Mystery of Chronic Cerebrospinal Venous Insufficiency: Identical Venographic and Ultrasound Findings in Patients with MS and Controls

Venous abnormalities as detected by sonography form the cornerstone of the hypothesis that they may be the cause of MS. These authors used high-resolution sonography to image neck vessels in 32 patients with MS and 10 controls using the Zamboni criteria. They found no differences were detected between patients and controls by using these criteria and that normal physiologic narrowing was very common in the internal jugular veins in controls. BACKGROUND AND PURPOSE: Stenosis of the internal jugular, azygos, and other veins detected by using intracranial and neck Doppler and B-mode sonography and confirmed by venography has been reported in MS with a high degree of sensitivity. This article reports the results of sonographic findings in patients with MS and controls, looking for evidence of the controversial entity chronic cerebrospinal venous insufficiency. Furthermore, the venographic appearance in controls is documented. MATERIALS AND METHODS: Thirty consecutive patients with definite MS and 10 controls had TCD and high-resolution Doppler sonography of the neck vessels by using the published sonography criteria of Zamboni et al. Those with 2 positive findings consented to undergo contrast digital subtraction venography for delineation of possible venous stenosis. Nine consecutive patients undergoing digital subtraction venography for petrosal venous sampling or parathormone sampling had images of their internal jugular veins obtained as part of their procedure, and they were assessed for stenosis. RESULTS: No patient with MS or control had 2 positive sonographic findings; therefore, none were subjected to venography. Of the 9 controls undergoing venography for other reasons, 6 had bilateral IJV narrowing of ≥50%, and 2 others had unilateral narrowing. CONCLUSIONS: No difference was detected between patients with MS and controls by using the objective sonographic criteria of Zamboni et al. Furthermore, normal physiologic narrowing is found very commonly in the internal jugular veins in healthy individuals. Nonblinded subjective sonographic assessment of the IJV may erroneously lead to venography, the findings of which may be misinterpreted due to the lack of widespread knowledge about the appearance of these veins in healthy individuals.

M ultiple sclerosis is an inflammatory disease of the central nervous system characterized by focal lymphocytic infiltration and demyelination. The cause is unknown, but the common hypothesis invoked is that environmental factors trigger the disease in individuals with complex genetic risk profiles. 1 There is an increased migration of lymphocytes across a breach in the bloodbrain barrier, triggering a cascade of events resulting in the development of an acute inflammatory demyelinating lesion. 1,2 The close relationship of plaques to deep venous structures has long been observed 3,4 and is demonstrable on MR imaging. 5,6 A body of controversial literature suggesting an alternative etiology has arisen in the past 5 years stating that the insult triggering demyelination is secondary to impaired cerebrospinal venous return. This evokes venous hypertension, blood-brain barrier breakdown, hemosiderin deposition, and a secondary immunemediated response with resultant demyelinating plaque. 7,8 This has been termed "chronic cerebrospinal venous insufficiency". 9 The proponents of this theory have devised 5 sonographic criteria when studying the internal jugular veins, vertebral veins, and intracranial veins. [9][10][11] Two of these sonographic criteria being abnormal correlated very specifically in patients with MS and was not found in controls (P value Ͻ.0001). These patients then underwent venography and were found to have uni-or bilateral IJV stenosis Ͼ50% in 91% of cases and azygos abnormalities in 86%, whereas a control group without MS, who had normal sonography findings, underwent venography and had no venographic stenosis. 9 There was 100% separation between patients without multiple sclerosis and those with the disease based on the sonographic criteria and venographic findings.
Furthermore, a registry of 65 patients treated predominantly by balloon angioplasty of the narrowed veins demonstrated improvement in their MS clinical outcome measures after treatment at 12 months, particularly in the relapsing-remitting form of the disease with a decrease in gadolinium-enhancing lesions seen on MR imaging. 12 This however is comparable with the improvement seen with placebos in published double-blind controlled trials of MS therapy. 13,14 Subsequent investigators have been unable to replicate the findings of Zamboni et al 9 by using sonography [15][16][17][18][19] or MR imaging, 20 with contradictory unsupportive data by using combined sonography and MRV imaging. 21 Furthermore, 2 other studies with positive Zamboni criteria did not have venographic correlative stenoses in early-onset 22 or long-standing MS. 19 However the findings of Zamboni et al have been replicated by other groups, [23][24][25] and it has been suggested that special sonographic training is required. 26 Finally, Simka et al 24 published an article noting that 90% of patients with MS studied with sonography had 2 Zamboni criteria for chronic cerebrospinal venous insufficiency but in 2012, they reported that compared with venography, sonography is of limited value and the diagnosis is best made directly by venography. 27 In view of the controversy, we undertook a nonblinded study selecting patients known to have MS and healthy controls for ECD and color TCD by using rigorous application of the original objective criteria of Zamboni et al. 9 Any patients with 2 abnormal criteria would consent to undergo venography. In addition, as originally described by Zamboni et al, 9 consecutive patients without MS who were undergoing retrograde IJV cannulation for parathyroid hormone or petrosal sinus sampling had formal venographic images of their IJV obtained. These images were compared with published images of patients reportedly with chronic cerebrospinal venous insufficiency.

