Aug 02 2010

CCSVI – The Importance of Replication

Scientists and skeptics are familiar with this pattern – a preliminary study suggests a wildly new understanding of a scientific or medical question. The scientific community is cautiously skeptical but interested. The press proclaims a stunning breakthrough, and the public is briefly fascinated. If the new discovery concerns a medical treatment, the community of those affected become fixated on the potential new “cure”, and many start demanding treatment based solely on the preliminary evidence. But then the wheels of research begin to grind and, more often than not (because that is the nature of discovery) the new idea turns out to be wrong – it fails the critical step of replication.

Then one of two things will happen: either the new idea or treatment will fade, becoming little more than a footnote in the history of science, or a subculture will persist in believing in the treatment and will dismiss contrary evidence and mainstream rejection as a conspiracy. Which course the new idea will take seems to depend largely on the original scientist – if they accept the new evidence and abandon their claims, it will likely fade. If they refuse to give up in the face of new evidence, then a new pseudoscience will likely be born.

We have seen this pattern play out with Laetrile, psychomotor patterning, cold fusion, and many other ideas.

Now we are in the midst of this pattern playing itself out yet again – with the Zamboni treatment for multiple sclerosis (MS). Dr. Zamboni is an Italian neurosurgeon whose wife has MS. He sought out to find a cure, and eventually discovered that patients with MS have a blockage in the venous drainage of their brains, a condition he termed “chronic cerebrospinal venous insufficiency” (CCSVI). He further claims that MS can be treated, even cured, by opening up the veins that drain the brain with either angioplasty or stenting. Many MS patients have been interested in this potential new treatment, and many have even undergone treatment.

The neurological community is, to say the least, skeptical – but interested. There are many red flags of crankery in CCSVI, however. First, it may make for good story telling, but it is a bit curious that a neurosurgeon set out to discover a treatment for MS and found a neurosurgical one – even though there was no prior suggestion that this would be the case. Further, we have decades of research that tell us that MS is primarily an autoimmune disease – the patient’s immune system is attacking their own central nervous system. It is true that inflammatory plaques tend to occur around veins, but that is explained by the fact that blood vessels allow immune cells access to the central nervous system.

And finally Zamboni found in his own research that his criteria for CCSVI has 100% sensitivity and 100% specificity for MS – which triggers the “too good to be true” red flag. It is rare (perhaps nonexistent) to find a test in medicine that is correct 100% of the time, with no false positives or false negatives. These kinds of results strongly suggest experimenter bias.

Zamboni presented his research at the latest meeting of the American Academy of Neurology. I was not present, but I have spoken to some colleagues who were – the buzz is that Zamboni’s presentation was not impressive, and he came off as a bit of a crank. This by itself does not mean much – it’s more about personality than data – but it’s not surprising given the pattern I have outlined above.

The only reasonable response to such claims is cautious interest, and that is exactly what the medical community has done. In general most neurologists acknowledge that Zamboni’s claims need to be replicated and researched. Perhaps he is on to something, even if he is overstating the claims. Perhaps the perivenous inflammation of MS leads to CCSVI which exacerbates MS. Maybe Zamboni has found a piece to the puzzle, even if he has not solved the whole thing in one fell swoop.

Well now we have two independent replications of Zamboni’s research published in the latest issue of the Annals of Neurology – and both are completely negative. The first is a German study by Florian Doepp et al, using ultrasound to test the CCSVI criteria in 56 MS patients and 20 controls. They found almost completely negative results (one MS patient met one criterion, but not the others) – no signs of venous blockage in the MS patients.

The second study is a Swedish study by SundstrÃm et al (“Venous and cerebrospinal fluid flow in multiple sclerosis – a case-control study.” Peter SundstrÃm, Anders WÃ¥hlin, Khalid Ambarki, Richard Birgander, Anders Eklund and Jan Malm. Annals of Neurology) – not yet available online. This study used MRI scanning to assess blood flow in the internal jugular vein in 21 MS patients and 20 controls, and also found no difference.


These two studies are not going to be the final word on CCSVI and MS, but they are very important in signaling a note of caution to patients and clinicians about this new hypothesis. The treatment for CCSVI is invasive and has serious risks, and should not be undertaken lightly. I agree with the current consensus that evidence for CCSVI is too preliminary to warrant treatment, especially given the risks.

I do wish that the media and public would learn the more general lesson here – new dramatic ideas in science, especially those that seem to go against established knowledge, are likely to turn out to be wrong when the dust settles. It is partly the job of the skeptical community to provide cultural memory of such events – so the next time a lone scientist or doctor claims to have made a revolutionary breakthrough that seems a bit dubious, it is the skeptics who will be there to say – remember Zamboni.

32 responses so far