This pilot study demonstrates that the presence and severity of CCSVI are associated with hypoperfusion of the brain parenchyma in patients with MS. In particular, a strong relationship was found between increased VHISS and decreased CBF in the GM, SGM and WM regions of the brain. No significant association was found in HC.
It has previously been demonstrated that MS patients show abnormal blood flow PWI patterns. These include increased MTT and decreased CBF and CBV within normal appearing WM and GM [4
]. Perfusion abnormalities in the normal appearing WM are present from the earliest stages of the disease. The GM perfusion changes seem to appear somewhat later in the disease [4
] and involve the thalamus, putamen and other SGM structures. PWI indices are also altered in both enhancing and nonenhancing lesions [12
]. The severity of the perfusion changes is more pronounced in progressive MS compared to relapsing forms of the disease [11
]. The hemodynamic abnormalities detected on PWI in patients with MS are currently interpreted as being a consequence of chronic inflammatory events related to local blood congestion and secondary hyperemia of the brain parenchyma [11
]. Furthermore, at this time, it is not clear whether reduced perfusion of the WM and GM in MS patients is a sign of vascular pathology or decreased metabolic demand [5
]. Alternatively, it can be hypothesized as the presence of a disorder that involves the major vasoactive substances. Norepinephrine, endothelin-1 and angiotensin II have been measured in their interaction with receptors only in the venous wall of the limbs, but not yet in CCSVI or in MS [16
]. Although hypoperfusion is consistently present in MS lesions and GM and WM, the formation of new lesions is preceded by hyperperfusion changes [17
]. Increased perfusion in the area of lesion formation could be a sign of vessel dilation mediated by proinflammatory cytokines.
An altered CBF pattern may be a consequence not only of local circulatory disturbances due to inflammatory mechanisms in acute or chronic phases, but instead could result from an outflow blockage situated far away from the lesions. CCSVI is a vascular condition described in MS patients that is characterized by stenoses caused by intraluminal defects such as web, septum, malformed valve or, rarely, by segmental hypoplasia/agenesis [1
]. Stenosing lesions of CCSVI have been classified among the truncular venous malformation in a consensus document [18
Therefore, CCSVI may impact local hemodynamics and overload microcirculation at places distant from the location of the mechanical stenosis, as in any condition of venous obstruction of the major trunks. Such a mechanism may lead to capillary hypertension and leakage, consistently contributing to inflammation [20
]. In this pilot study, we have shown a strong relationship between the severity of CCSVI and hypoperfusion in the WM, GM and SGM. There are other examples of overload of the cerebral venous circulation, albeit triggered by different mechanisms from those of CCSVI, leading to hemodynamic abnormalities similar to those reported in this study. For example, the picture of dural arteriovenous fistula, characterized by plaques very similar to those of MS as well as by retrograde cortical venous drainage, shows abnormal perfusion parameters of CBV, CBF and MTT [21
While we believe the present study provides important information about a poorly understood aspect of MS characteristics, it does have a number of limitations that motivate further work. First, the use of an FDR correction approach provides confidence in the overall body of results presented, but indicates that we expect approximately 5% of them to be incorrect. Thus, we can be confident that there is an overall relationship between VHISS and CBF in various areas of the brain, but need to further confirm individual regional findings in a more targeted study with a larger subject group. Second, when interpreting our perfusion results, it should be considered that absolute quantitation of CBF and CBV is challenging on dynamic contrast enhanced MRI. Some key parameters are estimated here and may not be correct. Nonetheless, the same MRI sequence, postprocessing steps and algorithm parameters were used for all subjects in all groups, so relative comparisons and correlations should still be reliable. Finally, several studies have reported that hypoperfusion of the brain parenchyma in MS patients advances with disease progression. It cannot be excluded that venous anomalies (CCSVI) may be secondary to reduced perfusion or that both are simply correlated with no direct causative relationship. Even if there were a causative relationship, the strongest r value we saw was 0.72, corresponding to a model explaining only about 50% of the overall variance seen. Thus, there may be other factors at work, and/or our measures may not be completely adequate to characterize individual hemodynamics. In either case, although we have established a relationship between CCSVI and reduced brain perfusion, the exact nature of that relationship remains uncertain and should be further investigated.
Several recent reports have presented evidence against the CCSVI hypothesis [23
]. Particularly, a study of 56 MS patients and 20 HC found no differences in cerebrospinal venous drainage using transcranial and extracranial Doppler imaging. However, in this study, there were significant deviations from the original Doppler methodology adopted in previous and in the present studies [1
]. The differences between the present study and other studies emphasize the need for a multimodal approach for the assessment of CCSVI.
Finally, accumulating evidence suggests that the increased energy demand of impulse conduction along excitable demyelinated axons and reduced axonal ATP production induce a chronic state of virtual hypoxia in demyelinated axons [27
]. In response to such a state, further hypoperfusion of brain parenchima, facilitated by venous outflow disturbances, may contribute to chronic necrosis of axons and mitochondrial dysfunction.