Friday, June 11, 2010

The Hypothesis

The Hypothesis

In my travels through life and the internets, I’ve recently had a bunch of people ask me exactly what it is I’m doing, and why I’m doing it. Looking back at my initial explanation post, I realize that it doesn’t really do any scientific justice to my thoroughly unscientific experiment, at least not when it comes to explaining and justifying my course of action through the statement of a hypothesis for testing.

So here‘s the hypothesis:

By eating no starch and no fiber, I will put my Crohn’s Disease into clinical remission by reducing or eliminating the activity of Klebsiella Pneumoniae in my gut.

Ok! But what the hell does that mean? Why do I think this will work? What’s the end game? Let’s start with the most compelling piece of theoretical science, move to the only significant tests of that science, and then on to how (and why) to test it yourself if you have Crohn’s, Ulcerative Colitis or IBD/IBS.


(In the spirit of credit where credit is due, I must thank AJ for telling me about Ebringer’s work and taking the time to talk to me about the implications. I owe him a great debt of gratitude! AJ is also attempting a low starch diet with success in treating his Crohn’s.)

Alan Ebringer, a professor of Immunology at King’s College, London, lead a research team studying ankylosing spondylitis (AS), which is a condition of auto-immune arthritis of the lower back. They discovered that patients with AS tended to have anti-bodies in their bodies to a ubiquitous bacterium called Klebisella Penumoniae (KP). KP is present in everyone’s intestines, but if everything is working properly your body’s immune system should not be required to control it. In AS patients, something was wrong with the way their guts and immune system were handling this bacteria.

Ebringer and his team eventually developed the following hypothesis. I am probably simplifying this through my ignorance of the full meaning of the terms in the papers, but I think I have the gist correct:

90% of their patients with AS produced the HLA-B27 antigen. KP produces a particular enzyme to break down starch for digestion which happened to partially mimic the structure of the HLA-B27 antigen, as well as the structure of several types of collagen in your body. When you eat a large bolus of starch, not all of it will be digested before your KP has a chance to eat some. The KP will merrily eat and reproduce until the starch is gone, at which point they slowly die back until the next meal.

If you have good gut permeability (that is low gut permeability), and little damage to your intestines from your diet, chances are that the products of bacterial action will never cross the tight junctions in your gut and into your body. If, however, you are at all sensitive to gluten, lectin, or any of the other common allergens in the modern diet, or you have recently suffered trauma to your gut, you may have a compromised gut which will allow minute amounts of KP and KP byproducts to by pass the gut wall. Your body’s immune system will respond like it does to any foreign protein – it will being producing anti-bodies which are coded to attack that protein.

But that’s the catch. If you are HLA-B27 positive, your immune response may be confused and attack not only the foreign proteins, but any protein which mimic them. You are now experiencing an auto-immune response, and in AS patients, this manifests itself as chronic inflammation of the spine and pelvic joints.

Ebringer and his team noticed an association between AS and IBS, so they decided to see how Crohnies fared in terms of KP anti-body counts. They discovered that Crohn’s sufferers also had very elevated KP anti-body activity, but they were predominately HLA-B27 negative. So they went looking and discovered that KP also mimics several kinds of collagen which are helpfully present in your intestinal wall: right where you get the colitis that characterizes Crohn’s! They concluded that people who are HLA-B27 positive have a high propensity to develop AS if they compromise their gut, while people who are HLA-B27 negative would get Crohn’s, Ulcerative Colitis, and the other inflammatory bowel disorders.

Ebringer and company developed what they call the “low-starch diet” and prescribed it to their AS patients. Unfortunately, funding for their research appears to have dried up and Ebringer himself has retired. Presumably someone will be along to pick up this line of inquiry, but until then we’re in limbo in terms of further theoretical advances, at least that I’ve seen. I make no claims to being all knowing and all seeing so maybe I’m pleasantly mistaken and there is a full blown clinical trial going on somewhere!

In the absence of that miracle, however, we are left with no published clinical tests run by Ebringer to establish whether or not the low-starch diet would work for AS or Crohn’s patients. Which leads us to the next question.


