Unruptured cerebral arteriovenous malformations are better off treated medically: the ARUBA trial

A cerebral arteriovenous malformation (AVM) is an abnormal set of direct connections between the arteries and veins in the brain. These can cause a variety of neurologic symptoms, especially if they are large, and especially if they rupture.

arteriovenous malformation in the great cerebral vein of Galen; from Wikipedia user Filip em via Dr Laughlin Dawes

Mohr et al. recently published the result of the ARUBA trial, which compared medical (i.e., medical treatment for symptoms as needed) to interventional (i.e., surgical) treatment of this condition.

Their intention-to-treat analysis favored event-free survival in the medical management (MM; red) group:

Screen Shot 2017-09-13 at 8.25.31 AM
Mohr et al 2017; doi: 10.1212/WNL.0000000000004532

The actually-treated analysis favored event-free survival in the medical management (MM) group even more strongly:

Screen Shot 2017-09-13 at 8.30.04 AM
Mohr et al 2017; doi: 10.1212/WNL.0000000000004532

The authors suggest on the basis of this data that a reasonable management approach for unruptured cerebral AVM is to wait to see if a hemorrhage occurs, which may be mild if it does occur, and only then consider surgical intervention.


Brain cooling following revival from cardiac arrest

Recently I took a bit of a dive into the literature surrounding controlled hypothermia for recovery from return of spontaneous circulation post-cardiac arrest.

I started with this recent editorial, which describes some of the RCT evidence showing improved neurologic outcome if you cool following revival (and possibly even mid-arrest), as well as the recent trial demonstrating equivalent outcomes between cooling to 33C and 36C. Here’s their takeaway:

We concur with the AAN experts that less is not more and cooling should be harder, better, faster, stronger, in the sense that neurologists should be hardliners who embrace cooling as a default mode for nearly all cardiac arrest survivors, making it harder to exclude patients, while using cooling techniques that are the better ones, starting as quickly as possible after ROSC, and that 338C is stronger than 368C.

I then watched this nice lecture by Nicola Robertson, who describes her work on cooling to prevent brain damage prevention in perinatal asphyxia. As far as I can tell, the research here has been extensive and explains more of the mechanisms of why cooling can be an effective treatment for hypoxia.

She includes in her talk a lovely image from Gunn et al. 2016, which is worth a thousand words:

Screen Shot 2017-07-17 at 9.50.39 AM
Gunn et al. 2016; NTP/EPP = markers of mitochondrial metabolism; Lactate = marker of cell death

The idea is that there are two phases to neurotoxicity in the presence of hypoxia. The first phase involves a primary energy failure. This is when cells die because they don’t have enough oxygen/energy. But surprisingly, this is often not that all that large effect in terms of long-term brain damage.

Often a larger factor comes in the second phase of neurotoxicity — the activation of molecular signaling cascades which tell the cells to die. This involves an evolutionary mechanism by which the body clears out poorly functioning cells if they aren’t working properly.

As it turns out, by cooling brain cells during the latent phase molecular signaling cascade, clinicians can slow down and mitigate the secondary activation damage, thereby improving long-term neurologic outcome.

Applying this in adults following cardiac arrest is still a rapidly evolving field, and it’ll be interesting to see how the field and evidence base evolves over time.

One of the major problems with applying cooling for neurologic injury in more diverse clinical settings is that cooling to low temperatures, such as 33C, often leads to excessive amounts of shivering. If the shivering problem can be overcome, it may receive more widespread use in different conditions.