I have been following the story of Dr. Zamboni, an Italian vascular surgeon who claims that multiple sclerosis (MS) is primarily caused by blockages in the veins that drain blood from the brain. This results in backup of blood in the brain, leading to inflammation around the blood vessels and MS. He sought to find the cause and cure for MS because his wife suffers from this disease – and he claims to have found one in his own specialty.
New ideas are presented in science and medicine all the time. This is healthy and necessary – we have to keep churning the pot so that new ideas can emerge and our thinking does not become calcified. But science is both a creative and destructive process, and most new ideas fall victim to the meatgrinder of research and peer-review. Ideally this process will take place mostly within the halls of science, and then those ideas that survive at least initial examination will start to penetrate the broader culture.
This is not what often happens today, however. With the internet and mass media, preliminary speculative studies are often presented to the public as if they are a stunning breakthrough. When the scientific community responds with their typical and completely appropriate skepticism, this may lead some to think that they are being stodgy or dogmatic, or even that a cover-up is in the works. The originator of the speculative claim is usually portrayed as a brave maverick, although sometimes the story can be framed as, “Brilliant scientist or dangerous crank? You decide.” When the topic is a new medical treatment, the stakes can be quite high. In this case many patients with progressive MS are seeking treatment with the so-called liberation procedure to treat the highly speculative CCSVI as an alleged cause for their MS.
This story has all the makings of the kind of scientific and medical drama the mass media loves. While the controversy rages, the science is quietly being done in the background, and the results are not heading in a favorable direction for Zamboni. A recent study, the largest to date, drives a further stake into the heart of CCSVI as a cause of MS.
First, let us consider how to approach Zamboni’s claims. His data suggests that nearly 100% of patient with MS have CCSVI (detectable blockages in the veins that drain the brain) while 0% of non-MS patients do. This kind of evidence is correlation only, and does not prove (even if it might suggest) causation. Before we leap to treatment, the cautious scientific approach is to first confirm the correlation with replication. If the correlation holds, then studies need to be done that can shed light on causation – does the pattern of correlation fit the hypothesis that CCSVI causes MS, rather than MS causing CCSVI or both correlating with some other factor. Finally, before treating CCSVI, we would need to study this treatment directly in specific types of MS.
Proponents of the liberation procedure are skipping over all these research steps, and then use anecdotal evidence to support claims of efficacy. This is a story we have seen before, and it usually does not turn out well. Getting back to the first step – how have attempts to replicate the correlation been going?
Last August I described the first four attempts at replication, three of which yielded negative results. Just last month I wrote a following up where I described three further studies of CCSVI – all negative. This month two more relevant studies have been published. The first compared 20 MS patients to 20 healthy controls, and found:
Only one healthy control and no MS patients fulfilled at least two criteria for CCSVI. Conclusions This triple-blinded extra- and transcranial duplex sonographic assessment of cervical and cerebral veins does not provide supportive evidence for the presence of CCSVI in MS patients. The findings cast serious doubt on the concept of CCSVI in MS.
Last week the largest CCSVI study was published, a study that enrolled 499 subjects, and compared MS patients to patients with other neurological disease (OND) and to healthy controls (HC). They found:
RESULTS: CCSVI prevalence with borderline cases included in the “no CCSVI” group was 56.1% in MS, 42.3% in OND, 38.1% in CIS, and 22.7% in HC (p < 0.001). The CCSVI prevalence figures were 62.5% for MS, 45.8% for OND, 42.1% for CIS, and 25.5% for HC when borderline cases were excluded (p < 0.001). The prevalence of one or more positive VH criteria was the highest in MS (81.3%), followed by CIS (76.2%), OND (65.4%), and HC (55.2%) (p < 0.001). CCSVI prevalence was higher in patients with progressive than in nonprogressive MS (p = 0.004).
CONCLUSIONS: Our findings are consistent with an increased prevalence of CCSVI in MS but with modest sensitivity/specificity. Our findings point against CCSVI having a primary causative role in the development of MS.
These findings are interesting – they do not entirely rule out a correlation between CCSVI and MS. However, the results are very ambiguous. There is a statistical correlation between MS and CCSVI, but there is also a correlation with other neurological diseases – with very different histories and probable causes than MS. CCSVI was also found in a quarter of healthy controls. So CCSVI is not specific to MS, and almost half of MS patients do not meet criteria for CCSVI.
Given the other negative studies, these results cannot be taken at face value but have to be put into context of the other research. At this time we can say that there might be a correlation, but it’s weak. It’s also still possible there is no correlation, and since there are some contradictory results more research would be helpful.
Even if there is a partial correlation, this study argues strongly against CCSVI being a significant cause of MS – if 44% of MS patients do not have it, and 42% of patients with OND do have CCSVI but not MS. This could mean that CCSVI only causes a subset of MS, or that it is a risk factor but not a direct cause. Or it could mean that MS (and apparently other diseases) cause CCSVI. This is plausible – we can imagine that the chronic inflammation caused by MS damages the veins over time resulting in CCSVI. It is even possible that this, in turn, will cause its own symptoms or worsen the MS and therefore treating it may be beneficial. This is all just speculation, however. In this case the phrase, “more research is needed” is appropriate.
One other recent study, that I have not written about previously, is worth mentioning. In this study researchers looked specifically at subjects at the very onset of their MS. If CCSVI causes MS then it should precede MS. They found no correlation, and concluded:
Our findings do not support a cause-effect relationship between CCSVI and pMS. Further studies are warranted to clarify whether CCSVI is associated with later disease stages and characterizes the progressive forms of MS.
With these latest studies the correlation between CCSVI and MS seems shaky – nonexistent to weak, but not entirely ruled out. That CCSVI is a significant cause of MS is even weaker. It cannot be ruled out as a late stage contributor, or a cause in a subset of MS patient, but neither is it established as a contributing cause at all, and the evidence is largely against it.
There is so far no controlled blinded studies of the liberation procedure in patients with CCSVI and MS. There is a controversy as to whether or not such studies would be ethical and appropriate. It would be getting ahead of the more basic research – we should determine that a phenomenon exists and is causative before studying a treatment of it. However, hype has generated great interest in the liberation procedure, and it is being done in various clinics. This is the Catch-22 that the modern information age has created for ethical medical researchers.
In a perfect world clinical trials of the liberation procedure would wait for more confirmatory studies, but we do not live in a perfect world. We may need to at least study those patients who are seeking out the treatment anyway, and provide useful data that future patients and practitioners can use to guide their decisions.