Oct 04 2007

The NEJM Takes On Lyme Quackery

I am a practicing neurologist in CT, which is practically at the epicenter of Lyme disease in the US (in fact, the disease is named after Lyme CT where it was first described). So I have seen my fair share of Lyme disease and a form a late Lyme disease known as chronic neurolyme. It is without a doubt a real and serious disease. However, it is also the basis for what I consider to be a fake disease, chronic Lyme. I was therefore very happy to see an excellent article in the New England Journal of Medicine critically analyze this troublesome fake syndrome.

Throughout history there has always been a number of popular fake diseases or syndromes – although the specifics have changed over time. Sometimes the disease is completely fictional, like electromagnetic hypersensitivity. Sometimes the diagnosis is really just a name being applied to a laundry list of non-specific symptoms, like “asthenia,” (a diagnosis popular a century ago). And often the fake diagnosis is based upon a real disease but overextends the diagnosis to incorporate those without the specific symptoms of the disease but who have nonspecific or vague symptoms. In this latter category I would place chronic fatigue syndrome, fibromyalgia, and chronic Lyme.

The phenomenon of the fake diagnosis is driven by several realities, most importantly that people often have symptoms for which there is no established diagnosis or where the diagnosis is undesirable. People often suffer from what we call “symptoms of life,” meaning common aches and pains that plague everyone, increasingly as we get older. Often people can get joint or muscle pain, headaches, fatigue, mild forgetfulness, and vague bowel symptoms without there being a specific underlying disease or disorder. The human machine is imperfect, it wears out over time, and these minor ailments are an unfriendly reminder of our frailty and mortality.

These symptoms can range from annoying to debilitating, and it is understandable that people want to minimize and avoid them. It is reasonable to seek a professional opinion when a new such symptom appears, and to seek symptomatic relief and also to pursue methods to stave off the onset of symptoms.

But sometimes people are not happy with this approach. For some reason they are convinced that there is a specific underlying disease attacking their body and this disease must be diagnosed and cured. Sometimes this is in fact the case, but when all such possibilities are ruled out and the symptoms are non-specific and do not point to a particular disease, the simple explanation of “wear and tear” is appropriate. Unwilling to accept this, many patients (often referred to as the “worried well”) create the demand for a label to attach to their symptoms of life, and thus are borne the wide variety of fake diagnoses to meet this demand.

Another source of this demand are those patients who do have a specific underlying disorder, but it is unrecognized, or they are unhappy with the answer. A common underlying problem is sleep disorders. Chronic sleep deprivation can cause fatigue, muscle aches, and even poor concentration and memory. Also, depression and anxiety can lead to the onset or exacerbation of common symptoms but some patients may find these diagnoses unacceptable or stigmatized, or may misinterpret them as a dismissal of their symptoms.

In the early to mid 20th century one of the more common fake diagnoses was tertiary syphilis. Obviously syphilis is a real disease, and it does have a chronic form that can involve the nervous system. It is also a great mimicker, meaning it can manifest in many ways. This is a good setup for a fake diagnosis – eventually any bizarre or non-specific syndrome was blamed on syphilis. It was the favorite diagnosis of quacks of the time, and the target of many snake oil claims.

It is probably not a coincidence that Lyme disease is also caused by a spirochete (a type of bacteria), as is syphilis, and that both diseases have a late neurological phase, and that both have a wide range of possible symptoms. What happens is that cases are reported where a patient has some weird syndrome and goes undiagnosed for a time and in the end it turns out they had syphilis (or Lyme disease) the whole time. This leads clinicians to think – “Hey, maybe my patient with a weird and undiagnosed list of symptoms also has syphilis/Lyme disease.” (Or, of course, the patient may follow this line of reasoning themselves.) It’s really that short a trip to the creation of a fake diagnosis. Throw in a little confirmation bias, a couple of logical fallacies, and you’re home.

With Lyme disease the story is very typical and instructive. Lyme is an infectious disease, so there should be some symptoms and signs of infection. There are also specific laboratory tests that can confirm the diagnosis – a Lyme titer, which looks for antibodies to Borrelia burgdorferi (the spirochete that causes Lyme in the US) or a Lyme western blot – a more specific antibody test. Finally, there are antibiotics to which B. burgdorferi is sensitive, and if treated aggressively and long enough should eradicate the infection.

