumed that all participants lost to follow‐up in the intervention group did not survive, and had a serious adverse event, and all those with missing outcomes in the control group survived, and had no serious adverse events.

We reported results from both scenarios.

When analysing continuous outcomes regarding missing data, a 'beneficial outcome' were the group mean plus two SDs of the group mean, and a 'harmful outcome' were the group mean minus two SDs of the group mean (Jakobsen 2014).

Summary of findings and assessment of the certainty of the evidence

We constructed summary of findings tables presenting all primary (death from all causes, and death or dependency) and secondary (proportion of participants with one or more serious adverse events, quality of life, rebleeding, delayed cerebral ischaemia, and hydrocephalus) outcomes. To assess the certainty of the evidence, we used the GRADE approach, which involves assessing the within‐study risk of bias, directness of evidence, heterogeneity of results, imprecision, and risk of publication bias (Guyatt 2008; Higgins 2021). We used the methods and recommendations described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021). We based the summary of findings tables on all prestated outcomes, and conclusions in part on the results of 'overall low risk of bias' trials. We justified all decisions to downgrade the certainty of evidence in footnotes, and we made comments to aid the reader's understanding of the review where necessary.

Results

Description of studies

We assessed included trials following the recommendations of the Cochrane Handbook of Systematic Review of Interventions (Higgins 2021), and according to the prepublished protocol (Maagaard 2018).

Results of the search

Our search yielded 28,060 records through searching Cochrane Stroke Register (231 records), CENTRAL (3205), MEDLINE (3019), Embase (16,938), Web of Science (2768), LILACS (73), CBM (703), CNKI (569), VIP (201), ClinicalTrials.gov (243), and WHO ICTRP (110). Of these, we excluded 2729 duplicates and 25,262 records based on title and abstract. We obtained 71 full‐text records for possible inclusion. Of these, we included three trials and there were five reports of included trials (see Characteristics of included studies table). We excluded 62 studies and there was one report of an excluded study (see Characteristics of excluded studies table). We found no ongoing trials. See Figure 1 for PRISMA flow chart.

1.

PRISMA flow diagram.

Included studies

We included three trials (Gathier 2018; Neil‐Dwyer 1985; Togashi 2015). Two trials randomised 61 participants with subarachnoid haemorrhage to induced hypertension versus no induced hypertension (Gathier 2018; Togashi 2015). One trial randomised 224 participants with subarachnoid haemorrhage to lowered blood pressure versus placebo (Neil‐Dwyer 1985). The trials were published between 1985 and 2018. None of the trials appeared to be sponsored by the industry, but Neil‐Dwyer 1985 did not mention conflicts of interest or describe how the trial was funded. All trials were at high risk of bias (Gathier 2018; Neil‐Dwyer 1985; Togashi 2015). See Characteristics of included studies table for a more detailed description of the included studies.

Induced hypertension versus no induced hypertension

One trial randomising 41 participants used fluids in case of hypovolaemia as well as various vasopressors to induce hypertension in the experimental group (21 participants). The goal was a maximum mean arterial blood pressure of 130 mmHg or a maximum systolic blood pressure of 230 mmHg. In the control group (20 participants), hypotension (mean arterial blood pressure below 80 mmHg) was avoided using low‐dose vasopressors. This trial was prematurely terminated due to lack of effect on overall cerebral perfusion and slow recruitment (Gathier 2018).

Another trial randomising 20 participants used a combination of normovolaemia, hypervolaemia, and vasopressors to induced hypertension in the experimental group (10 participants). The control group aimed for conventional blood pressure (10 participants) (Togashi 2015).

Lowered blood pressure versus no lowered blood pressure

One trial randomising 224 participants used beta‐blockers to lower blood pressure in the experimental group (111 participants; phentolamine plus propranolol or propranolol alone) and placebo in the control group (93 participants) (Neil‐Dwyer 1985).

Excluded studies

We excluded 62 studies; reasons are provided in the Characteristics of excluded studies table.

Studies awaiting classification

There are no studies awaiting classification.

Ongoing studies

We found no ongoing studies.

Risk of bias in included studies

The risk of bias evaluation for each of the risk of bias domains are presented in Figure 2 and Figure 3. All trials were at overall high risk of bias.

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

All included trials reported that they used random allocation. However, only two trials reported the method for randomising participants and utilised a randomisation‐sequence generation that we deemed at low risk of bias (Gathier 2018; Togashi 2015).

Blinding

One trial was open‐label and reported that participants and outcome assessors were blinded to treatment allocation, but that caregivers were aware of treatment allocation (Togashi 2015). One trial reported that the participants and outcome assessors were blinded to treatment allocation, but the treatment providers were unblinded (Gathier 2018). One trial used placebo and mentioned that the trial was 'double‐blinded', but provided no details on the method of blinding (Neil‐Dwyer 1985). None of the trials described how they attempted to adequately blind the participants, personnel, and outcome assessors. Blinding of blood pressure alterations remain a practical issue in randomised clinical trials but should nonetheless be attempted. A method for blinding stakeholders in blood pressure target investigations has been proposed previously and future trials should attempt to incorporate adequate blinding of participants, personnel, and outcome assessors (Grand 2018).

In addition to assessing the risk of bias for each trial, we also assessed the risk of bias for each outcome in regard to blinding of the outcome assessors (see Table 3 and Table 4).  Two trials reported blinded assessors carried out outcome assessment (Gathier 2018; Togashi 2015). In Gathier 2018, research nurses who were blinded to treatment allocation collected outcome measures. However, for serious adverse events and rebleeding, the principal investigator collected outcome measures, and it was not described if the principal investigator was blinded to treatment allocation (Gathier 2018). In Togashi 2015, the decision to test for delayed cerebral ischaemia with transcranial Doppler or CT was based on clinical assessments carried out by the treating physicians who were not blinded to treatment allocation. None of the trials described how they ensured adequate blinding of the outcome assessors. We judged the outcome at low risk of bias related to blinding of outcome assessment, if blinding of the outcome assessor was specifically mentioned for the outcome of interest. If blinding of the outcome assessor was not mentioned, it was judged at unclear risk of bias. In a trial, if an outcome measure was collected by personnel with no specific mention of blinding while other outcomes were collected by personnel with specific mention of blinding, then we judged the outcome measure collected by personnel with no specific mention of blinding to be at high risk of bias.

1. Definition and follow‐up of outcomes in trials comparing induced hypertension with placebo or no intervention.

Outcome	Trials (year)	Outcome reported in published articles	Definition/scale used	Primary, secondary, or explorative outcome in trial	Follow‐up period or day of assessment (after inclusion in trial)	Risk of bias: blinding of outcome assessor	Risk of bias: selective outcome reporting	Risk of bias: incomplete outcome data
Death from all causes	Togashi 2015	Yes	Not applicable	Secondary	6 months	Low risk of bias. Outcome assessors were blinded to treatment allocation.	Low risk of bias. Outcome reported.	Low risk of bias. No loss to follow‐up.
	Gathier 2018	Yes	Not applicable	Secondary	3 months	Low risk of bias. Outcome assessors were blinded to treatment allocation.	Low risk of bias. Outcome reported.	Low risk of bias. No loss to follow‐up.
Death or dependency	Togashi 2015	Yes	Modified Rankin scale	Primary outcome	6 months	Low risk of bias for both outcome measures as the outcome assessors were blinded to treatment allocation.	Low risk of bias. Outcome reported.	Low risk of bias. No loss to follow‐up.
			Extended Glasgow Outcome Scale	Secondary outcome
	Gathier 2018	Yes	Modified Rankin scale	Primary outcome	3 months	Low risk of bias. Outcome assessors were blinded to treatment allocation.	Low risk of bias. Outcome reported.	Low risk of bias. No loss to follow‐up.
Serious adverse events	Togashi 2015	Yes	Not specified	Secondary	Not specified	Low risk of bias. Outcome assessors were blinded to treatment allocation.	Unclear risk of bias. Protocol not available and trial not registered. Definition of outcome was not specified in article other than (quote) "we also monitored and recorded occurrences of serious adverse events and adverse events and their relationship to the study treatments."	Low risk of bias. No loss to follow‐up.
	Gathier 2018	Yes	Quote: "An SAE is any untoward medical occurrence in a patient or trial subject, which does not have a causal relationship with the treatment, and: (1) results in death; (2) is life threatening (at the time of the event); (3) requires prolongation of inpatients' hospitalization; (4) results in persistent or significant disability or incapacity; (5) is a new event of the trial likely to affect the safety of the subjects"	Not specified	Quote: "During hospital admission"	High risk of bias. Quote: "Outcome measures were obtained by a research nurses blinded for treatment allocation." However, "all serious adverse events (SAEs) were recorded during hospital admission by the principal investigator." Therefore, it is not documented that outcome assessment of serious adverse events was in fact blinded.	Low risk of bias. Outcome reported according to the ICH‐GCP.	Low risk of bias. No loss to follow‐up.
Quality of life	Togashi 2015	No	—	—	—	—	Unclear risk of bias. No protocol available and outcome not reported.	—
	Gathier 2018	Yes	Stroke Specific Quality of Life	Secondary	D‐90	Low risk of bias. Outcome assessors were blinded to treatment allocation.	Low risk of bias, as trial protocol or prospective trial registration was available.	Low risk of bias. No loss to follow‐up.
Rebleeding	Togashi 2015	No	—	—	—	—	Unclear risk of bias. No protocol available and outcome not reported.	—
	Gathier 2018	Yes	Included as a serious adverse event. Not specified if the diagnosis was clinical or radiological	Not specified	Quote: "During hospital admission"	High risk of bias. Quote: "Outcome measures were obtained by a research nurses blinded for treatment allocation." However, (quote) "all serious adverse events (SAEs) were recorded during hospital admission by the principal investigator." Therefore, it is not documented that outcome assessment of serious adverse events was in fact blinded.	Unclear risk of bias. Outcome reported, but it was not specified with the diagnosis was clinical or radiological.	Low risk of bias. No loss to follow‐up.
Delayed cerebral ischaemia	Togashi 2015	No	—	—	—	—	Unclear risk of bias. No protocol available and outcome not reported.	—
	Gathier 2018	Yes	Quote: "New focal deficit or a 2‐point reduction in the Glasgow Coma Scale. Clinical symptoms compatible with delayed cerebral ischaemia required confirmation by 2 of the following testing modalities: 1) transcranial Doppler; 2) computed tomography angiography; or 3) cerebral angiography."	Secondary	Not specified	High risk of bias. The treating physicians and nurses were not blinded to treatment allocation. Therefore, the clinical assessment leading to testing for delayed cerebral ischaemia was performed by personnel not blinded to treatment allocation.	Low risk of bias. Outcome reported and systematically assessed.	Low risk of bias. No loss to follow‐up.
Hydrocephalus	Not reported in any trials	—	—	—	—	—	—	—

ICH‐GCP: International Conference on Harmonisation of Good Clinical Practice.

2. Definition and follow‐up of outcomes in trials comparing lowered blood pressure with placebo or no intervention.

Outcome	Trials (year)	Outcome reported in published articles	Definition/scale used	Primary, secondary, or explorative outcome in trial	Follow‐up period or day of assessment (after inclusion in trial)	Risk of bias: blinding of outcome assessor	Risk of bias: selective outcome reporting	Risk of bias: incomplete outcome data
Death from all causes	Neil‐Dwyer 1985	Yes	Not applicable	Not specified	12 months	Unclear risk of bias.  Blinding of outcome assessor not mentioned.	Low risk of bias. Outcome reported.	Low risk of bias. 11 participants in intervention group and 9 participants lost to follow‐up in control group lost to follow‐up.
Death or dependency	Neil‐Dwyer 1985	No	Only death from all causes reported	—	—	—	—	—
Serious adverse events	Not reported in any trials	No	Only death from all causes reported	—	—	—	—	—
Quality of life	Not reported in any trials	—	—	—	—	—	—	—
Rebleeding	Not reported in any trials	—	—	—	—	—	—	—
Delayed cerebral ischaemia	Not reported in any trials	—	—	—	—	—	—	—
Hydrocephalus	Not reported in any trials	—	—	—	—	—	—	—

Incomplete outcome data

Incomplete outcome data did not impact the results related to induced hypertension versus control as none of the trials lost participants to follow‐up (Gathier 2018; Togashi 2015). In the trial assessing the effects of lowered blood pressure as the experimental intervention, 20/224 participants dropped out during the intervention period, but there was an equal drop‐out between groups (Neil‐Dwyer 1985).

No participants were lost to follow‐up in the two trials randomising participants to induced hypertension versus control. Therefore, we judged the risk of incomplete outcome data at low risk of bias for all outcomes (see Table 3) (Gathier 2018; Togashi 2015).

In the trial randomising participants to lowered blood pressure versus placebo, 11 participants in the lowered blood pressure group and nine participants in the placebo group were lost to follow‐up. However, the dropout was similar, and we judged the outcomes at low risk of bias related to incomplete outcome data (see Table 4).

Selective reporting

For induced hypertension versus control, only Gathier 2018 registered the trial prior to randomisation. However, both trials randomising participants to induced hypertension versus control reported on death from all causes and death or dependency. Therefore, we judged the trials to be at low risk of bias related to selective outcome reporting (Gathier 2018; Togashi 2015). Both the trials also reported serious adverse events (Gathier 2018; Togashi 2015). One trial reported quality of life and rebleeding (Gathier 2018). One trial reported delayed cerebral ischaemia (Togashi 2015). None of the trials reported on the incidence of hydrocephalus (Gathier 2018; Togashi 2015). We also assessed the risk of selective outcome reporting to each outcome in addition to each trial (see Table 3 and Table 4). Togashi 2015 did not define serious adverse event, making us judge this outcome result to be at unclear risk of bias related to selective outcome reporting (see Table 3). Only Gathier 2018 reported quality of life and rebleeding. We were unable to obtain protocols from Togashi 2015, making us judge selective outcome reporting related to quality of life and rebleeding as unclear (see Table 3).

For lowered blood pressure versus placebo, there was no protocol or trial registration available (Neil‐Dwyer 1985). Therefore, we judged the trial to be at unclear risk of bias related to selective outcome reporting. The trial only reported on death from all causes. Therefore, the data related to death or dependency only consisted of death from all causes.

Other potential sources of bias

Too few trials were included to enable an assessment of publication bias. We also assessed for‐profit bias, but none of the trials seemed to be sponsored by the industry or any other stakeholders with an interest in a given result.

