Introduction

A great debate has emerged: does CCSVI exist? Is it linked with MS?

Summarizing the current research on CCSVI is not simple – the results are wildly contradictory. Consider the following:

• CCSVI's existence and association with MS has received broad support from a wide range of studies conducted across three continents.^{1,2,3,4,5,6,7, and others}
  
• The existence of CCSVI has been refuted (and thus any association with MS dispelled) by a wide range of studies conducted in the U.S. and EU.^{8,9,10,11,12,13,14, and others}
  
• A recent but smaller group of studies suggests that CCSVI may exist, but that it may be a consequence of MS as opposed to a contributor.^15,7

As researchers, patients, and physicians, what do we make of these conflicting conclusions?

Clearly, given the contradictory research, no truly definitive statements about CCSVI can currently be made. However, a careful review of the studies does reveal important trends, and hints at where and how research should proceed.

Before moving on, we must clarify one issue:  This article does NOT address CCSVI treatment effectiveness.  Some may be disappointed by this, but the fact is that only two studies have explored CCSVI Treatment.^{16, 17} Both were “open label” (not scientifically conclusive), and both found similar results (that treatment may help, but that more research is needed before any conclusions can be drawn). Thus, no contradictions in treatment data currently exist.  Obviously, there is widespread dispute about CCSVI Treatment, but this dispute is not grounded in research, because science has not yet addressed the issue.

Hopefully, results from the several CCSVI clinical trials currently underway will emerge over the next twelve months, and then we will have some good treatment data to consider. Until then, we are left to consider research that does exist, and what it has to say about the presence of CCSVI and its relationship (or lack of relationship) to MS. To this end, however, we will consider how the two existing treatment studies factor into the debate over the existence of CCSVI and its connection to MS.

Lastly, CCSVI Alliance considers this a “living document” that will be updated and refined as additional research data becomes available, and as additional researchers and informed patients and physicians review and respond to the material.

Making Sense of the Research

To date, over 40 studies or articles investigating CCSVI have been published.  However, only a subset of these focuses on the heart of this debate:  does CCSVI exist, and if it does, is it associated with MS?

Given this subset of studies, a possible starting point is simply to count them up. Hence, Table 1 below summarizes relevant findings:

(Please note that for brevity’s sake we will refer to studies by the principal researcher’s last name and the year in which the study was conducted. Further, when a team/center has conducted multiple studies and all studies have found similar results (e.g. Zamboni and colleagues; Zivadinov and colleagues), we will only list that team’s work once.)

Table 1: Studies Measuring the Presence of CCSVI in MS Patients

Note: Negative results from Barachinni 2011 and from Doepp 2010 have been professionally disputed in the literature.

Unfortunately, simply counting the studies doesn’t tell us much: what does five studies supporting CCSVI in MS versus seven studies refuting CCSVI in MS actually mean?

More importantly, when looking carefully at each of these studies, we see vast differences not just in results, but in the methods and technologies used to measure CCSVI, the type of study conducted (open label, Blinded/controlled), the very definition of the phenomenon being measured (CCSVI), and even in conclusions drawn from the same results.  In fact, the closer we look at the research, the more clouded the picture becomes.

Nonetheless, despite this complexity, we can make some sense of this research.   Here’s how: we will conduct a meta-analysis. A meta-analysis is a scientific process where results from multiple studies are selected, vetted, combined, and analyzed.  A meta-analysis can help reveal the “big picture" (or trends) that emerge from a group of studies that may, individually, report conflicting results.

The objective of our meta-analysis will be to determine whether CCSVI exists, and if so, whether it is associated with MS . To this end, the meta-analysis will give us a formal process to review current research and discern any meaningful patterns. More specifically, we our meta-analysis process will:

1. Define “inclusion criteria” to help determine what types of research can and can’t answer our question. (Does CCSVI exist and, if so, is it associated with MS?)
2. Seek out and identify relevant research that can answer our question.
3. Review and analyze eligible research, noting and accounting for any biases that may impact the reliability of the study’s results.
4. Analyze and “roll up” results from all eligible studies to discern any “big picture” trends.
5. Review and discuss the trends. Describe what research is needed to clarify issues that remain uncertain, and note problems or shortcomings in research which have contributed to the current confusion.

Please see our note in the Appendix for more information on biases our methodology.