But the Lyme phenomenon, like syphilis before it, has taken on a life of its own, propelled by patients in search of a diagnosis and by misguided or unscrupulous clinicians. There are many people walking around today with the label of Lyme disease who do not have, and never had, any signs or symptoms specific for infection or of Lyme, who have a negative antibody titer, and who have not responded typically to antibiotics. Then how can anyone say they have Lyme? Good question.

Physicians who have made this diagnosis have fallen prey to the “invisible dragon” fallacy – or systematically rendering a hypothesis (in this case a diagnosis) unfalsifiable. What they are saying, in essence, is that their patient has an atypical presentation of Lyme that lacks all the specific signs, that it is seronegative (the antibody tests are negative) and it is resistant to standard treatment. I submit that Occam’s razor would favor the alternative hypothesis that the patient simply does not have a Lyme infection. This NEJM review article also comes to the same conclusion.

The story of Lyme, however, has been made more complex by the modern quack innovation of the fake diagnostic test to support the fake diagnosis. For example, there are labs that will run their own Lyme serological tests that show Lyme where none exists. Their tests have not been validated, or they use low thresholds for positivity that are guaranteed to cause false positives.

Another example is the SPECT scan – single positron emission computed tomography. This is a scan that creates a map of blood flow to the brain. It has legitimate uses, but some Lyme true believers can see in the blobs of computer generated color the telltale signs of neurolyme disease – again without any validation. How more difficult is it for the public to distinguish scientific medicine from this slick imitation, when the purveyors of fake diagnoses have laboratory tests and fancy imaging studies to back them up?

Before one is tempted to claim that making such a fake diagnosis causes no harm and serves to comfort the patient, or that it is worth a try to treat for possible Lyme even in the absence of clear evidence – let me assure you that there is real harm. First, making a fake diagnosis often results in the premature end to the search for the real diagnosis. I have seen patients settle on the diagnosis of Lyme disease when in fact they had something else entirely – something that therefore went untreated.

Also, the treatment for chronic Lyme is far from benign. It usually involves months or years of chronic oral or intravenous antibiotics, with real risks and side effects.

It is also important to point out that the question of whether or not co-called chronic Lyme disease responds to long term antibiotics is a separate and specific question that has been addressed by research. The NEJM study also reviewed this evidence and concludes that antibiotic treatment for chronic Lyme is not effective.

This is a very important distinction. I often teach my students that whether or not a patient has a specific diagnosis is often not definitively answerable, and may be a matter of definition. But further, it is actually a proxy question. The real, and far more important question, is whether or not a patient meeting certain objective criteria would benefit from a specific treatment. So we can put aside the question of whether or not patients can have atypical seronegative treatment-resistant chronic Lyme disease and ask instead, do such people respond to any treatment. The answer, according to this review, is no.

It is also worth pointing out that there does seem to exist a real, if rare, post-Lyme syndrome – people who have persistent symptoms after treatment for Lyme disease. These represent the vast minority of patients diagnosed with chronic Lyme, and they don’t respond to antibiotic treatment either.

It remains to be seen what effect, if any, this review published in a prestigious journal by recognized experts will have on the chronic Lyme subculture. I predict none. The evidence and arguments were all already there. Also, the chronic Lyme true believers are a self-selective group that have already proven resistant to logic and evidence.

But this article is an excellent reference for the practicing physician or patients suffering from chronic symptoms. It also makes my job easier, at least in reference to this specific form of quackery.

Perhaps the most important lesson in this whole affair (and the one most likely to be overlooked) is that it demonstrates that it is possible for so many patients and practitioners to fool themselves for so long. Often belief in a phenomenon is used to justify the existence of the phenomenon, using the “where there is smoke there is fire” logic. But often the smoke is just an illusion, or an artifact of sloppy thinking. The story of chronic Lyme adds one more historical example of this, a cautionary tale for anyone putting forward their own pet unfalsifiable hypothesis.


This article is also reviewed by Mark Hoofnagle over at denialism.

15 responses so far