There are two “tests” of this theory available, one strong, one weak. I’ll lead with my low card here.

The Specific Carbohydrate Diet was created by Sidney Haas in the 1920s in an attempt to address celiac, Crohn’s and other digestive disorders. It was fairly successful, but it was supplanted once the actual mechanism behind celiac was discovered. This probably buried Haas in the footnotes of celiac disease and the effectiveness of the diet for Crohn’s was lost until Elaine Gottschall wrote Breaking the Vicious Cycle. Since the book was published, many people suffering from various bowel disorders have reported success if they were very strict in following the dietary recommendations Gottschall laid out.

I call this my weak evidence because anecdotes are not data. There are also a substantial number of people who fail on the SCD – adherents might say that’s because the failures couldn’t stick to the program, a view which I tend to agree with, but which must be taken into account when applying proper skepticism. SCD also lacks a specific mechanism for how it treats the diseases it appears to alleviate, which makes it less useful when one is trying to synthesize disparate sources into an overall hypothesis or worldview. So the SCD is intriguing, but barring clinical trials will remain only intriguing for the mainstream.

For our purposes, the collective anecdote of the SCD community provides a partial test of the hypothesis. The SCD also allows invert sugar (honey) and nuts in the form of nut flours in the diet. I’m somewhat curious about the nut flours, since many nuts appear to have a fair amount of starch in them, and leery of the sugar for other reasons. I wonder if some of the SCD failures have been caused by people falling into the candy cigarettes trap of trying to fake themselves out with “bread” made from walnut flour.

On to the second test of the hypothesis. Wolfgang Lutz was an early low-carbohydrate diet promoter. He practiced in Austria in the 1970s, and wrote a book entitled Leben Ohne Brot, or in English, Life without Bread. The book is fairly standard low-carb fare until the end, where the authors detail how various medical conditions responded to low-carbohydrate diets.

When I hit this part in the book, I started skimming until I hit the gastrointestinal disorder section. That’s when I discovered a truly jaw-dropping graph for any Crohn’s sufferer. If I weren’t suddenly inspired to obey copyright law, I would reproduce the original here, but instead I’ll just describe the results: Lutz claimed an 80% remission rate after six months of nothing more complicated than low carb dieting.

That’s either complete bull or a very powerful, if overly broad, confirmation of the hypothesis. People following Lutz’s diet are allowed 6 “bread units” a day – basically, the amount of carbohydrate in six slices of bread, or 72 grams – and so one can imagine that the diet of the 80% who were successful contained plenty of starch. If so, then the hypothesis would actually be falsified, since the starch and fiber restrictions would be shown to be irrelevant, or at least overly strict. But that information is not forthcoming in the book and I doubt that it would be possible to obtain the food logs of patients from the 70s at this point.

We can’t fully trust the Lutz results because they were not verified via clinical trial. The patient records of a practicing doctor can be considered very powerful arguments in the justification of a trial, but are not themselves as compelling as the results of a well run trial. So again, we must remain skeptical.

Thankfully, we don’t have to leave it at that.

How (and why) to test this yourself

The essence of science is the testing of ideas by experiment. Here we have an eminently testable idea, one you can test in your very own home. You don’t need a particle collider to change your diet and observe the results. Here’s how to run the test:

1) Start eating according to the low-starch diet outlined by Ebringer. This may not be readily available on the internet so I’ll reproduce the basics here:

a. Eliminate grains, sugar, all starchy vegetables and legumes (basically anything that isn’t green and crunchy), fruits and nuts.

b. Replace the eliminated foods with fatty meat and non-starchy vegetables.

Your goal is to completely get rid of all the starch in your diet. All of it – don’t compromise, don’t wimp out, don’t pine away for an ice cream cone and cheat. Your mindset should be that of a boxer stepping into the ring or a marathoner taking the first step in the race – you are in a fight here, a fight where you don’t get to make many mistakes, a fight which quite literally might determine how the rest of your life plays out.