CCSVI Meta Analysis - Getting Reliable Information from the Studies We Have

Finding Studies for Review

Table 1 above lists all studies listed in PubMed (as of June 1, 2011), that self-describe as answering the question of whether CCSVI exists, and whether it is or is not related to MS. Because PubMed is by no means exhaustive, we have cross-checked our list with informal lists maintained by various on-line CCSVI research tracking sites, and have confirmed that PubMed currently lists most studies that are available in English. (This is likely because CCSVI research is comparatively new and most research has been conducted by either U.S. or E.U. research sites.)

Note: A very large group of studies that either describe specific aspects of CCSVI, or attempt to measure ancillary/downstream effects of CCSVI, are not included in our analysis because they are not directly investigating our research question: does CCSVI exist, and, if so, is it associated with MS.  Please see our note on Inclusion Criteria in the Appendix for more information.

Inclusion  Criteria

Clearly, a study’s self-described assessment of its merit is not sufficient to determine whether it actually does address our objective (Does CCSVI exist, and if so, is it associated with MS?).  Studies that use incompatible approaches, conflicting or inadequate definitions of what is being studied, or unreliable diagnostic means, cannot in fact answer our question.

Thus, we have defined a very specific set of criteria for evaluating each study and determining whether each study does in fact address our question. In short, we begin by evaluating whether the “methods and means” used by each study are sufficient for addressing whether CCSVI exists and if it is related to MS. (“Methods” refers to the soundness of the study’s scientific methodology. “Means” refers to the suitability of the diagnostic technologies and processes used to acquire the study’s results.)

Specifically, our inclusion criteria are outlined as follows:

The only published formal definition of CCSVI is Dr. Zamboni’s 5-part technical definition of CCSVI.  Studies reporting results that do not define CCSVI using Dr. Zamboni's five part technical definition simply cannot be compared against studies that do. Thus, studies not using the  5-part definition of CCSVI cannot meet the inclusion criteria and will not be included in the analysis.

Please note: We recognize that Dr. Zamboni’s definition of CCSVI has been questioned. However, no other researchers have proposed an alternative definition, and no research has cast doubt specifically on Dr. Zamboni’s definition. Thus, the only way to systematically measure CCSVI must adhere to the only existing definition of CCSVI. In other words, research not employing Dr. Zamboni’s criteria is not measuring CCSVI; it is measuring something else instead. This will remain true until the current definition is changed, or until other measures/definitions of CCSVI are proposed and corroborated. (For more discussion of these issues, please see Definitions in the Appendix.)

At present, the only scientifically validated method for measuring CCSVI via the 5-part definition is a very specific use of Doppler sonography (DS). Importantly, the reliability and reproducibility of measuring CCSVI via Doppler sonography has been demonstrated in three distinct studies across two continents.^18,19,20 Hence, we will exclude research that does not measure CCSVI via Doppler Sonography.

We appreciate that requiring DS to measure CCSVI is limiting; however, until alternative definitions of CCSVI are proposed, and methods for measuring CCSVI via alternative means are identified, we have no choice but to move forward with the only existing validated processes and definitions.

Note: Excluding Magnetic Resonance and venographic research does NOT mean that MRI/MRV and venography are invalid tools for investigating CCSVI. To the contrary – MRV (especially via the Haacke-Hubbard protocol) and venography are essential and irreplaceable tools for investigating CCSVI.  However, MRV and venography simply cannot measure CCSVI via Dr. Zamboni’s 5-part technical definition. To learn more, please see our notes on MRV/Venography in the Appendix.

• Studies not using healthy controls for comparative purposes must be excluded.

Simply put, the question of CCSVI's relationship to MS cannot be clearly addressed without using a healthy control group for comparison.

Note: For those concerned about the exclusion of studies not using healthy controls, please see our note on Healthy Controls in the Appendix.

Next, we can apply our inclusion standards to the research listed in Table 1, and determine which research is NOT eligible for inclusion (that is, which research cannot reasonably address our central question: Does CCSVI exist, and is it related to MS?). Research not meeting our inclusion requirements includes:

• Simka, 2010 (No healthy control group was included; the study cannot verify whether CCSVI is common in all people or just in MS patients.)

• Wattjes, 2010 (Used MRV. MRV studies, particularly in this case because no validated MR protocol was used, cannot measure CCSVI via the 5-part technical definition of CCSVI.)

