I started this blog on May 15th. On May 24th came my first post about my trials and tribulations with trying to get a celiac diagnosis. Well, LOOK AT THIS:

Pic of the day - At Last!!!

OK, now between May and now, there have been a lot of changes. The big one has been that my primary care doctor moved to another country. That is kind of a biggie when it comes to moving forward with any kind of health questions. 😉

I was assigned to a new doc, and wasn’t any too sure about this, since I knew nothing about the person. I was thinking about possibly switching clinics, since I knew more primary care folk I don’t trust at my current clinic than folks I do trust, and there are a couple who I wouldn’t trust to throw me a lifesaver if I fell overboard in a boat. They would either think they knew a better way, or they’d be busy asking other people’s advice and not listening to me. So I’ve been judiciously reserving judgment.

I asked around a lot. Many people told me that the doc I’d been assigned to was mentored by my previous primary care doc (who is the first doc I trusted and who has earned that trust by both listening and saving my life a couple times). People told me the new doc is geeky, like me; that he’s interested in e-patients, like me; that he’s open to using email for communication about minor issues; etcetera. So I made a “new patient / establish care” appointment, which takes a while to get. That was today.

I was very careful preparing for the appointment. After all, I have longstanding complicated unresolved health issues. I’m a participant in this PGEN study, and some docs really aren’t into that. And I’m aggressively self-educated about my own health issues. I’m a medical librarian. I’m a really GOOD medical librarian. I specialize in systematic review searches. I have both co-taught on this topic with Cochrane folk, and am on Cochrane review teams. Docs will tell you there is no worse patient than another doctor, or another healthcare provider. Well, they might make an exception to that guideline for me, and if not, I’m darn close to being just as awful.

A few months ago one of the docs I work with was chatting with me. I was rambling on at fairly high speed about some of the new research in her area of interest, or something like that. She hadn’t seen it yet. When we changed topics, I did the same thing in the new topic. She just looked at me with a combination of humor, horror, and awe in her eyes, and asked, “What kind of doctor has to deal with YOU as a patient?” I told her, and she said, “Thank goodness! He can handle it!” Of course, that was before I found out he was moving to another country. Oh. A new doctor? Oh. Uh oh.

For this first meeting, I established my priorities, and clustered them: celiac/skin; MTHFR/folate; osteoporosis/HRT. I made notes in my phone. I printed off a highly selected subset of pages from the genetic studies. I brought copies of a couple other articles just in case. I researched the available expertise on celiac in the system. I hunted down the newest clinical guidelines for diagnosis of celiac. I reviewed the guidelines. I reviewed the articles by the only in-system clinician who remotely comes close to being a celiac “expert.” I made two copies of everything I planned to share with him. I went through the list, & cut it in half. I then hid my copies, and my extra back up single articles. I debated bringing my box of vitamins, but finally decided to. The nurse was glad, which made it ok.

The doc had a student with him today, an MD/PhD fellow. She and I had a lovely chat, but it became rapidly clear that this was too complicated for the amount of time available, so she passed the baton along to him. Because of the time crunch, the conversation was rapidfire, with lots of interruptions back and forth. It wasn’t the best communication I’ve ever had with a doc, but it was far far from the worst, also.

His top priority was the osteoporosis. That’s fine. So we talked about bisphosphonates for a while, and why I don’t want to take them. It’s a risk/benefit issue. I’m not persuaded that I’m badly enough off for it to be worth it to me to take them. I could bore you with the details, but to make it short, this is an area where I’m not an expert, but I know a lot of folk who are, and I’m able to talk about bisphosphonates moderately intelligently. What was important for me was for the doctor to believe that. I was able to show that I’d read some of the same literature that he had, and this helped (I hope) establish some sort of credibility baseline for me. (The genomic content came into play a few moments later, with the doc deciding that the potential value of the HRT is just not worth the potential risk of DVT, given my genetic scan results, and he has me tapering off it.)

Then he was willing to take a look at the genomic info I’d selected out. Some of these we’ll have to deal with later. He and I agreed that most of what healthcare currently knows about MTHFR is suspect and insubstantial. He asked why I’d decided it was important. I pointed out some superficial similarities between my body type and the more severe form of MTHFR; showed him my brief distillation of Pubmed searches on emerging trends in MTHFR research; and then emphasized that a major driver was my constant craving for foods high in folate. Since he was already doing bloodwork, he agreed to add in the baseline test for a few vitamins and minerals.

