À partir d’une série prospective de 86 patients opérés d’une arthroplastie totale de hanche de première intention non cimentée, nous avons étudié la conformité de la programmation et cherché à préciser si les divergences constatées conduisaient à des défauts de réalisation. Nous avons évalué l’efficacité avec laquelle la programmation permettait de restituer les différents paramètres en terme de positionnement des implants. Pour cela, nous avons comparé les implants prévus et les implants définitifs. Nous avons comparé l’architecture de la hanche prothésée avec celle de la hanche saine (position du centre de rotation, latéralisation du fémur, bras de levier fémoral (offset), inégalité de longueur). La longueur du col a été correctement prévue dans 75 % des cas, la taille de la cupule dans 62 % et la taille de la tige dans 64 %. La conformité exhaustive de la programmation était de 37 %. L’offset de la tige défini en préopératoire n’a jamais été modifié au cours de l’intervention. À plus ou moins 5mm près, une implantation idéale de la prothèse a été obtenue dans 60 % des cas, la hauteur du centre de rotation a été reproduite dans 81 % des cas et sa latéralisation dans 84 % des cas. La latéralisation du fémur a été reproduite dans 75 % des cas et l’offset de la hanche dans 66 %. Une égalisation de longueur a été obtenue dans 85 % des cas. La planification permet de déterminer de manière fiable l’offset de la tige à implanter. La prédictibilité de la taille des cupules non cimentées est plus faible, mais n’a pas d’influence déterminante sur la restitution de l’anatomie de la hanche dans notre expérience. La restitution de la longueur est obtenue en ajustant le réglage de la longueur du col en peropératoire. Nous recommandons de planifier sur un col moyen afin de pouvoir modifier facilement de taille du col au cours de l’intervention. La disponibilité de pivots disposant d’un offset variable permet de restituer au mieux la longueur et la latéralisation de la hanche prothésée. La chirurgie assistée par ordinateur pourrait palier aux imperfections de la planification préopératoire et apporter un confort au chirurgien.

For hip prosthesis surgery, the challenge is to obtain optimal function of the instrumented hip but also to eliminate any limb length discrepancy, correct the femur offset and guarantee the center of rotation of the hip joint. Preoperative planning for total hip arthroplasty (THA) enables determination of the appropriate length for the prosthetic neck and the size and eventually the type of implants to use. From a prospective series of 86 patients who underwent first-intention THA for implantation of a noncemented prosthesis, we studied the precision of the outcome as function of the preoperative planning. We also ascertained whether the preoperative planning was sufficient to provide the measurements necessary for correct implant position.

We analyzed a prospective series of patients who underwent first-intention THA from January 2004 through January 2006. To be eligible for inclusion, patients could not have a THA of the contralateral hip. The series was composed of 58 females and 28 males, mean age 70.2 years (range 45–93). The reasons for THA were primary degenerative disease (n=76) and aseptic osteonecrosis (n=10). The contalateral hip was intact and free of osteoarthritis with an anatomic presentation considered to be normal. The standard X-ray protocol included an anteroposterior view of the pelvis in the upright position and 10° internal rotation obtained preoperatively and three months postoperatively. All radiographic measurements were made by the same investigator using a manual nondigitalized technique. We compared planning parameters (pivot size and type, length of the neck, and size of the cup) with the final outcome in order to determine the compliance with the preoperative planning. All operations were performed in the lateral supine position under general anesthesia and by the same surgeon. The posterolateral Moore approach was used. All implants were press fit without cement, both for the cup and for the femoral piece.

All planning parameters selected for study (offset, size of the head and the cup, length of the neck) were available for 32 hips, giving an overall conformity of 37%. The length of the neck was as planned in 75% of hips, the size of the cup in 62% and the size of the femoral stem in 64%. The offset defined preoperatively was never changed during the operation. Ideal implantation (±5mm for all criteria selected for study) was obtained in 60% of hips; the height of the center of rotation was reproduced in 81% and the lateralization in 84%. Femur lateralization was reproduced in 75% of the hips and hip offset in 66%. Leg length discrepancy was avoided in 85% of the patients.

Preoperative planning reliably predicts the final offset of the implanted femoral stem. It is more difficult to predict the size of a press fit cup but in our experience the difference does not greatly affect restitution of the hip anatomy. We readily changed the length of the neck during the operation if necessary and have found that the leg length has been better with this approach. This leads to the observation that all of the planning parameters are not fully accurate because of the magnification effect, anatomic conditions, or possible defective execution. While the overall rate of conformity was low, looking at the results for each element separately provided a useful element for each phase of the operation. We recommend planning a medium length neck so it can be easily changed during the operation. The availability of offset measurements is particularly important to control hip lateralization and leg length. Current advances in computer-assisted surgery should be helpful in improving the imperfections of preoperative planning.