• Sunstrom, 2010 (Used MRV. MRV studies, particularly in this case because no validated MR protocol was used, cannot measure CCSVI via the 5-part technical definition of CCSVI.)

• Yamout, 2010. (No healthy control group was included; the study cannot verify whether CCSVI is common in all people or just in MS patients.)

• Krogias, 2010. (Unfortunately, we are unable to find a full English language translation of Krogias, 2010, so a full review has not yet been performed. However, per the English language abstract, it appears that while Krogias provides theoretical arguments against CCSVI, this very small study itself does not use healthy controls (and appears unblinded). Thus, per abstract review only - and subject for re-examination after study translation - Krogias 2010 cannot be included.)

Finally, we note that one study (Barachinni, 2011) proved difficult to assess: it used both DS and venography for diagnosis, but came to different results based on the respective technologies. Lacking an ideal solution, we will allow both the positive and negative results from Barachinni, 2011 (for more information, please see the Barachinni discussion in the Appendix).

Moving forward, we can now update Table 1 to include only research that passes eligibility requirements for legitimately answering the question “Does CCSVI Exist, and if so, is it related to MS?”:

Table 2: Validated (Eligible) Studies Measuring the Presence of CCSVI in MS Patients

Next, using the studies in Table 2, we proceed to the next phase of the meta-analysis: determining which of the eligible studies offer more convincing results and which offer less convincing results. In other words, while all studies in Table 2 have met the inclusion requirements, they are not all “equal.” Important factors separate them and strongly suggest that results from some studies should be given greater weight than others (here, “weight” reflects the probable predictability/accuracy of the study’s results).

To weigh the studies in Table 2, we consider each study with respect to three factors known to impact the accuracy and/or reliability of research results: 

1. The size of the study.
2. Whether the researchers were trained in CCSVI diagnosis (to minimize bias effects).
3. Whether the study used a blind (more reliable) or open label (less reliable) protocol to minimize bias effects.

Study Size

The larger the sample size of a study, the more suggestive the results (assuming all other variables and biases are equal). Given the studies in Table 2 above, we can rank the size of each study based on the number of subjects involved:

Largest Study:

• Zivadinov (BNAC), 2011 [289 MS patients/163 Healthy Controls; This study contains more than 3 times the number of MS subjects than any other study.]

Intermediate-sized Studies

• Centonze, 2011    [84 MS/56 HC]
• Zamboni, 2009     [65 MS/265 HC]
• Barachinni, 2011  [50 CIS-MS/50 HC]

Smaller Studies:

• Mayer, 2011 [20 MS/20HC]
• Doepp, 2010 [56 MS/20 HC]
• Al Omari, 2010 [25 MS/25 HC]

Next, given that CCSVI diagnostic accuracy has been very clearly tied to training¹⁸, we can prioritize all studies in Table 2 based on the diagnostic experience of the investigators:

More Reliable (investigators had hands-on training in CCSVI diagnosis via Doppler ultrasound. Training information was obtained from study’s methodological and/or discussion section):

• Zamboni, 2009
• Zivadinov, 2011
• Centonze, 2011

Less Reliable (higher probability of errors in the results):

• Doepp, 2010
• Al Omari, 2010
• Barachinni, 2011
• Mayer, 2011

Lastly, we can quickly separate out the studies that used a blinded/controlled protocol (more reliable) from those that did not (less reliable).

Blinded/controlled studies

• All studies except Al Omari, 2010

Unblinded (less reliable)

• Al Omari, 2010

Putting the Picture Together

Finally, we can compile all this information into a single comprehensive snapshot, as seen in Graph 1 below, which provides a comprehensive view of all eligible studies, weighted for accuracy/predictability, and plotted on two dimensions:

1) Does CCSVI exist?
2) Does CCSVI contribute to MS?

Results

Given Graph 1 above, several key points become apparent.

