-

December 22nd, 2012

Chaos in the Diagnosis of C diff, and Dogs are Amazing Creatures

If you’re confused about the best way to diagnose C diff these days, welcome to the club. There are all kinds of tests out there, and no uniform approach between labs. Our lab actually does three tests — and will do a fourth (the classic cytotoxicity assay) if you request it.

The result? Chaos, confusion, and lots of queries.

For an example, here’s an email from a colleague:

Hey Paul — my patient had diarrhea after antibiotics, so I sent a C diff. Not sure what to tell her about the result, which I’ve pasted below. (She’s better now by the way with no treatment.)
Audrey

Specimen Type: STOOL
C. DIFFICILE ANTIGEN/TOXIN ASSAY – Final
ANTIGEN:POSITIVE; TOXIN:NEGATIVE
RESULTS PENDING PCR ANALYSIS.
TOXIGENIC CLOSTRIDIUM DIFFICILE QUALITATIVE BY PCR – NEGATIVE

My response:

Audrey — the lab does two rapid tests initially — antigen and toxin assay.  She had a positive antigen (indicating the presence of the organism) but a negative toxin (not all C diff produces toxin).  So the PCR was done as a more sensitive and specific test for toxin, and this was negative. Strongly suggests she does not have active C diff. No treatment required.  Paul

Yes, this is tricky. Some labs go right to PCR as the first step, but this is more expensive — but it sure would eliminate the confusion.

Or we could just start using dogs. As demonstrated in this paper, and summarized in Journal Watch ID by Stephen Baum, a trained dog does a terrific job of diagnosing people with C diff, even without a specimen:

A 2-year-old male beagle was trained to identify the odor of toxin-producing C. diff and to sit or lie down on detection of this scent. In preliminary testing involving 50 positive and 50 negative stools, the dog’s sensitivity and specificity were 100% …[In further testing done in the hospital], the dog correctly identified 25 of the 30 case-patients (sensitivity, 83%) and 265 of the 270 controls (specificity 98%).

The most amazing thing about this study is not only the performance of the poop-sniffing dog, but the fact that he most likely enjoyed doing it. He’s a dog, after all.

Don’t believe me? Just Watch these two video clips. First, one from the BMJ paper (let’s hope that the beagle took his HIPAA training class seriously):


And second, this Cat vs. Dog short, which clearly demonstrates (around 1 minute in) why a dog is eager to help us diagnose C diff:

7 Responses to “Chaos in the Diagnosis of C diff, and Dogs are Amazing Creatures”

  1. Cathy says:

    Can a dog like Cliff get sick from coming into contact with C Diff? Can he spread the bacteria?

    Meantime: go, dog, go!

    Thanks for answering my questions –

    Cathy

  2. cs says:

    The dog is amazing and adorable, but I can’t believe the people in the lab were not wearing gloves!!

  3. Paul Sax says:

    Can a dog like Cliff get sick from coming into contact with C Diff? Can he spread the bacteria?

    1) Nope — C diff shouldn’t be a problem for Cliff, just like it’s not a problem for most of us who aren’t taking antibiotics (it’s a real opportunist); 2) Dogs can only spread the bacteria if they get it on them! Presumably these researchers were not letting the patients with C diff come into physical contact with Cliff, then having him mingle with the others …

    Paul

  4. Carlos del Rio says:

    Would love to have Cliff available to round with me. Wouldn’t it be fun that the ID team would now include a dog! We would be loved by many in the hospital.

  5. Sarah says:

    Fantastic post. I’d never even heard of C Diff before but what a remarkable, innovative, and adorable way to diagnose it!

  6. Loretta S says:

    When I originally read this story in Journal Watch, I thought, wow, that’s a dog’s dream job! :) Of course, I sent a link to all my family members, most of whom refused to open the link, because “that’s gross”.

  7. Christian Ramers says:

    Paul,
    This story is phenomenal, heartwarming, cute, and in a strange way quite innovative. I’d definitely like to see a cost-effectiveness study putting Cliff up against antigen, culture, and PCR methods. It would be hard for Hospital Administration to say no! I do wonder how long canine ID fellowships will be though, and whether we will run into duty hour restrictions.
    Christian

Leave a Reply

Note: This is a moderated forum. By clicking on the "Submit Comment" button below, you agree to abide by the NEJM Journal Watch Terms of Use.

HIV Information: Author Paul Sax, M.D.

Paul E. Sax, MD

Editor-in-Chief

NEJM Journal Watch HIV/Aids Clinical Care

Biography | Disclosures | Summaries

Learn more about HIV and ID Observations.