Effects of interventions

See: Table 1; Table 2

Induced hypertension

Death from all causes

Two trials randomising 61 participants reported death from all causes. Fixed‐effect meta‐analysis showed no evidence of a difference between induced hypertension and control at three months' follow‐up (RR 1.60, 95% CI 0.57 to 4.52; P = 0.38; Analysis 1.1). Visual inspection of the forest plot and I² statistic (I² = 0%) showed no signs of heterogeneity. TSA showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more (no TSA graph produced). No participants were lost to follow‐up. Therefore, we did not assess the impact of incomplete outcome data for this result. This outcome result was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

1.1. Analysis.

Comparison 1: Death from all causes, Outcome 1: Death from all causes closest to 3 months

The analysis of death from all causes at maximal follow‐up showed a similar result (RR 1.60; 95% CI 0.57 to 4.38; P = 0.38; Analysis 1.2).

1.2. Analysis.

Comparison 1: Death from all causes, Outcome 2: Death from all causes at maximal follow‐up

Death or dependency

Two trials randomising 61 participants reported on death or dependency. Fixed‐effect meta‐analysis showed no evidence of a difference between induced hypertension and control at three months' follow‐up (RR 1.29, 95% CI 0.78 to 2.13; P = 0.33; Analysis 2.1). Visual inspection of the forest plot and I² statistic (I² = 0%) showed no signs of heterogeneity. TSA showed that we had insufficient information to confirm or reject our relative risk reduction of 20% or more (no TSA graph produced). No participants were lost to follow‐up. Therefore, we did not assess the impact of incomplete outcome data for this result. This outcome result was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

2.1. Analysis.

Comparison 2: Death or dependency, Outcome 1: Death or dependency closest to 3 months

The analysis of death or dependency at maximal follow‐up showed a similar result (RR 1.29, 95% CI 0.78 to 2.13; P = 0.33; Analysis 2.2).

2.2. Analysis.

Comparison 2: Death or dependency, Outcome 2: Death or dependency at maximal follow‐up

Serious adverse events

Two trials randomising 61 participants reported on serious adverse events. Fixed‐effect meta‐analysis showed no evidence of a difference between induced hypertension and control at three months' follow‐up on our adjusted threshold of statistical significance (RR 2.24, 95% CI 1.01 to 4.99; P = 0.05; Analysis 3.1). Visual inspection of the forest plot and I² statistics (I² = 0%) showed no signs of heterogeneity. TSA showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more (no TSA graph produced). No participants were lost to follow‐up; therefore, we did not assess the impact of incomplete outcome data for this result. This outcome result was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

3.1. Analysis.

Comparison 3: Serious adverse events, Outcome 1: Serious adverse events closest to 3 months

We post‐hoc assessed all individual serious adverse events. We found no evidence of a difference between induced hypertension and control for any individual serious adverse event.

The analysis of serious adverse events at maximal follow‐up showed a similar result (RR 2.24, 95% CI 1.01 to 4.99; P = 0.05; Analysis 3.2).

3.2. Analysis.

Comparison 3: Serious adverse events, Outcome 2: Serious adverse events at maximal follow‐up

Quality of life

One trial reported quality of life (Gathier 2018). Therefore, meta‐analysis could not be conducted. Quality of life was reported using the Stroke Specific Quality of Life Scale, where a higher score indicates a better quality of life. In the induced hypertension group, quality of life at follow‐up was at a median of 47 points (interquartile range 35 to 55). In the control group, quality of life at follow‐up was at a median of 49 points (interquartile range 35 to 55). TSA showed that we had insufficient information to confirm or reject our predefined MD of the SD divided by '2' (Figure 4). This outcome was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

4.

Trial Sequential Analysis graph of quality of life. The cumulative z‐curve (the blue line) did not cross any of the monitoring boundaries (futility, benefit, or harm) and did not reach the diversity‐adjusted required information size. Hence, there was not enough information to confirm or reject important intervention effects. MD: mean difference.

Rebleeding

One trial reported rebleeding (Gathier 2018). Therefore, meta‐analysis could not be conducted. In the induced hypertension group, rebleeding occurred in 1/21 (5%) participants. In the control group, rebleeding occurred in 2/20 (10%) participants. Fisher's exact test showed no evidence of a difference between induced hypertension and control (P = 1.0). TSA showed that we had insufficient information to confirm or reject our predefined relative risk reduction of 20% or more. This outcome was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

The analysis of rebleeding at maximal follow‐up showed a similar result (Fisher's exact test P = 1.0).

Delayed cerebral ischaemia

One trial randomising 20 participants reported on delayed cerebral ischaemia (Gathier 2018). Therefore, meta‐analysis could not be conducted. In the induced hypertension group, delayed cerebral ischaemia occurred in 0/10 participants. In the control group, delayed cerebral ischaemia occurred in 1/10 (10%) participants. Fisher's exact test showed no evidence of a difference between induced hypertension and control (P = 1.0). TSA showed that we had insufficient information to confirm or reject a relative risk reduction of 20% or more (no TSA graph produced). This outcome was at high risk of bias (see Table 3). The certainty of evidence was very low due to high risk of bias and imprecision.

The analysis of delayed cerebral ischaemia at maximal follow‐up showed a similar result (Fisher's exact test P = 1.0)

Hydrocephalus

Neither of the included trials reported on hydrocephalus.

Subgroup analyses

None of our predefined subgroup analyses could be conducted.

Lowered blood pressure

Only one trial randomising 224 participants assessed lowered blood pressure versus placebo (Neil‐Dwyer 1985). Therefore, no meta‐analyses could be conducted on lowered blood pressure in people with subarachnoid haemorrhage.

Death from all causes

One trial analysing 204 participants reported death from all causes. When assessing death from all causes closest to three months' follow‐up (28 days), 13 participants in the lowered blood pressure group died and 21 participants in the placebo group died. Fisher's exact test showed no evidence of a difference between groups (P = 0.058). TSA showed that we had insufficient information to confirm or reject a relative risk reduction of 20% or more (no TSA graph produced). This outcome was at high risk of bias (see Table 4). The certainty of evidence was very low due to high risk of bias and imprecision.

The analysis of death from all causes at maximal follow‐up using Fisher's exact test also showed no evidence of a difference between lowered blood pressure and control (P = 0.084)

Death or dependency

No usable data was reported regarding dependency. Only data regarding death from all causes could be included in the analysis of death or dependency. Therefore, the analysis of death or dependency would show similar results as the death from all causes outcomes as assessed at both closest to three months' follow‐up and at maximal follow‐up.

Serious adverse events

No serious adverse events were reported. Only data regarding death from all causes could be included in the analysis of serious adverse events. Therefore, the analysis of serious adverse events would show similar results as the death from all causes outcomes as assessed at both closest to three months' follow‐up and at maximal follow‐up.

Quality of life

The study reported no data regarding quality of life.

Rebleeding

The study reported no data regarding rebleeding.

Delayed cerebral ischaemia

The study reported no data regarding rebleeding.

Hydrocephalus

The study reported no data regarding hydrocephalus.

Subgroup analyses

No subgroup analyses could be conducted due to the lack of trials regarding blood pressure lowering in people with subarachnoid haemorrhage.

Discussion

Summary of main results

We included a total of two trials randomising 61 participants to induced hypertension versus no intervention (Gathier 2018; Togashi 2015), and one trial randomising 224 participants to lowered blood pressure versus placebo (Neil‐Dwyer 1985). All trials were at high risk of bias. The certainty of the evidence according to GRADE was very low for all outcomes.

Induced hypertension versus control

Meta‐analyses showed no evidence of a difference between induced hypertension and control on death from all causes, death or dependency, and serious adverse events. Only one trial reported quality of life, therefore, we were unable to perform meta‐analysis. Only one trial reported rebleeding and delayed cerebral ischaemia, therefore, we were unable to perform meta‐analyses of these outcomes, but Fisher's exact test showed no evidence of a difference between induced hypertension and no intervention on rebleeding and delayed cerebral ischaemia. Trial sequential analyses showed that we had insufficient information to confirm or reject our predefined minimal important intervention effects regarding death from all causes, death or dependency, quality of life, delayed cerebral ischaemia, and rebleeding.

None of the trials randomising participants to induced hypertension versus control reported on hydrocephalus. No subgroup analyses could be conducted for trials randomising participants to induced hypertension versus control.

Lowered blood pressure versus placebo

One trial randomised participants to lowered blood pressure versus placebo. Therefore, meta‐analysis could not be performed for any outcomes. The trial only reported on death from all causes. We could not assess death or dependency, quality of life, rebleeding, delayed cerebral ischaemia, and hydrocephalus. Fisher's exact test showed no evidence of a difference between lowered blood pressure and control on death from all causes. TSA showed that we had insufficient information to confirm or reject our predefined minimal important intervention effect regarding death from all causes. No subgroup analyses could be conducted for trials randomising participants to lowered blood pressure versus control.

Overall completeness and applicability of evidence

We only identified two trials assessing induced hypertension and only one trial assessing lowered blood pressure in people with subarachnoid haemorrhage. It is thought‐provoking that we identified so few trials, considering that alteration of blood pressure in people with subarachnoid haemorrhage has been practice for more than three decades (Connolly 2012; Diringer 2011; Kosnik 1976; Steiner 2013). International guidelines recommend lowering the blood pressure when the systolic blood pressure exceeds 160 mmHg (Connolly 2012; Diringer 2011), or 180 mmHg (Steiner 2013), before occluding the aneurysm. No recommendations exist regarding postoperative blood pressure management in people with subarachnoid haemorrhage unless delayed cerebral ischaemia occurs. The guidelines either recommend inducing hypertension in people with delayed cerebral ischaemia (Connolly 2012; Diringer 2011), or refrain from making such a recommendation due to a lack of evidence (Steiner 2013). The findings presented in this systematic review might be highly influenced by random error ('play of chance') due to lack of information.

Induced hypertension

The trials assessing induced hypertension included few participants and were both at high risk of bias. All outcomes, except for quality of life, were also at high risk of bias. There was a lack of information to confirm or reject minimal important intervention effects for all outcomes. The reporting of serious adverse events was inadequate. Therefore, it is highly problematic to make any inferences about the use of induced hypertension in people with acute subarachnoid haemorrhage. One trial reported rebleeding and quality of life. No trials reported hydrocephalus.

Lowered blood pressure

The single trial assessing induced hypotension in people with subarachnoid haemorrhage included few participants, was at high risk of bias, and only reported usable data for death from all causes. Therefore, the effect estimates for death or dependency and serious adverse events were identical to that of death from all causes. There was a lack of information to confirm or reject minimal important intervention effects for all outcomes. No inferences can be made regarding the beneficial and harmful effects of our other predefined patient‐important outcomes.

Quality of the evidence

None of the trials randomising participants to induced hypertension versus control blinded the participants or personnel. Two trials randomising participants to induced hypertension versus control described blinding of outcome assessors (Gathier 2018; Togashi 2015). The remaining trial did not report adequate blinding. The lack of adequate blinding, and the high risk of bias in general, increases the risk that our results overestimate the beneficial effects and underestimate the harmful effects (Hróbjartsson 2014; Savovic 2018). Masking blood pressure remains a practical issue in randomised clinical trials, but a method for blinding stakeholders in blood pressure trials has been described and future trials should attempt to incorporate blinding of the involved participants, personnel, and outcome assessors (Grand 2018).

The certainty of the evidence was very low for all outcome measures primarily due to imprecision and high risk of bias. We had between 0.23% and 11.8% of the information needed in the meta‐analyses on induced hypertension to be conclusive about the results as estimated by trial sequential analyses. We only had 7.7% of the information needed to be conclusive on lowered blood pressure as estimated by TSA.

Potential biases in the review process

We attempted to avoid any biases in the review process by using a thorough search strategy to search all databases created by an information specialist. We also predefined our methodology in our protocol published prior to literature searching (Maagaard 2018). We included trials regardless of publication year, language, setting, and status without regard to participant age, sex, and comorbidities.

Agreements and disagreements with other studies or reviews

One previous systematic review found no beneficial effects on inducing hypertension in people with subarachnoid haemorrhage for the treatment or prevention of delayed cerebral ischaemia (Veldeman 2016). We also found no beneficial effects of inducing hypertension in people with subarachnoid haemorrhage on death from all causes, death or dependency, serious adverse events, quality of life, rebleeding, or delayed cerebral ischaemia.

Guidelines recommend lowering the blood pressure in people with hypertension in the presence of an unsecured aneurysm and induce hypertension in people with delayed cerebral ischaemia, but we found no evidence from randomised clinical trials validating these recommendations (Connolly 2012; Diringer 2011; Steiner 2013).

Authors' conclusions

Implications for practice.

Based on the current evidence, for both induced hypertension and lowered blood pressure, there is a lack of information needed to confirm or reject minimal important intervention effects on patient‐important outcomes.

Implications for research.

There is an urgent need for randomised clinical trials at low risk of bias assessing the beneficial and harmful effects of altering blood pressure in people with acute subarachnoid haemorrhage.

Sparse evidence exists regarding interventions for altering blood pressure in people with subarachnoid haemorrhage. Currently, the international guidelines are primarily informed by observational studies, case reports, and expert opinion. None of trials randomising participants to induced hypertension versus control blinded the participants and personnel. The two trials randomising participants to induced hypertension versus control mentioned blinding of outcome assessors, but neither of them described how they ensured adequate blinding of the outcome assessors. Blinding stakeholders in blood pressure trials remains a practical issue, but should nonetheless be attempted. Future randomised trials should use the SPIRIT statement for their design and the CONSORT statement for their reporting. Moreover, the trials should use methods allowing for blinding of blood pressure and report on patient‐centred outcomes such as mortality, serious adverse events, quality of life, rebleeding, hydrocephalus, and delayed cerebral ischaemia.

