A battle over the link between strep and OCD

Today I introduced a class of students to the notion that a strep infection could give a child obsessive compulsive disorder (OCD). The idea was first widely publicized in the memoir “Saving Sammy: Curing the Boy Who Caught OCD“, and briefly recapped in this clip from the Today Show:

I am already on record as identifying this as the beginning of the next major revolution in our understanding of the brain. My students, however, did not find it any more revolutionary than the idea of the neuron. Several of them were in fact far more shocked at the notion that some doctors would reject the idea that strep could trigger a mental illness.

From a certain perspective, it does seem like common sense: Your brain is part of your body. Your immune system is part of your body. Sometimes your immune system goes wonky and attacks parts of your body, such as in rheumatoid arthritis or multiple sclerosis. In the case of multiple sclerosis, it’s even an attack on neurons, including the ones in the optic nerve in the brain. The current model of mental illness places a root for OCD in the functioning of an individual’s neurons, somewhere in the brain. So why not a strep infection that triggers an immune response that mistakenly but selectively attacks those neurons in the brain, creating what looks like OCD?

The argument was indeed so full of common sense that even after pointing out a human tendency toward irrationally holding on to beliefs, and reminding myself of the truism that “scientific change happens in the obituary pages”, I had to go out and check.

First, although awareness of PANDAS has grown since Saving Sammy was first published in 2010, claims about its existence remain controversial. Two years later, the director of the National Institutes of Mental Health wrote about resistance and controversy to the idea of infections causing mental illness, despite the hundred-year history since syphilis was recognized to cause forms of psychosis. In a magazine article from June 2013 – aptly titled “The Feverish Debate” – a Canadian doctor who treated a PANDAS patient asked not to be identified by name; the author comments that “names of physicians who will even consider the diagnosis are traded among families, in a manner reminiscent of how women tracked down abortionists in the ’60s”. And in another news article from just April of 2014, four months ago, one neurologist went on the record declaring “No one has established that it’s true…No one has established what the infection is that brings it out or which antibodies bring it out”.

So, controversy and resistance remain; but how can trained doctors resist such a clear and compelling idea? Aside from usual human irrationality, I think we can find reasons in three places:

1. Not all cases are textbook, and complex cases make PANDAS easier to dismiss.

Upon close inspection, Sammy’s case is not quite as clear-cut as the usual narrative (which I myself am guilty of using for simplicity) of “boy caught strep, boy caught OCD, boy took antibiotics, boy no longer had strep, boy no longer had OCD, all future instances of OCD were linked to strep infections and cured by antibiotics”. That would be a truly textbook case. Sammy’s case has a few complications – he never showed any strep symptoms, it took some work to find the right antibiotics to cure the OCD, and it took a long-term course of serious antibiotics not the usual 10-day pills for strep to do – but still shows strep infection and OCD symptoms dovetailing together.

But as most doctors know, “classic” or “textbook” presentations of a disease are not as common as the average person might think. Look at other stories in the PANDAS brigade, and room for doubt will creep in. Here, you will find children whose first bout of OCD might have coincided with a strep infection, but whose symptoms did not go away when the strep infection did. Also, children who could not be cured with mere antibiotics, and instead required their entire blood supply to be flushed clear of all antibodies (called “plasmapheresis”, bearing some similarities to dialysis) and their immune systems suppressed.

When the relations between strep infection and mental illness symptoms are not precisely in sync, it becomes easier for any skeptic to dismiss the notion as wishful thinking or mere coincidence. To quote “The Feverish Debate”, which recounts the tale of a very different boy named Sammy who struggled with PANDAS:

Depending on where you stand on PANDAS, Sammy’s story can have one of two arcs. In the first, a little boy gets sick, a doctor recalls an obscure disorder, a mother presses for help, treatment is given, and the boy recovers. In the other, a little boy gets sick, no single treatment works consistently and quickly, and the symptoms wax and wane but resolve over time—by coincidence, right after a tonsillectomy.

Because strep infections are common, it’s virtually guaranteed that some children will have a strep infection around the time they first develop symptoms, just like some children will have a strep infection when they are injured in a car accident. The one doesn’t necessarily have anything to do with the other. Because our population is quite large and a million-to-one chance will still happen 313 times, some children may well have waxing and waning OCD symptoms that happen to coincide with strep infections and treatments. It’s as much a fluke of statistics as likely and as meaningless as the fact that two different boys named Sammy became poster children for the same disease.