We have a long talk about the skin issues and antibiotics and dermatitis herpetiformis and rosacea and celiac … what to do, and why a diagnosis matters if it doesn’t change my determination to remain gluten free. My computer crashed so I’m going to just wind this up now and perhaps continue another day. The end point is that the genomic tests DID change my healthcare practice. I walked away with a piece of paper that says I have a celiac diagnosis; we are removing the HRT; and I have the docs OK to start folate supplementation once I have my labs drawn. Pretty good starting point!


8 thoughts on “AT LAST!!

  1. Pingback: It’s … Complicated. (A.K.A. “MTHFR Deficiency, Sequel”) | PGen Participant

  2. Since this blog entry was written 1.5 years ago, I don’t know if you’re still uncertain about your celiac/gluten sensitivity diagnosis. If you are, send me an email, and I can give you the rs numbers on your 23andMe results that will tell you if you have celiac disease and possibly gluten sensitivity.

    The criteria that 23andme uses to establish if someone has celiac disease is very limited. There are additional SNPs that should be considered in the determination. 23andMe tests for those SNPs, but they don’t look at them in saying whether someone has celiac or not. 23andMe gives an explanation of how their health report for celiac is generated, and they only consider the two most common SNPs for celiac as indicating celiac disease. One has to look up their raw data on 23andMe (using the Browse Raw Data feature) and then do a search on these particular SNPs to get a comprehensive determination.

      • According to, these are the risk SNPs for the classic celiac conditions:
        DQ 2.5 = rs2187668 (T). DQ7 = rs4639334 (A). DQ8 = rs7454108 (G)

        One form of non-celiac gluten intolerance (NCGS) is HLA DQB1*0202. The SNPs necessary for DQ 2,2 are rs2395182 (T), rs7775228 (G), and rs4713586 (T). If rs2395182 (T), rs7775228 (G), and rs47135586 (G), then the person is DQ4, which is not a gluten intolerant condition.

        23andMe tests all of the above SNPs except rs4713586, which is the SNP that excludes someone from the category of being DQ 2,2.

        Most of the CD patients who do not carry DQ2.5 or DQ8, carry half of the DQ2.5 or DQ2.2 (that is either HLA-DQA1*05 or DQB1*0202) suggesting that carrying part of the risk molecules has functional implications for the risk of CD.

        I did a genetic test with in 12/2011, and they determined I was homozygous for HLA DQB1*0202. 23andMe says I am homozygous for rs2395182 and rs7775228, so the two different testing methods agree. I am 67 now and stopped eating any gluten-containing food in 12/2011. My life-long joint aches have disappeared as well as any bloating. I was diagnosed with osteoporosis 10 years ago, but I haven’t done any follow-up scans since 2008, so I don’t know if that’s changed any.

        I believe there are additional SNPs that could be considered NCGS. And of course, just because someone has the mutated SNPs doesn’t mean that the SNPs have been expressed and caused a gluten intolerance.

      • Curious. 23andMe calculated my increased risk on a different collection of SNPs, with only the HLA DQA1 in common. I blogged them here
        * HLA-DQA1 (SNP rs2187668)
        * 4q27 (SNP rs6822844)
        * 3p21 (SNP rs6441961)
        * 3q28 (SNP rs9851967)

        Of the additional celiac SNPs you mention, my raw data didn’t include one at all, and the other was fine. For the NCGS SNPs you mentioned, one of those also doesn’t show up in my raw data, and the other two are both homozygous TT.

        I’m starting to see how I and my son can have ~4-5x the normal risk of celiac, and my Dad can have 17x!

        I found this helpful page on SNPedia:

        I went ahead and dug through the raw data for ALL of the ones mentioned, but they didn’t always say which [AGCT] is the risk.

        rs13119723 AA
        rs13119723 (G) AA
        rs2187668 (A) CT
        rs2187668 (T) CT
        rs2305764 AG
        rs2395182 TT
        rs2395182 (T) TT
        rs3184504 CT
        rs4639334 > not found
        rs4713586 (T) > not found
        rs6441961 CT
        rs6822844 (G) GG
        rs7454108 TT
        rs7775228 TT
        rs7775228 (G) TT
        rs917997 CC
        rs9851967 CT
        rs9851967 CT

      • I don’t know if anyone has taken issue with the Monsuur article I linked to, but their statistics for successful diagnosis using their criteria seemed compelling. 23andMe seems to have used a very different approach.

