Education News and Reviews

Read our latest blog articles and Journal Club reviews, catch up on our Quants or post your own article

Latest education news from CCCEG

Quisque eleifend elit a mauris malesuada rutrum. Quisque fermentum nibh in libero gravida auctor.


Cerebral Microbleeds and the Effect of Intensive Blood Pressure Reduction on Hematoma Expansion and Functional Outcomes. A Secondary Analysis of the ATACH-2 Randomized Clinical Trial. 

Presented by Dr Jonathon Slater


In patients with intracerebral haemorrhage (ICH) is there an interaction between underlying small vessel disease (number of microbleeds and location) and intensive systolic blood pressure (SBP) control which may lead to a worse functional outcome or significant haematoma expansion?


Systolic blood pressure control in spontaneous ICH is frequently managed on or about 140mmHg (range 120mmHg – 160mmHg) and differs depending on the aetiology, location, and other medical comorbidities. The 120mmHg – 160mmHg range is within current research guidelines and recommendations. 


Specifically, there is little known about cerebral microbleeds. It is known that these microbleeds are remnants of prior micro-haemorrhages at the level of arterioles and capillaries and that such events can be visualised on T2 weighted (blood-sensitive) GRE MRI scans. In addition, these microbleeds are markers of underlying cerebral small vessel disease usually representing hypertensive arteriopathy.

There are two competing theories as to the impact of these microbleeds in ICH: 

  1. The fragility of cerebral microbleeds may increase the risk of haematoma expansion (and therefore increase the burden of injury).
  2. The vessels become more resilient over time and are therefore somewhat protective against further expansion. 

Haematoma expansion occurs early – usually within the first few hours and as much as a 10% increase can have a devastating effect. The jury is still out on the level of blood pressure reduction that provides the greatest benefit for reducing hematoma expansion. There is however, greater research, evidence, and understanding of the ‘general principals and management of ICH’:

  1. ICH is graded (Lisk) into 4 categories:
    1. 0 = No Intraventricular haemorrhage (IVH).
    1. 1 = Blood in the 3rdventricle or <1/3 of 1 lateral ventricle
    1. 2 = Blood in <1/2 of both lateral ventricles or 2/3 of 1 ventricle
    1. 3 = 1 ventricle completely filled or 2 more than 50% filled
  • Poorer outcomes for ICH are frequently seen with
    • Haematoma expansion 
    • ICH volume >60mls
    • IVH extension
    • ICH in certain areas (eg: cerebellum – requiring a high neurosurgical intervention; pons – 100% mortality) 
    • Low GCS <5 (and less researched)
  • Studies tend to remove those at highest risk – ICH has an on or about 20% mortality at 24 hours. The major RCTs exclude on or about 9 in 10 patients and as such generalisation/applicability of the results may be difficult. 
  • Early treatment of ICH has a positive impact – for instance, patients <70 years of age, with small ICH (<5mls), without IVH, and treated within 2.5 hours were shown to have reduced haematoma expansion and a ‘favourable’ OR for poor outcome of 0.28 at 90 days (FAST trial). The journal club today will look at the interplay of rapid initial treatment, prevention of haematoma expansion, and the effect of cerebral microbleeds. 
  • There are often difficult and competing risks – intracerebral pressure (ICP) in traumatic ICH may limit blood pressure targets and where there is suspicion of elevated ICP, CPP of 60-80mmHg is recommended and relaxing the lowering of BP targets may be needed.
  • Blood pressure is frequently elevated in acute non-traumatic ICH. Most research suggests 140mmHg is a reasonable systolic target although the INTERACT2 trial provides a cautionary tail that lower blood pressure (SBP <140mmHg) may improve functional outcomes. Ordinal analysis of Rankin scores (rather than ‘favourable’ vs ‘unfavourable’ scores) and quality of life scores were improved in the more intensive therapy group. However, many patients received mannitol (ICP monitoring data was not provided) and the use of specific antihypertensives not available in all countries may reduce the external validity. Further, whilst the use of multiple different antihypertensive agents was pragmatic in the trial, there may be some benefit from specific antihypertensives via non blood pressure pleiotropic effects that may have affected the results. 
  • The ATACH-2 trial showed no difference between Intensive and ‘Standard’ blood pressure treatment strategies for the first 24 hours on functional outcome at 3 months or haematoma expansion. The study ended early due to futility. Greater renal adverse events at 7 days were noted in the intensive therapy group.
  • Commonly used medications include, but are not limited to labetalol, hydralazine, nicardipine, clevidipine, and phentolamine.