1. The overall weight of the research trends toward suggesting that CCSVI exists and is associated with MS.
   
   1. Two of the three largest studies (particularly BNAC 2011, which included by far the largest number of patients, together with data validating their diagnostic technique) support CCSVI in MS.
2. The debate over the presence of CCSVI in MS remains unsettled.
   1. Results from Centonze, 2011, produce an intriguing counterweight to Zivadinov, 2011 and Zamboni, 2009. Centonze, 2011, raise legitimate questions about the presence of CCSVI in MS.
3. As a whole, the results do not address the question of whether CCSVI may play a contributing role in MS or may be a consequence of MS.
   1. Results from Zamboni 2009 and Al Omari 2010 hint that CCSVI may play a causative role in MS.
   2. All results suggesting that CCSVI doesn't exist (Mayer 2011, Doepp 2010, Barachinni 2011, Centonze 2011) portray CCSVI as having no causative role in MS.
   3. BNAC 2011 suggests that CCSVI does not have a "primary" causative role in CCSVI. However, BNAC does note that "the role of CCSVI in contributing to or being a consequence of MS progression cannot be excluded and should be further investigated."⁷

Discussion

New theories often enter the scientific discourse amid a flurry of conflicting data. In fact, inconsistencies or gaps in the data often spur the development of new theories to begin with.

Further, during early stages of research, results may be particularly confusing. Consider, for example, the decades-long debate and starkly differing research results on the link between smoking and lung cancer. Thus, we should not be surprised that research on CCSVI has produced conflicting results.

Studying CCSVI is complicated. Diagnosis requires expertise in a particular application of Doppler sonography and, in the best cases, catheter venography, vascular science and neurology, together with a deep understanding of CCSVI theory.  It is, however, precisely this complexity that has allowed us to make some sense of the data:  by excluding studies whose methods or means were technically insufficient to determine whether CCSVI exists, we have simplified the research landscape and made clear at least three important trends:

1. The preponderance of evidence trends toward suggesting that CCSVI does exist and is correlated with MS.
2. Due to conflicting data from well-designed studies, the debate over the existence of CCSVI is not settled.
3. At present, research focusing on the presence of CCSVI does not clearly address whether CCSVI is a contributor to, a consequence of, or perhaps even unrelated to, MS.

Most apparent in Graph 1 is the weight of evidence provided by Zivadnov, 2011, which remains by far the largest, most well documented, and arguably most accurate study to date. With results falling very much in the middle of the debate, Zivadinov 2011 casts doubt on the degree of association between CCSVI and MS reported in Zamboni, 2009, as well as on the failure to find any association between MS and CCSVI as reported by Doepp 2010, Mayer 2011, and Barrachini, 2011. (It should be noted that results from Doepp, 2010 and Barrachini, 2011 have also been professionally disputed in the literature.)

Given these results, CCSVI begins to appear similar to many other factors associated with MS. Like genetic markers, geographic region, sunlight/vitamin D levels, exposure to certain viruses, or smoking, we see a higher prevalence in MS patients, but of course not in all MS patients. Further, as with all these factors, we see varying degrees of the same in healthy controls.

While these results suggest that CCSVI is not the sole cause of MS, the research has in no way excluded that CCSVI may (or may not) play a contributing role in both disease emergence and/or progression.

Moreover, while both published CCSVI treatment studies have been open label, both have nonetheless documented a plausible association between CCSVI treatment and alleviation of symptoms for at least some MS patients. This data, while suggestive only, adds an additional dimension to the notion that CCSVI and MS may indeed be related.

Given these findings, it becomes clear that the urgent need for well-designed studies continues.

HOWEVER, new research must avoid the missteps of past efforts. Specifically, we urge the scientific community to use the methods and means proven to produce accurate results (or, better still, to propose and validate more reliable alternative approaches to measuring CCSVI). Specifically:

1. Until a new definition of CCSVI is proposed, future studies not relying on the current 5-part definition of CCSVI simply cannot make credible research claims about the existence (or lack thereof) of CCSVI.  
   
   We appreciate the limitations of the current 5-part definition, but, at present, it is the only definition available.
2. Until non-ultrasound methods for documenting CCSVI are validated and available, studies not employing Doppler ultrasound (DS) cannot make credible claims about the existence (or lack thereof) of CCSVI. Research "proving" that CCSVI does or doesn't exist via techniques not shown to reliably measure CCSVI is a waste of everyone's time.  
   
