New MRI Technique Can Detect Diabetes

MonkeyBoy writes "Researchers at Joslin Diabetes Center and Massachusetts General Hospital have unveiled a new magnetic nanoparticle based magnetic resonance imaging technique that can detect diabetes even before clinical symptoms. In mice they were able to take non-invasive images of pancreatic inflammation and its reversal for type 1 diabetes. Full article is available as a PDF from Pubmedcentral. Will we see rapid translation of these pre-clinical observations to prediction and/or stratification of type 1 diabetes and treatment of individuals with the disease? This would provide a crucially needed early predictor of response to therapy. As an added bonus it looks like the analysis was done on a Linux box too."

Comments

Open Source Medicine?

[Khyber]

Just reading the summary, it says the box might run on Linux?

This is a good thing, now if only the pharmaceutical companies would make alot of their stuff open to the public, maybe we'd all not suffer nearly as badly due to our wallets not being fat enough.

Fullfilling MRI's promise (sort of)

[IvyKing]

The original push for MRI (from Damadian) was to locate cancers - the main benefit is eliminating a lot of exploratory surgeries. In this case it eliminates the need for biopsies (which, from TFA, are rarely performed due to the risk). Very nice work.

This is an interesting application of magneto-immunoassay - using the change in magnetic properties to determine if there has been a reaction.

MRI

[mchawi]

I have heard from several people that have had an MRI that (a) it is almost impossible to get scheduled in any reasonable timeframe for one and (b) they are 'incredibly expensive'.

For something like diabetes that can be diagnosed in other ways, I don't see a normal doctor or health insurance company prescribing an MRI. Health care is one of the areas where capitalism is really the wrong motivation (because our long term health is not their concern - just short term costs).

I guess I'm just cynical, but I see some of these great advances being almost useless to the majority of people because they simply can't afford it or don't even have healthcare.

I keep hoping for things to change though :)

Re:MRI

[rcolquhoun]

I have heard from several people that have had an MRI that (a) it is almost impossible to get scheduled in any reasonable timeframe for one and (b) they are 'incredibly expensive'.


So is diabetes.

I am a type 1 diabetic and it costs thousands perhaps 10's of thousands per year to manage(if you include side effects, lost opportunities etc). This will be the case for the rest of my life(hopefully 50+ years, although diabetes has a major effect in shortening it).

If it can be prevented by a series of MRI's + drugs or any other treatment for that matter that costs less than say $100k if will be economic for most pre-type 1 patients(ie children) to undertake.

  - Robert

Extremely preminary data

[martalli]

As a doctor, I had a few observations about this
stusdy. These researchers demonstrated that an
MRI can find evidence of diabetes in mice before
the development of clinical symptoms. How does
this relate to people?

1. Practical: Type I diabetes typically develops in
children, who usually must be sedated and
observed for an MRI. This could double the cost
of the MRI (typically $1-2k in US). Also, we
don't know how long these findings are present
before diabetes develops ... how often would
this need to be done?

2. The Incidence of type I diabetes is about
0.4% among people without relatives with
type I. First degree relatives have a 5-30%
risk... Certain genes predispose to type I
diabetes.... maybe this could be used in certain
people.

3. Since there is no such early diagnosis, we
don't really know who we might treat this
pre-type I state. Certainly we could make a
few reasonable guesses, but any benefit is
still in doubt until there are human trials.

I think this will definitely be more handy when
tricorders are perfected!