Over many years, I have I have written letters "on spec" to researchers, surgeons and others who might take-up these ideas. I have tried many different ways to get the basic ideas across in a short letter. I am adding a selection of these here. They will all be on this one page with the most recent the top.
Professor Nikolai Bogduk Dept of Clinical Research University of
Newcastle New South Wales Australia. Posted July 2005 Reposted 8th
Nov 2005 Reposted April 2006
First sent July 2005 as email
Professor Nikolai Bogduk
Dept of Clinical Research
University of Newcastle
New South Wales
Posted July 2005
Reposted 8th Nov 2005
Reposted April 2006
UK Registered Chiropractor
Dear Nikolai Bogduk.
I am an engineer and an independent researcher into low back pain -- and now a chiropractor too. I have on my desk the first, third and fourth editions of your book, The Clinical Anatomy of the Lumbar Spine and Sacrum. I used your first edition and your previous papers as the basis of my submission to Spine in 1987. (Not published). In my own 1992 book I referred to yours as "the definitive book containing all the correct muscle and ligament connections in the lumbar spine".
It is therefore very pleasing to me to see, in your third and fourth editions, the way in which your ideas "home in on" access by the nucleus to the nerves in the posterior annulus as being in some way the centre of the low back problem. This has been the centre of my hypothesis since 1987.
You cover this in considerable detail in your "low back pain" chapters all of which I am very happy with. Accepting that mechanism, many new questions are raised. The two most important to me are:
-- why should these nerves cause such initially paralysing and ultimately persistent pain?
-- what can we do to correct the situation or avoid it in the first place?
Taking the second question first: my hypothesis is that the L4 /5 and L5 S disks should remain wedge shaped at all times even in full flexion. It would seem likely that this will tend to keep the nucleus away from the nerves in the posterior annulus. (I hope you agree that these two joints are, in some way, the centre of the low back pain problem.) To support my suggestion on this wedging, may I quote an e-mail I sent recently to another researcher.
( ps his paper is at http://www.mubabol.ac.ir/Babol/19-L.htm#3
Yours is the only paper that I have seen which specifically looks at the wedge angle of all the lumbar discs. If we look just at the lowest two discs your wedge angles are 14° for L4 5 and 21° for L5 S. I believe that your figures are correct because they are more or less the same as my measurements. I was able to examine the x-rays of the group of English people selected by Piercy etc as being free of back problems for their paper in Spine 1984. Professor van Niekerk (of Jonk and van Niekerk 1961) was also able to obtain for me some x-rays of the lumbar spine of tribal living Africans, which had very similar angles of wedging (slightly greater).
If we now look at the flexion mobility of these joints in the above paper from Spine in 1984 we have 13° for L4 5 and 9° for L5 S. These figures are again confirmed by various papers including Adams and Hutton in Spine 1981. In other words in full flexion these joints will not flex beyond parallel sided and will always maintain a wedge shape with the wide end anterior (for people who are chosen as free of back pain in England, Iran or South Africa!)
(I wonder if it is significant that an x-ray survey of 100 healthy individuals would probably not be permitted now in England or Australia!).
In your book, you do remark on the wedge shape of the L5 S disk in the lordosis chapter but you very specifically include the L4 5 disk as similar to the other lumbar discs. Your picture on page 2 shows the L4 5 parallel sided and the L3 4 is wedge shaped. This is also true of the outline that has graced the front cover of each of your editions.
This tendency to show the L4-5 disk as parallel sided is fairly universal. It applies to wall charts, chiropractic books and many other publications. I have always suspected that this is because x-rays of back sufferers are so much more readily available than lateral x-rays of people with no back problem.
I believe that both the L4 5 and the L5 S. discs have to be looked upon as different from other lumbar discs and always wedge shaped in the "normal" spine. (Normal used here in the sense of ‘totally as designed’ even though it may only apply to a minority of those of us living a western lifestyle.) In these discs the wedge shape is always maintained even in full flexion. (I admit that the situation for L4 5 is less clear than for L5 S, but surely this fits well with the general belief that L4 5 is so often the trigger for a back problem.)
Do I have any evidence that the pain behaves in the way that I suggest? i.e. that the pain switches on when flexion occurs beyond parallel sided and can be switched off by extension.
