Le plan pelvien antérieur ou plan de Lewinnek est couramment utilisé comme plan de référence pour guider la navigation informatique en vue de l'implantation d'une cupule de prothèse totale de hanche (PTH).

Des clichés radiographiques de bassin de profil strict en qualité argentique de 106 sujets ont servi de matériel d'étude. Ces 106 sujets se répartissaient en : 82 porteurs d'une PTH (40 ayant eu une luxation et 42 ayant une PTH stable choisis par tirage au sort) et 24 sujets témoins indemnes de pathologie. Dix-neuf des 42 patients porteurs d'une PTH stable ont eu un cliché avant et après la pose de l'arthroplastie, les 24 témoins ont eu un cliché de profil en position couchée et debout.

Trente-huit pour cent des sujets avaient en orthostatisme une orientation du plan pelvien antérieur en dehors d'une fourchette de ± 5 ° par rapport à la verticale et 13 % au-delà d'une fourchette de ± 10 °. Le passage de la position debout à la position couchée influençait de manière significative (p = 0,002) l'orientation du plan pelvien antérieur avec des variations extrêmes comprises dans une fourchette de plus de 20 °, la moitié de la population variant de plus de 7 °.

Les variations de l'orientation du plan pelvien antérieur par rapport à la verticale nous conduisent à penser qu'il ne s'agit pas d'une référence fiable pour l'orientation de la cupule d'une PTH puisque l'on risque de se tromper 1 fois sur 3 de plus de 10 °, soit la moitié de la valeur de l'antéversion anatomique de l'acétabulum. Une assistance informatique à la pose d'une PTH sans imagerie préopératoire doit, soit bénéficier d'un repère plus fiable, soit s'affranchir d'un plan de référence en utilisant une approche cinématique de la hanche opérée.

The anterior pelvic plane, also called the Lewinnek plane, is commonly used as the reference plane to guide imageless computer assisted surgery for total hip arthroplasty (THA) because this plane is considered to be globally vertical in the standing position. To our knowledge, no study has evaluated this hypothesis or the potential variations in orientation as a function of gender, position of the subject, or THA insertion. The purpose of this work was to examine these different hypotheses in a radio-clinical study.

The orientation of the anterior pelvic plane was measured in relation to the vertical plane on plain lateral x-rays of the pelvis in the standing position. X-rays were studied for 106 patients: 1) 82 patients with a THA (40 with at least one dislocation, 42 with a stable hip selected randomly, 19 with a standing lateral x-ray before and after arthroplasty) and 24 control subjects for whom lateral images were obtained in the supine and standing positions to assess potential position-related changes in orientation.

The orientation of the anterior pelvic plane was not affected by gender or age. The anterior pelvic plane formed an angle greater than 5° with the vertical plane in 38% of patients and more than 10° in 13%. The orientation of the anterior pelvic plane was not significantly different between the study groups (control versus THA) nor between the THA groups (stable versus dislocated). The supine position modified significantly the orientation of the anterior pelvic plane which changed on average from 1.20° to -2.25°; the change was greater than 7° in twelve subjects. Implantation of a THA did not modify signi-ficantly the orientation of the anterior pelvic plane in the standing position for the 19 subjects [the variations were small (-1° to 7° on average, range -21° to 8°) but greater than 5° for 7 of 19 subjects].

Most teams use the anterior pelvic plane to guide computer-assisted navigation, considering that this plane is vertical in the standing position. Our findings show however that this is not true for 38% of subjects with a margin of error of 10°, i.e. about half of the anatomic anteversion of the acetabulum. Moving to the standing position would produce a significant variation in the orientation of the anterior plane of the pelvis. This is a source of error which has not been integrated into most imageless navigation systems. Similarly variations in the position of the pelvis from the standing to sitting and supine positions which can produce impingement or dislocation have not been taken into consideration.

Variations in the orientation of the anterior pelvic plane in relation to the vertical would suggest that this plane is not a reliable reference. To our knowledge, there is no reliable reference which can be easily identified during the operation which would take into account variations in the position of the pelvis. We thus believe it would be preferable to attempt to operate without a reference plane, relying on a more kinematic approach to guide computer-assisted implantation of the THA cup.