Three Steps To Avoiding AVS Failures

[Guest guest]

In the wake of my failed adrenal vein sampling at Atlanta's Piedmont Hospital last month, I went looking for some explanations of why the failure happened. What I found was a depressing rehash of the usual reasons for American medicine's notoriously high rate of medical error, but I have narrowed it down to three factors that can be critical for others who are considering the procedure.

First let me thank Dr Grim for opening my eyes to the failure. As I commented once before, I am reasonably confident that I and/or my endocrinologist would have noticed that the numbers could not possibly have represented success, once the initial excitement of receiving the results had passed. I, for one, I require complete peace and quiet for certain insights to bubble to the surface. I came home and posted in the flush of excitement and optimism. My past experiences should have warned me that all is not always what it seems, but I took it for granted that my doctors would not have presented the results as successful without properly vetting them. This incident does emphasize the importance of experience, though. In one glimpse, Dr Grim saw that the results did not pass the " sniff test. " Other doctors with more than a glimpse of the results failed to recognize what the numbers plainly told.

I certainly haven't heard anything from the doctors involved in the failure. I informed them of three independent reasons the test could not possibly be construed as a success. All I have heard in response is crickets. Nobody owns a failure in U.S. medicine. Alden who? There is really no incentive to own it. Everyone got paid. For the radiologist, it was a frustrating hour out of a drizzly day. It is only for me that it was a careless, life-altering waste of my sole opportunity for this procedure. We maintain our illusion of the superiority of American medicine by comparing our best performances and achievements to everyone else's average performance. We gloss over failure as unrepresentative. Like poverty and ignorance, it goes largely unseen. It doesn't enter into our self-image as a nation.

I explained to my doctors that (1) a world-renowned expert deemed the test to have failed the immediate sniff test because the right adrenal and inferior vena cava numbers were too close; (2) more formally speaking, while the requisite 4x gradient existed between the peripheral cortisol and the supposed right adrenal cortisol, the results lacked the requisite 2x gradient between the right adrenal cortisol and the inferior vena cava cortisol, showing less than 3% instead; and (3) on the basis of fluid dynamics alone, if my right adrenal were completely inert, one might plausibly though surprisingly have seen a coincidence wherein the right adrenal aldosterone exactly matched the peripheral aldosterone, but chances were at least thousands-to-one against a correct result correctly yielding the exact match my results showed between the right adrenal aldosterone and the inferior vena cava aldosterone. Clearly the right adrenal sample was really an IVC sample. Response, as mentioned above: crickets.

The failure could have been averted by adherence to well-documented best practices. I identify the three most important best practices that were ignored in my procedure.

Part of the problem is that AVS has just recently become the procedure du jour for interventional radiologists. Everyone wants to be able to say, " Oh yes, I've done a few of those. " Indeed that is almost the exact response a number of radiologists outside my circle of caregivers have given on hearing that I underwent the procedure. AVS has spread enough to appear on the radar of interventional radiology departments across the country but not enough for most to have accumulated any significant experience in it. Many interventional radiologists have probably just attended a seminar and watched some video of the procedure -- and are itching to get the opportunity to try one.

NUMBER ONE: Virtually every well-cited article emphasizes the critical importance of channeling all procedures to one or at most two practitioners within the group. This did not happen in my case. My endo had painstakingly chosen a radiologist very experienced in AVS. I have since come to doubt that assessment, but it didn't matter because the radiologist was switched at the last moment. I was informed as the procedure was beginning. The substitute doctor said not to worry, that " We all do it. " Actually, that was his non-answer to my question, " How many of these have you done? " So in hindsight I am guessing I may have been his first. Now this is a group with at least 24 radiologists. And they all do lots of AVS and are expert in it? I don't think so. This was a clear and fateful failure to follow documented best practices. If you are going to have an adrenal vein sampling done locally, wherever you live, you must ensure the procedure will be done by the one person (or one of two) who handles ALL AVS procedures for the practice. Get it in writing. (As this procedure becomes more common, it will eventually become impossible to get it scheduled with the Mayo or NIH specialists who have performed hundreds of them.)

NUMBER TWO: Most of the experience-based articles I've read emphasize the importance of the radiologist studying all prior venography before beginning the procedure. I know that I was not asked to sign an authorization for release of medical records before my procedure, so it seems most likely my radiologist did not study my adrenal CT images taken just a couple of months earlier at a hospital about 3 miles away. Here again we see overconfidence, the most common source of medical error in the U.S., in the assumption that everything would become clear in real-time fluoroscopy.

NUMBER THREE: I've read a number of reports emphasizing the importance of using rapid intra-procedural cortisol assay during the adrenal vein sampling to ensure that the catheter is actually sampling right adrenal blood before locking in the aldosterone sample. One cannot simply rely on the experience of the radiologist. No matter how experienced the radiologist, a certain percentage of people have anomalous venous configurations in which the visually most plausible vein will not be the vein that is actually draining most of the adrenal output. Daunt's excellent article from Australia's Greenslopes Private Hospital (Adrenal Vein Sampling: How To Make It Quick, Easy, and Successful) describes and systematizes the varieties of less-common configurations seen in the nearly 1,000 procedures done over a ten-year period at that hospital. An article by Mengozzi et al at the University of Torino details a careful analysis of AVS performed both with and without real-time cortisol testing. Those authors observe that some people have more than half a dozen visually plausible veins to sample, of which normally only one will yield correct results. They further calculate a comparative cost of USD 55,000 over all AVS in their study for the addition of rapid cortisol assay versus USD 220,000 in AVS failures if a vein had been chosen without rapid cortisol assay. The cost argument is compelling. Drs. Auchus et al at the University of Texas make a very similar argument, pointing out that most radiology groups cannot or will not devote a single radiologist to the procedure and that it is therefore essential to use rapid cortisol assay. And as I note above, even WITH a single sub-specialist doing all AVS in a practice, there are venous configurations where fluoroscopy alone will lead to the wrong choice of vein. You are likely to encounter resistance from experienced AVS practitioners who think they're " too good to use training wheels. " But if you have an unusual venous configuration, you could well become one of even these most experienced doctors' occasional failures. Insist on the rapid cortisol assay.

With these three best practices:

1 channeling all AVS to one or at most two practitioners in a practice

2 studying all previous adrenal venography

3 using rapid cortisol assay during the procedure to assure correct selection and cannulation

....it should be possible to get correct results in nearly any major metropolitan area. But you have to negotiate for these things ahead of time and get them in writing.

I foolishly relied on the assumption that a premier hospital like mine would adhere to best practices. Yet my radiologist overconfidently ignored point 1 and 3 for certain and probably point 2 of these requirements, and thus my result became a predictable and avoidable failure. You cannot rely on " name " and " reputation " alone. You must insist on adherence to best practices. You may very well get only one chance at this procedure, unless you are able to pay $10,000 - $30,000, plus external costs (additional thousands in my case), out of pocket for a do-over -- or pay a lawyer to wrest one from your insurance company. I said earlier that " nobody owns a failure. " But actually, that's not quite accurate. The truth is that we, the patients, inevitably own the medical system's failures.

One simply must be a royal PITA to avoid becoming another statistical casualty in the assembly-line world of American medicine.

Alden G