   Importantly, we are in no way denying the value of MR and Venographic technologies. Indeed, both are potentially superb at documenting the characteristics of the venous anatomy. Moreover, both are invaluable tools for treating physicians. However, at present, MR and venography simply cannot measure CCSVI via the 5-part definition of CCSVI.
3. Studies not using technicians specifically trained in CCSVI diagnosis cannot be relied on to provide accurate results. Peer reviewed published data supporting this conclusion has been widely available since early 2010.^18,19,20  
   
   Research "proving" that CCSVI does or does not exist that is published by researchers who have not been specifically trained in CCSVI diagnosis should be given little weight (and, potentially, are a waste of everyone's time and money).
4. The methods and means for diagnosing CCSVI must be improved.  
   
   The current Doppler Ultrasounds process is prone to error and difficult to reproduce. This places considerable time and resource burdens on researchers looking to conduct legitimate investigations of CCSVI.
   
   Those seeking to advance CCSVI research must focus on creating a diagnostic approach that is more clearly defined, easier to reproduce, and less reliant on diagnostician training. Moreover, the relationship between Doppler diagnosis of CCSVI, MR investigations of CCSVI, and catheter venographic investigations of CCSVI must be better understood.

--------------------------------        End of Main Text       ---------------------------------------

APPENDIX

(Additional and Supporting Information (linked from the text above)

• Biases:  TheCochrane Handbook describes research bias as “selection bias, performance bias, detection bias, attrition bias, reporting bias and other biases that do not fit into these categories.” We note some selection bias in our sample (eligible studies used patients with differing MS diagnosis, and did not always use a MS diagnosis based on the same measures). Also, for lack of resources, we have not accounted for attrition bias or reporting bias. Please see our Note on Methodology (below) for more detail.

• Note on Methodology:  Those familiar with research methodologies will recognize that our process (and some of our terminology) is roughly similar to a Cochrane analysis.  However, we need to make clear: the present analysis isNOTa Cochrane Analysis. A qualified Cochrane analysis is an immense and thorough task, requiring multiple resources, substantial funding, rigorous adherence to theCochrane methodology, and independent Cochrane review. Due to resource constraints, we simply cannot perform a full Cochrane analysis. Moreover, it is not clear that sufficient data exists to make such an endeavor worthwhile. Thus, while we will rely on some of the fundamentals of a Cochrane analysis (which are themselves a codified and tested set of best practices for scientific analysis), we are definitively NOT presenting our meta-analysis as a "Cochrane" analysis.

Additionally, most meta-analyses rely heavily on statistical analysis, yet you will find that the present analysis does not. The bias present in many of the studies is extensive, and there is no rational (statistical) method for weighting that bias (e.g. all studies not using researchers trained in CCSVI diagnosis are suspect, as research has shown that diagnosis requires extensive training). Thus, the amount of data from fully qualified studies is relatively narrow, and we found that a graphical representation of the data (as found in Graph 1 above) was more effective and informative than a statistical analysis.

• Inclusion Criteria. Two particular types of studies have been intentionally excluded from our eligibility list:
  • Studies aimed at characterizing CCSVI. Several of Dr. Zamboni's team's studies, and as well as several of Dr. Zivadinov's team's studies fall into this category (also, e.g. Dr. Lee). This research is aimed at itemizing and describing specific features of CCSVI. This is compelling information, but it rests on the assumption that CCSVI exists, and thus does not directly answer our objective.
  • Studies measuring hypothesized "downstream" effects of CCSVI, or "corroborating" secondary features of CCSVI. For example, Worthington 2011¹⁰ and colleagues suggest that their failure to find supposedly CCSVI-predicted CSF Ferrtin and flow/pressure anomalies in MS patients refutes the existence of CCSVI. The problem with this type of research is two-fold: first, Worthington's results are contradicted by other research. Second, and perhaps more importantly, it's arguable that CCSVI (to the extent that it does exist) has not yet been sufficiently described to make this degree of prediction (the complexity of the science, together with the early stage of understanding, only suggest certain "downstream" consequences, but cannot fully "predict" them yet).
  • Thus, we have chosen to focus only on studies that directly envision/measure CCSVI.
• Definitions of CCSVI. Presently, the only formally published definition of CCSVI (in peer reviewed scientific journals) is Dr. Zamboni's 5-part definition. Hence, studies claiming to measure or document CCSVI must use this definition. Importantly, this does not mean that Dr. Zamboni's 5-part definition of CCSVI is the only possible definition, or even the best definition (in fact, it has not even been independently validated). It simply means that it is, at present, the ONLY definition, and thus the only guide researchers have.Clearly, many researchers are not satisfied with Dr. Zamboni’s definition. Even relatively straightforward measurement issues need clarification (for example, the relationship between venous anatomy and specific Doppler findings needs to be elaborated).  Without question, researchers would welcome an alternative definition, and/or additional methods for measuring CCSVI. However, we must reiterate again: the only definition is Dr. Zamboni’s 5-part definition, and thus we must either accept it, or refute the definition itself. Presently, and rather surprisingly, no one has published a well-constructed refutation of the definition of CCSVI. The most interesting discussions on definition have been raised by Barrachhini, 2011¹¹.

