New findings here - does this mean we can kill ongoing Parkinsons with antibiotics or is the damage already done by the time the disease presents itself?
While this looks like an intriguing bit of epidemiology, the evidence presented here is extremely weak. To take one particular quote from the abstract:
Their data shows there might be anywhere from a 72% decrease to a 20% increase in Parkinson’s disease in patients. IMO, the data is too weak to say that there was a real decrease.
In the larger context of research, there’s a “file drawer” effect where negative results never see the light of day. That means there’s a significant bias for publishing false positives. Without more research, I’m going to assume that this is a statistical fluke.
(Of course, I work in a field where I can rely on large effect sizes and large samples that are trivial to collect, so it’s not really fair for me to judge the epidemiologists…)
Couldn’t they be jumping to conclusions with the infection pathway theory?
Maybe it is something else related but not an infection pathway.
This post made me check out the Wikipedia article on the vagus nerve(s). Looking through the “Clinical significance” section … those nerves seem to have fingers in a lot of physiological pies.
Depression. Obesity. Heart arrhythmia. Tinnitus. Epilepsy. Parkinson’s. Probably more.
That’s astounding. Seems that severing those nerves would be hugely risky., yet is has been done and looks to be sometimes used in place of gastric bypass and such. I wonder how much of this stuff is tried-and-true medicine, and how much is experimental?
More here - study findings seem to be taken seriously
There is a school of thought in autism, mostly consigned to the woo/nonsense/gullible parent file, that some GI disruptions are related to the severity of the deficits. Many kids on the spectrum have “delicate tummies” and proponents claim that a carefully monitored diet (often lapping over into things like gluten-free and nonsense diets of the week) can show improvements in responsiveness and attention.
Maybe there is something to this idea. But waiting until there’s more than one questionable study is probably a good idea.
I think you nailed it. The results aren’t even statistically significant. And that’s after studying tens of thousands of patients.
Beyond that, although the results may be consistent with bacterial spread up the vagus, I would think there are other explanations as well. For example, PD is characterized by a deficiency of the neurotransmitter dopamine in a part of the brain called the substantia nigra. And that leaves the effect of acetylcholine in the substantia nigra unopposed. Since the vagus is a major source of acetylcholine outflow from the brain, maybe cutting the vagus somehow makes that imbalance worse.
At best, the study results are hypothesis generating. In fact, unless I am missing something, I’m surprised it was published in one of the better neurology journals.
Remarkable that the article survived review (apparently not statistical) without strong comments on limited findings). Given that the intervals contain the null value for the hazard ratio (1), these are null findings, which are consistent with a lack of correlation. However, if the point estimates are consistently below 1, the implication is that there is some sort of risk reduction in there, somewhere.
Key here is replication.
To clarify: as someone who does statistical review for candidate papers, I would be sure to make very strong notes about a paper piling up statistically nonsignificant results and then still claiming an association . Very surprised that a reputable journal would or this one through with the language that it used. Unless they hVe done GR CIs hidden somewhere that have upper limits below 1, no dice.
As I said just above, at best the study would seem to be “hypothesis generating”.
If you’re convinced that Parkinson’s “starts in the gut”, then this latest study might fan the flames of your enthusiasm.
Unfortunately we are far from being able to conclude that “the gut” is indeed the source of the disease, that an infectious agent is the trigger, or that the putative agent does in fact travel the Vagus Superhighway.
Edited to fix
Unless they have some hidden hazard ratio CIs with upper bounds below 1, no dice. I could see an issue of replication or hypothesis generation if exploratory analysis yielded statistically significant findings, but there doesn’t seem to be much here.
So to recap:
-
Statistically this is perhaps “suggestive” that vagotomy may reduce future risk of PD but is far from conclusive that it does so.
-
A fairly conclusive finding would be interesting and would generate multiple hypotheses. Vagotomy results in all kinds of stuff (higher heart rate and a lack of increased heart rate in response to low oxygen, for example) and parasympathetic tone dysfunction (of which the vagus nerve is part of) is already related to dementia is ways other than the gut. In fact one neurologist I know makes sure those training under him include need to urinate frequently as part of their neurological review of symptoms as that is a frequent early sign of PD. Increased salivation (also parasympathetic) is also an early sign.
-
Lastly if the very long jump to infectious agent traveling from the gut is made assuming that it is bacterial is another even longer leap. Interesting suggestive work recently that PD may in fact be a prion disorder.
The numbers reported in the abstract are not even suggestive, the point estimates are weak, and completely ignorable given the limits of the confidence intervals.
It’s as if the authors and reviewers agreed to ignore the numbers and just go with the gut hypothesis.
This is a quibble but there is nothing magical about the 95% CI. That is an arbitrary interval, less strong than 99% and stronger than 90%.
I personally would accept this as very weakly suggestive.
Yes, if for no other reason than the numbers all seem to be going the same direction (albeit none on its own being statistically significant).
Here’s a recent article from Cracked.com about why you should be suspicious of medical journal articles with results not confirmed in several other articles:
File away in your mind results like this and wait for there to be confirmation from the general medical community. If you go to a doctor and demand to be get a treatment based on this one article, the doctor will tell you that there isn’t enough evidence for it yet. He probably won’t even have heard of it. If you’re super-lucky he might know of studies for it that are being conducted currently that you might get into. Like any study, it’s iffy what would happen to you. You may get better and you may get worse. Otherwise, wait for a general medical acceptance of any new treatment.
I’ve always wondered whether many modern disease are caused by infections. Anything that appears mostly randomly in individuals and can’t be tied to a specific failure, is it an infection?
Huntingdon’s, hemophilia, sickle cell, etc. can all be tied to specific genetic deficiencies, and their appearance is predictably tied to the genetic lottery and parents.
Parkinson’s, Alzheimer’s, Type I diabetes, autism, schizophrenia - many of these sorts of diseases appear randomly, with no specific tie to a single cause. I wonder if they could be the long-term result of some infection, much as post-polio occurs many decades after the initial infection, or syphilis returns to cause senility.
You’re saying they had a gut feeling about it?
The hypothesis in context is that a GI infection travels up the vagus nerve to the brain, so yeah, gut.