Introduction: In patients with gonarthrosis secondary to a femoral or a tibial mal union, the technical problems are different according to the localization and the importance of the deformity, the presence of boneless, the cutaneous and ligamenteous status and the degree of preoperative motion.

Matériel et méthodes: Between 1995 and 2003, 34 TKR have been performed in patients with mal unions either post trauma (26 cases) either secondary to surgery (osteotomy with hypercorrection).

There were 21 males and 13 females. The average age was 63 years (38 to 77)

The mal union was localized to the femur (9 cases) or the tibia (23 cases) or to the both femur and tibia (2 cases). The deformity was variable : varus, valgus, flessum, recurvatum or rotationnal mal union. IKS scoring, HKA, MFA and MTA angles were evaluated pre and post operatively. 11 cases of intra articular mal unions, secondary to epiphyseal fractures were operated : a TKR posterostabilized (9 cases) or constrained (2 cases) was performed.

In the extra articular mal unions (23) the technique depended on the degre of intraosseous deformity : medial or lateral release or osteotomy performed when the intra osseous deformity was more than 10°. TKR was associated with an osteotomy in one time surgery in 5 femoral mal unions and 12 tibial deformities.

Results: The average follow up was 8 years (4 to 13 years). Complications consisted in 5 phlebitis, 2 superficial skin necrosis, 4 stiff knees (flexion less than 80°). There was no infection in this short serie. The average IKS score was 65 before and 163 after operation. The average flexion was 83° preoperatively and 98° after surgery. Average HKA angle was 167° pre and 182° post operatively in the varus deformities. In the valgus deformity it was 191° pre and 181° post surgery.

Discussion: Average IKS scoring is less good in post traumatic mal unions than in the habitual TKR specially because of the motion : the knee is often stiff preoperatively and remain often stiff postoperatively. A quadriceps release is sometimes indicated either during the TKR either in a second time. Constrained implants (constrained condylar knee or rotating hinge) are necessary in some cases of medial or lateral insufficiency of the collateral ligament.

Hip-spine relationships should be better investigated in THP as lumbo-sacral orientation in the sagittal plane plays a critical role in the function of the hip joints. Lateral X-rays showing spine and hips together in standing, sitting or squatting positions characterize the adaptations of the sagittal balance and the functionnal interactions between hips and spine.

Acetabular cup implantation has to be planned for frontal inclination, axial anteversion, and sagittal orientation. The later refers to the sacro-acetabular angle, key-point in the spine – hip relationships, and that is redefined by the surgeon at the time of implantation.

Usual standard CT-sections are biased for evaluating acetabular anteversion. The conventional CT procedure does not refer to the pelvic bony frame and. the measured anteversion is a projected angle on a transverse plane, depending on the pelvic adaptation in lying position. This measured angle is often considered as anatomical anteversion, leading to some confusion. Therefore this angle is only a “functional” supine anteversion, reflecting the anterior opening angle of the acetabulum in a specific position. According to the sagittal orientation of the pelvis, the true functional acetabular orientation can virtually be assessed in various postures from adjusted CT-scan sections.

The EOS^™ low irradiation 2D-3D X-ray scanner is an innovative technology already used for global evaluation of the spine. This technology allows simultaneously “full body” frontal and lateral X-rays with the patient in standing, sitting or squatting positions; a tridimensionnal patient specific bone recontruction can be performed and the cup anteversion can be directly assessed according to the position.

We investigated the lumbo-pelvic parameters influencing the tridimensionnal orientation of the acetabulum. We compared the data obtained for real postural situations using the EOS^™ system and the measures from plane X Rays and classical CT scan cuts replicating standing, and sitting positions.368 patients with cementless THP were involved in a prospective follow-up protocol. Sacral slope and pelvic tilt, incidence angle, acetabular frontal and sagittal inclination were evaluated on AP and lateral standard XRays. Functionnal anteversion of the cup has been measured using a previously described protocol with CTscan cuts oriented according to standing and sitting sacral slope. The mean difference between CTscan and EOS^™ system was 4,4° with comparable accuracy and reproductibility.

Sacral slope decrease in sitting position was linked to anteversion increase (38,8° SD 5,4°). Sacral slope increase in standing position was linked to lower ante-version (31,7° SD 5,6°). The anatomical acetabular anteversion, the frontal inclination, and the sagittal inclination were functional parameter which significantly varied between the standing, sitting, and lying positions. We noticed that the acetabular parameters in lying position highly correlated to the one in standing position, while poorly correlated with the one in sitting position. The difference between the lying and the sitting positions was about 10°, 25°, and 15° for the cup anteversion (CA) and the frontal and sagittal inclinations (FI,SI) respectively. The poor correlation between the lying and sitting positions suggests that the usual CT scan protocol is biased and not fully appropriate for investigating the cases of posterior THP dislocation and subluxation, which happen in sitting position. On the contrary, a strong correlation was observed between lying and standing measurements with all the acetabular parameters (CA,FI, SI), suggesting that the classical CT assessment of the cup anteversion remains an interesting source of information in case of anterior THP

Each patient is characterized by a morphological parameter, the incidence angle. High incidence angle is linked to low acetabular anteversion, increasing the instability risk and anterior impingement in sitting and squatting position; higher anteversion angles are observed in low incidence angle patients, leading to more internal rotation of the hip in any position.

Lumbo-sacral orientation in the sagittal plane influences the tridimensionnal orientation of the acetabulum, especially for anteversion. Aging of the hip-spine complex is linked to progressive pelvic posterior extension. Impingement phenomenons, orientation of stripe wear zones and some instability situations can be interpreted according to those data.

This study points out the opportunity to adjust the CT scan sections to the sacral slope in functional position for properly investigating the orientation of the acetabular cup, mainly in case of posterior dislocation.

In addition, the mobility of the lumbo-sacral junction could be a crucial parameter in the mechanical functioning and the stability of a THP due to its impact on sacral slope and pelvic tilt. Therefore we also recommend doing dynamic lateral radiographs of the lumbo sacral junction in standing and sitting position for planning a THP implantation in order to detect stiff lumbosacral junction or sagittal pelvic malposition.

Anterior approaches have been suggested for THP revision in order to reduce dislocation rate. However, the exposure is considered to be more strenuous. The goal of the study was to evaluate if anterior approach in lateral position may improve the exposure.

From 2005 to 2007, 47 patients underwent THP revision, 34 times on the acetabular side, 2 times on the femoral side and 11 patients had a bipolar revision. Mean age was 64 years and mean BMI was 23. Patients were positioned on the lateral side and had an antero-lateral approach. During the femoral procedure, the leg was placed in a sterile bag stuck on the lateral side in order to optimize the exposure by positioning the femur in adduction and posterior translation.

Acetabular and femoral exposures were achieved correctly in all the cases allowing to perform all the revisions using this technique and no additional approach was needed in any patient. Antero-posterior femorotomies were performed in 7 patients for stem replacement and cement extraction, without any specific complication. Early post-operative anterior dislocations occurred in 2 patients who underwent monopolar cup revision. Dislocation was explained by an excessive anteversion of the remaining stems. 2 patients had an incomplete and transitory sciatic deficiency due to excessive posterior translation of the femoral head in the sciatic notch.

Using this technique, THP revision seems to be achievable even in complicated cases requiring stem revision and femorotomy. Dislocation rate was low; however a larger cohort is needed to confirm these preliminary results.