MATERIALS AND METHODS
Western Australia has a multiple sclerosis registry comprising Ͼ1000 patients, and an experienced neurologist 2 selected consecutive patients on the basis of being seen in the clinic with availability, consent, typical MS phenotypes with an EDSS score of Ͼ3.5 MRI, and exclusion of any coexistent medical or surgical history. These patients were offered and consented to TCD and ECD assessment of the IJV and vertebral veins. The studies were conducted by 2 experienced sonographers 3 (with Ͼ30 year's collective ECD and TCD experience) in conjunction with an experienced interventional neuroradiologist (with Ͼ15 years' ECD, TCD, and angiographic experience) by using the original criteria of Zamboni et al 9 : 1) Reflux in the IJVs and/or vertebral veins in sitting and supine postures (Ͼ.88 seconds) 2) Reflux in the deep cerebral veins (Ͼ.5 seconds) 3) High-resolution B-mode evidence of IJV stenoses and/or a CSA of the IJV in the supine position of Ͻ0.3 cm 2 4) Flow not Doppler-detectable in the IJVs and/or vertebral veins 5) Paradoxical increase in the size of the CSA of the IJV when going from a supine to an erect position.
Ten volunteers also underwent ECD and TCD after informed consent was obtained. We used a 7.5-MHz linear vascular array transducer (Toshiba, Tokyo, Japan) for ECD and a 2-MHz sector array transducer (Toshiba) with color TCD, looking at the basal vein of Rosenthal, which appears to be the most likely structure demonstrated on previous studies. Direction of flow and reflux was analyzed. The images were obtained and were available on a PACS network. They were formally reviewed by the neuroradiologist. The sonographer's findings and the neuroradiologist's report were compared. If there was a discrepancy, the patient was recalled for consensus imaging with the sonographer and neuroradiologist both present.
Consecutive patients who were undergoing jugular cannulation and venography for petrosal sinus sampling or parathormone sampling of the IJV were studied for stenosis (Ն50%) or the presence of bypass collaterals above a stenosis. Both are said to be important venographic findings in chronic cerebrospinal venous insufficiency. 9
All 30 patients had ECD of the neck veins, and 18 consented to concomitant or interval TCD. Ten healthy volunteers were also studied (6 women/4 men; mean age, 37.4 years; range, 25-54 years) with TCD and ECD.
No single patient or volunteer had 2 abnormal findings, the stated requirement for a sonography-based diagnosis of chronic cerebrospinal venous insufficiency. 9 Two volunteers had 1 abnormal finding of a CSA of the IJV Ͻ.3 cm 2 in the supine position but without a paradoxical increase in the CSA in the erect position (Fig 1). One patient with secondary-progressive MS had a small paradoxical increase in the left CSA from the supine to the erect position, but nothing else (Fig 1). This patient also had TCD findings that were normal.