Understanding whether cerebral microbleeds require greater blood pressure control will improve our appreciation of both the pathology and management of ICH with concurrent small vessel disease. More broadly speaking it will also help smaller centres to align their practices with tertiary and quaternary referral centres. 


  1. Is there an interaction between underlying small vessel disease and intensive systolic blood pressure (SBP) control for functional outcome and haematoma expansion?
  2. Blood pressure targets in ICH and exceptions to general rule
  3. The high level of practice variation
  4. Differences between the ATACH-2 and INTERACT2 trials


No! As we will discuss at the journal club, cerebral microbleeds did not appear to have an impact on functional outcomes or mortality. Additionally, their presence was not associated with haematoma expansion or response to blood pressure management. 


Presented by Dr Luke Terrett


Study Question:

In adults with TBI and ICP > 20 mmHg, despite “stage 1 treatments,” does hypothermia (32°C – 35°C) improve outcome compared with standard care?

How does this relate to our practice on NCCU?

Traumatic Brain Injury is almost the raison d’être of NCCU.

Elevated ICP is a frequent finding that requires prompt management in order to prevent further brain injury.

Therapeutic hypothermia (TH) has been shown to lower ICP and is a central part of our protocolised management of patients with TBI and elevated ICP.

TH has been shown to improve neurologic outcomes in comatose survivors of out-of-hospital cardiac arrest. The original 2 studies (HACA and Bernard) from 2002 targeted 32-34°C but in a more recent study (TTM trial), 36°C has been shown to be as good as 33°C. Different population and different indication.

Frankly, removing TH would go against our almost religious approach to the management of TBI – perhaps that is the reason why this study sits so uncomfortably. But, in our business, our thoughts should always be discordant and our actions should always make us uncomfortable. For this reason it is one of the most important studies in the context of how manage patients in Cambridge.

What do we currently know about this area?

TH may have a range synergistic neuroprotective effects:

  • It lowers the cerebral metabolic rate
  • Decreases excitotoxic neurotransmitter release
  • Decreases free radical formation
  • Decreases sustained electrical depolarisations
  • Inhibits proinflammatory and apoptotic pathways

TH lowers ICP perhaps through the following mechanisms, but we really don’t know:

  • Decreased inflammation
  • Decreased vasogenic edema
  • Decreased cerebral blood volume

However, TH is not without risks:

  • Coagulopathy/platelet dysfunction
  • Immunosuppression
  • Cardiac dysrhythmias
  • Hypotension
  • Pneumonia
  • Insulin resistance
  • Cold diuresis and electrolyte depletion
  • Decreased catecholamine responsiveness

Have a look at  Nature Reviews Neurology this blog.

Why was this study needed?

In the setting of TBI, multiple older studies showed mixed results. A 2014 systematic review and meta-analysis found possible benefit for therapeutic hypothermia in TBI. A well-designed RCT was needed to definitively answer the question.

At journal club we should discuss the following:

  • Hypothermia was applied quite early (after 5 min of sustained ICP > 20), seems a bit quick, see our letter.
  • Initial interventions were modest, yet safe, where does the risk of TH sit with this?
  • The study did not use other stage 2 treatments in hypothermia group unless persistent ICP elevation
  • What about how TH was achieved, it was quickly down to the target range 32-35°C, why not use an incremental, stepwise approach and achieve the minimum temp decrease required to control ICP?
  • What about all that fluid?
  • Sadly the trial stopped early after 387/600 patients enrolled, is this a problem?


Should we change practice on NCCU

Not based on the results of this study. They applied therapeutic hypothermia very differently than we do in the NCCU at Cambridge: They used it very early and prior to basic interventions, such as osmotherapy. Although basic is a word we use when we don’t know the harm a treatment really casues.

A trial that more closely mirrors our pattern of practice is needed prior to considering a change.

The next big RCT on TH in TBI, POLAR, has just finished enrolling patients, will this help us? Let’s see at journal club.


Further Reading

BTF guidelines

The Bottom Line 

Nature Reviews Neurology – Hypothermia for acute brain injury

SNACC Review