By far the most important bit of evidence for me is provided by the Mackenzie system of therapy. Although I am qualified as a chiropractor, I am trained in Mackenzie by the Mackenzie Institute UK and I use it in most back pain cases. In the trial published in the BMJ in 1990, neither manipulative physiotherapy nor chiropractic was particularly effective in the early stages of an acute LBP attack. This is exactly when Mackenzie is most effective. It can completely stop the attack in its tracks. If the person is careful with sitting and uses my pelvic supports when driving they can be back to square one quite quickly. My interpretation is that the Mackenzie extension exercise squeezes the nucleus back to the anterior of the disk where it should be all the time in L 4 5 or 5 S.
I know that Mackenzie has a slightly dubious reputation inside orthopaedic circles. I assume this is because they see those cases where the nucleus squeezes in the wrong direction! In daily therapy, with normal back pain and no evidence of prolapse to start with, it is very effective.
I have always looked upon the "peripheralisation" part of Mackenzie as rather a red herring so I was happy to read in your low back pain chapter that more pain stimulation in a disk can send the pain further down the leg.
Although the Mackenzie extension exercise works well I find it even more effective to apply the extension force directly to the pelvis, rather than using the legs as lever. With the patient supine I put a cushion under the knees to take the tension out of the psoas and then I lift the iliac crest on each side with both my hands. I pull upwards and towards the patient's feet to apply an extension plus tension to the lumbo sacral junction area. For a light person where I can apply sufficient force this is much more effective than standard Mackenzie. Sometimes I, and the patient, can feel the nucleus oozing back to where it should be!
In the standard Mackenzie extension exercise, the psoas muscle will be working against the extension of L5 S and L4 5. This must be one of the important functions of the psoas muscle although I have not seen it mentioned in published papers. The line of action of the psoas lies anterior to the discs of L4 5 and L5S and will therefore apply a flexion force to these joints. This would occur at the completion of the running stride as the extension of the hip joint applies a direct tension to the psoas. The L4 5 and L5 S joints need this protection because they have such an extended shape in the normal standing position and would otherwise be damaged by running or other activities, which tend to take the hip joint into hyperextension.
I am pleased to see that you have extended the scope of your book to the pelvis and particularly the sacroiliac joints. I always say that osteopaths and chiropractors make most of their money by loosening off the sacroiliac joints. If a patient has had a low back problem for several months or years and the location of the pain seems to be the sacroiliac, buttock or thigh, I can be very confident that a few normal chiropractic treatments will solve the problem. This is where chiropractic did so well in that 1990 trial.
What exactly locks up in the sacroiliac joint is not so clear to me. I have modified my plastic demonstration pelvis with nylon cord rather than pins to hold the pubic symphysis together and to simulate the interosseous sacroiliac ligament and the sacro-spinous ligament. If I twist one side of this in relation to the other then one ASIS (anterior superior iliac spine) will be about one inch higher than the other with the iliac crests level. Twisting the opposite way gives the opposite situation. This is with the simulating nylon cord pretty tight.
This is exactly the same situation that I find with the human pelvis when the person has no problem. When there is a problem, the pelvis always seems to be stuck in the twisted position. Usually there is sensitivity over the sacroiliac joint so I had assumed that the locking was associated with some muscles that crossed the joint. Unfortunately your book does not mention any muscles crossing the sacroiliac joint so this is an ongoing question for me.
I had hoped that you would show that the laminar fibers of the multifidus below S1 had evolved to cross and strengthen the joint. (Where have they gone? Don't they exist? Page 102 Fig 9.4A). This would have been nice because the locking of the sacroiliac joint would then have been of exactly the same nature as the locking of any other spinal joint as recognised by the chiropractor and osteopath. It is these fibers, and correspondingly the rotatores in the thoracic spine, which I believe are responsible for the sublesation (I don’t use the term ‘chiropractic subluxation’ because it is a wrong use of a medical term. See ).
I was also interested to find recently that the Australian writers on core stability concentrate on these parts of the multifidus in their book ‘Therapeutic Exercise for Spinal Segmental Stabilization in Low Back Pain’.
I still have one contribution to the first half of the question ‘what can we do to correct the situation or avoid it in the first place?’
