Inhaled anesthesia increases plaque burden in mouse model of Alzheimer's

We all know that inhaled anesthesia is over the short-term impairs neurological function; that is sort of the point using it for surgery.

However, a debate exists about whether inhaled anesthetics have long-term neurological consequences as well. In light of that debate Bianchi et al, publishing in the journal Neurobiology of Aging, have shown that in a mouse model of Alzheimer's disease some anesthetics increase histological measures of severity. Before you get excited though, let me tell you that the results are mixed.


An important point to understand in this paper is that mice do not naturally get Alzheimer's disease. To use them for research, experimenters have created a wide variety of genetically modified mice that recapitulate parts of the disease, but these mice are not 100% identical to each other or to the disease itself. The mouse model that they use in this study is a pretty common one called Tg2576 that possesses a mutation called the Swedish APP mutation.

APP or Amyloid Precursor Protein is a protein that we know is involved in Alzheimer's because one of defining pathological features in the brains of patients is the formation of what are called amyloid plaques. Amyloid plaques are aggregates of protein in the brain, and they are formed partly from a breakdown product of APP -- called Abeta. In a study of a family in Sweden that had very early onset Alzheimer's disease, a mutation in APP was discovered that causes them to make much more Abeta than normal. This is called the Swedish mutation. To make the Tg2576 mouse strain the Swedish-mutated APP gene was added to normal mice. This makes the mice get amyloid plaques and go into cognitive decline late in life, sort of like Alzheimer's disease.

Biachi et al

In this study the researchers wanted to see if anesthetics could accelerate cognitive decline in aged normal mice or mice that are prone to this Alzheimer's like disease, Tg2576 mice.

Bianchi et al exposed aged normal control mice and aged Tg2576 mice (about 1 year) to clinically relevant concentrations of two common inhaled anesthetics -- isofluorane and halothane. Then they tested the aged animals in a variety of memory tasks. Finally, they sacked the animals and looked at the number of amyloid plaques in their brains.

So that you can know what they were looking for with respect to plaques, here is a picture from the brains of a control mouse and a Tg2576 mouse (click to enlarge). The picture on the left is the brain of a aged normal mouse. The picture on the right is the brain of a Tg2576 mouse (I don't think that either were exposed to anesthetics). The amyloid plaques have been immunohistochemically-stained brown:


What the researchers found when they count the number of plaques was quite interesting. In the Tg2576 mice, the plaque burden showed a statistically significant increase for one of the anesthetics -- halothane. No significant change was observed in the normal mice. The results are below (both charts are for the Tg2576 mice, the normal mice are not shown):

i-51c4e5abb58c4fc641dc97cd30e10e24-plaque burden.jpg

The behavioral data on the other hand showed a different story. When we look at the behavior data, neither the Tg2576 mice nor the normal mice showed any deficit with respect to memory tasks. Also, the researchers sought to measure the numbers of cells actually undergoing death -- using a process called caspase-3 staining. The researchers found no significance increase in neuronal death in either the Tg2576 or the normal mice.

It would appear that though the plaque burden is increasing, this not creating a significant deficit.


This paper is interesting for several reasons, but it is also difficult to interpret:

  • 1) We know that amyloid plaques are a consequence of Alzheimer's diseases, but the most current theory (this is really up for debate, just so you know) does not attribute the neuronal death to the plaques themselves.

    Rather, the problem appears to be what are called Abeta oligomers. Remember that Abeta forms the plaques. Well, there is an intermediate stage of aggregation where several Abeta monomers join together to form oligomers. We believe that the oligomers are the actual toxic species because their concentrations correlate best with the decline that is observed in Alzheimer's patients (and for some other reasons).

    In that context, more plaques is difficult to interpret. The authors imply that the anesthetics also increase the concentration of oligomers -- I think they say that article is in press. Also, they cite research that states that inhaled anesthetics increase Abeta oligomerization in vitro.

    The important part that I trying to say here is that amyloid plaque is an indicator of the disease, it is not the disease itself.

  • 2) This point of plaques being an indicator is emphasized by the fact that no deficit was observed in the behavior of the tested animals. It could be that the anesthetics are increasing the plaque burden, but that is not enough to cause a measurable progression of the disease. Again, these mouse models are imperfect.

    Also, the behavioral tests may be imperfect. The researchers admit that performance on these tests is not a great indicator of the cognitive decline in this mouse model. Maybe if there were better tests we would see an effect.

  • 3) It is intriguing to me that the effect is anesthetic-specific. That suggests that the chemistry involved might be quite complicated.
  • 4) I would not run out and stop having necessary surgeries as a consequence of this research.

    It is important to recognize all the myriad health risks associated with putting something in your body that wasn't there before. Many times these risks are unexpected.

    However, surgery is always a cost-benefit calculation, and in most cases the consequences of not having the surgery are far more dire than the possible side-effects of anesthetic. The effect sizes that they are seeing here are really not that large, so I don't think that they should figure largely in our cost-benefit analysis when considering surgery.

  • 5) Finally, it will be interesting to see where this goes in the context of post-operative cognitive dysfunction. We know that patients can get cognitive dysfunction after surgery, and we know that age is a risk factor for this. However, it was never quite clear why. I have heard it also attributed to micro-strokes or hypoxia during the surgery. The effects of anesthetic might also have a role.

More like this

Also, the behavioral tests may be imperfect.

Since we know that, under stress, individuals prioritize input/output channels differently (perhaps according to brain type and/or previous damage), we would have to be able to test a mouse's vision, hearing, sense of touch, taste, and smell, as well as its communication skills (eg., Do tested mice continue to tell their control buddies where the chocolate chip cookies are in their maze?).

Are there any longterm studies of pocd in people with known brain types, for example, people with various CTD, AVM, MS, and/or Asperger's syndromes?

surgery is always a cost-benefit calculation

Don't the POCD studies suggest that pushing unnecessary surgery to people with certain markers in their personal and family histories is, at best, unethical? Or, more strongly put, that simply not explaining possible cognitive costs to any surgical candidate is unethical?

What I'm asking is, isn't it reckless at best to tell, for example, a joint replacement candidate that his decision to have the surgery need only be guided by his own perception of pain - especially when his pain could be controlled with non-injurious pain-control methods and meds (under non-injurious pain meds, I would include whole cannabis (not analogs) but not steroids, because the former strengthens bone development, rather than disrupts it like the latter)? Shouldn't the patient be allowed to decide what he is willing to risk in exchange for pain-free mobility? (Or, in a more striking scenario, for improved cosmetic gains?)

Or, on the other hand, couldn't the POCD studies also suggest that non-chemical ways of anesthetizing patients, which are used in Eastern medicine, should be explored further?

Isn't it ultimately true that, no matter how you interpret Bianchi et al, anesthesia is doing harm unnecessarily?

[btw, nice entry, and say hello to dr. jesse weinberger for me - I met him when I was working at the Met museum.]

I think that surgical intervention is an option that patients can feel uncomfortable refusing if it's offered by their clinician. For instance, Patient A: 80 year old man, reasonable surgical candidate, widower, 6cm (and growing) abdominal aortic aneurysm (not suitable for anything other than open surgical repair). Has been offered surgery, refused it. He knows the risks of rupture and has chosen to make the best of time left. How hard must it have been to make that decision? I bet his family would have tried to persuade him to have to op too.

Sorry, very off subject, but was on my mind. New to this bloggin malarchy and enjoying the freedom of expression!