To confirm our clinical experience, this work wanted to verify, on a sample of 50 patients habitually classified from clinical signs as algo-dystrophic syndromes of the temporo-mandibular joint (TMJ), that modifications of occlusion by the wearing of a bite brings about an immediate modification of some stabilometric parameters.

      In fact, just after the beginning of the wearing of the bite one observes:

   Whatever be the interpretation of these facts, these results show clearly that the postural control reacts immediately to an instrumental modification of the occlusion.

   Daily clinical observation shows, if we accept to observe it, that a modification of occlusion (dento-dental) induces evident postural modifications (1, 21) and reciprocally (4, 16). But in order to become more convincing these modifications have to be demonstrated by means of an instrumentation which prevents observer from doing any clinical test.

   I already (3) put forward the interest of posturographic recordings to precise the eventual relationship between occlusal and postural modifications, and to follow up such patients. I gave some examples of this interest of posturographic recordings, but as demonstrative as they could have been, these clinical stories were not a proof.

   In order to establish the reality of this relationship between occlusion and posture, I undertook verifying if the modification of occlusion by a bite (and of mandibular position and of muscular balance corresponding to the new occlusion) brings about an objective reply from the fine postural control system.

    This first work is concerned only with the immediate repercussions.

   Objectivity was obtained by using a stabilometric platform, according to the standards (equipment, recording conditions and situations) of the 'Association Française de Posturologie' (8).With this apparatus, postural control from each subject can be characterized by values of several parameters which have been studied and standardized for several years by this Association, so that a subject can be classified as normal or not from a statistical point of view. (In our standard recording conditions, the position of the center of pressure of the subject on the platform is not very different from the projection of his center of gravity (9) on his support basis). Sampling frequency of the signal was 5 Hz; one minute recording; signal was filtered by a fourth order filter with a frequency cut at 2.5 Hz.

   All patients, suffering from an occlusal problem, presented a syndrome of postural dysfunction whose the occlusal origin was systematically verified according to the Meersseman (6) technic. After a first usual postural clinical examination, a thin piece of bristol (or salivary cotton according to the eventual loss of vertical dimension of occlusion) was interposed between arcades, then, after some swallowings, the same postural examination was made to see if there is a systematic variation of the postural tone in relation with the interposition; this variation must disappear when the piece of bristol is removed.

   Each patient was recorded three times, at some minutes interval:

   A statistical analysis was made of all parameters habitually studied in stabilometry, in the eyes-open and eyes-closed conditions (8.9): X-mean and and Y-mean positions, length of stabilogram, statokinesigram area, VFY and LFS parameters, Romberg's quotient, amplitude spectra of postural sway (first thirty values of the FFT, after application of the Hamming's window and normalization of the signal).
These parameters were compared, in matched samplings, by the Student's t-test.

   As the distribution of area is log-normal, comparison of different areas was made by their quotient, A1/A2, and not by soustraction, A1-A2.

   50 subjects, 43 females and 7 males (proportion habitually observed in this category of patients (13)), mean age 40 years, range from 19 to 71 years (younger subjects were excluded) were examined. Nine patients were excluded from the cohort because their recordings were not reproducible.

   The wearing of bite modified the muscle tone of the external rotators of inferior limbs for all patients (with a doubt for one subject).

   Recordings are reproducible for the 41 selected patients. Except the Y-mean position, it does not exist any significant difference for any of the parameters, including for the amplitude spectra in the frequency band 0.02/0.4 Hz, characteristic of the postural control.

   There is a significant difference (p<0.05) between areas, in the eyes-open situation (fig. 1) before (UI2) and after (MI) the wearing of the bite. This difference does not exist in the eyes-closed situation.

   Comparison between recordings UI2 and MI shows a statistically significant difference (p<0.05) between amplitude spectra in the frequency band 0.3 Hz for forward-backward postural sway in the eyes-open condition (fig. 2). This difference does not exist in the eyes-closed condition.

   There is no statistically significant difference among these 41 patients between values of parameters of the two first recordings, except the Y-mean position. But it is very well known that this particular parameter is not reproducible among normal subjects.

   There is a statistically significant difference between areas in situation UI2 and MI, in the eyes-open condition. The quotient of these areas is beyond the confidence limits at 95% of the distribution of random variations for ten subjects (fig. 1). The wearing of bite modifies immediately precision of postural control for one of four patients.