The very notion of having to correct parents’ incorrect interpretation of meaning in statistical coincidence is no doubt a default and entrenched behavior in the 21st century doctor. Among other things, they will have turned to it fairly frequently to counter the pernicious and persistent belief that vaccines cause autism, which sticks around in part because, since vaccines are administered around the age when autism symptoms become more apparent, some children will develop symptoms of autism around the time they have recently received a vaccine.

2. Complex cases also make PANDAS harder to study.

As with vaccines and autism, there is also a scientific study to back up PANDAS-doubters’ seemingly irrational beliefs. This was a longitudinal study, published in 2002, that tracked children with OCD who were classified either as being potentially strep-related or not strep-related. Over the course of several years, there were very few clear links observed between strep infection and worsening OCD symptoms – and when they were observed, they were just as likely to occur in the children who were not supposed to have strep-related OCD. Here, skeptics can say, once you start to bring careful study to this so-called PANDAS instead of relying on anecdote, the coincidences fall apart.

But the study does has flaws – significant enough that they have been pointed out by the original lead author of the study, James Leckman. At the heart of the problem is how you decide if a child has PANDAS or non-strep-related OCD. There may be some children who just happened to have strep around the time their symptoms began, but developed OCD for other reasons, who get counted as “PANDAS” – but who we wouldn’t expect to show any worsening with future strep infections. On the opposite side, there may be children (like the original Sammy) who never showed strep symptoms, who were incorrectly classified as having traditional OCD, but who would get worse with a new strep infection. Likewise, there may be children whose strep infections start OCD symptoms but aren’t necessary for maintaining them; the symptoms might persist when no infection is detectable (like the second Sammy, whose OCD may have been maintained by asymptomatic tonsillitis). Suddenly finding a clear and direct link between infections and symptoms is not so easy.

3. Changes in our understanding of PANDAS share a major characteristic with pseudoscience.

If anything in that last paragraph struck you as a very convenient definition change on my part so the idea of PANDAS couldn’t be disproven, well, you have a new insight into the feelings of those who object to PANDAS. One of the major characteristics of a “pseudoscience” – something that pretends to be science but really, really isn’t – is that it makes itself unfalsifiable by always changing the definition or moving the goalposts whenever the data are against it. As in, I insist there are fairies in my kitchen. You investigate thoroughly and find no evidence. I declare that the fairies moved to the living room. You investigate the entire house. I declare that the fairies head for greener pastures whenever skeptics are around. And so on.

Unfortunately, it can be hard to distinguish between the moving goalposts of pseudoscience and the legitimate shifts in our definitions and understanding of a new disease. PANDAS made its first appearance in the medical journals less than 20 years ago; it’s younger than my students. Even in the miraculous 21st-century, We do not develop a full-blown and perfect understanding of a disease as soon as we identify it well enough to give it a name. Research for a new disease is slow, as only a few people will have heard about it and funding will be scarce; as the results trickle in, that new information will necessitate changes in what we think is going on. Unfortunately, those changes might appear convenient and seem like goalpost-shifting to the skeptics. The fact that PANDAS (pediatric autoimmune neurological disorder associated with strep) is being renamed to PANS (pediatric acute-onset neuropsychiatric syndrome) is intended to reflect our current understanding better – much as the change from “shell shock” to “post-traumatic stress disorder” better reflected the variety of patients and causes – but could be read as a retreat saying, “well, not just strep, and not always strep” so the lack of connection to strep specifically can be excused.

4. PANDAS and its treatment probably push a few M.D. buttons.

There are two things I have no doubt many doctors have gotten sick of: patients demanding antibiotics for illness that have nothing to do with bacteria, as if antibiotics are the new “take two aspirin and call me in the morning”; and patients diagnosing themselves on WebMD in what is apparently now known as “cyberchondria“. PANDAS will hit both buttons: parents learn about it from the Internet or word-of-mouth, insist that their self-education is more trustworthy than the doctor’s hard-earned M.D., and want antibiotics to treat their children. Everything else about the science and history of this disease aside, I’m not sure how many doctors will be in an open and receptive frame of mind after being hit with that double-whammy.

Knowing these reasons that doctors might resist the idea of PANDAS/PANS is, of course, only the first step; the next step is persuading them to drop that resistance and consider the idea more fully. Given that strep and other infections are being linked to bipolar disorder and possibly even self-injury behaviors, and that “immune dysregulation” is one of the latest theories proposed to explain autism, that consideration may not come a moment too soon.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s