2) Continue this diet strictly for at least six months. Work with your doctor on your medications; most aren’t too happy if you just drop them cold turkey, and if you are on steroid therapy quitting cold turkey can be very dangerous to your health. I’ll say that again: if you are taking predisone or another steroid, do not quit cold turkey!

3) Let us know how it turned out!

Many people respond to the diet proposed with horror. No more cookies? No more cake? No more bread? Are you insane?

My response is simple: no, I’m not insane. You give up all of those things, but you also might get to throw your pills away forever. You get to take your health back forever. Rather than cycling through flare-ups and getting progressively more disquieting reports on your colonoscopies until you eventually end up with a colostomy bag, you can take control of your future and maybe avoid that fate. You can avoid a huge increase in risk of prostate and bowel cancers. You can avoid the late night sessions sitting cramped on the toilet wondering what the hell is going on and when it’s going to end, and the creeping despair inherent in the knowledge that tomorrow you’re going to have to pop several grams of medication just to stay alive.

I can’t promise anyone that this diet will work. I still don’t know if it will work. I can say with certainty that I do not have any symptoms of Crohn’s – I have no pain or diarrhea, I’m gaining rather than losing weight, I feel great – but even if my next colonscopy is so amazing that doctor thinks I lied about the diagnosis I still can’t say with certainty that the diet did it. I’ll probably live my whole life wondering. But at least that life won’t be lived as a slave to a bottle of pills, and maybe, if I’m lucky, enough people will try this diet and diets like it that the mainstream will take notice.

So if you’re still with me at this point, I sincerely hope you give it a shot. By way of offering support, feel free to shoot me an email and ask for help or let me know how things are going. I'm always interested in hearing from people who have tried a dietary approach - whether it worked or it didn't.

Good luck!


  1. Hey PFW--

    Josh, here, from PaNu forum. I did try the Gottschall SCD about ten years ago and followed it RIGOROUSLY. However, it did not help. I did eat lots of chestnut bread, though--lots. So there's one anecdotal report, for what it's worth.

    I am pondering going the Ebringer path. If so, I will post my results here. I am pretty much in U.C. remission (with azathioprine), so tracking results won't be simple...However, if I take baselines of fatigue, gas, and other symptoms, I think there would be measurable improvement if the diet were to "work."

    I don't take lightly the decision to do the diet. But, rest assured, it I do it, I will do it 100%.

  2. I'll take another anecdotal data point over nothing any day. And if you do try the diet, awesome and good luck :)

    I was in the same boat: remission via drugs before trying the diet. I stayed on the meds as I started the diet and more or less immediately went into flare up mode. I went from 160 on the scale down to 150 over a month or two, crapping my brains out all the time. Then things slowed down. By December I was back up to 160. Now I'm up at 170 with normal bowel movements being the norm unless I gorge on dairy - and no meds.

    So best of luck if you do give it a shot. I think the fact that you are in remission actually gives you a better chance of succeeding than if you were in an active flare - you're starting from a better position and so should be able to survive the adaptation period. Since I started in remission, my goal was maintaining that remission via diet rather than medications, so even if I had stayed right where I was it would have been a win. As it is though I actually improved.

    Also feel free to borrow my tracking spreadsheet and hack it up however you like. I find weight and bowel movement frequency/type to be a decent Crohn's indicator. I think if I had to do it all over again I would just use the Bristol Stool Scale to score each BM rather than the scheme I came up with on my own.

  3. Minor nitpick from a German nitpicker: It should read "Leben ohne Brot" not "Leben Ohne Bröt".

  4. This is why you should double check yourself when remembering high school German vocabulary lists. Fixed!

  5. Have you considered a fecal transplant? It might be hard to find a doc to do it, but it's apparently very effective. Protocol is something like laxatives to clear out the bowel contents, and then bacterial colonies from a non-Crohn's sufferer are placed into your lower intestine by coloscope.

  6. This comment has been removed by a blog administrator.

  7. This comment has been removed by a blog administrator.

  8. This comment has been removed by a blog administrator.

  9. This comment has been removed by a blog administrator.