History

Protocol first published: Issue 8, 2018

Appendices

Appendix 1. CENTRAL search strategy

#1    MeSH descriptor: [Subarachnoid Hemorrhage] explode all trees    613
#2    MeSH descriptor: [Intracranial Hemorrhages] this term only    260
#3    MeSH descriptor: [Intracranial Aneurysm] this term only    519
#4    MeSH descriptor: [Rupture, Spontaneous] this term only    125
#5    #3 and #4     30
#6    MeSH descriptor: [Aneurysm, Ruptured] this term only    154
#7    MeSH descriptor: [Brain] explode all trees    11154
#8    MeSH descriptor: [Meninges] explode all trees    300
#9    #7 or #8     11432
#10    #6 and #9     11
#11    ((subarachnoid or arachnoid) near/6 (haemorrhage* or hemorrhage* or bleed* or blood*)):ti,ab,kw  (Word variations have been searched)    1686
#12    MeSH descriptor: [Vasospasm, Intracranial] this term only    161
#13    ((cerebral or intracranial or cerebrovascular) near/6 (vasospasm or spasm)):ti,ab,kw  (Word variations have been searched)    433
#14    SAH:ti,ab,kw  (Word variations have been searched)    577
#15    #1 or #2 or #5 or #10 or #11 or #12 or #13 or #14     2134
#16    MeSH descriptor: [Antihypertensive Agents] explode all trees    8024
#17    (antihypertens* or anti‐hypertens*):ti,ab,kw  (Word variations have been searched)    17144
#18    ((blood pressure or hypertens*) near/5 (lower* or reduc* or decreas* or control*)):ti,ab,kw  (Word variations have been searched)    29224
#19    ((high or elevat* or rais*) near/2 blood pressure):ti,ab,kw  (Word variations have been searched)    3828
#20    MeSH descriptor: [Vasodilator Agents] explode all trees    3803
#21    vasodilat*:ti,ab,kw  (Word variations have been searched)    10381
#22    (acetylcholine or adenosine or adrenomedullin or alprostadil or amiodarone or amlodipine or amrinone or amyl nitrite or bencyclane or bepridil or betahistine or bradykinin or calcitonin or celiprolol or chromonar or colforsin or cromakalim or cyclandelate or diazoxide or dihydroergocristine or dihydroergocryptine or dilazep or diltiazem or dipyridamole or dyphylline or enoximone or ergoloid mesylates or erythritol or erythrityl tetranitrate or felodipine or fenoldopam or flunarizine or hexobendine or hydralazine or iloprost or isosorbide dinitrate or isoxsuprine or isradipine or kallidin or khellin or lidoflazine or mibefradil or milrinone or minoxidil or molsidomine or moxisylyte or nafronyl or nebivolol or niacin or nicardipine or nicergoline or nicorandil or nicotinyl or nifedipine or nimodipine or nisoldipine or nitrendipine or nitric oxide or nitroglycerin or nitroprusside or nonachlazine or nylidrin or oxprenolol or oxyfedrine or papaverine or pentaerythritol tetranitrate or pentoxifylline or perhexiline or phenoxybenzamine or pinacidil or pindolol or polymethyl or prenylamine or propranolol or sildenafil citrate or sodium azide o rsuloctidil or tadalafil or theobromine or theophylline or thiouracil or tolazoline or trapidil or trimetazidine or verapamil or vincamine or xanthinol):ti,ab,kw  (Word variations have been searched)    48309
#23    {or #16‐#22}     82811
#24    MeSH descriptor: [Calcium Channel Blockers] explode all trees    2929
#25    (calcium near/3 (inhibit* or antagonist? or block*)):ti,ab,kw  (Word variations have been searched)    6520
#26    (amlodipine or amrinone or bencyclane or bepridil or cinnarizine or conotoxins or diltiazem or felodipine or fendiline or flunarizine or gallopamil or isradipine or lidoflazine or magnesium sulfate or mibefradil or nicardipine or nifedipine or nimodipine or nisoldipine or nitrendipine or perhexiline or prenylamine or tiapamil hydrochlorideverapamil or omega‐agatoxin iva or omega‐conotoxin gvia or omega‐conotoxins):ti,ab,kw  (Word variations have been searched)    13524
#27    {or #24‐#26}     16438
#28    MeSH descriptor: [Angiotensin‐Converting Enzyme Inhibitors] explode all trees    4113
#29    MeSH descriptor: [Angiotensin Receptor Antagonists] explode all trees    2141
#30    ((angiotensin or dipeptidyl) near/3 (convert* or enzyme or inhibit* or recept* or block* or antagon*)):ti,ab,kw  (Word variations have been searched)    13028
#31    (((ace or renin) near/3 inhibit*) or ACEI):ti,ab,kw  (Word variations have been searched)    4830
#32    (alacepril or altiopril or benazepril or captopril or ceronapril or cilazapril or delapril or enalapril or fosinopril or idapril or imidapril or lisinopril or moexipril or moveltipril or pentopril or perindopril or quinapril or ramipril or spirapril or temocapril or trandolapril or zofenopril or aliskiren or remikiren):ti,ab,kw  (Word variations have been searched)    9677
#33    (KT3‐671 or candesartan or eprosartan or irbesartan or losartan or olmesartan or tasosartan or telmisartan or valsartan):ti,ab,kw  (Word variations have been searched)    6366
#34    (abitesartan or azilsartan or candesartan or elisartan or embusartan or eprosartan or forasartan or irbesartan or losartan or milfasartan or olmesartan or saprisartan or tasosartan or telmisartan or valsartan or zolasartan):ti,ab,kw  (Word variations have been searched)    6426
#35    {or #28‐#34}     21182
#36    MeSH descriptor: [Adrenergic Antagonists] explode all trees    5810
#37    (andrenergic near/3 (alpha or antagonist*)):ti,ab,kw  (Word variations have been searched)    1
#38    ((andrenergic or alpha or receptor*) near/3 block*):ti,ab,kw  (Word variations have been searched)    12131
#39    (alfuzosin or bunazosin or doxazosin or metazosin or neldazosin or prazosin or silodosin or tamsulosin or terazosin or tiodazosin or trimazosin):ti,ab,kw  (Word variations have been searched)    2761
#40    (beta near/3 (antagonist* or receptor* or adrenergic? block*)):ti,ab,kw  (Word variations have been searched)    11229
#41    (acebutolol or adimolol or afurolol or alprenolol or amosulalol or arotinolol or atenolol or befunolol or betaxolol or bevantolol or bisoprolol or bopindolol or bornaprolol or brefonalol or bucindolol or bucumolol or bufetolol or bufuralol or bunitrolol or bunolol or bupranolol or butofilolol or butoxamine or carazolol or carteolol or carvedilol or celiprolol or cetamolol or chlortalidone cloranolol or cyanoiodopindolol or cyanopindolol or deacetylmetipranolol or diacetolol or dihydroalprenolol or dilevalol or epanolol or esmolol or exaprolol or falintolol or flestolol or flusoxolol or hydroxybenzylpinodolol or hydroxycarteolol or hydroxymetoprolol or indenolol or iodocyanopindolol or iodopindolol or iprocrolol or isoxaprolol or labetalol or landiolol or levobunolol or levomoprolol or medroxalol or mepindolol or methylthiopropranolol or metipranolol or metoprolol or moprolol or nadolol or oxprenolol or penbutolol or pindolol or nadolol or nebivolol or nifenalol or nipradilol or oxprenolol or pafenolol or pamatolol or penbutolol or pindolol or practolol or primidolol or prizidilol or procinolol or pronetalol or propranolol or proxodolol or ridazolol or salcardolol or soquinolol or sotalol or spirendolol or talinolol or tertatolol or tienoxolol or tilisolol or timolol or tolamolol or toliprolol or tribendilol or xibenolol):ti,ab,kw  (Word variations have been searched)    16881
#42    {or #36‐#41}     32359
#43    MeSH descriptor: [Thiazides] explode all trees    2540
#44    MeSH descriptor: [Sodium Chloride Symporter Inhibitors] explode all trees    403
#45    MeSH descriptor: [Sodium Potassium Chloride Symporter Inhibitors] explode all trees    79
#46    ((loop or ceiling) near/3 diuretic*):ti,ab,kw  (Word variations have been searched)    642
#47    (amiloride or benzothiadiazine or bendroflumethiazide or bumetanide or chlorothiazide or cyclopenthiazide or furosemide or hydrochlorothiazide or hydroflumethiazide or methyclothiazide or metolazone or polythiazide or trichlormethiazide or veratide or thiazide?):ti,ab,kw  (Word variations have been searched)    6712
#48    (chlorthalidone or chlortalidone or phthalamudine or chlorphthalidolone or oxodoline or thalitone or hygroton or indapamide or metindamide or s‐1520 or s1520 or se‐1520 or se1520):ti,ab,kw  (Word variations have been searched)    1248
#49    {or #43‐#48}     8204
#50    (alphamethyldopa or amodopa or dopamet or dopegyt or dopegit or dopegite or emdopa or hyperpax or hyperpaxa or methylpropionic acid or dopergit or meldopa or methyldopa or methyldopate or medopa or medomet or sembrina or aldomet or aldometil or aldomin or hydopa or methyldihydroxyphenylalanine or methyl dopa or mulfasin or presinol or presolisin or sedometil or sembrina or taquinil or dihydroxyphenylalanine or methylphenylalanine or methylalanine or alpha methyl dopa):ti,ab,kw  (Word variations have been searched)    919
#51    (reserpine or serpentina or rauwolfia or serpasil):ti,ab,kw  (Word variations have been searched)    335
#52    (adesipress or arkamin or caprysin or catapres* or catasan or chlofazolin or chlophazolin or clinidine or clofelin* or clofenil or clomidine or clondine or clonistada or clonnirit or clophelin* or clonidine or dichlorophenylaminoimidazoline or dixarit or duraclon or gemiton or haemiton or hemiton or imidazoline or isoglaucon or klofelin or klofenil or m‐5041t or normopresan or paracefan or st‐155 or st 155 or tesno timelets):ti,ab,kw  (Word variations have been searched)    3633
#53    MeSH descriptor: [Hydralazine] explode all trees    329
#54    (hydralazin* or hydrallazin* or hydralizine or hydrazinophtalazine or hydrazinophthalazine or hydrazinophtalizine or dralzine or hydralacin or hydrolazine or hypophthalin or hypoftalin or hydrazinophthalazine or idralazina or 1‐hydrazinophthalazine or apressin or nepresol or apressoline or apresoline or apresolin or alphapress or alazine or idralazina or lopress or plethorit or praeparat):ti,ab,kw  (Word variations have been searched)    616
#55    {or #50‐#54}     5325
#56    #23 or #27 or #35 or #42 or #49 or #55     113875
#57    MeSH descriptor: [Hypertension] this term only and with qualifier(s): [Chemically induced ‐ CI]    281
#58    MeSH descriptor: [Intracranial Hypertension] this term only    150
#59    MeSH descriptor: [Blood Pressure] this term only    25961
#60    MeSH descriptor: [Arterial Pressure] this term only    320
#61    (triple H or triple‐H or HHH or hyperdynamic):ti,ab,kw  (Word variations have been searched)    1928
#62    ((increase* or raise*) near/5 (blood or plasma or fluid) near/5 volume):ti,ab,kw  (Word variations have been searched)    553
#63    ((increase* or raise* or maintain*) near/5 blood pressure):ti,ab,kw  (Word variations have been searched)    6389
#64    ((reduce* or decreas* or lower*) near/5 blood viscosity):ti,ab,kw  (Word variations have been searched)    280
#65    (hypervolemi* or hypervolaemi* or (induc* near/5 hypertension) or hypertensive or hemodilut* or haemodilut*):ti,ab,kw  (Word variations have been searched)    16034
#66    MeSH descriptor: [Blood Volume] this term only    779
#67    MeSH descriptor: [Plasma Volume] this term only    300
#68    MeSH descriptor: [Blood Viscosity] this term only    422
#69    MeSH descriptor: [Hemodilution] this term only    391
#70    MeSH descriptor: [Fluid Therapy] this term only    1625
#71    MeSH descriptor: [Blood Substitutes] this term only    84
#72    MeSH descriptor: [Plasma Exchange] this term only    235
#73    MeSH descriptor: [Plasma Substitutes] this term only    587
#74    MeSH descriptor: [Hydroxyethyl Starch Derivatives] this term only    601
#75    MeSH descriptor: [Dextrans] this term only    508
#76    MeSH descriptor: [Polygeline] this term only    67
#77    MeSH descriptor: [Povidone] this term only    238
#78    MeSH descriptor: [Albumins] explode all trees    7011
#79    MeSH descriptor: [Sodium Chloride] this term only    2237
#80    MeSH descriptor: [Saline Solution, Hypertonic] this term only    497
#81    MeSH descriptor: [Colloids] explode all trees    6517
#82    (hemodilut* or haemodilut*):ti,ab,kw  (Word variations have been searched)    1049
#83    (blood near/3 (expand* or expansion or exchange* or substitut* or replace* or remov* or dilut*)):ti,ab,kw  (Word variations have been searched)    2288
#84    (plasma near/3 (expand* or expansion or exchange* or substitut* or replace* or remov* or dilut*)):ti,ab,kw  (Word variations have been searched)    1805
#85    (hetastarch or pentastarch or hydroxyethy lstarch or hydroxyethyl‐starch or dextran* or polygeline or povidone or albumin or salineor colloid*):ti,ab,kw  (Word variations have been searched)    13852
#86    VIH:ti,ab,kw  (Word variations have been searched)    31
#87    {or #57‐#86}     72685
#88    MeSH descriptor: [Vasoconstrictor Agents] explode all trees    1720
#89    MeSH descriptor: [Calcium Channel Agonists] this term only    47
#90    MeSH descriptor: [Vasopressins] this term only    459
#91    MeSH descriptor: [Arginine Vasopressin] this term only    358
#92    ((vasoconstrict* near/3 (agent* or agonist*)) or vasopress*):ti,ab,kw  (Word variations have been searched)    4701
#93    (calcium channel near/3 (agonist* or activat*)):ti,ab,kw  (Word variations have been searched)    102
#94    MeSH descriptor: [Angiotensins] explode all trees    989
#95    (angiotensin or arginine vasopressin or dihydroergotamine or ephedrine or epinephrine or ergotamine or etilefrine or felypressin or lypressin or mephentermine or metaraminol or methoxamine or methysergide or midodrine or nordefrin or norepinephrine or octopamine or ornipressin or phenylephrine or racepinephrine or sumatriptan or synephrine or vasopressins or vasotocin):ti,ab,kw  (Word variations have been searched)    28237
#96    MeSH descriptor: [Dopamine Agents] explode all trees    1748
#97    (dopamin* near/3 (agent* or agonist* or effect*)):ti,ab,kw  (Word variations have been searched)    3102
#98    (alnespirone or apomorphine or brexpiprazole or bromocriptine or cabergoline or cy 208‐243 or dihydrexidine or dihydroergocornine or dihydroergocryptine or dihydroergotamine or dihydroergotoxine or dironyl or docarpamine or dopexamine or ergocryptine or fananserin or fenoldopam or ibopamine or lisuride or mesulergine or metergoline or "n 0437" or naxagolide or ncq 298 or pd 158771 or pergolide or piribedil or pramipexole or preclamol or quinagolide or quinelorane or quinpirole or ropinirole or stepholidine or talipexole):ti,ab,kw  (Word variations have been searched)    3404
#99    MeSH descriptor: [Adrenergic Agonists] explode all trees    3183
#100    MeSH descriptor: [Sympathomimetics] this term only    288
#101    sympathomimet*:ti,ab,kw  (Word variations have been searched)    1013
#102    (adrenerg* near/3 (agent* or agonist* or effect*)):ti,ab,kw  (Word variations have been searched)    9108
#103    (albuterol or clenbuterol or clonidine or dexmedetomidine or dobutamine or epinephrine or ergotamine or etilefrine or fenoterol or formoterol fumarate or guanabenz or guanfacine or hexoprenaline or isoetharine or isoproterenol or isoxsuprine or medetomidine or mephentermine or metaproterenol or metaraminol or methoxamine or methyldopa or midodrine or naphazoline or nebivolol or norepinephrine or nylidrin or octopamine or oxyfedrine or oxymetazoline or phenylephrine or phenylpropanolamine or prenalterol or procaterol or racepinephrine or ritodrine or salmeterol xinafoate or synephrine or terbutaline or tretoquinol or xamoterol or xylazine):ti,ab,kw  (Word variations have been searched)    28619
#104    {or #88‐#103}     56031
#105    MeSH descriptor: [Adrenal Cortex Hormones] this term only    2491
#106    MeSH descriptor: [Glucocorticoids] this term only    4419
#107    MeSH descriptor: [Receptors, Glucocorticoid] this term only    113
#108    glucocorticoid*:ti,ab,kw  (Word variations have been searched)    7294
#109    (beclomethasone or betamethasone or budesonide or clobetasol or desoximetasone or dexamethasone or diflucortolone or flumethasone or fluocinolone acetonide or fluocinonide or fluocortolone or fluorometholone or fluprednisolone or flurandrenolone or fluticasone propionate or melengestrol acetate or methylprednisolone or paramethasone or prednisolone or prednisone or triamcinolone):ti,ab,kw  (Word variations have been searched)    36017
#110    (alclometasone or amcinonide or ciclesonide or clobetasone butyrate or clocortolone or dichlorisone acetate or diflorasone or difluprednate or drocinonide phosphate potassium or fluocortin butyl ester or fluperolone acetate or fluprednidene acetate or medrysone or prednicarbate or rimexolone or terofenamate):ti,ab,kw  (Word variations have been searched)    747
#111    {or #105‐#110}     40519
#112    #87 or #104 or #111     151162
#113    #56 or #112     215761
#114    #15 and #113     652

