Purpose of the study: The purpose of this study was to confirm long-term changes in frontal alignment after wedge osteotomy(even for with an «ideal» postoperative wedge angle of 3–6°), that the frontal alignment is correlated with functional degradation and also with femorotibial skeletal torsion.

Material and methods: A non-consecutive retrospective series of 70 patients aged 57.5 on average at surgery for medial open-wedge tibial osteotomy were reviewed at 10–25 years. Goniometry measurements were obtained in the upright position after healing. Tibial and femoral torsion values were measured on the CT scan. Functional outcome at last follow-up was noted good, fair or poor.

Results: Postoperatively 80% percent of the knees presented frontal realignment within the 3–6° range. At last follow-up frontal alignment had changed on average 10° for 40% of knees. The change in frontal alignment resulted from a deterioration of the medial or lateral joint space and in 80% was associated with poor functional outcome. Knees which preserved valgus of 3–6° at last follow-up had statistically better results than the rest of the series. There was a correlation between valgus frontal misalignment and femoral torsion greater than 14° (anteversion) and between varus frontal misalignment and femoral torsion less than 14°. There thus appeared to be a linear correlation between postoperative changes in the correction and femoral torsion.

Conclusion: Good functional outcome of open wedge tibial osteotomy is correlated with stability of the axial correlation over time. Achieving postoperative valgus of 3–6° does not appear to be sufficient for stable axial correction. To achieve long-term preservation of the axial correction, it would be preferable to modulate the postoperative correction according to the degree of femoral torsion.

Purpose: Alumina is a bioinert ceramic used for total hip arthroplasty as an alternative to metal-on-polyethylene bearings which can wear producing massive osteolysis and loosening. The purpose of this retrospective analysis was to examine the Ceraver combination implant which uses a cemented smooth titanium femoral stem, a 32 mm alumina head, and a cemented alumina cup.

Material and methods: Between December 1979 and February 1983. 104 total hip arthroplasties were performed in 81 patients, mean age 57.8 years (2.1–70.9). The main indication was primary degenerative disease (71 hips). The Postel Merle d’Aubigné score was used for clinical assessment. Plain x-rays were used to establish the actuarial survival curves using the Harris criteria for radiological loosening for the cup and the Massin criteria for the femoral piece. Periprosthetic femoral and acetabular osteolysis were noted. Histological samples taken during revision procedures were analysed.

Results: Six hips with suppuration were not retained for analysis. The clinical scores for the other 98 hips were, at last follow-up: excellent in 34, very good in 21, good in 16, fair in 21, and poor in 6. Mean follow-up was eleven years and reached 18 years for 38 hips. Fracture of the alumina head (n=1), aseptic certain radiographic loosening of the cup (n=24), probable radiolographic loosening of the cup (n=12), and certain radiographic loosening of the femoral piece (n=3) were noted. Revision was required for 23 hips for replacement of the cup (n=23), the head (n=12), or the femoral stem (n=1). There were no cases of massive radiographic osteolysis. The histological examination of surgical specimens obtained at revision were normal in all cases (very moderate aseptic foreign body reaction). Excepting the cases of suppuration, the estimated actuarial survival without revision at 20 years was 61.4% (57.1% for he radiographic cup loosening criteria and 95.2% for the radiographic femoral implant criteria).

Discussion: This analysis confirms the long-term biotolerance of the alumina-alumina bearing despite the poor maintenance of the cemented alumina cup. It also confirms the good maintenance of the cemented smooth titanium femoral stem.

Conclusion: Cup anchorage must be improved to use the alumina-alumina bearing which does not cause osteolysis nor histological reactions.

Purpose: It was demonstrated in 1986 that to obtain a good radioclinical result at 10–13 years after valgus tibial osteotomy for the treatment of medial femorotibial osteoarthritis that the frontal valgus at this follow-up had to be 3–6°. In 1995, it was demonstrated that the side of deterioration in knees initially aligned between 2° varus and 2° valgus or with genu valgum (≥ 3° valgus) depended on the tibiofemoral axis: a positive index (tibial torsion greater than femoral torsion) favouring medial femorotibial deterioration and progressive varisation, and a negative index favouring lateral femorotibial deterioration and progressive valgisation. Can the post-osteotomy valgus be modified by the tibiofemoral index and prevent obtaining ideal correction at 10–13 years?

Material and methods: Forty-five knees with femortibial deterioration of the medial compartment were treated between 1987 and 1990 by tibial medial opening osteotomy for valgisation. Functional outcome in the 45 knees was assessed at a mean follow-up of 11 years (range 10–13 years). Postoperative frontal axis after healing and frontal axis at last follow-up was measured by goniometry in the standing position for all knees. A scan in the torsion position was obtained for 36 knees to measure the tibiofemoral index.

Results: At maximum follow-up, outcome was good in 58% of the knees, fair in 24%, and poor in 18%, differences which were not statistically different. Frontal axis changed with time. Among the 36 knees which had been realigned correctly (3–6° valgus) after healing, four exhibited an increase in valgus beyond 6° and five lost valgus passing below 3°. But ideal valgus was achieved at last follow-up for three of five knees which had been undercorrected, Among the 38 knees for which a torsion scan was available, 33 were correctly realigned postoperatively and 22 were well aligned at last follow-up. There was no statistical difference between knees with good, fair, or poor outcome among the 33 knees well corrected postoperatively (3–6° valgus). There was however a statistical difference between the good (64%), fair (27%), and poor (9%) functional results among knees with ideal valgus at last follow-up (p = 0.03).