• MRV and Venography: MRV is an essential and irreplaceable tool for investigating CCSVI. MRV is the only tool that can image the overall venous anatomy, giving researchers and physicians essential information about an individual's anatomy. Venography is at least equally important. Venography, often described as the "gold standard" for picturing venous anomalies that are alleged to be associated with CCSVI, is almost universally used by physicians while examining a patient's venous anatomy and while treating CCSVI.

However, it is currently impossible to use MRV or venography to measure CCSVI per the 5-part definition of CCSVI. This is because the definition requires venous blood flow measurements in various patient postures (from upright to supine), but MRV and venography cannot measure a patient in the upright position). Moreover, while many treating physicians are using the Haacke-Hubbard MRV protocol, most published research has not.

• Note on Healthy Controls: Studies not using health controls are not "invalid" studies. To the contrary – the characteristics and severity of CCSVI and its relationship to MS can all be investigated without using healthy controls. However, to answer our principal question – Does CCSVI Exist, and Is It Related to MS? – we must compare the presence (or lack thereof) of CCSV in MS patients against the presence (or lack thereof) of CCSVI in healthy controls.
• Baracchini: Using DS, Baracchini show a statistically significant association between CCSVI and MS. Using venography, they show no association either between CCSVI and MS, or between their own DS and venography results. We acknowledge that Baracchini would object to including both the positive and negative results, as Barachinni et al. believe that their data, as a whole, argue against the existence of CCSVI. However, because Barachinni's own data offers differing conclusions (as noted in the journal in which the results where published), we will accept both. Moreover, Baracchini failed to document any internal validation of their methods, and thus, to researchers reviewing their data, it is unclear whether their DS data is more, or less, valid than their venographic data.

References

1. Zamboni P, Galeotti R, Menegatti E, Malagoni AM, Tacconi G, Dall'Ara S, Bartolomei I, Salvi F. Chronic cerebrospinal venous insufficiency in patients with multiple sclerosis. J Neurol Neurosurg Psychiatry. 2009 Apr;80(4):358.

2. Zamboni P, Consorti G, Galeotti R, Gianesini S, Menegatti E, Tacconi G, Carinci F. Venous collateral circulation of the extracranial cerebrospinal outflow routes. Curr Neurovasc Res. 2009 Aug;6(3):204-12.

3. Al-Omari MH, Rousan LA. Internal jugular vein morphology and hemodynamics in patients with multiple sclerosis. Int Angiol. 2010 Apr;29(2):115-20.

4. Simka M, Kostecki J, Zaniewski M, Majewski E, Hartel M, et al. Endovascular treatment for chronic cerebrospinal venous insufficiency: is the procedure safe? Phlebology. 2010 Dec;25(6):286-95.

5. Zivadinov R, Schirda C, Dwyer MG, et al. Chronic cerebrospinal venous insufficiency and iron deposition on susceptibility-weighted imaging in patients with multiple sclerosis: a pilot case-control study. Int Angiol. 2010 Apr;29(2):158-75.

6. Menegatti E, Genova V, Tessari M, Malagoni AM, Bartolomei I, Zuolo M, Galeotti R, Salvi F, Zamboni P. The reproducibility of colour Doppler in chronic cerebrospinal venous insufficiency associated with multiple sclerosis. Int Angiol. 2010 Apr;29(2):121-6.

7. Zivadinov R, Marr K, Cutter G, Ramanathan M, Benedict RH, Kennedy C, Elfadil M, Yeh AE, Reuther J, Brooks C, Hunt K, Andrews M, Carl E, Dwyer MG, Hojnacki D, Weinstock-Guttman B. Prevalence, sensitivity, and specificity of chronic cerebrospinal venous insufficiency in MS. Neurology. 2011 Jul 12;77(2):138-44.