Appendix 2. MEDLINE Ovid search strategy

1. Subarachnoid Hemorrhage/

2. intracranial hemorrhages/ or cerebral hemorrhage/

3. Intracranial Aneurysm/

4. Rupture, Spontaneous/

5. 3 and 4

6. Aneurysm, Ruptured/

7. exp brain/ or exp meninges/

8. 6 and 7

9. ((subarachnoid or arachnoid) adj6 (haemorrhage$ or hemorrhage$ or bleed$ or blood$)).tw.

10. Vasospasm, Intracranial/

11. ((cerebral or intracranial or cerebrovascular) adj6 (vasospasm or spasm)).tw.

12. subarachnoid haemorrhage.tw.

13. 1 or 2 or 5 or 8 or 9 or 10 or 11 or 12

14. exp Antihypertensive Agents/

15. (antihypertens$ or anti‐hypertens$).tw.

16. ((blood pressure or hypertens$) adj5 (lower$ or reduc$ or decreas$ or control$)).tw.

17. ((high or elevat$ or rais$) adj2 blood pressure).tw.

18. exp Vasodilator Agents/

19. vasodilat$.tw.

20. (acetylcholine or adenosine or adrenomedullin or alprostadil or amiodarone or amlodipine or amrinone or amyl nitrite or bencyclane or bepridil or betahistine or bradykinin or calcitonin or celiprolol or chromonar or colforsin or cromakalim or cyclandelate or diazoxide or dihydroergocristine or dihydroergocryptine or dilazep or diltiazem or dipyridamole or dyphylline or enoximone or ergoloid mesylates or erythritol or erythrityl tetranitrate or felodipine or fenoldopam or flunarizine or hexobendine or hydralazine or iloprost or isosorbide dinitrate or isoxsuprine or isradipine or kallidin or khellin or lidoflazine or mibefradil or milrinone or minoxidil or molsidomine or moxisylyte or nafronyl or nebivolol or niacin or nicardipine or nicergoline or nicorandil or nicotinyl or nifedipine or nimodipine or nisoldipine or nitrendipine or nitric oxide or nitroglycerin or nitroprusside or nonachlazine or nylidrin or oxprenolol or oxyfedrine or papaverine or pentaerythritol tetranitrate or pentoxifylline or perhexiline or phenoxybenzamine or pinacidil or pindolol or polymethyl or prenylamine or propranolol or sildenafil citrate or sodium azide o rsuloctidil or tadalafil or theobromine or theophylline or thiouracil or tolazoline or trapidil or trimetazidine or verapamil or vincamine or xanthinol).tw.

21. or/14‐20

22. exp Calcium Channel Blockers/

23. (calcium adj3 (inhibit$ or antagonist? or block$)).tw.

24. (amlodipine or amrinone or bencyclane or bepridil or cinnarizine or conotoxins or diltiazem or felodipine or fendiline or flunarizine or gallopamil or isradipine or lidoflazine or magnesium sulfate or mibefradil or nicardipine or nifedipine or nimodipine or nisoldipine or nitrendipine or perhexiline or prenylamine or tiapamil hydrochlorideverapamil or omega‐agatoxin iva or omega‐conotoxin gvia or omega‐conotoxins).tw.

25. or/22‐24

26. exp Angiotensin‐Converting Enzyme Inhibitors/ or exp Angiotensin Receptor Antagonists/

27. ((angiotensin or dipeptidyl) adj3 (convert$ or enzyme or inhibit$ or recept$ or block$)).tw.

28. (((ace or renin) adj3 inhibit$) or ACEI).tw.

29. exp enalapril/

30. (alacepril or altiopril or benazepril or captopril or ceronapril or cilazapril or delapril or enalapril or fosinopril or idapril or imidapril or lisinopril or moexipril or moveltipril or pentopril or perindopril or quinapril or ramipril or spirapril or temocapril or trandolapril or zofenopril or aliskiren or remikiren).tw.

31. exp losartan/

32. (KT3‐671 or candesartan or eprosartan or irbesartan or losartan or olmesartan or tasosartan or telmisartan or valsartan).tw.

33. or/26‐32

34. adrenergic antagonists/ or exp adrenergic alpha‐antagonists/ or exp adrenergic beta‐antagonists/

35. (andrenergic adj3 (alpha or antagonist$)).tw.

36. ((andrenergic or alpha or receptor$) adj3 block$).tw.

37. (alfuzosin or bunazosin or doxazosin or metazosin or neldazosin or prazosin or silodosin or tamsulosin or terazosin or tiodazosin or trimazosin).tw.

38. (beta adj3 (antagonist$ or receptor$ or adrenergic? block$)).tw.

39. (acebutolol or adimolol or afurolol or alprenolol or amosulalol or arotinolol or atenolol or befunolol or betaxolol or bevantolol or bisoprolol or bopindolol or bornaprolol or brefonalol or bucindolol or bucumolol or bufetolol or bufuralol or bunitrolol or bunolol or bupranolol or butofilolol or butoxamine or carazolol or carteolol or carvedilol or celiprolol or cetamolol or chlortalidone cloranolol or cyanoiodopindolol or cyanopindolol or deacetylmetipranolol or diacetolol or dihydroalprenolol or dilevalol or epanolol or esmolol or exaprolol or falintolol or flestolol or flusoxolol or hydroxybenzylpinodolol or hydroxycarteolol or hydroxymetoprolol or indenolol or iodocyanopindolol or iodopindolol or iprocrolol or isoxaprolol or labetalol or landiolol or levobunolol or levomoprolol or medroxalol or mepindolol or methylthiopropranolol or metipranolol or metoprolol or moprolol or nadolol or oxprenolol or penbutolol or pindolol or nadolol or nebivolol or nifenalol or nipradilol or oxprenolol or pafenolol or pamatolol or penbutolol or pindolol or practolol or primidolol or prizidilol or procinolol or pronetalol or propranolol or proxodolol or ridazolol or salcardolol or soquinolol or sotalol or spirendolol or talinolol or tertatolol or tienoxolol or tilisolol or timolol or tolamolol or toliprolol or tribendilol or xibenolol).tw.

40. or/34‐39

41. exp Thiazides/

42. exp sodium chloride symporter inhibitors/ or exp sodium potassium chloride symporter inhibitors/

43. ((loop or ceiling) adj3 diuretic$).tw.

44. (amiloride or benzothiadiazine or bendroflumethiazide or bumetanide or chlorothiazide or cyclopenthiazide or furosemide or hydrochlorothiazide or hydroflumethiazide or methyclothiazide or metolazone or polythiazide or trichlormethiazide or veratide or thiazide?).tw.

45. (chlorthalidone or chlortalidone or phthalamudine or chlorphthalidolone or oxodoline or thalitone or hygroton or indapamide or metindamide or s‐1520 or s1520 or se‐1520 or se1520).tw.

46. or/41‐45

47. (alphamethyldopa or amodopa or dopamet or dopegyt or dopegit or dopegite or emdopa or hyperpax or hyperpaxa or methylpropionic acid or dopergit or meldopa or methyldopa or methyldopate or medopa or medomet or sembrina or aldomet or aldometil or aldomin or hydopa or methyldihydroxyphenylalanine or methyl dopa or mulfasin or presinol or presolisin or sedometil or sembrina or taquinil or dihydroxyphenylalanine or methylphenylalanine or methylalanine or alpha methyl dopa).tw.

48. (reserpine or serpentina or rauwolfia or serpasil).tw.

49. (adesipress or arkamin or caprysin or catapres$ or catasan or chlofazolin or chlophazolin or clinidine or clofelin$ or clofenil or clomidine or clondine or clonistada or clonnirit or clophelin$ or clonidine or dichlorophenylaminoimidazoline or dixarit or duraclon or gemiton or haemiton or hemiton or imidazoline or isoglaucon or klofelin or klofenil or m‐5041t or normopresan or paracefan or st‐155 or st 155 or tesno timelets).tw.

50. exp hydralazine/

51. (hydralazin$ or hydrallazin$ or hydralizine or hydrazinophtalazine or hydrazinophthalazine or hydrazinophtalizine or dralzine or hydralacin or hydrolazine or hypophthalin or hypoftalin or hydrazinophthalazine or idralazina or 1‐hydrazinophthalazine or apressin or nepresol or apressoline or apresoline or apresolin or alphapress or alazine or idralazina or lopress or plethorit or praeparat).tw.

52. or/47‐51

53. 21 or 25 or 33 or 40 or 46 or 52

54. hypertension/ci or intracranial hypertension/ or blood pressure/ or arterial pressure/

55. (triple H or triple‐H or HHH or hyperdynamic).tw.

56. ((increase$ or raise$) adj5 (blood or plasma or fluid) adj5 volume).tw.

57. ((increase$ or raise$ or maintain$) adj5 blood pressure).tw.

58. ((reduce$ or decreas$ or lower$) adj5 blood viscosity).tw.

59. (hypervolemi$ or hypervolaemi$ or (induc$ adj5 hypertension) or hypertensive or hemodilut$ or haemodilut$).tw.

60. blood volume/ or plasma volume/ or blood viscosity/

61. hemodilution/ or fluid therapy/

62. blood substitutes/ or plasma exchange/ or exp plasma substitutes/

63. hetastarch/ or dextrans/ or poly geline/ or povidone/ or exp albumins/ or sodium chloride/ or saline solution, hypertonic/ or expcolloids/

64. (hemodilut$ or haemodilut$).tw.

65. (plasma adj3 (expand$ or expansion or exchange$ or substitut$ or replace$)).tw.

66. (blood adj3 (expand$ or expansion or exchange$ or substitute$ or replace$ or remov$ or dilut$)).tw.

67. (hetastarch or pentastarch or hydroxyethy lstarch or hydroxyethyl‐starch or dextran$ or polygeline or povidone or albumin or salineor colloid$).tw.

68. VIH.tw.

69. or/54‐68

70. exp vasoconstrictor agents/ or calcium channel agonists/

71. vasopressins/ or arginine vasopressin/

72. ((vasoconstrict$ adj3 (agent$ or agonist$)) or vasopress$).tw.

73. (calcium channel adj3 (agonist$ or activat$)).tw.

74. exp angiotensins/

75. (angiotensin or arginine vasopressin or dihydroergotamine or ephedrine or epinephrine or ergotamine or etilefrine or felypressin or lypressin or mephentermine or metaraminol or methoxamine or methysergide or midodrine or nordefrin or norepinephrine or octopamine or ornipressin or phenylephrine or racepinephrine or sumatriptan or synephrine or vasopressins or vasotocin).tw.

76. exp Dopamine Agents/

77. (dopamin$ adj3 (agent$ or agonist$ or effect$)).tw.

78. (alnespirone or apomorphine or brexpiprazole or bromocriptine or cabergoline or cy 208‐243 or dihydrexidine or dihydroergocornine or dihydroergocryptine or dihydroergotamine or dihydroergotoxine or dironyl or docarpamine or dopexamine or ergocryptine or fananserin or fenoldopam or ibopamine or lisuride or mesulergine or metergoline or "n 0437" or naxagolide or ncq 298 or pd 158771 or pergolide or piribedil or pramipexole or preclamol or quinagolide or quinelorane or quinpirole or ropinirole or stepholidine or talipexole).tw.

79. exp Adrenergic Agonists/

80. exp Sympathomimetics/

81. sympathomimet$.tw.

82. (adrenerg$ adj3 (agent$ or agonist$ or effect$)).tw.

83. (albuterol or clenbuterol or clonidine or dexmedetomidine or dobutamine or epinephrine or ergotamine or etilefrine or fenoterol or formoterol fumarate or guanabenz or guanfacine or hexoprenaline or isoetharine or isoproterenol or isoxsuprine or medetomidine or mephentermine or metaproterenol or metaraminol or methoxamine or methyldopa or midodrine or naphazoline or nebivolol or norepinephrine or nylidrin or octopamine or oxyfedrine or oxymetazoline or phenylephrine or phenylpropanolamine or prenalterol or procaterol or racepinephrine or ritodrine or salmeterol xinafoate or synephrine or terbutaline or tretoquinol or xamoterol or xylazine).tw.

84. or/70‐83

85. adrenal cortex hormones/ or exp glucocorticoids/

86. Receptors, Glucocorticoid/

87. glucocorticoid$.tw.