DISCUSSION
Chronic cerebrospinal venous insufficiency is an entity proposed to be associated with MS and said to be detected in 90% of patients with MS by the use of ECD and TCD as a screening tool, 9,24 with promising treatment proposed by transvenous angioplasty or stent placement primarily of the IJV and, less often, the azygos vein. 12,28 This treatment is reported to be performed with low morbidity as an outpatient procedure, 29 with improvement in the patient's symptoms, though the restenosis rate in angioplasty of the IJV approaches 50% at 6 months. 12 The sonographic data have been confirmed by some groups 24,30 and are unable to be reproduced at all by others. [15][16][17][18]31 MR imaging venographic data have been mixed, with the sonographic findings not being replicated. 20,32 A combined sonography and MRV study 21 was not able to replicate the findings in 40 patients with MS. A recent MR imaging study by Garaci et al 33 demonstrated no significant specific interaction between the presence of chronic cerebrospinal venous insufficiency and MS in measured hemodynamic changes of mean transit time, cerebral blood flow, or volume.
We undertook a study with experienced vascular sonographers, a vascular interventional neuroradiologist, and an MS neurologist and were unable to find a single case on sonography fulfilling the criteria of Zamboni et al 9 for chronic cerebrospinal venous insufficiency. Moreover, the number of positive sonographic findings was higher in a small healthy control group.
How is this so? Studying the criteria carefully, one finds the following: The above criteria are objective and relatively easy to perform. Nevertheless, the basis of these arbitrary values is, at best, sketchy.
Reflux of Ͼ.88 seconds being abnormal is a derived figure and arbitrarily determined by using a study looking at IJV insufficiency. 15,34 The validity of this supposition has been questioned by others. 15 Likewise, the figure of 0.5-second reflux in the deep cerebral veins being abnormal appears to be derived from lower limb varicose vein incompetence studies. 35 Menegatti et al 11 stated, "Stenosis of the IJVs with a CSA of 0.4 cm 2 or less was measured by Lichtenstein et al." 36 This appears to be used as justification of the value of 0.3 cm 2 being abnormal. In the original article by Lichtenstein, 36 this finding was not described as an abnormality or stenosis but merely asymmetry in patients requiring central venous catheterization in an intensive care unit setting. The European Society of Neurosonology has published a position statement declaring that there is no scientifically sound validation for each of the 5 criteria quoted by Zamboni et al for the diagnosis of chronic cerebrospinal venous insufficiency. 37 The problem of a lack of blinding of the sonographer in this area of research has been widely discussed. 38 One of the subcriteria of Zamboni et al 9 is high-resolution B-mode evidence of IJV stenosis. This allows very subjective analysis of the IJV and valvular morphology, and the literature on chronic cerebrospinal venous insufficiency is littered with descriptions of IJV valvular abnormalities, 24,30 describing the presence of an inverted valve, a membranaceous or netlike septum, flap, immobility, annulus, double channels, anomalous orientation of the valve leaflets, or ectasia of the carotid bulb causing IJV narrowing. 24,30,39 Sixty-seven percent of patients with MS have these abnormal subjective findings compared with 17.5% of healthy individuals; these are the most common sonographic abnormalities found. 39 It has been suggested that special training is required 26,39 to facilitate detection, and it would appear that these subjective valvular findings are a central tenet in the technique.
Simka et al 27 subjected 58 patients with MS to ECD and catheter venography and concluded that ECD was relatively insensitive in detecting all venous abnormalities compared with a 92.2% incidence of IJV stenosis detected by venography and that the Although IJVs are commonly accessed for central lines and interventional procedures, detailed knowledge of the normal diagnostic venographic appearance of the IJV and azygos system is limited. The anatomy is variable, [41][42][43][44] and we hypothesize that any person subjected to venography is highly likely to have findings of apparent narrowing in these vessels. The reality is that any noninvasive test that purports to accurately differentiate patients with MS from those without MS on the basis of a supposed vascular stenosis will result in a "confirmatory" venogram, which may be misconstrued as detecting pathologically narrowed veins. This inevitably clears the way for endovascular treatment of this "abnormality." The essential missing step is that the knowledge of the normal venographic appearance of the IJV and its tributaries is absent from the investigation and discussion.
The safety of endovascular angioplasty and, to a lesser extent, stent placement has been reported in the setting of chronic cerebrospinal venous insufficiency and MS, 45 but the efficacy of treatment and the vascular pathogenesis on the basis of venous hypertension remains highly controversial. 46 The overwhelming evidence is that it is unlikely that a thickened jugular valve is responsible for this disease; and the absence of MS in patients with dural fistula, radical neck dissection, idiopathic intracranial hypertension, and other known causes of significant venous hypertension has also been noted. 46 Most important, the normal appearance of the head and neck veins needs to be widely disseminated to prevent inadvertent ascription of a pathologic process to these findings. The label of chronic cerebrospinal venous insufficiency as a medical condition needs to be questioned very closely.

CONCLUSIONS
We were unable to separate healthy controls from patients with MS on the basis of objective sonographic criteria specific for   chronic cerebrospinal venous insufficiency. Furthermore, venography in 8 of 9 consecutive control patients demonstrated findings identical to those published in chronic cerebrospinal venous insufficiency articles claiming to be pathologically abnormal in patients with MS.