If in a particular case a person's severe or persistent low back pain seems to be caused by flexion of one of the lowest to two joints beyond this parallel sided limit then the lack of a supraspinous ligament at this level would seem to be important. The posterior ligamentous system is presumably adequate but works in a significantly different way. The insertion of a prosthetic supraspinous ligament between the spinous process of L4 and some point on the sacrum might be very effective in switching off the pain and "curing" an "unstable" joint before serious disk damage had occurred.
I have done a preliminary but fairly detailed document on this suggestion at
Now to second part of that question: "or avoid it in the first place". The association of sitting, particularly civilised sitting, goes back a long way in the literature and human experience. However this new geometrical limit on the flexion of L4 5 and L5 S makes the association with civilised sitting so much more specific.
The most important point here is that backrests, and particularly lumbar support, do not limit the flexion of the lumbo sacral junction, they increase it. This was clearly shown in the tabulations in the paper in Spine in 1979 " The Influence of Backrest Inclination and Lumbar Support on Lumbar Lordosis" (Spine 4: 52-58, 1979 by Anderson, Nachemson et al.).
That paper clearly showed that lumbar support flexed the L5 S joint in comparison with slumped sitting and that increased lumbar support flexed it more. This corresponds to mechanical common sense but I have not found any reference to this obvious point in any literature (and the authors didn’t notice it either!)
Putting these two points together makes it is unsurprising that driving is the one activity that shows up in any epidemiological survey of back pain incidence. My personal starting point from Easter 1981 is that car seats should be designed to support the pelvis and not the lumbar spine.
I do of course have a patent on pelvic support in seating and you might assume, as I did, that I would become rich beyond the dreams of avarice when this argument was accepted. Sadly the patent, applied for in 1982 or three, expires next year so that dream has long since faded and with it any unreasonable bias on my part! May I sum up the car seat situation by quoting from another letter of mine?
“The only clear statistical association is with driving cars. This was first shown by Kelsey and Hardy in 1970 and has been confirmed by many studies since. Would we expect this mechanically? The answer is definitely yes. Car seats tip the thighs and therefore the pelvis further back than almost any other seat. Car seats incorporate far more direct support for the lumbar spine than any other seats. We sit still in a car seat for far longer than we ever sit anywhere else. This mechanical situation could not be better designed to cause the problem outlined above.
If we combine Kelsey and Hardy's work with the surveys by Mark Porter, Professor of Ergonomics at Loughborough University UK it appears that a high mileage car driver is about twelve times as likely to have a serious low back problem as anyone who never uses a car at all.
This suggests a far higher correlation between driving cars and low back pain than between smoking and lung cancer. Comparison of these two surveys also suggests that the situation is far worse for the high mileage driver in the late nineties than it was in the late sixties. Is this compatible with the mechanical argument proposed?
Yes. It is more than compatible. It's predictable. In the eighties and nineties the angle of the sitting surface of car seats was tilted further backwards in order to counter the problem of slipping under the seat belt in an accident. At the same time lumbar support was introduced generally into car seats. The combination of these two is bound to increase the flexion force in the spine at the lowest spinal joints. We should expect the statistical situation to be worse and judging by these published statistics it is”
Believe it or not, this was meant to be a short e-mail. I am sure that you also find that writing a document of this type raises new questions that need to be addressed but that is an infinite road so I had better get back to my first important question:
-- why should these nerves cause such initially paralysing and ultimately persistent pain?
There is something illogical about asking why in a case like this. After all, we can never prove that evolution has worked in a particular way at some time in the distant past. Maybe it is because I am an engineer and also because back pain has played such an important part in my life that I need an answer to this question even if I can’t prove it.
My explanation is as follows:
In your book your cover in detail the deterioration of the disk, which can eventually lead to prolapse. You quote the experiments, which show that compression of a disk does not result in prolapse almost regardless of the circumstances. There is one case, however, that you do not seem to cover.
This is the case of compression combined with flexion beyond the normal flexion limit of that disk.
This is the situation studied by Adams and Hutton to win the Volvo Prize in 1981. (Your reference 317 page 200) I accept that this is not the normal mode of failure or prolapse but the question is: why not?
Adams and Hutton showed that hyper-flexion of a healthy disk combined with compression would, in most cases, result in sudden prolapse of the nucleus into the neural canal or intervertebral foramen. This seems to correspond to mechanical common sense. If the flexion is sufficient then the posterior annulus will be in vertical tension despite the compression of the whole disk. It does not seem surprising that it is unable to contain the large hydrostatic pressure in the nucleus while simultaneously being torn vertically.