88. (beclomethasone or betamethasone or budesonide or clobetasol or desoximetasone or dexamethasone or diflucortolone or flumethasone or fluocinolone acetonide or fluocinonide or fluocortolone or fluorometholone or fluprednisolone or flurandrenolone or fluticasone propionate or melengestrol acetate or methylprednisolone or paramethasone or prednisolone or prednisone or triamcinolone).tw.

89. (alclometasone or amcinonide or ciclesonide or clobetasone butyrate or clocortolone or dexamethasone 21‐phosphate or dichlorisone acetate or diflorasone or difluprednate or drocinonide phosphate potassium or flumethasone pivalate or fluocortin butyl ester or fluperolone acetate or fluprednidene acetate or medrysone or prednicarbate or rimexolone or terofenamate).tw.

90. or/85‐89

91. 69 or 84 or 90

92. 53 or 91

93. Randomized Controlled Trials as Topic/

94. Random Allocation/

95. Controlled Clinical Trials as Topic/

96. control groups/

97. clinical trials as topic/ or clinical trials, phase i as topic/ or clinical trials, phase ii as topic/ or clinical trials, phase iii as topic/ or clinical trials, phase iv as topic/

98. double‐blind method/

99. single‐blind method/

100. Placebos/

101. placebo effect/

102. cross‐over studies/

103. randomized controlled trial.pt.

104. controlled clinical trial.pt.

105. (clinical trial or clinical trial phase i or clinical trial phase ii or clinical trial phase iii or clinical trial phase iv).pt.

106. (random$ or RCT or RCTs).tw.

107. (controlled adj5 (trial$ or stud$)).tw.

108. (clinical$ adj5 trial$).tw.

109. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.

110. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.

111. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.

112. (cross‐over or cross over or crossover).tw.

113. (placebo$ or sham).tw.

114. trial.ti.

115. (assign$ or allocat$).tw.

116. controls.tw.

117. or/93‐116

118. 13 and 92 and 117

Appendix 3. Embase search strategy

1. subarachnoid hemorrhage/    
2. subarachnoid space/    
3. exp brain hemorrhage/ or brain artery aneurysm rupture/    
4. exp intracranial aneurysm/ or carotid artery aneurysm/    
5. rupture/ or blood vessel rupture/    
6. 4 and 5    
7. aneurysm/ or aneurysm rupture/    
8. exp brain/ or exp meninx/    
9. 7 and 8    
10. ((subarachnoid or arachnoid) adj6 (haemorrhage$ or hemorrhage$ or bleed$ or blood$)).tw.    
11. ((cerebral or brain or intracranial or cerebrovascular) adj5 (vasospasm or spasm)).tw.    
12. sah.tw.    
13. 1 or 2 or 3 or 6 or 9 or 10 or 11 or 12    
14. exp antihypertensive agent/ or antihypertensive therapy/    
15. (antihypertens$ or anti‐hypertens$).tw.    
16. ((blood pressure or hypertens$) adj5 (lower$ or reduc$ or decreas$ or control$)).tw.    
17. ((high or elevat$ or rais$) adj2 blood pressure).tw.    
18. exp vasodilator agent/    
19. vasodilat$.tw.    
20. (acetylcholine or adenosine or adrenomedullin or alprostadil or amiodarone or amlodipine or amrinone or amyl nitrite or bencyclane or bepridil or betahistine or bradykinin or calcitonin or celiprolol or chromonar or colforsin or cromakalim or cyclandelate or diazoxide or dihydroergocristine or dihydroergocryptine or dilazep or diltiazem or dipyridamole or dyphylline or enoximone or ergoloid mesylates or erythritol or erythrityl tetranitrate or felodipine or fenoldopam or flunarizine or hexobendine or hydralazine or iloprost or isosorbide dinitrate or isoxsuprine or isradipine or kallidin or khellin or lidoflazine or mibefradil or milrinone or minoxidil or molsidomine or moxisylyte or nafronyl or nebivolol or niacin or nicardipine or nicergoline or nicorandil or nicotinyl or nifedipine or nimodipine or nisoldipine or nitrendipine or nitric oxide or nitroglycerin or nitroprusside or nonachlazine or nylidrin or oxprenolol or oxyfedrine or papaverine or pentaerythritol tetranitrate or pentoxifylline or perhexiline or phenoxybenzamine or pinacidil or pindolol or polymethyl or prenylamine or propranolol or sildenafil citrate or sodium azide o rsuloctidil or tadalafil or theobromine or theophylline or thiouracil or tolazoline or trapidil or trimetazidine or verapamil or vincamine or xanthinol).tw.    
21. or/14‐19    
22. exp calcium channel blocking agent/    
23. (calcium adj3 (inhibit$ or antagonist? or block$)).tw.    
24. (amlodipine or amrinone or bencyclane or bepridil or cinnarizine or conotoxins or diltiazem or felodipine or fendiline or flunarizine or gallopamil or isradipine or lidoflazine or magnesium sulfate or mibefradil or nicardipine or nifedipine or nimodipine or nisoldipine or nitrendipine or perhexiline or prenylamine or tiapamil hydrochlorideverapamil or omega‐agatoxin iva or omega‐conotoxin gvia or omega‐conotoxins).tw.    
25. or/22‐24    
26. exp angiotensin antagonist/    
27. angiotensin converting enzyme inhibit$.tw.    
28. (angiotensin adj3 (receptor antagon$ or receptor block$)).tw.    
29. (((ace or renin) adj3 inhibit$) or ACEI).tw.    
30. (alacepril or altiopril or ancovenin or benazepril or captopril or ceranapril or ceronapril or cilazapril or deacetylalacepril or delapril or derapril or enalapril or enalaprilat or epicaptopril or fasidotril or fosinopril or foroxymithine or gemopatrilat or idapril or imidapril or indolapril or libenzapril or lisinopril or moexipril or moveltipril or omapatrilat or pentopril$ or perindopril$ or pivopril or quinapril$ or ramipril$ or rentiapril or saralasin or s nitrosocaptopril or spirapril$ or temocapril$ or teprotide or trandolapril$ or utibapril$ or zabicipril$ or zofenopril$).tw.    
31. (abitesartan or azilsartan or candesartan or elisartan or embusartan or eprosartan or forasartan or irbesartan or losartan or milfasartan or olmesartan or saprisartan or tasosartan or telmisartan or valsartan or zolasartan).tw.    
32. (KT3‐671 or candesartan or eprosartan or irbesartan or losartan or olmesartan or tasosartan or telmisartan or valsartan).tw.    
33. or/26‐32    
34. exp adrenergic receptor blocking agent/    
35. (andrenergic adj3 (alpha or antagonist$)).tw.    
36. ((andrenergic or alpha or receptor$) adj3 block$).tw.    
37. (alfuzosin or bunazosin or doxazosin or metazosin or neldazosin or prazosin or silodosin or tamsulosin or terazosin or tiodazosin or trimazosin).tw.    
38. (beta adj3 (antagonist$ or receptor$ or adrenergic? block$)).tw.    
39. (acebutolol or adimolol or afurolol or alprenolol or amosulalol or arotinolol or atenolol or befunolol or betaxolol or bevantolol or bisoprolol or bopindolol or bornaprolol or brefonalol or bucindolol or bucumolol or bufetolol or bufuralol or bunitrolol or bunolol or bupranolol or butofilolol or butoxamine or carazolol or carteolol or carvedilol or celiprolol or cetamolol or chlortalidone cloranolol or cyanoiodopindolol or cyanopindolol or deacetylmetipranolol or diacetolol or dihydroalprenolol or dilevalol or epanolol or esmolol or exaprolol or falintolol or flestolol or flusoxolol or hydroxybenzylpinodolol or hydroxycarteolol or hydroxymetoprolol or indenolol or iodocyanopindolol or iodopindolol or iprocrolol or isoxaprolol or labetalol or landiolol or levobunolol or levomoprolol or medroxalol or mepindolol or methylthiopropranolol or metipranolol or metoprolol or moprolol or nadolol or oxprenolol or penbutolol or pindolol or nadolol or nebivolol or nifenalol or nipradilol or oxprenolol or pafenolol or pamatolol or penbutolol or pindolol or practolol or primidolol or prizidilol or procinolol or pronetalol or propranolol or proxodolol or ridazolol or salcardolol or soquinolol or sotalol or spirendolol or talinolol or tertatolol or tienoxolol or tilisolol or timolol or tolamolol or toliprolol or tribendilol or xibenolol).tw.    
40. or/34‐39    
41. exp thiazide diuretic agent/    
42. ((loop or ceiling) adj3 diuretic$).tw.    
43. (amiloride or benzothiadiazine or bendroflumethiazide or bumetanide or chlorothiazide or cyclopenthiazide or furosemide or hydrochlorothiazide or hydroflumethiazide or methyclothiazide or metolazone or polythiazide or trichlormethiazide or veratide or thiazide?).tw.    
44. (chlorthalidone or chlortalidone or phthalamudine or chlorphthalidolone or oxodoline or thalitone or hygroton or indapamide or metindamide or s‐1520 or s1520 or se‐1520 or se1520).tw.    
45. or/41‐44    
46. 21 or 25 or 33 or 40 or 45    
47. hypertension/ or blood pressure/    
48. (triple H or triple‐H or HHH or hyperdynamic).tw.    
49. ((increase$ or raise$) adj5 (blood or plasma or fluid) adj5 volume).tw.    
50. ((increase$ or raise$ or maintain$) adj5 blood pressure).tw.    
51. ((reduce$ or decreas$ or lower$) adj5 blood viscosity).tw.    
52. (hypervolemi$ or hypervolaemi$ or (induc$ adj5 hypertension) or hypertensive or hemodilut$ or haemodilut$).tw.    
53. blood viscosity/ or blood volume/    
54. hemodilution/    
55. blood substitute/ or plasma substitute/    
56. sodium chloride/    
57. h?emodilut$.tw.    
58. (plasma adj3 (expand$ or expansion or exchange$ or substitut$ or replace$ or remov$ or dilut$)).tw.    
59. (blood adj3 (expand$ or expansion or exchange$ or substitut$ or replace$ or remov$ or dilut$)).tw.    
60. (hetastarch or pentastarch or hydroxyethy lstarch or hydroxyethyl‐starch or dextran$ or polygeline or povidone or albumin or salineor colloid$).tw.    
61. VIH.tw.    
62. or/47‐61    
63. exp vasoconstrictor agent/    
64. exp calcium channel stimulating agent/    
65. exp vasopressin derivative/    
66. ((vasoconstrict$ adj3 (agent$ or agonist$)) or vasopress$).tw.    
67. (calcium channel adj3 (agonist$ or activat$)).tw.    
68. exp angiotensin derivative/    
69. (angiotensin or arginine vasopressin or dihydroergotamine or ephedrine or epinephrine or ergotamine or etilefrine or felypressin or lypressin or mephentermine or metaraminol or methoxamine or methysergide or midodrine or nordefrin or norepinephrine or octopamine or ornipressin or phenylephrine or racepinephrine or sumatriptan or synephrine or vasopressins or vasotocin).tw.    
70. exp dopamine receptor stimulating agent/    
71. (dopamin$ adj3 (agent$ or agonist$ or effect$)).tw.    
72. (alnespirone or apomorphine or brexpiprazole or bromocriptine or cabergoline or cy 208‐243 or dihydrexidine or dihydroergocornine or dihydroergocryptine or dihydroergotamine or dihydroergotoxine or dironyl or docarpamine or dopexamine or ergocryptine or fananserin or fenoldopam or ibopamine or lisuride or mesulergine or metergoline or "n 0437" or naxagolide or ncq 298 or pd 158771 or pergolide or piribedil or pramipexole or preclamol or quinagolide or quinelorane or quinpirole or ropinirole or stepholidine or talipexole).tw.    
73. exp adrenergic receptor stimulating agent/    
74. sympathomimet$.tw.    
75. (adrenerg$ adj3 (agent$ or agonist$ or effect$)).tw.    
76. (albuterol or clenbuterol or clonidine or dexmedetomidine or dobutamine or epinephrine or ergotamine or etilefrine or fenoterol or formoterol fumarate or guanabenz or guanfacine or hexoprenaline or isoetharine or isoproterenol or isoxsuprine or medetomidine or mephentermine or metaproterenol or metaraminol or methoxamine or methyldopa or midodrine or naphazoline or nebivolol or norepinephrine or nylidrin or octopamine or oxyfedrine or oxymetazoline or phenylephrine or phenylpropanolamine or prenalterol or procaterol or racepinephrine or ritodrine or salmeterol xinafoate or synephrine or terbutaline or tretoquinol or xamoterol or xylazine).tw.    
77. or/63‐76    
78. exp corticosteroid/    
79. glucocorticoid receptor/    
80. glucocorticoid$.tw.    
81. (beclomethasone or betamethasone or budesonide or clobetasol or desoximetasone or dexamethasone or diflucortolone or flumethasone or fluocinolone acetonide or fluocinonide or fluocortolone or fluorometholone or fluprednisolone or flurandrenolone or fluticasone propionate or melengestrol acetate or methylprednisolone or paramethasone or prednisolone or prednisone or triamcinolone).tw.    
82. (alclometasone or amcinonide or ciclesonide or clobetasone butyrate or clocortolone or dichlorisone acetate or diflorasone or difluprednate or drocinonide phosphate potassium or fluocortin butyl ester or fluperolone acetate or fluprednidene acetate or medrysone or prednicarbate or rimexolone or terofenamate).tw.    
83. or/78‐82    
84. 62 or 77 or 83    
85. 46 or 84    
86. Randomized Controlled Trial/ or "randomized controlled trial (topic)"/    
87. Randomization/    
88. Controlled clinical trial/ or "controlled clinical trial (topic)"/    
89. control group/ or controlled study/    
90. clinical trial/ or "clinical trial (topic)"/ or phase 1 clinical trial/ or phase 2 clinical trial/ or phase 3 clinical trial/ or phase 4 clinical trial/    
91. Crossover Procedure/    
92. Double Blind Procedure/    
93. Single Blind Procedure/ or triple blind procedure/    
94. placebo/ or placebo effect/    
95. (random$ or RCT or RCTs).tw.    
96. (controlled adj5 (trial$ or stud$)).tw.    
97. (clinical$ adj5 trial$).tw.    
98. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.    
99. (quasi‐random$ or quasi random$ or pseudo‐random$ or pseudo random$).tw.    
100. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.    
101. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.    
102. (cross‐over or cross over or crossover).tw.    
103. (placebo$ or sham).tw.    
104. trial.ti.    
105. (assign$ or allocat$).tw.    
106. controls.tw.    
107. or/86‐106    
108. 13 and 85 and 107

