Purpose: After implantation of a total knee arthroplasty (TKA) for osteoarthritis, early demineralisation of the superior tibial epiphysis occurs with modification of bone mineral density (BMD) in the two compartments. The long-term trend is not known to date. We report the results of 38 prostheses followed prospectively with densitometric measurements at minimum five years.

Material and methods: This prospective analysis included 38 TKA implanted for primary degenerative disease in patients with a mean age of 70±4 years at implantation, 60% women. Clinical assessment (IKS) and radiography (HKA) as well as osteodensitometry were recorded before surgery, at six months, one year, two years and five years. Bone mineral status was assessed using the densitometry of the femoral neck before surgery and at five years for all patients. The DEXA method was used for each knee on the anteroposterior film. Seven zones were defined around the tibial implant, in particular two under the medial and lateral plateaus, and under the stem. We studied changes in each zone over time. The alpha risk was set at 5%. Inter- and intraobserver reproducibility was 2.9% and 2.8% respectively.

Results: At five years follow-up, the mean IKS function score was 85±19 and the knee score was 918. The mean mechanical axis was 180±2° with symmetrical distribution. There were no progressive lucent lines. The BMD of the homolateral femoral neck did not change significantly (0.763 g/cm2 at inclusion and 0.750 g/cm2 at five years), unlike the natural evolution in a reference population (women -4.5%, men -2.4%). The mean BMD in the seven zones decreased significantly (11.6%, p< 0.0001. Mean BMD was 0.936 g/cm2 at inclusion and 0.863 g/cm2 at six months), 0.823 g/cm2 at five years. BMD decreased very strongly from 0 to 6 months (−6.51%, p< 0.0001) then more slowly to the end of the first year (−3%) and finally declining regularly, but non-significantly, at a slower rate from 1 to 5 years. Study of the seven zones showed a difference in changes in the BMD between the medial, lateral and stem zones. The two medial zones decreased significantly from 6.33% to 6.18% especially during the first year (−2.06% and −2.09%) and more moderately from 1 to 5 years (−1.6%, −2.65%). The lateral zones showed a greater average decline in BMD (−10.5%, −8.92%) between 0 and 5 years: −8.57% and −6.75% during the first year then at a slower rate. The greatest loss in BMD was found under the stem; −14.3% at five years. Here again, between 0 and 6 months the decline was rapid: −8.09%. It reached −12.74% at one year then varied little, −1% and −2% between 1 and 5 years.

Conclusion: 1) Bone remodelling under the tibial base plate occurs early after implantation of a TKA. It occurs during the first year (especially the first six months). 2) Remodelling is more pronounced laterally than medially (good realignment but persistent varus stress with greater stress on the medial than lateral side). 3) The greatest loss in BMD occurs under the stem. 4) BMD of the femoral neck remains stable, unlike the evolution observed in a reference population.

Purpose: The risk of recurrence and progression to chronic instability after a first episode of anteromedial shoulder dislocation is high in young patients. Risk assessment has varied in published reports but is constantly high in subjects aged less than 25 years. The injuries occurring during the first episode are poorly identified and rarely treated. We thus propose an arthroscopic assessment for young subjects with sports activities to identify lesions and achieve stabilisation after the first dislocation. The purpose of this work was to report the lesions observed and present our surgical protocol.

Material and methods: Between February 2002 and March 2003, we included fifteen patients in a prospective study. All patients were aged 17–25 years at the first episode of traumatic anteromedial dislocation of the shoulder. The patients were informed of the “usual” orthpaedic treatment and of the risk of recurrence. We proposed an arthroscopic assessment of their lesions and concomitant treatment. All patients accepted this therapeutic alternative. All procedures were performed by the same operator within ten days of dislocation. Patients were immobilised for 21 days with an elbow to body brace followed by rehabilitation in an outpatient setting, avoiding external rotation for 21 days. The Duplay score was determined.

Results: In this prospective series of patients, we identified a haematoma, a Malgaigne notch, and disinsertion of the anteroinferior capsulolabral complex in all patients. We were unable to find any glenoid or ligament injury on the humerus. The cuff was intact in all patients except one who had a deep wound of the supraspinatus. Lesion suture with resorbable anchors was satisfactory in 14 patients. We have not observed any recurrences. Physical examination did not disclose any apprehension and there has been no case of altered external rotation (< 5).

Conclusion: Considering the major risk of recurrent dislocation after a first episode in these young patients, we have studied an alternative to orthopaedic treatment. All patients accepted the proposed arthroscopic treatment. All patients presented capsulolabral detachment which was easily treated. At last follow-up, all patients have recovered a pain-free stable shoulder. This was a small series with a short follow-up so these results must be considered with caution. They are nevertheless very encouraging.

Purpose: The position of the femoral implant in external rotation remains a controversial issue. It can be determined using bone landmarks (Whiteside line, parallel to the biepicondylar axis, 3° external rotation from the posterior condylar plane). For the last seven years, we have related femoral rotation to the orientation of the tibial cut in order to ensure good femorotibial stability in flexion using specific instruments (Cores®). This prospective study was conducted to examine the position of the femoral implant determined with this method and to measure the position from bone landmarks.

Material and methods: Twenty consecutive patients were included in this study. Bilateral computed tomographic measurements were made before and after surgery. Joining 8mm/8 slices were obtained for the femoral necks and 5mm/3 slices for the knees. The angle of femoral torsion was defined in two ways: the first by the angle formed between the axis of the femoral neck (on two superimposed slices) and the tangent to the most posterior part of the femoral condyles; the second by the angle formed between the epidondylar line and the posterior condylar line.

Results: The preoperative scans demonstrated that the angle between the biepicondylar line and the posterior condylar line was 5.8±1.5°. Using Cores®, led to an external rotation of the femoral implant to 2.7±0.6°. The postoperative scans demonstrated that the angle between the biepicondylar line and the posterior condylar prosthetic play was a mean 3.3°. The measurements using the femoral neck were less precise, with, in one case, an external rotation of 5°. The patella was well balanced postoperatively (irrespective of the external rotation position of the femoral implant).

Discussion: The angle of about 6° between the biepicondylar line and the posterior plane of the condyles has also been reported by others (Beaufils, Matsuda). To obtain a rectangular space in flexion, the posterior condyle cuts are more important medially than laterally. We found a correlation between the correction provided by the specific instrument set and the difference in the posterior condyle cuts, demonstrating the intraoperative precision of Cores®. It is difficult to orient the femoral piece parallel to the biepicondylar axis. This study demonstrates that there always remains 2 to 3° of inclination of the biepi-condylar axis from the posterior condylar plane.

Conclusion: The positioning the femoral implant parallel to the biepicondylar line leads to inducing an important external rotation. While using 3° rotation systematically would reduce the risk of internal malrotation, we feel it better to adapt the rotation to each individual knee depending on the anatomic presentation. Cores® enables positioning the femoral implant in external rotation as a function of the ligament balance obtained in flexion after peripheral tension is applied. This enables avoiding medial femorotibial laxity in flexion.