The shape and muscle connections of the human lumbo sacral region can produce exactly these conditions. Many of the most important muscles have their origin on the prominence of the sacrum or the posterior iliac spines. However one evaluates the "hinge point" of the lower lumbar spine on flexion, it will be above (more rostral than) these muscle connection points. To the eye of an engineer, this whole structure is an "over-centre mechanism". Beyond a certain angle of flexion, on lifting, the line of action of the muscles will get so close to the hinge that the structure will collapse. The flexion of the lowest discs will increase rapidly while compression remains high. The result will be prolapse by hyperflexion as shown by Adams and Hutton.
Your reference 69 on page 114 does not pick up this point because it covers only the normal range of motion. Figure 4 B in that paper shows L4 5 and L5 S parallel sided in full flexion. This corresponds exactly to the criterion suggested earlier in this e-mail. It seems good evolutionary design that total muscle moments remain approximately constant within the normal range of motion. However there is nothing that can be done mechanically beyond this range to avoid the "over-centre effect". The only solution is to stop the person using the spine in that range by a warning pain that is initially paralysing and very persistent.
I suggest that it is the coincidence, that muscle offsets become inadequate in flexion as L4 5 and L5 S flex through the parallel sided shape, that has allowed such an effective warning pain to evolve so easily by such a simple mechanism. (The shape of the sacrum is of course defined more by the need to maintain an adequate birth canal than any other consideration.)
I used your papers on spinal musculature to do my own detailed analysis of the effect of flexion in 1987. This paper (The Obstetric Reason for Lordosis and the Implications for Lifting and Low Back Pain-unpublished) is available at This included flexion of L4 5 and L5 S to a wedge shape of 10 degrees, wide end posterior, and confirmed the “instant evaluation” above.
It would have been nice if the warning pain had evolved to apply only when flexion was excessive and to switch off completely when the safe shape had been restored. As we all know this is not the case. The characteristics of the pain are what one would expect of the mechanism that you describe. The nucleus ‘leaks’ to the nerves in the annulus and there is obvious difficulty in correcting the situation quickly. I have listed various characteristics at such as the following:
6) The pain is most severe when it first occurs and can take a long time
to die away.
10) In some cases (possibly in a lot of cases) the shape can have been wrong from
childhood. In that case the phase of severe warning pain will
not have occurred and the person will simply have a ‘weak’ back with a
tendency to bouts of ache due to sitting, bending or lying in bed.
11) Usually this gets worse with age.
I enjoyed your new chapter on instability because, I think, your views are exactly the same as mine.
The description ‘unstable’ is often used to describe the whole situation of an acute back. The spine seems to have lost the stability that we normally rely on and the sufferer has to hang onto things for stability. Somehow this use of "unstable" seems to have been taken by biomechanists to refer to the spine with its passive components for stability. But the spinal column with only the passive components doesn't have any stability. As a column, it isn't even straight for heaven's sake!
The stability of the spine relies completely on the muscles. Pushing on a person's spine doesn't distort it at all whether they have a back problem or not and regardless of whether they are standing or lying down relaxed. It is the muscles that give this stability or rigidity. The spine only bends as a part of the movement of the body under muscular control. (I have always suspected that this is why Anderson, Nachemson et al. didn't notice the effect of lumbar support in their 1979 paper in Spine. The instantaneous effect of lumbar support is very small. Civilised sitting however is not an instantaneous effect. It applies for hours and hours and years and years.)
Many radiographic surveys have shown that one totally degenerated disk is often asymptomatic so long as the spinal muscle system is working normally. The instability referred to by any clinician is a muscular effect and is always associated with pain. I would refer to it as a part of the "shape warning pain".
I am comparing your criticisms and quite strong words in this chapter on instability with your comments on internal disk disruption in your back pain chapter. I think you see far more value in fully understanding the latter. So do I.
In my 24 years of studying low back pain I have spent much time developing chairs and car seats. I thought that this would lead on to a wide acceptance of these basic points about the spine. In fact it hasn't done anything of the kind. I have therefore decided to approach researchers like yourself in the hope of working towards a more general understanding of low back pain and some real solutions. I hope that you will see some value in the suggestions that I have made in this e-mail and that you might have some ideas on how we could take cooperation further. I hope that I will hear from you.