Appendix 4. Web of Science search strategy

# 73    #72 AND #59 AND #3     
# 72    #71 OR #70 OR #69 OR #68 OR #67 OR #66 OR #65 OR #64 OR #63 OR #62 OR #61 OR #60     
# 71    TS=controls     
# 70    TS=(assign* or allocat*)     
# 69    TI=trial     
# 68    TS=(placebo* or sham)     
# 67    TS=(cross‐over or cross over or crossover)     
# 66    TS=((singl* or doubl* or tripl* or trebl*) NEAR/5 (blind* or mask*))     
# 65    TS=((control or experiment* or conservative) NEAR/5 (treatment or therapy or procedure or manage*))     
# 64    TS=(quasi‐random* or quasi random* or pseudo‐random* or pseudo random*)     
# 63    TS=((control or treatment or experiment* or intervention) NEAR/5 (group* or subject* or patient*))     
# 62    TS=(clinical* NEAR/5 trial*)     
# 61    TS=(controlled NEAR/5 (trial* or stud*))     
# 60    TS=(random* or RCT or RCTs)     
# 59    #58 OR #34     
# 58    #57 OR #53 OR #44     
# 57    #56 OR #55 OR #54      
# 56    TS=(alclometasone or amcinonide or ciclesonide or clobetasone butyrate or clocortolone or dichlorisone acetate or diflorasone or difluprednate or drocinonide phosphate potassium or fluocortin butyl ester or fluperolone acetate or fluprednidene acetate or medrysone or prednicarbate or rimexolone or terofenamate)     
# 55    TS=(beclomethasone or betamethasone or budesonide or clobetasol or desoximetasone or dexamethasone or diflucortolone or flumethasone or fluocinolone acetonide or fluocinonide or fluocortolone or fluorometholone or fluprednisolone or flurandrenolone or fluticasone propionate or melengestrol acetate or methylprednisolone or paramethasone or prednisolone or prednisone or triamcinolone)     
# 54    TS=glucocorticoid*     
# 53    #52 OR #51 OR #50 OR #49 OR #48 OR #47 OR #46 OR #45    # 52    TS=(albuterol or clenbuterol or clonidine or dexmedetomidine or dobutamine or epinephrine or ergotamine or etilefrine or fenoterol or formoterol fumarate or guanabenz or guanfacine or hexoprenaline or isoetharine or isoproterenol or isoxsuprine or medetomidine or mephentermine or metaproterenol or metaraminol or methoxamine or methyldopa or midodrine or naphazoline or nebivolol or norepinephrine or nylidrin or octopamine or oxyfedrine or oxymetazoline or phenylephrine or phenylpropanolamine or prenalterol or procaterol or racepinephrine or ritodrine or salmeterol xinafoate or synephrine or terbutaline or tretoquinol or xamoterol or xylazine)     
# 51    TS=(adrenerg* near/3 (agent* or agonist* or effect*))     
# 50    TS=sympathomimet*     
# 49    TS=(alnespirone or apomorphine or brexpiprazole or bromocriptine or cabergoline or cy 208‐243 or dihydrexidine or dihydroergocornine or dihydroergocryptine or dihydroergotamine or dihydroergotoxine or dironyl or docarpamine or dopexamine or ergocryptine or fananserin or fenoldopam or ibopamine or lisuride or mesulergine or metergoline or "n 0437" or naxagolide or ncq 298 or pd 158771 or pergolide or piribedil or pramipexole or preclamol or quinagolide or quinelorane or quinpirole or ropinirole or stepholidine or talipexole)     
# 48    TS=(dopamin* near/3 (agent* or agonist* or effect*))     
# 47    TS=(angiotensin or arginine vasopressin or dihydroergotamine or ephedrine or epinephrine or ergotamine or etilefrine or felypressin or lypressin or mephentermine or metaraminol or methoxamine or methysergide or midodrine or nordefrin or norepinephrine or octopamine or ornipressin or phenylephrine or racepinephrine or sumatriptan or synephrine or vasopressins or vasotocin)     
# 46    TS=(calcium channel near/3 (agonist* or activat*))     
# 45    TS=((vasoconstrict* near/3 (agent* or agonist*)) or vasopress*)# 44    #43 OR #42 OR #41 OR #40 OR #39 OR #38 OR #37 OR #36 OR #35         
# 43    TS=(hetastarch or pentastarch or hydroxyethy lstarch or hydroxyethyl‐starch or dextran* or polygeline or povidone or albumin or salineor colloid*)     
# 42    TS=(plasma near/3 (expand* or expansion or exchange* or substitut* or replace* or remov* or dilut*))     
# 41    TS=(blood near/3 (expand* or expansion or exchange* or substitut* or replace* or remov* or dilut*)):     
# 40    TS=(hemodilut* or haemodilut*)     
# 39    TS=(hypervolemi* or hypervolaemi* or (induc* near/5 hypertension) or hypertensive or hemodilut* or haemodilut*)     
# 38    TS=((reduce* or decreas* or lower*) near/5 blood viscosity)     
# 37    TS=((increase* or raise* or maintain*) near/5 blood pressure)     
# 36    TS=((increase* or raise*) near/5 (blood or plasma or fluid) near/5 volume)     
# 35    TS=(triple H or triple‐H or HHH or hyperdynamic)     
# 34    #33 OR #28 OR #24 OR #18 OR #12 OR #9     
# 33    #32 OR #31 OR #30 OR #29     
# 32    TS=(hydralazin* or hydrallazin* or hydralizine or hydrazinophtalazine or hydrazinophthalazine or hydrazinophtalizine or dralzine or hydralacin or hydrolazine or hypophthalin or hypoftalin or hydrazinophthalazine or idralazina or 1‐hydrazinophthalazine or apressin or nepresol or apressoline or apresoline or apresolin or alphapress or alazine or idralazina or lopress or plethorit or praeparat)
# 31    TS=(adesipress or arkamin or caprysin or catapres* or catasan or chlofazolin or chlophazolin or clinidine or clofelin* or clofenil or clomidine or clondine or clonistada or clonnirit or clophelin* or clonidine or dichlorophenylaminoimidazoline or dixarit or duraclon or gemiton or haemiton or hemiton or imidazoline or isoglaucon or klofelin or klofenil or m‐5041t or normopresan or paracefan or st‐155 or st 155 or tesno timelets)     
# 30    TS=(reserpine or serpentina or rauwolfia or serpasil)     
# 29    TS=(alphamethyldopa or amodopa or dopamet or dopegyt or dopegit or dopegite or emdopa or hyperpax or hyperpaxa or methylpropionic acid or dopergit or meldopa or methyldopa or methyldopate or medopa or medomet or sembrina or aldomet or aldometil or aldomin or hydopa or methyldihydroxyphenylalanine or methyl dopa or mulfasin or presinol or presolisin or sedometil or sembrina or taquinil or dihydroxyphenylalanine or methylphenylalanine or methylalanine or alpha methyl dopa)     
# 28    #27 OR #26 OR #25     
# 27    TS=(chlorthalidone or chlortalidone or phthalamudine or chlorphthalidolone or oxodoline or thalitone or hygroton or indapamide or metindamide or s‐1520 or s1520 or se‐1520 or se1520)     
# 26    TS=(amiloride or benzothiadiazine or bendroflumethiazide or bumetanide or chlorothiazide or cyclopenthiazide or furosemide or hydrochlorothiazide or hydroflumethiazide or methyclothiazide or metolazone or polythiazide or trichlormethiazide or veratide or thiazide?)     
# 25    TS=((loop or ceiling) near/3 diuretic*)     
# 24    #23 OR #22 OR #21 OR #20 OR #19     
# 23    TS=(acebutolol or adimolol or afurolol or alprenolol or amosulalol or arotinolol or atenolol or befunolol or betaxolol or bevantolol or bisoprolol or bopindolol or bornaprolol or brefonalol or bucindolol or bucumolol or bufetolol or bufuralol or bunitrolol or bunolol or bupranolol or butofilolol or butoxamine or carazolol or carteolol or carvedilol or celiprolol or cetamolol or chlortalidone cloranolol or cyanoiodopindolol or cyanopindolol or deacetylmetipranolol or diacetolol or dihydroalprenolol or dilevalol or epanolol or esmolol or exaprolol or falintolol or flestolol or flusoxolol or hydroxybenzylpinodolol or hydroxycarteolol or hydroxymetoprolol or indenolol or iodocyanopindolol or iodopindolol or iprocrolol or isoxaprolol or labetalol or landiolol or levobunolol or levomoprolol or medroxalol or mepindolol or methylthiopropranolol or metipranolol or metoprolol or moprolol or nadolol or oxprenolol or penbutolol or pindolol or nadolol or nebivolol or nifenalol or nipradilol or oxprenolol or pafenolol or pamatolol or penbutolol or pindolol or practolol or primidolol or prizidilol or procinolol or pronetalol or propranolol or proxodolol or ridazolol or salcardolol or soquinolol or sotalol or spirendolol or talinolol or tertatolol or tienoxolol or tilisolol or timolol or tolamolol or toliprolol or tribendilol or xibenolol)     
# 22    TS=(beta near/3 (antagonist* or receptor* or "adrenergic? block*"))     
# 21    TS=(alfuzosin or bunazosin or doxazosin or metazosin or neldazosin or prazosin or silodosin or tamsulosin or terazosin or tiodazosin or trimazosin)     
# 20    TS=((andrenergic or alpha or receptor*) near/3 block*)     
# 19    TS=(andrenergic near/3 (alpha or antagonist*))     
# 18    #17 OR #16 OR #15 OR #14 OR #13     
# 17    TS=(abitesartan or azilsartan or candesartan or elisartan or embusartan or eprosartan or forasartan or irbesartan or losartan or milfasartan or olmesartan or saprisartan or tasosartan or telmisartan or valsartan or zolasartan)     
# 16    TS=(KT3‐671 or candesartan or eprosartan or irbesartan or losartan or olmesartan or tasosartan or telmisartan or valsartan)     
# 15    TS=(alacepril or altiopril or benazepril or captopril or ceronapril or cilazapril or delapril or enalapril or fosinopril or idapril or imidapril or lisinopril or moexipril or moveltipril or pentopril or perindopril or quinapril or ramipril or spirapril or temocapril or trandolapril or zofenopril or aliskiren or remikiren)     
# 14    TS=(((ace or renin) near/3 inhibit*) or ACEI)     
# 13    TS=((angiotensin or dipeptidyl) near/3 (convert* or enzyme or inhibit* or recept* or block* or antagon*))     
# 12    #11 OR #10     
# 11    TS=(amlodipine or amrinone or bencyclane or bepridil or cinnarizine or conotoxins or diltiazem or felodipine or fendiline or flunarizine or gallopamil or isradipine or lidoflazine or magnesium sulfate or mibefradil or nicardipine or nifedipine or nimodipine or nisoldipine or nitrendipine or perhexiline or prenylamine or tiapamil hydrochlorideverapamil or omega‐agatoxin iva or omega‐conotoxin gvia or omega‐conotoxins)     
# 10    TS=(calcium near/3 (inhibit* or antagonist? or block*))     
# 9    #8 OR #7 OR #6 OR #5 OR #4     
# 8    TS=(acetylcholine or adenosine or adrenomedullin or alprostadil or amiodarone or amlodipine or amrinone or amyl nitrite or bencyclane or bepridil or betahistine or bradykinin or calcitonin or celiprolol or chromonar or colforsin or cromakalim or cyclandelate or diazoxide or dihydroergocristine or dihydroergocryptine or dilazep or diltiazem or dipyridamole or dyphylline or enoximone or ergoloid mesylates or erythritol or erythrityl tetranitrate or felodipine or fenoldopam or flunarizine or hexobendine or hydralazine or iloprost or isosorbide dinitrate or isoxsuprine or isradipine or kallidin or khellin or lidoflazine or mibefradil or milrinone or minoxidil or molsidomine or moxisylyte or nafronyl or nebivolol or niacin or nicardipine or nicergoline or nicorandil or nicotinyl or nifedipine or nimodipine or nisoldipine or nitrendipine or nitric oxide or nitroglycerin or nitroprusside or nonachlazine or nylidrin or oxprenolol or oxyfedrine or papaverine or pentaerythritol tetranitrate or pentoxifylline or perhexiline or phenoxybenzamine or pinacidil or pindolol or polymethyl or prenylamine or propranolol or sildenafil citrate or sodium azide o rsuloctidil or tadalafil or theobromine or theophylline or thiouracil or tolazoline or trapidil or trimetazidine or verapamil or vincamine or xanthinol)     
# 7    TS=vasodilat*     
# 6    TS=((high or elevat* or rais*) near/2 blood pressure)     
# 5    TS=(("blood pressure" or hypertens*) near/5 (lower* or reduc* or decreas* or control*))     
# 4    TS=(antihypertens* or anti‐hypertens*)     
# 3    #2 OR #1     
# 2    TS=((cerebral or intracranial or cerebrovascular) near/6 (vasospasm or spasm))     
# 1    TS=((subarachnoid or arachnoid) near/6 (haemorrhage* or hemorrhage* or bleed* or blood*))    

Appendix 5. LILACS search strategy

Latin American and Carribean Health Sciences Literature

1. hemorragia subaracnoidea

Appendix 6. CBM search strategy

1. 全部字段= 蛛网膜下腔出血 or 颅内出血 or 脑出血 or 动脉瘤破裂 or 血管痉挛

2. 全部字段=抗高血压药 OR 抗高血压 OR 血压升高 OR 血管扩张剂 OR 钙阻滞剂 OR 钙拮抗剂 OR 血管紧张素转化酶抑制剂 OR 血管紧张素受体拮抗剂 OR 肾上腺素能拮抗剂 OR 糖皮质激素

3. 1 and 2

Appendix 7. CNKI search strategy

1. SU  =('蛛网膜下腔出血'+'颅内出血'+'脑出血'+'动脉瘤破裂'+'血管痉挛')

2. 在结果中检索：SU  =('抗高血压药'+'抗高血压'+'血压升高'+'血管扩张剂'+'钙通道阻滞剂'+'钙通道拮抗剂'+'血管紧张素转化酶抑制剂'+'血管紧张素受体拮抗剂'+'肾上腺素能拮抗剂'+'糖皮质激素')

3. 1 and 2

Appendix 8. VIP search strategy

1. 1. U= 蛛网膜下腔出血 OR 颅内出血OR 脑出血OR 动脉瘤破裂OR 血管痉挛

2. 2. U=(抗高血压药 OR 抗高血压 OR 血压升高 OR 血管扩张剂 OR 钙阻滞剂 OR 钙拮抗剂 OR 血管紧张素转化酶抑制剂 OR 血管紧张素受体拮抗剂 OR 肾上腺素能拮抗剂 OR 糖皮质激素)

