Article Outline
To the Editor:
I would like to congratulate Symons et al1 on the extraordinary experiments the group did and reported on in the October 2002 issue of JMPT. Their painstaking work has gone to the heart of the matter with regard to the forces induced during cervical spinal manipulative therapy (SMT) and associated vertebral artery (VA) damage. However, upon reading the article a number of times, it occurred to me that there are quite a few issues yet to be resolved, and as such, the authors’ conclusions are at best premature. Additionally, their conclusions contradict their own words when, for example, they state that “we cannot interpolate these results into a living system” and then make the bold claim 5 sentences later that “under normal circumstances, a single, typical (high-velocity, low-amplitude) SMT thrust is very unlikely to tear or otherwise mechanically disrupt the VA.” This concluding statement is especially disturbing, considering that the most common site of VA injury from SMT (item 1, below), as well as (potentially) the most forceful cervical manipulation (item 2, below) have yet to be studied!
In order of importance then, I would like to identify those items that must be investigated before these investigators can make any public generalized claims of safety in regard to the forces induced during cervical SMT:
1.The sections of the VA (distal C0-1 loop and proximal [C6] loop) that all measurements were carried out on are of secondary concern. The most important section that must be measured is the C1-2 section, especially at the C2 foramen. This is the primary area of injury indicated by Terrett2 and Haldeman et al,3 as well as by the very recent work of Haynes et al.4 In fact, the distal C0-1 loop might even be mechanically protected because of the natural slack in this curved section and the small motions that occur at C0-1.
2.In terms of patient positions, the authors showed that the greatest strain occurs to the contralateral VA during rotation (ie, maximal right VA strain during left cervical rotation). They also showed that the greatest strains to the VA occurred on the side of thrust. Finally, they showed that comparatively, rotational manipulations created more strain than rotation-lateral flexion or more pure lateral flexion manipulations. What they failed to measure was the strain induced to the thrust side VA when the neck is fully rotated contralaterally (ie, right VA strains during thrust while the neck is rotated into left rotation) and a rotary manipulation is applied. I would suggest that this is the most forceful manipulation of all (and maybe one of the most common used by practitioners), and until this is measured and used as our baseline to address forces used in manipulation, we are simply deluding ourselves and misleading others. Also, the suggestion that high values of strain recorded on the ipsilateral VA (thrust side) are an “artifact” because of hand placement is a hollow argument. It makes complete sense that the nearer the tissue is to the area receiving the thrust, the greater the force that is applied to it, especially during rotational types of thrusts. Therefore, in all probability, this force is not an artifact, but a “true fact.”
3.The type of force that was induced to determine the point of mechanical failure was longitudinal stretching of the entire length of the VA (from the subclavian artery to the foramen magnum). First, using the entire length of the VA would give an exaggerated measurement because the VA is tethered as it passes through the transverse foramen; therefore, the best approximation would be to measure a section from C3 to the foramen magnum. This would presumably result in much lower failure points. Second, Haynes et al4 have strongly suggested that this type of stretching force is not the force that causes damage during SMT; instead, it is a compression force at the C2 foramen. Given this, ultimate failure point measurements during longitudinal stretching probably are not a good indicator of the forces required to disrupt the VA during SMT.
4.Postrigor 80- to 99-year-old cadavers are probably not able to achieve the ranges of motion (ROM) possible seen in chiropractic offices when testing and manipulations are done. The presented strain values measured in ROM and vertebrobasilar insufficiency testing are therefore probably too low and not representative of those actually occurring in life. In fact, the data collected during thrust procedures are probably artificially low for the same reason.
5.The authors themselves comment on the “large individual variations in the behavior of the VA.” Absolute failure forces ranged from 4.2N to 18N, a difference of over 4 times (an over 400% difference!). Similarly, failure strains ranged from 31% to 75%, a difference of over 2.4 times (140% difference). Moreover, all of the failure strains measured were probably artificially low as well due to the age of the cadavers, which the authors readily admit.
6.There are a number of smaller questions that need to be answered, such as: Did the ultrasound gel increase the elasticity of the VA during the experiment? Were there any microscopic signs of arteriopathy (this is an important area of speculation) that could have influenced these measurements?
Given the limitations as noted above, any broad, generalized conclusions about the safety of SMT and the strength of the VA cannot be made at this time. In the interest of good science and public safety, I strongly suggest that the authors revise their conclusions to something such as the following:
In studies on cadavers of older persons, the greatest strains measured on the VA were found on the side contralateral to rotation. Additionally, the greatest strains measured during manipulative thrusts were found on the VA on the thrust side. Finally, in terms of manipulation direction, the greatest strains measured were during rotational types of manipulations. Although these strains have not been studied in live patients, in the interest of patient safety, practitioners of high-velocity, low-amplitude manipulation procedures would be prudent to avoid such patient positions and rotary manipulations until further investigations have been performed.
Back to Article Outline
References
-
Symons BP, Leonard T, Herzog W.
Internal forces sustained by the vertebral artery during spinal manipulative therapy.
J Manipulative Physiol Ther. 2002;25:504–510
-
Terrett A.
Current concepts in vertebrobasilar complications following spinal manipulation. West Des Moines (IA): NCMIC Group; 2001;
-
Haldeman S, Kohlbeck FJ, McGregor M.
Unpredictability of cerebrovascular ischemia associated with cervical spine manipulation therapy.
Spine. 2002;27:49–55
-
Haynes MJ, Lesley AC, Melsom A, Mastaglia FL, Milne N, McGeachies JK.
Vertebral arteries and cervical rotation (modeling and magnetic resonance angiography studies).
J Manipulative Physiol Ther. 2002;25:370–383
