Index Home About Blog
From: "Howard McCollister" <>
Subject: Re: Sigmoidoscopy/Serology for IBD detection?
Date: 10 Jul 2004 14:29:56 -0500
Message-ID: <40f042fe$0$56371$>

"Brad Edwards" <> wrote in message
> I'm curious, have any of you had a sigmoidoscopy to diagnose Ulcerative
> Colitis (UC)?
> The procedure is similar to the colonoscopy, but only shows 1/3 of the
> colon. The advantages are that it only takes 10 minutes, is safer,
> doesn't require much prep, no sedation, and can be done in the doctors
> office. Biopsies may also be taken.
> UC almost always involves the lower colon and rectum, so it can be
> diagnosed with a sigmoidoscopy. I haven't compiled the numbers, but it
> seems that if a patient is showing symptoms of an IBD then a combination
> of pANCA / ASCA (IgA/IgG) serological tests with a flexible
> sigmoidoscopy would catch a majority of IBD cases. This combination is
> the most convenient, and could potentially spare people from undergoing
> more elaborate tests.
> When I saw a gastroenterologist I asked him about the test, and was
> willing to do it right then and there, but he refused. I don't recall
> the specific reason why, but I think he said something about how it'd be
> useless since he wouldn't be able to tell what type of inflammation it
> was if he saw any. But since biopsies are possible, I don't understand
> the problem.

Visualization of the whole colon and terminal ileum are important in
distinguishing ulcerative colitis from Chrohn's disease and other, more
non-specific colitides. Biopsies *might* be helpful *if* they show any signs
of granulomatous change, but you still wouldn't know if it were Chrohn's
disease or ulcerative colitis, and those biopsies usually aren't helpful.
And the association of ulcerative colitis with colon cancer makes
colonoscopy all the more important.

By and large, sigmoidoscopy is a useless test.


From: "Howard McCollister" <>
Subject: Re: Sigmoidoscopy/Serology for IBD detection?
Date: 10 Jul 2004 21:41:56 -0500
Message-ID: <40f06b6d$0$346$>

"Brad Edwards" <> wrote in message
> I think that if you saw obvious signs of UC with the sigmoidoscopy, that
> it could help rule out Crohn's.

What would "obvious signs of UC" be? How does ulcerative colitis look
different from Chrohn's disease of the colon?

Really, I hope you can tell me, because I've done close to 10,000
colonoscopies and I don't think I could tell the difference on


From: "Howard McCollister" <>
Subject: Re: Sigmoidoscopy/Serology for IBD detection?
Date: 11 Jul 2004 08:25:07 -0500
Message-ID: <40f13f17$0$43462$>

"Brad Edwards" <> wrote in message

> Continuous lesions? There are a lot of clues based on statistics that
> could help you differentiate one form from another, though it would be
> somewhat of an art and not always accurate. But that's not what I'm here
> to debate, I would've preferred you respond to everything I said besides
> that one sentence.
> I'm not doubting the inefficacy of sigmoidoscopy for differentiating
> IBDs. What I am interested in, is analyzing the choices in diagnostics.
> The chances of complications involving endoscopy is much higher with
> colonoscopy vs sigmoidoscopy.
> The current trend seems to be highly in favor of jumping into the
> colonoscopy without bothering with less invasive procedures. My feelings
> on why this is going on is that from the physicians perspective, why
> should they have to piece together incomplete data from various less
> elaborate tests and attempt to use their brain to make an assessment
> when they can simply do a colonoscopy and dive right in, physically
> seeing the whole picture. I can understand why this would be tempting
> when you have tons of patients and you don't want to take the time to
> feel around in the dark, hoping to find the light switch.
> You've done 10,000 colonoscopies, so you've had what, around 30
> perforations? That's pretty good, but what if nothing was found by doing
> those colonoscopies, and they go home with a perforated colon and an IBS
> diagnosis. Why place a patient in a higher risk than necessary? If it is
> unknown whether or not they have an IBD, what is wrong with using
> simpler approaches to first determine this prior to diving in with the
> big guns?

Doing a colonscopy for ulcerative colitis is nothing more that applying the
most accurate tool for someone who comes in with cramping and bloody
diarrhea. Continuous lesions? How would you know unless you do a
colonoscopy? I've seen many cases of inflammation in the rectum and sigmoid
colon, with skip lesions elsewhere and out of the reach of the

Perforations? I've had two, both in the sigmoid colon. The sigmoid is the
toughest part of the colon to traverse, and by FAR the most painful. Once
you get by the sigmoid colon, the risk of perforation is very, very low.
Your perforation assertion doesn't hold water, especially relative to the
information the extra 4 feet of colon might give you.

The problem with your light switch analogy is that we're really not talking
about turning on the light. We're talking about someone's health, and
possibly colon cancer. You want doctors to forego the most accurate test
available for the diagnosis of colon pathology so they can indulge in some
sort of inductive game. You want your doctor to gather as much
non-colonoscopic information as possible, then make a guess and treat the
patient based on probabilities. You suggest, therefore, that going all the
way and getting a colonoscopy is somehow "cheating", like skipping ahead in
your math book to get the answer. When you "piece together incomplete data"
from a variety of "less elaborate tests", your risk of misdiagnosis goes way
up. It has nothing to do with being lazy, or too busy. How much work do you
think is involved? How much brain power is required to piece together such
data? It's what doctors get paid to do every single day.

The gain from "piecing together incomplete data" simply to avoid colonscopy
does not justify the risk of misdiagnosis.


Index Home About Blog