3. 3. 1 and 2

Appendix 9. ClinicalTrials.gov search strategy

US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov) 

( Blood Pressure OR Blood Volume OR Hypertensive OR Hypertension OR Vasodilator OR Vasoconstrictor OR Calcium Channel OR Adrenergic OR Angiotensin OR Thiazides ) AND INFLECT EXACT "Interventional" [STUDY‐TYPES] AND ( Subarachnoid Hemorrhage OR Intracranial Hemorrhages OR Intracranial Aneurysm OR Vasospasm ) [DISEASE]

Appendix 10. WHO ICTRP search strategy

World Health Organization International Clinical Trials Registry Platform
(apps.who.int/trialsearch);

Condition: Subarachnoid Hemorrhage OR Intracranial Hemorrhages OR Intracranial Aneurysm OR Vasospasm
Intervention: Blood Pressure OR Blood Volume OR Hypertensive OR Hypertension OR Vasodilator OR Vasoconstrictor OR Calcium Channel OR Adrenergic OR Angiotensin OR Thiazides

Recruitment status is: ALL
Phases are: ALL

Hide synonyms 
‐ BLEEDING INTRACRANIAL, BRAIN‐‐HEMORRHAGE, HAEMORRHAGE INTRACRANIAL, HAEMORRHAGE;INTRACRANIAL, HEMORRHAGE INTRACRANIAL, HEMORRHAGE, INTRACRANIAL, HEMORRHAGE;INTRACRANIAL, HEMORRHAGES, INTRACRANIAL, HEMORRHAGIC STROKE, INTRACRANIAL BLEED, INTRACRANIAL BLEEDING, INTRACRANIAL HAEMORRHAGE, INTRACRANIAL HAEMORRHAGE NOS, INTRACRANIAL HAEMORRHAGE, UNSPECIFIED, INTRACRANIAL HEMORR NOS, INTRACRANIAL HEMORRHAGE, INTRACRANIAL HEMORRHAGE (DIAGNOSIS), INTRACRANIAL HEMORRHAGE (DISORDER), INTRACRANIAL HEMORRHAGE NOS, INTRACRANIAL HEMORRHAGE, NOS, INTRACRANIAL HEMORRHAGE, UNSPECIFIED, INTRACRANIAL; HEMORRHAGE, UNSPECIFIED INTRACRANIAL HAEMORRHAGE, UNSPECIFIED INTRACRANIAL HEMORRHAGE, intracranial hemorrhages ‐ ACUTE CEREBROVASCULAR DISEASE, HAEMORRHAGE SUBARACHNOID, HAEMORRHAGE;SUBARACHNOID, HEMORRHAGE SUBARACHNOID, HEMORRHAGE, SUBARACHNOID, HEMORRHAGE; SUBARACHNOID, HEMORRHAGE;SUBARACHNOID, HEMORRHAGES, SUBARACHNOID, HEMORRHAGIC STROKE, INTRACRANIAL HEMORRHAGE, MENINGEAL HEMORRHAGE, OTHER SUBARACHNOID HAEMORRHAGE, SAH ‐ SUBARACHNOID HAEMORRHAGE, SICKLE CELL ANEMIA, SUBARACHNOID BLEEDING, SUBARACHNOID HAEMORRHAGE, SUBARACHNOID HAEMORRHAGE FROM ANTERIOR COMMUNICATING ARTERY, SUBARACHNOID HAEMORRHAGE FROM BASILAR ARTERY, SUBARACHNOID HAEMORRHAGE FROM CAROTID SIPHON AND BIFURCATION, SUBARACHNOID HAEMORRHAGE FROM INTRACRANIAL ARTERY, UNSPECIFIED, SUBARACHNOID HAEMORRHAGE FROM MIDDLE CEREBRAL ARTERY, SUBARACHNOID HAEMORRHAGE FROM OTHER INTRACRANIAL ARTERIES, SUBARACHNOID HAEMORRHAGE FROM POSTERIOR COMMUNICATING ARTERY, SUBARACHNOID HAEMORRHAGE FROM VERTEBRAL ARTERY, SUBARACHNOID HAEMORRHAGE NOS, SUBARACHNOID HAEMORRHAGE, UNSPECIFIED, SUBARACHNOID INTRACRANIAL HAEMORRHAGE, SUBARACHNOID INTRACRANIAL HEMORRHAGE (DISORDER), subarachnoid hemorrhage ‐ ANEURYSM INTRACRANIAL, ANEURYSM OF CEREBRAL ARTERY, ANEURYSM OF CEREBRAL ARTERY (DIAGNOSIS), ANEURYSM, INTRACRANIAL, ANEURYSM;ARTERY;CEREBRAL, ANEURYSMS, INTRACRANIAL, BERRY ANEURYSM, BRAIN ANEURYSM, CEREBRAL ANEURYSM, CEREBRAL ANEURYSM, NONRUPTURED, CEREBRAL ARTERIAL ANEURYSM, CEREBRAL ARTERIAL ANEURYSM (DISORDER), CEREBRAL ARTERY ANEURYSM, CRANIAL ANEURYSM, intracranial aneurysm ‐ ANGIOSPASM, PERIPHERAL VASOCONSTRICTION, SPASM; VASCULAR, SYNDROME; VASOSPASTIC, VASCULAR CONSTRICTION, VASCULAR CONSTRICTION (FUNCTION), VASCULAR CONSTRICTION, FUNCTION (OBSERVABLE ENTITY), VASCULAR CONSTRICTION, NOS, VASCULAR SPASM, VASCULAR; SPASM, VASOCONSTRICTION, VASOCONSTRICTIONS, VASOSPASTIC; SYNDROME, vasospasm ‐ AGENTS, VASOPRESSOR, AGONISTS, VASOACTIVE, VASOACTIVE AGONISTS, VASOPRESSOR, VASOPRESSOR AGENTS, VASOPRESSOR DRUGS, VASOPRESSORS, VASOPRESSORS (MEDICATION), vasoconstrictor ‐ CA CHANNEL, CA CHANNEL ANTAG RECEPT, CA CHANNEL BLOCK RECEPT, CA CHANNELS, CALCIUM ION CHANNEL, CALCIUM ION CHANNELS, CHANNELS, VOLTAGE‐DEPENDENT CALCIUM, ION CHANNEL CA, ION CHANNEL, CALCIUM, ION CHANNELS, CALCIUM, RECEPT CA CHANNEL ANTAG, VDCC, calcium channel ‐ ARTERIAL PRESSURE, ARTERIAL PRESSURES, ARTERIAL TENSION, ARTERIAL TENSIONS, HYPERTENSION, PRESSURE, ARTERIAL, PRESSURE, ARTERIAL BLOOD, PRESSURE, BLOOD, PRESSURES, ARTERIAL, PRESSURES, ARTERIAL BLOOD, SAP‐SYSTEMIC ARTERIAL PRESSURE, SYSTEMIC ARTERIAL PRESSURE, SYSTEMIC ARTERIAL PRESSURE (OBSERVABLE ENTITY), TENSION, ARTERIAL, TENSIONS, ARTERIAL, blood pressure ‐ WITH HIGH BLOOD PRESSURE, hypertensive ‐ 3‐02 HYPERTENSIVE DISEASES, ABDOMEN SWELLING, BLOOD PRESSURE HIGH, BLOOD PRESSURE, HIGH, BLOOD PRESSURE; HIGH, BLOOD PRESSURES, HIGH, CARDIO/PULM: HYPERTENSIVE DISORDER, HBP, HIGH BLOOD PRESSURE, HIGH BLOOD PRESSURE DISORDER, HIGH BLOOD PRESSURES, HIGH BP, HIGH; ARTERIAL TENSION, HIGH; BLOOD PRESSURE, HTN, HYPERPIESIA, HYPERPIESIS, HYPERTENSIVE CARDIOVASCULAR DISEASE OR SYNDROME, HYPERTENSIVE DISEASE, HYPERTENSIVE DISEASE, NOS, HYPERTENSIVE DISEASES, HYPERTENSIVE DISEASES (I10‐I15), HYPERTENSIVE DISORDER, HYPERTENSIVE DISORDER, SYSTEMIC ARTERIAL (DISORDER), HYPERTENSIVE VASCULAR DISEASE, hypertension ‐ VOLUME BLOOD, VOLUME, BLOOD, VOLUMES, BLOOD, blood volume ‐ 815‐816 VASODILATING AGENTS, VASODILATING AGENT, VASODILATING AGENT (PRODUCT), VASODILATING AGENT, NOS, VASODILATING AGENTS, VASORELAXANTS, vasodilator ‐ angiotensin ‐ adrenergic ‐ 722 THIAZIDE DIURETICS, BENZOTHIADIAZINE DIURETICS, DIURETICS, BENZOTHIADIAZINE, DIURETICS, THIAZIDE, THIAZIDE, THIAZIDE DIURETIC, THIAZIDE DIURETIC (PRODUCT), THIAZIDE DIURETIC (SUBSTANCE), THIAZIDE DIURETIC, NOS, THIAZIDE DIURETICS, THIAZIDE DIURETICS (MEDICATION), THIAZIDE‐RELATED DIURETIC, thiazides

Appendix 11. Assessment of risk of bias in included studies

Random sequence generation

• Low risk of bias: if sequence generation was achieved using computer random number generator or a random numbers table. Drawing lots, tossing a coin, shuffling cards, and throwing dice were considered adequate if performed by an independent adjudicator.

• Unclear risk of bias: if the method of randomisation was unclear.

• High risk of bias: if the allocation sequence was not randomised.

Allocation sequence concealment

• Low risk of bias: if the allocation of participants was performed by a central independent unit, on‐site locked computer, identically looking numbered sealed opaque envelopes, drug bottles, or containers prepared by an independent investigator. There must have been no risk of the investigator knowing the sequence.

• Unclear risk of bias: if the trial was classified as randomised but the allocation concealment process was unclear.

• High risk of bias: if the allocation sequence was known to the investigators who assigned participants.

Blinding of participants and personnel

• Low risk of bias: if the participants and the personnel were blinded to treatment allocation, and this was described.

• Unclear risk of bias: if the procedure of blinding was unclear.

• High risk of bias: if blinding of participants and personnel was not performed.

Blinding of outcome assessment

• Low risk of bias: if the trial investigators performing the outcome assessments were blinded to the intervention.

• Unclear risk of bias: if the procedure of blinding was unclear.

• High risk of bias: if blinding of outcome assessment was not performed.

Incomplete outcome data

• Low risk of bias: 1. the numbers and reasons for dropouts or withdrawals in all groups were clearly reported, and may have been described as similar in both groups, and the trial handled missing data appropriately in intention‐to‐treat analyses using proper methodology, e.g. multiple imputations, or 2. there were no or few dropouts or withdrawals. Generally, we considered the trial at low risk of bias due to incomplete outcome data if the number of dropouts was 5% or less; however, this level was regarded as guiding, not definitive.

• Unclear risk of bias: the numbers and reasons for withdrawals and dropouts were not clearly stated.

• High risk of bias: the pattern of dropouts could be described as being different in the two intervention groups, or the trial used improper methodology in dealing with the missing data, e.g. last observation carried forward.

Selective outcome reporting

• Low risk of bias: a protocol was published before or at the time the trial was begun, and the outcomes defined in the protocol were reported on. If there was no protocol or the protocol was published after the trial had begun, reporting of the primary outcomes classified this domain at low risk of bias.

• Unclear risk of bias: if there was no protocol and the primary outcomes were not reported on.

• High risk of bias: if the outcomes in the protocol were not reported on.

Other (for‐profit bias)

• Low risk of bias: if the trial was not financed by a company that might have had an interest in a given result.

• Unclear risk of bias: if there was no description of how the trial was financed.

• High risk of bias: if the trial was financed or had other involvement by a company that might have had an interest in a given result.

Overall risk of bias

We assessed the overall risk of bias of each trial as:

• low risk of bias: the trial was classified as overall 'low risk of bias' only if all of the bias domains described in the paragraphs above were classified as low risk of bias.

• high risk of bias: the trial was classified as overall 'high risk of bias' if any of the bias risk domains described in the paragraphs above were classified as 'unclear' or 'high risk of bias'.

Appendix 12. Trial sequential analysis input

Induced hypertension

Death from all causes at 3 months' follow‐up

Alpha‐spending boundary input: incidence in the control arm 13.33%, relative risk reduction 20%, alpha 3.33%, beta 20%, diversity 0%

Effect measure: risk ratio

Model: fixed‐effect

The alpha‐spending boundary was ignored due to too little information in the analysis. Therefore, a trial sequential analysis‐adjusted confidence interval could not be calculated. Hence, there was not enough information to confirm or reject important intervention effects.

Death or dependency at 3 months' follow‐up

Alpha‐spending boundary input: incidence in the control arm 36.66%, relative risk reduction 20%, alpha 3.33%, beta 20%, diversity 0%

Effect measure: risk ratio

Model: fixed‐effect

The alpha‐spending boundary was ignored due to too little information in the analysis. Therefore, a trial sequential analysis‐adjusted confidence interval could not be calculated. Hence, there was not enough information to confirm or reject important intervention effects.

Serious adverse events at 3 months' follow‐up

Alpha‐spending boundary input: incidence in the control arm 20.00%, relative risk reduction 20%, alpha 1.67%, beta 20%, diversity 0%

Effect measure: risk ratio

Model: fixed‐effect

The alpha‐spending boundary was ignored due to too little information in the analysis. Therefore, a trial sequential analysis‐adjusted confidence interval could not be calculated. Hence, there was not enough information to confirm or reject important intervention effects.

Quality of life at 3 months' follow‐up

Alpha‐spending boundary input: mean difference 7.4 (the standard deviation of 14.8 in the trial divided by '2'), variance 221.25, alpha 1.67%, beta 20%, diversity 0%

Effect measure: mean difference

Model: fixed‐effect

The alpha‐spending boundary did not cross any of the monitoring boundaries and did not reach the required information size. Hence, there was not enough information to confirm or reject important intervention effects.

Rebleeding at 3 months' follow‐up

Alpha‐spending boundary input: incidence in the control arm 10.00%, relative risk reduction 20%, alpha 1.67%, beta 20%, diversity 0%

Effect measure: risk ratio

Model: fixed‐effect

The alpha‐spending boundary was ignored due to too little information in the analysis. Therefore, a trial sequential analysis‐adjusted confidence interval could not be calculated. Hence, there was not enough information to confirm or reject important intervention effects.