   This result is confirmed by the spectral analysis: a statistically significant difference appears in the frequency band 0.3 Hz when patients wear a bite (fig. 2). Such modification of the amplitude spectrum is habitually considered as a sign of a difference in the functioning of the fine postural control system (9).

   It is worth noting that these two effects of the bite, on area and on amplitude spectrum, appear only in the eyes-open situation but not in the eyes-closed situation. It has been known for a long time (19) that mandibular input has some relations with the functioning of the oculomotor system; in this way our results are coherent with anterior observations. Moreover, there is a topological connection between pathways of oculomotor proprioception and mandibular inputs, from the Gasser ganglion to brainstem, though, till now, no anatomical connection was demonstrated between these two pathways.

   For most of the patients one observes an abnormal peak (9) in the frequency band 0.2 Hz of the amplitude spectrum (fig. 3). This peak is found in the two recordings, UI1 and UI2, when it exists, and can not therefore be considered as an artefact. It is more acute in the eyes-closed condition and on the amplitude spectrum of left-right postural sway. This 0.2 Hz peak is related to ventilatory rhythm and appears when normal functioning of the spine is impaired (9). An interpretation which is coherent with dental clinic that notices almost constantly cervical pain in patients suffering from TMJ problems.

   Studies on the postural control show a great difference between static and dynamic situations. Nashner´s platform is well suited for studying postural control in dynamic conditions as a control system. But in static conditions what we call the fine postural control system, is not a simple one, it behaves as a dynamic non-linear system (5, 18) and inputs are at the limits of discontinuity. That is why it seems better not to study it as a simple control system by opening its feedback loops.

   It was necessary first to verify that recordings were reproducible in this particular subject sample, emotionalism being able to modify successive recordings (2, 7.12, 13, 14, 23, 25). This analysis shows that for nine on fifty patients (18%), recordings are not reproducible. Therefore we had to record in this order UI1 then UI2 and then MI, with the same procedures between recordings, the alone change between UI2 and MI being the pose of bite, therefore the modification of the relationship of occlusion.

   Of course another experience remains to be done with a sampling of patients with no TMJ pathology. But the present experimentation demonstrates the first step of our argument: modifications of occlusion brings about a change in the functioning of the fine postural control system.

   If another experience shows that the same holds true for a not-TMJ population this will be only an extension of this demonstration. But then it will be necessary to be preoccupied with postural modifications brought about by our works in mouth, and occluso-dentists will have to control their works by this kind of recordings.

   In the opposite case, these same therapists will be encouraged to practice such recordings that, positives, would be then the pathognomonic sign of occlusal problems.

   One may always discuss the choice and the realization of the protheses used during this experimentation, a gutter. This objection does not appear as determinant. Never mind if the gutter is good and improves the postural control or if the gutter is bad and worsens it, the objective of the protocol being to put in obviousness that a change of occlusion modifies immediately the regulation of the postural system. But now a work has to be done on the choice of the right gutter thanks to stabilometry.

   A complementary study with recordings made at different periods remains to be done, it will bring about information as for the stability of induced modifications. But a great deal of factors intervene in the integration of necessary information for the good functioning of the fine postural control (cf. 10). A particular and more difficult protocol is needed for this kind of complementary study. Before undertaking it, it appeared indispensable, in a first step, to demonstrate immediate repercussions of the modification of the occlusion.

   Meersseman´s procedure allows us to determine patients whose postural system is affected by modifications of occlusion.

   Among these patients and when their stabilometric recordings are reliable, one observes immediat and statistically significant changes of statokinesigram areas, in the eyes-open situations, in relation with a modification of occlusion. A similar study (21), on another recording platform, showed similar results. A recent work (20) using the Fukuda´s stepping test ends to concordant conclusions.

   It remains to try understanding why areas sometimes decrease sometimes increase, that is to say why precision of the fine postural control system (24) is either improved or worsened by the wearing of a bite.

   We must not to confuse precision of postural control observed by recordings and modifications of muscles tone showed by clinical tests.

   Many works remain to be done, but according to this work´s results it is no longer possible to contest that the fine postural control system is objectively concerned by occlusion.