The variation between the postoperative and last follow-up goniometry data exhibited a statistical correlation with the tibiofemoral index (p = 0.0005). If the index was less than 13°, most of the knees showed an increase in valgus (13 out of 19 knees); if valgus was greater than or equal to 13°, valgus was lost (for 12 of 19 knees).

Conclusion: To have the best chance of obtaining a good functional result 10 to 13 years after tibial osteotomy for valgisation, the valgus at this follow-up must be between 3° and 6°. But to achieve this valgus, the postoperative valgus must be modulated in relation to the tibiofemoral index. For an index ≥ 13°, the postoperative valgus should be pushed towards 6°; for an index < 13°, valgisation should aim at achieving a 3° postoperative valgus or less.

Leakage after simple suture repair of rotator cuff tears depends on the overall preoperative fatty degeneration index (FDI) of the muscles and preoperative fatty degeneration (FD) of the infraspinatus. When the FDI is = 2, cuff leakage is always observed after repair. The risk of recurrent tears of the supraspinus is high if the FD of the infraspinatus is > 1. However if the FDI is very low or nil, the rate of recurrent tears is 15%. These tears can be explained by tension on sutures in macroscopically and histologically abnormal tendons.

Purpose: The purpose of this study was to determine whether repair of rotator cuff tears using sutures without tension after removal of abnormal tendon stumps, a technique requiring tendon plasty, can improve the leakage rate.

Material and methods: Total repair of 24 rotator cuff tears was performed without tension on the sutures after resection of at least one centimetre of the supraspinatus tendon stump. This consecutive series was studied prospectively. The tears involved the supraspinatus and the infraspinatus in ten cases, all three tendons in 13 cases and only the supraspinatus in one case. Mean preoperative FDI was 1.18 (0.5–2.16); pre-operative FD of the infraspinatus was a mean 1.19 (0–2). The supraspinatus stump was resected from the trochiter in 14 cases, on the apex of the head of the humerus in seven and facing the glenoid cavity in three. Repair required translation advancement of the supraspinatus in 24 cases, translation advancement of the infraspinatus in six (associated with a rhomboid flap) and a trapezeal flap in three cases. Postoperative leakage was assessed with arthroscan (n=23) and ultrasonography (n=1) at one year.

Results: No leakage was observed in 20 of the 24 cuffs (83%). Recurrent tear of the supraspinatus was observed in three cases and in the supraspinatus and infraspinatus in one case each. The FDI of cuffs with recurrent tears (1.31) was not significantly greater than the FDI of cuffs without leakage (1.15) (p = 0.085). Preoperative FD of the infraspinatus of cuffs with recurrent tears (1.5) was significantly higher than that for cuffs without leakage (1.12) (p = 0.16). For the supraspinatus and the subscapularis, there was no significant difference between preoperative fatty degeneration of cuffs with recurrent tears and cuffs without leakage. The number of repaired tendons had a statistically significant effect (p = 0.012) on postoperative leakage: 23% of the recurrent tears after repair of three tendons showed no leakage compared with 9% after repair of one or two tendons (there was no significant difference for preoperative FDI, p = 0.33).

Discussion: Resection of macroscopically abnormal tendon stumps which requires tendinomuscular plasty, gives better anatomic results than simple suture. For an equivalent FDI, this allows fewer cases of leakage (83% without leakage compared with 50% after simple suture). Results were also better for an equivalent number of tendons repaired: 77% and 50% for three tendons and 91% and 55% for two tendons respectively).

Conclusion: Despite the almost constant need for plasty, rotator cuff repair using sutures without tension after resection of macroscopically abnormal tendon stumps gives, for an equivalent preoperative degree of fatty degeneration and an equivalent number of tendon repairs, better anatomic results than simple suture.

Purpose of the study: Surgical treatment of degenerative spondylolisthesis generally requires spinal fusion. Arthrodesis can be achieved via an anterior or posterior approach. Over the last ten years, minimally invasive methods have been developed to limit operative trauma secondary to the anterior approach. There have however been few studies comparing outcome after this new technique with classical open surgery. The aim of this work was to compare spinal fusion achieved with a cage implanted retroperitoneoscopically with conventional screw-plate fixation using open lombotomy.

Material and methods: Sixty two patients with degenerative spondylolisthesis (L3-L4, L4-L5, L5-S1) were reviewed: 39 had had conventional spinal fusion (group 1) and 23 had undergone a minimally invasive procedure (group 2). Clinical (Beaujon score) and radiographic assessment was recorded at last follow-up. All patients in group 1 were operated on by the same surgeon via lombotomy (L3-L4 or L4-L5 fusion) or subumbilical laparotomy (L5-S1 fusion). An intersomatic graft was combined with screw-plate fixation in all cases. Patients in group 2 were also operated on by the same surgeon (different from group 1) who used a retroperitoneoscopic approach (L3-L4 or L4-L5 fusion). An intersomatic cage was filled with cancelous bone and screwed into the intersomatic space.

Results: Mean follow-up was four years in group 1 and two years in group 2. Mean Beaujon score improved from 8 to 17 in group 1 and from 9 to 16.5 in group 2. All patients achieved bone healing at last follow-up without secondary displacement or disassembly of the osteosynthesis. There were no neurological complications and no infections of the operative site. One patient in group 2 developed a vascular complication requiring conversion to classical lombotomy. Blood loss was 1100 ml in group 1 and 200 ml in group 2.

Conclusion: Anterior fusion with implantation of an intersomatic cage via retroperitoneoscopic access is a reliable and effective method for the treatment of degenerative spondylolisthesis. It reduced postoperative morbidity but must obviously comply with classical indications for lumbar fusion.