8. Doepp F, Paul F, Valdueza JM, Schmierer K, Schreiber SJ. No cerebrocervical venous congestion in patients with multiple sclerosis. Ann Neurol. 2010 Aug;68(2):173-83.

9. Krogias C, Schröder A, Wiendl H, Hohlfeld R, Gold R. "Chronic cerebrospinal venous insufficiency" and multiple sclerosis: critical analysis and first observation in an unselected cohort of MS patients. Nervenarzt. 2010 Jun;81(6):740-6.

10. Worthington V, Killestein J, Eikelenboom MJ, Teunissen CE, Barkhof F, Polman CH, Uitdehaag BM, Petzold A. Normal CSF ferritin levels in MS suggest against etiologic role of chronic venous insufficiency. Neurology. 2010 Nov 2;75(18):1617-22.

11. Baracchini C, Perini P, Calabrese M, Causin F, Rinaldi F, Gallo P. No evidence of chronic cerebrospinal venous insufficiency at multiple sclerosis onset. Ann Neurol. 2011 Jan;69(1):90-9. doi: 10.1002/ana.22228.

12. Mayer CA, Pfeilschifter W, Lorenz MW, Nedelmann M, Bechmann I, Steinmetz H, Ziemann U. The perfect crime? CCSVI not leaving a trace in MS. J Neurol Neurosurg Psychiatry. 2011 Apr;82(4):436-40.

13. Worthington V, Killestein J, Eikelenboom MJ, Teunissen CE, Barkhof F, Polman CH, Uitdehaag BM, Petzold A. Normal CSF ferritin levels in MS suggest against etiologic role of chronic venous insufficiency. Neurology. 2010 Nov 2;75(18):1617-22.

14. Centonze D, Floris R, Stefanini M, Rossi S, Fabiano S, Castelli M, Marziali S, Spinelli A, Motta C, Garaci FG, Bernardi G, Simonetti G. Proposed chronic cerebrospinal venous insufficiency criteria do not predict multiple sclerosis risk or severity. Ann Neurol. 2011 Jul;70(1):52-9. doi: 10.1002/ana.22436.

15. Yamout B, Herlopian A, Issa Z, Habib RH, Fawaz A, Salame J, Wadih A, Awdeh H, Muallem N, Raad R, Al-Kutoubi A. Extracranial venous stenosis is an unlikely cause of multiple sclerosis. Mult Scler. 2010 Nov;16(11):1341-8.

16. Zamboni P, Galeotti R, Menegatti E, Malagoni AM, Gianesini S, Bartolomei I, Mascoli F, Salvi F. A prospective open-label study of endovascular treatment of chronic cerebrospinal venous insufficiency. J Vasc Surg. 2009 Dec;50(6):1348-58.e1-3.

17. Malagoni AM, Galeotti R, Menegatti E, Manfredini F, Basaglia N, Salvi F, Zamboni P. Is chronic fatigue the symptom of venous insufficiency associated with multiple sclerosis? A longitudinal pilot study. Int Angiol. 2010 Apr;29(2):176-82.

18. Menegatti E, Genova V, Tessari M, Malagoni AM, Bartolomei I, Zuolo M, Galeotti R, Salvi F, Zamboni P. The reproducibility of colour Doppler in chronic cerebrospinal venous insufficiency associated with multiple sclerosis. Int Angiol. 2010 Apr;29(2):121-6.

19. Hojnacki D, Zamboni P, Lopez-Soriano A, Galleotti R, Menegatti E, Weinstock-Guttman B, Schirda C, Magnano C, Malagoni AM, Kennedy C, Bartolomei I, Salvi F, Zivadinov R. Use of neck magnetic resonance venography, Doppler sonography and selective venography for diagnosis of chronic cerebrospinal venous insufficiency: a pilot study in multiple sclerosis patients and healthy controls. Int Angiol. 2010 Apr;29(2):127-39.

20. Zivadinov R, Galeotti R, Hojnacki D, Menegatti E, Dwyer MG, Schirda C, Malagoni AM, Marr K, Kennedy C, Bartolomei I, Magnano C, Salvi F, Weinstock-Guttman B, Zamboni P. Value of MR venography for detection of internal jugular vein anomalies in multiple sclerosis: a pilot longitudinal study. AJNR Am J Neuroradiol. 2011 May;32(5):938-46. Epub 2011 Apr 7.