Discussion

This "congenital theory" has been criticized in recent years (Kunnert19, Simon 37, Horeau 13, Yoslow 42), but is still favoriced by some autors. There are three arguments in favour of a congenital origin of lumbar osseous bridges:

25 of the 59 existing reports on bony bridges are classified by the authors as "congenital".

4. The fact that most of the lesions have been found during routine X-ray examinations for various other conditions, and not through follow-up of transverse process fractures.

5. The lack of known trauma as a causative argument (diagnosis by exclusion).

6. The suggestion that osseous bridges are a variation of the lumbar spine simulating the os sacrum ("sacralization"); this argument being partly supported by the occasional association with other congenital variants.

Ad 1) One result of our 2 studies does seem to favour the congenital theory: we had approximately 8 times more incidental findings than findings through follow-up of transverse process fractures. A comparison of the two studies can be misleading. No search for accidental findings was undertaken in the first "trauma" series. The second, random collective has a higher average age. Since an acquired or innate osseous bridge remains for a lifetime, the incidence of accidental findings naturally increases together with the age of the collective.

Ad 2) The suspected trauma occurred on average about 11 years prior to the diagnosis of the osseous bridge in 29 of the documented cases. (about 13 yrs prior in 17 own cases.)

The detailed history is often difficult, especially in the radiological routine. Unrecalled previous trauma is usually not the foremost problem facing the patient and his physician. Most of our patients revealed new details during second or third visits.

This Problem of consistent patient history can be combined with the role of "minimal trauma". The lumbar transverse process is vulnerable to fractures even from minimal trauma. In 1898, Thiem (40) explained that such fractures may occur after heavy lifting (indirect trauma). Stress fractures or fatigue fractures are described on both L3 transverse processes (Laarmann 20). It must be suggested that the patient may be unaware that he or she has fractured this part of the vertebral column.

We do not know the incidence of bone formation following such indirect injuries. It may be lower than after violent traumata, with more spread of bone tissue, but must be suggested as a possibility.

Rupture of muscles often leads to a Scoliosis convex to the ipsilateral side.

Exceptions to this rule were considered argument against the traumatic etiology (2);

The concavity of Scoliosis to the pathological side was considered a possible sign of congenital osseous bridging. Now it is not certain whether these often complex injuries permit such a simple role.

4 proven cases in our study have direct violent traumata with clear fractures. We suggest the incidence of osseous bridging is greater in serious than in mild trauma.

Not only fractures of transverse processes, but also contusions of the back with hematoma of the soft tissues will lead to paravertebral osseous bridges (traumatic myositis ossificans). The muscles damaged here are the Quadratus Lumborum, the Psoas Major, the Multifidus Lumborum and the Intertransversarii. While the bleeding in the intertransversarii muscles plays the most significant role, the spread of bone cells may also be important.

Ad 3) The hypothesis of "Sacralization" is not supported by embryological data. The transverse processes develop after the differentiation of vertebral parts in the 6th-7th embryonal week, and are first visible in the 7th or 8th month of gestation. The apophyses of those processes only develop after the differentiation of the other parts of the vertebrae. In the 3rd month, blood vessels grow in the apophyses and initiate ossification.

Other congenital variations (Spina bifida occulta S1, Lumbalisation of S1, Sacralization of L5 and Dorsalization of L1) are no more frequently found in our series than in the general population.

The congenital theory does not adequately explain why only 8 osseous bridges (in the literature) occurred in female patients.

The terms "Sacralization and Lumbalization" should be strictly reserved for the vertebrae adjacent to the lumbral-sacral junction. There seems to be no reason why process of Sacralization should omit L5 and S1, but in the majority of osseous bridges it is so. They are most typically located between L3 and L4, exactly where transverse process fractures are most often found.

We find additional arguments to support the traumatic theory:

- 10 cases in literature and 4 of our own cases prove to be undoubtedly caused by injury by radiological follow-up. Their final x-ray appearance very closely resembles that of the bridges believed to be of congenital etiology. The morphology of fracture healing is so varied that the value of special types remains unclear.

- Remarkably enough, the youngest patient with an intertransverse osseous bridge described in the literature was 23 years old, and in our series 20 years of age. The congenital theory does not explain this lack of occurrence in children and adolescents. One would expect a Scoliosis concave to be involved in congenital bridges as well as cross unions acquired in childhood (Gilsanz).

- One of the strongest arguments for the traumatic theory is the fact of "intervertebral disc degeneration and osteoarthric changes".

If the intervertebral junction was blocked by an osseous bridge the entire life, the occurrence of such changes could not be explained. Cross unions of transverse processes must have been formed during the lifetime, after degenerative changes have happened. Congenital tying of the intervertebral connection would save it from attrition.

We agree with the opinion of Yoslow (42), that "most (,if not all) of the osseous bridges are traumatic in origin".

It remains uncertain whether or not congenital cases exist.-

Including those taken from the literature, a total of ??81 cases were observed.

Together they give a clear picture of the distribution by age and sex, as well as orientation, bilaterality, position, form, size, and number of bridged vertebral segments. However, some observations are missing. For the future, most attention should be paid to the following types of osseous bridges:

3. Traumatic etiology proven by charting cases from the time of an accident (particularly in the case of osseous bridges originating form minimal fractures or only hematoma)

4. Observations combined with bony variants

5. The described O-shape /very regular formation with no longer recognizable transverse process)

6. Concavity of Scoliosis to the pathological side

7. Bilaterality in the same segment

8. Observations in siblings

9. Patients under the age of 20 years

Points 1 - 3 remain rare observations.

Points 4 - 7 would be not easy to explain with the traumatic theory described above.