        I should have also noted that my raw data didn’t show results for the DQ7 SNP on 23andMe.

        I went through your list of SNPs found at SNPedia. My two known GS221 +/+ SNPs for NCGS DQB1*0202 showed up there, no surprise. I was -/- for all but one of the other SNPs, rs917997, IL18RAP, which is weakly linked to CD. I’m CC, which appears to be extremely rare.

        SNPedia has the wrong risk allele for rs6822844. The population diversity data for shows that over 94% of all populations studied had the GG allele. I’d say T must be the risk allele. But I didn’t read the PubMed links.

        When I’m trying to figure out if I have a risk allele or not, I always look at the population diversity graphics at dbSNP, which can be accessed at the dbSNP Lookup Feature associated with every SNP listed by 23andMe. If the population diversity for a particular allele in the European population is 10% or less, I figure there’s usually a risk associated with it. I haven’t seen my theory confirmed by anyone though.

        So do you know what your HLA-DQ A1 and B1 groups are? Do you have yourself pigeonholed into a CD category yet?

  3. Looking at the population diversities given for each SNP at dbSNP, these are the alleles with a very minor proportion in European populations, and thus what I would consider to be the risk alleles. I put the risk in parenthesis, and your alleles to the right:
    rs13119723 (G) AA
    rs2187668 (A=T) CT
    rs2305764 (C) AG
    rs2395182 (G) TT
    rs3184504 (T) CT
    rs4639334 > not found
    rs4713586 (T) > not found
    rs6441961 (C) CT
    rs6822844 (T) GG
    rs7454108 (C) TT
    rs7775228 (G) TT
    rs917997 (C) CC
    rs9851967 (T) CT
    I differ from what you indicated for the risk allele for rs2395182 and rs6822844. A considerable majority of the European population sampled has the alleles you had indicated as the risk alleles for these two SNPs. Doesn’t make sense that those would be the risk alleles and not the alleles with very minor representation. SNPedia has some bad information.

    For rs2187668, you do have the risk allele. This is the HLA-DQA1. The population frequency of your CT allele in European populations is .159 and .183, out of 1, of the two Euro populations samples. According to the Monsuur et. al study I linked to, this would be sufficient to classify you as DQ 2.5, the most common form of classic celiac disease.

    The population statistics don’t support the other 3 risk alleles 23andMe gives for you as being risk factors. For Euros, the CT allele for rs3184504 is the most common, with rates of .455 to .583. For rs6441961, CT has rates of .3 to .6. For rs9851967, CT is .3 to .57 in the populations given. No Euros samples were listed for that one. I’m beginning to understand why people questioned 23andMe’s health reports. I don’t understand how they arrived at their risk statistics.

    As you probably know, the blood antibody tests for celiac disease don’t show positive until nearly all of the intestinal villas have atrophied. It’s really a disservice that doctors rely on them to rule out CD. The “gold standard” intestinal biopsies aren’t much better at catching CD before there is considerable damage. You might have fared better diagnostically with a test of a stool sample such as Enterolabs offered.

    But genetic tests rule supreme, in my opinion. My Enterolab stool test showed me one point below the cutoff level for tissue transaminase antibodies. It was only 6 years later when I ran their genetic test that I had confirmation of my NCGS. And avoiding gluten has eliminated a lot of joint pain for me. Having a definitive test of NCGS makes it so much easier for me psychologically to avoid wheat.

  4. People homozygous AA (or TT) for rs2187668, defined to be part of the classic HLA DQ2.5 celiac disease, are exceedingly rare, according to dbSNP. The authors of the article you linked to on 3/12/13,, didn’t have enough AAs to do a case study on them. They concluded that just having one risk allele, the A (or T), is sufficient for someone to be celiac.

    However, many people heterozygous for rs2187668 do not have celiac. The authors said that people heterozygous for the rs2187668, the HLA-DQA1, often have another predisposing SNP, particularly the rs9275141, risk factor T I believe, which tilts them into celiac disease. 23andMe does test for that one.

    They also mention the possibility of other rare HLA DQB1 SNPs. Since your symptoms are unusual, if you don’t have the rs9275141 risk SNP, perhaps you have one of those rare DQB1 SNPs.

    I tried to find where they identified the risk allele for rs6822844, but I couldn’t.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s