J Manipulative Physiol Ther 1999 (Jul); 22 (6): 363–367 ~ FULL TEXT
Licht PB, Christensen HW, Hoilund-Carlsen PF
Department of Clinical Physiology and Nuclear Medicine,
Odense University Hospital, Denmark.
BACKGROUND: A number of studies have investigated vertebral artery flow velocity. Because perfusion relates to the volume of blood flowing through the vessel, this parameter is of great importance when vertebral artery hemodynamics are investigated. We could not find any such Doppler studies in the literature, possibly because of known errors with previous techniques. New advanced color-coded duplex sonography has since been validated and may be used with confidence for volume flow investigations.
OBJECTIVE: To use advanced color-coded duplex sonography to investigate volume flow through the vertebral arteries during cervical rotation, as well as before and after spinal manipulation therapy.
DESIGN AND SETTING: A randomized controlled study at a university hospital vascular laboratory.
PARTICIPANTS: Twenty university students.
RESULTS: Volume blood flow through the vertebral arteries does not change with cervical rotation or after spinal manipulation therapy.
CONCLUSION: This appears to be the first in vivo Doppler study on human vertebral artery volume blood flow. Our results indicate that in symptom-free subjects there is no change in vertebral artery perfusion during rotation in spite of significant changes in flow velocity. This finding, as well as the observed changes in flow velocity reported by others, may be explained by a positional change in the vertebral artery diameter. In addition, we have investigated volume blood flow in the vertebral arteries before or after spinal manipulation therapy but found no significant changes.
From the FULL TEXT Article:
The vertebral artery has received much attention by the chiropractic profession because of its intimate relationship with the cervical vertebrae and because of a concern of cerebrovascular complications after spinal manipulation therapy. Previous Doppler sonographic studies have measured blood flow velocity, but this is the first study to investigate vertebral artery volume flow during cervical rotation. Our results suggest that, in symptom-free subjects, volume flow does not change with cervical rotation to either side.
It is known that most of the cervical rotation takes place at the atlanto-occipital level.  During rotation, the ipsilateral atlantoaxial joint is fixed, but the contralateral joint slides forward and down.  This causes the vertebral artery to stretch, kink, and narrow because it is fixed in the surrounding transverse foramina, paravertebral muscles, and fibrous ligaments. [6, 22, 23] Such narrowing has been observed in dynamic angiographic studies in cadavers [1, 4, 21, 24] and in vivo in selected patients with positional vertebrobasilar insufficiency. [23, 25-33] Magnetic resonance angiography also has been used,  and most investigators report a narrowing of the contralateral vertebral artery, [23, 26-29, 32, 34] but it has also been demonstrated in the ipsilateral vertebral artery. [25, 30-33, 35] In some cases rotation resulted in occlusion of the artery. [27, 28, 30, 33, 34] In symptom-free subjects, documentation is scarce, but similar findings have been postulated, although no angiographic evidence was published. [28, 36] On the other hand, lack of rotational obstruction has been observed in dynamic angiography in spite of positional vertebrobasilar insufficiency.  Throughout the literature it seems implicit that such anatomic narrowing of the vertebral artery results in decreased flow through the artery, but we postulate that this assumption may not necessarily be correct. In general, volume flow (Q) through any vessel may remain constant in spite of a dramatic decrease in the cross-sectional area (A) if there is a compensatory increase in the mean blood flow velocity (vmean) because Q = vmean × A. Similarly, an increase in cross-sectional area may be compensated by a decreased mean flow velocity. This is illustrated by an example from daily life: if one squeezes the tip of a water hose (ie, reduces the cross-sectional area) the volume flow through the hose remains the same but the flow velocity increases, and thus the jet of water reaches further. Both an increase and a decrease of vertebral artery blood flow velocity have been observed with noninvasive Doppler techniques during cervical rotation. [5-12] We have previously discussed how these results are conflicting and how this may be explained by different Doppler techniques or different levels of insonation.  We measured flow velocities midway between the origin of the artery from the subclavian artery and its disappearance into the foramen of the sixth transverse cervical process, a location that we presumed did not participate in the rotation of the head, and thus a location where the cross-sectional area presumably did not change.  We did not measure the lumen diameter in our previous study, but the observed changes might be explained by a change in cross-sectional area. Other studies have examined the vertebral artery at higher levels where the diameter of the artery is known to change during rotation, that is, above the transversal plane through the atlas [5, 10] or at levels C1–C2  and C3–C5. [8-9] The observed changes in flow velocity in these studies are likely to be the result of a change in the cross-sectional area. When the vertebral artery is completely obstructed during rotation, flow will terminate. However, it follows from the above that during incomplete obstruction the volume flow does not necessarily decrease.
Positional narrowing of the vertebral artery has been observed by various causes of extrinsic compression such as cervical osteophytes, [31-32, 35] scar tissue after thyroidectomy,  fibrous bands from the longus colli or scalene muscle [27, 29] sometimes forming a narrowed scalenovertebral angle,  neurofibroma,  thickened atlanto-occipital membrane,  displaced disk material,  or atlantoaxial joint instability. 
It is claimed that most instances of positional vertebral artery occlusion do not produce symptoms [1, 31, 36] because of adequate collateral circulation through the opposite vertebral artery. Symptoms are unlikely to appear unless the opposite vertebral artery is either hypoplastic or occluded. In this investigation both vertebral arteries were patent in all subjects. We have only found a single study documenting a causal relationship between a compressive agent and symptoms of vertebrobasilar insufficiency.  In this the authors demonstrated by single photon emission computed tomography a markedly reduced regional cerebral blood flow during an attack of vertebrobasilar insufficiency with angiographically verified compression of the vertebral artery. However, indirect evidence of such causal relationship is suggested by the fact that surgical decompression of compressive agents have been reported to improve or relieve symptoms of positional vertebrobasilar insufficiency. *Otherwise, a clear causal relationship has rarely been established. 
In addition to cervical rotation we measured volume flow through the vertebral artery before and after spinal manipulation therapy of a biomechanical dysfunction and found no change in vertebral artery perfusion. Similarly, in our previous study we found no change in peak flow velocities.  As discussed previously we are aware that volume flow was not measured during the actual manipulation, but 3 minutes later. Again, this was necessary to ensure the blinding of the study, but theoretically there could have been a short-lived change in volume flow. If so, the effect apparently had leveled off completely within 3 minutes.
This is the first in vivo Doppler study on human vertebral artery volume blood flow during cervical rotation, as well as before and after spinal manipulation therapy. Our results indicate that in symptom-free subjects there is no change in vertebral artery perfusion during rotation in spite of significant changes in flow velocity. This finding, as well as the observed changes in flow velocity reported by other investigators may be explained by a positional change in the vertebral artery diameter and, hence, the cross-sectional area. In addition, we have investigated volume blood flow in the vertebral arteries before or 3 minutes after spinal manipulation therapy but found no significant changes.