Delayed cerebral ischaemia at 3 months' follow‐up

Alpha‐spending boundary input: incidence in the control arm 10.00%, relative risk reduction 20%, alpha 1.67%, beta 20%, diversity 0%

Effect measure: risk ratio

Model: fixed‐effect

The alpha‐spending boundary was ignored due to too little information in the analysis. Therefore, a trial sequential analysis‐adjusted confidence interval could not be calculated. Hence, there was not enough information to confirm or reject important intervention effects.

Data and analyses

Comparison 1. Death from all causes.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Death from all causes closest to 3 months	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	1.60 [0.57, 4.52]
1.2 Death from all causes at maximal follow‐up	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	1.60 [0.57, 4.52]

Comparison 2. Death or dependency.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Death or dependency closest to 3 months	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	1.29 [0.78, 2.13]
2.2 Death or dependency at maximal follow‐up	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	1.29 [0.78, 2.13]

Comparison 3. Serious adverse events.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Serious adverse events closest to 3 months	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	2.24 [1.01, 4.99]
3.2 Serious adverse events at maximal follow‐up	2	61	Risk Ratio (M‐H, Fixed, 95% CI)	2.24 [1.01, 4.99]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Gathier 2018.

Study characteristics
Methods	Multicentre, single‐blind, randomised trial with masked outcome assessment
Participants	SAH with an aneurysmatic bleeding pattern Aged > 18 years Delayed cerebral ischaemia based on a decrease of ≥ 1 point on the Glasgow Coma Scale or the development of new focal neurological deficits according to NIHSS, diagnosed by a neurologist, neurosurgeon, or intensivist, unless the deterioration did not reflect delayed cerebral ischaemia as evaluated by the treating physician, or both decrease on GCS and development of new focal neurological deficits  Exclusion criteria: co‐existing severe head injury, perimesencephalic haemorrhage, history of ventricular cardiac rhythm disorders or heart failure, moribund, pregnancy, likely transfer to another hospital, and no informed consent
Interventions	Experimental group (n = 21): induced hypertension. Hypertension induced with fluids, if hypovolaemia was suspected as well as vasopressors. Positive inotropic agents could be added in case of low cardiac output. Vasopressors could be: noradrenaline, dopamine, phenylephrine, or terlipressin. Maximum mean arterial blood pressure aimed to be 130 mmHg or maximum systolic blood pressure aimed to be 230 mmHg. If participants showed no improvement in their neurological deficits within 24 hours, vasopressors were discontinued. Induced hypertension was continued for ≥ 48 hours in participants who showed improvement in their neurological deficits Control group (n = 20): no hypertension. Therapy at the discretion of the treating physician. Hypotension (mean arterial blood pressure < 80 mmHg) was prevented by low‐dose vasopressors Co‐interventions: all participants received oral nimodipine and aimed for neutral fluid balance
Outcomes	Primary outcome Proportion of participants with poor outcome 3 months after randomisation (defined as modified Rankin Scale of 4, 5, or death) Secondary outcomes Treatment failure (proportion of participants in the experimental group who did not achieve clinical improvement regarding symptoms of delayed cerebral ischaemia Mortality 30 days after SAH Activities of daily living 3 months after randomisation Quality of life 3 months after randomisation (Stroke Specific Quality of Life Scale) Anxiety and depression 3 months after randomisation (Cognitive Failure Questionnaire) Functional outcome 12 months after randomisation (modified Rankin scale) Complications related to insertion of central venous catheter or intra‐arterial catheter Intracranial complications related to induced hypertension including cardiac rhythm disorders, low cardiac output state, and cardiac ischaemia Direct medical costs of used healthcare resources compared between the 2 intervention groups 12 months after randomisation
Notes	Sources of support: none reported other than the support for the author CS Gathier CS Gathier supported by the Dutch Heart Foundation (grant 2009B046) and the Brain Foundation Netherlands (grant 2009(1)‐72). The other authors report no conflicts of interest Follow‐up closest to 3 months:  3 months Maximal follow‐up: 3 months The trial was prematurely terminated  due to lack of an effect on overall cerebral perfusion and slow recruitment
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Initially performed using sealed opaque envelopes, but was later changed to web‐based randomisation with stratification for center with maximum random block size of eight." Judgement: probably done.
Allocation concealment (selection bias)	Low risk	Quote: "… web‐based randomisation with stratification for center with maximum random block size of eight." Judgement: probably done.
Blinding of participants and personnel (performance bias) all outcomes	High risk	Quote: "Clinical improvement within 24 hours was judged by the unblinded treating clinicians."
Blinding of outcome assessment (detection bias) all outcomes	Low risk	Quote: "Outcome measures were obtained by research nurses blinded for treatment allocation." Judgement: probably done.
Incomplete outcome data (attrition bias) all outcomes	Low risk	Quote: "One patient randomised to the hypertension group did not receive treatment due to previously unknown cardiomyopathy." Comment: but the participant was included in intention‐to‐treat analyses.
Selective reporting (reporting bias)	Low risk	Protocol obtained, all outcomes reported.
Other bias	Low risk	No notable conflicts of interest.

Neil‐Dwyer 1985.

Study characteristics
Methods	Prospective, randomised, double‐blind, placebo‐controlled
Participants	Aged 15–65 years, admitted within 48 hours of SAH confirmed by lumbar puncture Exclusion criteria: moribund, hypertension, ischaemic heart disease, congestive heart failure, or obstructive airways disease
Interventions	224 participants entered into trial The first 118 participants received both phentolamine 20 mg three‐hourly and either propranolol 80 mg eight‐hourly or placebo tablets The last 106 participants received either propranolol long action 320 mg once daily or placebo In the latter group, 50 participants were randomised so that 2 out of 3 participants received propranolol Co‐interventions: 6 participants received tranexamic acid, 18 received dexamethasone, anticonvulsants were administered to participants with seizure and as prophylactic treatment after surgery Treatment was continued for 3 weeks
Outcomes	Neurological assessment on day 28 and at 1 year after SAH. Participants were graded as follows. Good outcome: no major neurological deficit, have returned to work or capable of working Poor outcome: major neurological deficit, incapable of returning to work
Notes	Sources of support: not reported No mention of conflicts of interest Follow‐up closest to 3 months: 28 days Maximal follow‐up: 1 year
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No mention of randomisation method.
Allocation concealment (selection bias)	Unclear risk	No mention of allocation concealment.
Blinding of participants and personnel (performance bias) all outcomes	Low risk	Quote: "Treatment was double‐blind, only the anaesthetists has access to the code." Judgement: probably done.
Blinding of outcome assessment (detection bias) all outcomes	Unclear risk	Not mentioned.
Incomplete outcome data (attrition bias) all outcomes	Low risk	11 participants from intervention group and 9 from control group withdrew.
Selective reporting (reporting bias)	Unclear risk	No protocol could be obtained. Only death from all causes reported.
Other bias	Unclear risk	No mention of conflicts of interest.

Togashi 2015.

Study characteristics
Methods	Open‐label, randomised, controlled pilot trial using a 2‐way factorial design
Participants	Men and women aged ≥ 18 years with aneurysmal SAH of any clinical grade documented on CT scan treated with clipping or coiling within 72 hours of bleeding Exclusion criteria: history of traumatic SAH; non‐aneurysmal SAH; presence of an unsecured intracranial aneurysm at risk of rupture; delayed referral with clipping or coiling after 72 hours of initial bleeding; undetermined time of symptom onset; intracranial hypertension (intracranial pressure > 25 mmHg); history within the past 6 months or physical findings on admission of decompensated congestive heart failure; acute or recent myocardial infarction; cardiac arrhythmia; second‐ or third‐degree atrioventricular block; chronic renal failure requiring dialysis; suspected or confirmed pregnancy; non‐English speaking; prior Alzheimer's, stroke, degenerative condition, cerebral tumour, or mental retardation; severe terminal disease with < 6 months' life expectancy; refusal of informed consent
Interventions	Experimental group (n = 10): normovolemiae or hypervolemiae and augmented blood pressure (systolic blood pressure goal 140–220 mmHg) using noradrenaline or phenylephrine Control group (n = 10): normovolemiae or hypervolemiae and conventional blood pressure (systolic blood pressure goal 120–160 mmHg) using noradrenaline or phenylephrine to increase blood pressure and labetalol or hydralazine to decrease blood pressure
Outcomes	Primary outcome Modified Rankin Scale at 6 months as assessed by the extended Glasgow Outcome Scale and neurobehavioural performance using various neuropsychological tests Secondary outcomes Changed in transcranial doppler velocities Occurrence of delayed cerebral ischaemia New stroke on CT scan Serious adverse events Adverse events
Notes	Sources of support: the Anesthesiology and Critical Care Outcome Research Network (ACCORN) and departmental seed funds The authors reported no personal, financial, or institutional conflicts of interest Follow‐up closest to 3 months: 10 days Maximal follow‐up: 6 months
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Computer‐generated random number list." Judgement: probably done.
Allocation concealment (selection bias)	Low risk	Quote: "Allocation was concealed in sequential sealed envelopes." Judgement: probably done.
Blinding of participants and personnel (performance bias) all outcomes	High risk	Quote: "Single‐blinded, as participants were not aware of the treatment allocation; however, caregivers were aware of the study algorithms." Judgement: probably done.
Blinding of outcome assessment (detection bias) all outcomes	Low risk	Quote: "The outcome evaluation was objective in nature and the assessments were collected by investigators blinded to the treatment allocation." Judgement: probably done.
Incomplete outcome data (attrition bias) all outcomes	Low risk	Quote: "No patients who dropped out post‐randomisation or were lost to follow‐up at six months." Judgement: probably done.
Selective reporting (reporting bias)	Low risk	No protocol available, but death from all causes and death or dependency.
Other bias	Low risk	No conflicts of interest.

CT: computed tomography; n: number of participants; NIHSS: National Institutes of Health Stroke Scale; SAH: subarachnoid haemorrhage.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
An 2005	Participants did not have SAH.
Bao 2004	Participants did not have SAH.
Berge 2014	Participants did not have SAH.
Bi 2017	Participants did not have SAH.
Chang 2013a	Intervention not aimed at altering arterial blood pressure.
Chang 2013b	Participants did not have SAH.
Chen 2010	Participants did not have SAH.
Chu 2013	Participants did not have SAH.
Dong 2014	Participants did not have SAH.
Dong 2015	Participants did not have SAH.
Egge 2001	Triple‐H therapy (haemodilution, hypervolaemia, hypertension) vs no intervention; therefore, not strictly induced hypertension vs no intervention.
Fan 2003	Participants did not have SAH.
Fang 2003	Participants did not have SAH.
Frontera 2010	Not randomised.
Gan 2012	Participants did not have SAH.
Gomis 2010	Methylprednisolone as experimental intervention; this has a different mode of action and is not aimed at controlling blood pressure.
Graf 1974	Not true randomisation. Physician preference could place participants in either of the intervention groups.
Graf 1981	Not true randomisation. Physician preference could place participants in either of the intervention groups
Hu 1998	Participants did not have SAH.
Hu 2017	Participants did not have SAH.
Huang 2010	Participants did not have SAH.
Huang 2017	Participants did not have SAH.
Jia 2018	Participants did not have SAH.
Li 2013	Participants did not have SAH.
Li 2014	Participants did not have SAH.
Li 2015	Not randomised.
Liao 2017	Participants did not have SAH.
Lin 1996	Participants did not have SAH.
Lin 2010	Participants did not have SAH.
Liu 2014	Participants did not have SAH.
Li Y 2013	Wrong intervention.
Ma 2014	Participants did not have SAH.
MacDonald 2008	Clazosentan as experimental intervention; this has a different mode of action and is not aimed at controlling blood pressure.
MacDonald 2011	Clazosentan as experimental intervention; this has a different mode of action and is not aimed at controlling blood pressure.
MacDonald 2012	Clazosentan as experimental intervention; this has a different mode of action and is not aimed at controlling blood pressure.
Nan 1991	Not randomised.
Pei 2012	Participants did not have SAH.
Polderman 2014	No data regarding SAH participants.
Potter 2009	Participants did not have SAH.
Qi 2017	Participants did not have SAH.
Reinert 2004	Both experimental and control groups received antihypertensive drugs.
Shi 2006	Participants did not have SAH.
Shi 2008	Participants did not have SAH.
Song 2004	Participants did not have SAH.
Song 2005	Participants did not have SAH.
Tan 2001	Participants did not have SAH.
Tan 2017	Participants did not have SAH.
Tanaka 2015	Participants did not have SAH.
Tao 2014	Participants did not have SAH.
Unknown 2009	Participants did not have SAH.
Wang 2004	Not randomised.
Wang 2006a	Participants did not have SAH.
Wang 2006b	Participants did not have SAH.
Wang 2016	Participants did not have SAH.
Wang 2020	Non‐haemorrhagic intracranial aneurysm.
Wijdicks 1990	Retrospective observational study.
Wu 2006a	Participants did not have SAH.
Wu 2006b	Participants did not have SAH.
Yano 1993	Not randomised.
Zall 1990	Both experimental and control groups received antihypertensive drugs (adenosine vs nitroglycerin).
Zhang 1962	Not randomised.
Zhu 2005	Participants did not have SAH.

SAH: subarachnoid haemorrhage.

Differences between protocol and review

There were no differences between the protocol and the review.

Contributions of authors

MM: screened literature, extracted data, performed statistical analyses, assessed risk of bias and GRADE, and wrote the first draft of the Cochrane Review.

WKK: screened literature, extracted data, performed risk of bias assessments, and amended the Cochrane Review.

CO: contributed with invaluable comments, aided in risk of bias assessments, and amended the Cochrane review

CG: contributed with invaluable comments and methodological insights, and amended the Cochrane Review.

JCJ: assisted with interpretation of statistical analyses and writing of the final version of the Cochrane Review.

All authors agreed on the final version of the Cochrane Review.

Sources of support

Internal sources

• Copenhagen Trial Unit, Denmark

  The review was supported by the Copenhagen Trial Unit through wages paid to Christian Gluud and Janus Christian Jakobsen. The funding for the Copenhagen Trial Unit comes from the Danish Finance Act, which had had no influence on the review process.

External sources

• No external sources of support., Denmark

  No external sources of support.

Declarations of interest

MM: none.

WKK: work as a health professional: Neurology Residency, University Hospital Rigshospitalet, Copenhagen, Capital Region, Denmark (MD; during specialisation in neurology).

CO: support for attending meetings or travel, including sabbaticals and study tours: conference invitation, Merck Sharp & Dohme (funds paid to author). Ownership of stock shares or stock options: shareholder, Novo Nordisk (funds paid to author). Work as a health professional: medical doctor, private practice in Copenhagen (May 2020 to January 2021); medical doctor, Department of Internal Medicine, Herlev Hospital, Denmark

CG: none.

JCJ: none.

New