Purpose: Collapse or necrosis of the head of the humerus after fracture of the proximal humerus can be an indication for shoulder arthroplasty. The poor results classically reported have led to a search for factors predictive of the anatomic and functional outcome after arthroplasty for fracture sequelae.

Material and methods: Among the 221 Aequalis prostheses implanted for the treatment of sequelae after fracture of the proximal humerus, 137 (62%) developed post-traumatic avascular osteonecrosis of the humeral head with a deformed callus of the tuberosity. Head tilt was in a valgus position in 83 shoulders and in a varus position in 54. Mean age was 61.49 years. The initial fractures were subtuberosity fractures in 20% of the cases, three-fragment fractures in 32%, and cephalobituberosity fractures with four fragments in 48% of the cases.Twenty-five percent of the patients had undergone initial osteosynthesis. The rotator cuff was repaired in 4.5%, and two osteotomies of the less tuberosity and four osteotomies of the greater tuberosity were performed at implantation. Pre- and postoperative clinical and functional outcomes were assessed with the Constant score and a function index composed of eleven usual movements.

Results: Mean follow-up was 44 months (24–104), with a mean 42° gain in anterior elevation and a 29° gain in external rotation. The gain in the Constant score was +32 points (mean score 61), and +43% with the weighted score. The four subscores (pain, motion, activity, force) improved two-fold. Analysis of the eleven usual movements demonstrated recovery in 88% of the patients. The subjective satisfaction index was 86%. The mean Constant score improved significantly more after total arthroplasty (67 points) than after humerus prosthesis (55 points). Complications (per- or postoperative mechanical problem, infection, neurological disorder) or the need for revision were unfavourable elements.

Discussion: Deformation and deviation of the tuberosities, especially the greater tuberosity, often leads to osteotomy during the implantation procedure. In this series, osteotomies were exceptional and functional outcomes showed that deviated tuberosities could be preserved without having an unfavourable effect on functional prognosis. Much on the contrary, the absence of a tuberosity osteotomy simplified the operative procedure and produced much better functional outcome than observed in earlier studies. The rate of complication for secondary prosthetic implantation is not negligible (15%) and a simplified procedure without osteotomy is a useful criterion. Furthermore, rehabilitation may be started earlier after implantation when it is not retarded by osteotomy bone healing, found to be an unfavourable factor.

Conclusion: Implantation of a shoulder prosthesis after collapse or necrosis of the head of the humerus after proximal fracture with varus or valgus impaction has provided good functional outcome without tuberosity osteotomy since the deformation of the tuberosity is generally well tolerated.

Purpose: Consolidation of the tuberosity is the key to success of shoulder arthroplasty for fracture. The purpose of this study was to assess the number and causes of tuberosity complications in order to find solutions for this problem.

Material and methods: This retrospective multicentric study included 334 shoulder prostheses implanted for fracture between 1991 and 2000. Two different prostheses were used: 300 standard Aequalis prostheses implanted between 1991 and 1997 (mean follow-up four years) and 31 Aequalis Fracture prosthesis between 1999 and 2000 (mean follow-up nine months). Radiological results were assessed on the postoperative and last follow-up x-rays. The Constant score was used for clinical assessment.

Results: For the 300 standard prostheses, the Constant score was 54 points with active anterior elevation = 104°. For the 31 fracture prostheses, the Constant score was 58 points with active anterior elevation = 114°. According to the operator’s assessment, 49% of the postoperative radiological results were fair or poor and objectively 35% of the tuberosities were poorly positioned with the standard prosthesis and 22% with the fracture prosthesis. Twenty-six percent of the good or poorly positioned tuberosities migrated secondarily with the standard prosthesis and 10% with the fracture prosthesis. Statistically significant prognostic factors limiting tuberosity complications were: satisfactory initial osteosynthesis with correct prosthesis height and retroversion facilitated by use of the fracture system, rehabilitation in a specialised centre, relative immobilisation during the first postoperative month limiting exercises to balancing movements which divided the number of secondary migrations by two compared with immediate moblisation (14% versus 27%).

Discussion: A precise analysis of the radiograms revealed a very high rate of tuberosity complications (50%). There has been little study of these complications which are underestimated in the literature. The Aequalis fracture prosthesis can reduce these tuberosity complications two-fold. Postoperative immobilisation also reduces two-fold tuberosity migrations. These observations are against the early passive motion advocated by Neer. Finally, the quality of tuberosity fixation is crucial for success. The surgeon must concentrate on this element, searching to achieve a perfectly positioned prosthesis on the peroperative x-ray.

Purpose: Neer modified the Bankart operation, adding a reinforcement crossing the capsule anteriorly on the humeral side. The purpose of this study was to report results after more than two years.

Material and methods: Between 1991 and 1998, 77 patients underwent surgery for traumatic anteroinferior instability. Clinical and radiological outcome was reviewed in 64 of this patients by an observer different from the operator at a mean follow-up of 45 months (24–120 months). Patients with a unique anterior reinforcement were excluded from the analysis. The patients were generally young (mean 27 years) with sports activities (89%). Recurrent dislocation was observed in 39 patients, subdislocation in seven and painful and unstable shoulders in seven. Ten patients had an associated hyperlaxity, defined by elbow-to-body external rotation greater than 85%, according to the SOFCOT criteria. Three patients had had a prior procedure for a coracoid bone block.

Results: According to the Duplay score: outcome was excellent in 27 cases, good in 22, fair in nine and poor in six. Mean delay to return to former occupational activity was four months; it was seven months for sports activity. The deficit in external rotation was 3.4° on the average. Ten patients had persistent apprehension. Recurrence was observed in seven patients (11%) a mean 25 months after the operation (seven days to six years) as dislocation in two and subdislocation in five and due to trauma in five cases. Young age, hyperlaxity, high-risk sports, an important humeral notch, major capsular distension, and a high number of dislocations or subdislocations were the factors associated with recurrence. According to the Samilson criteria, pre-osteoarthritic lesions of the gleno-humeral joint were present in two cases preoperatively (one grade I and one grade II) and in eight cases postoperatively (four grade I, three grade II, and one grade III).

Discussion, conclusion: The Bankart operation as modified by Neer does not produce a stiff joint as is thought by many, probably due to the upper-lower capsular retention rather than lateral-medial retention. Nevertheless, the stability results are less satisfactory than generally reported for coracoid stop procedures.

Purpose: Cyclope syndrome is described in the literature as a postoperative complication of arthroscopic anterior cruciate ligament (ACL) grafts, leading to permanent flexion. The discovery of this syndrome in ten patients before reconstruction of the ACL led us to revisit the pathophysiology.

Material and methods: Among 250 candidates for ACL grafts, ten presented a positive but dull Lachman-Trillat sign with permanent flexion greater than 10°. The rotation click was negative in two and dull in eight. KT 200 measured differential laxity greater than 4 mm in all. Nine patients were reviewed at consultation, one patient had recently undergone another operation. Clinical and radiographic findings recorded in the patient’s files and operation reports were reviewed by two observers different from the operator.

Results: Arthroscopy revealed a partial tear of the ACL in three cases, a scarred ACL nourished by the PCL in five and a full thickness tear of the ACL in two. There was a fibrous barbell nodule inserted on the tibia in all cases, a characteristic feature of cyclope syndrome. The nodule was interposed between the femur and tibia at extension and was resected in all cases. Pathology reported a ligamentoid structure undergoing fibrous organisation. At last follow-up, greater than two years for nine patients, the IKDC rating was A for six patients and B for three patients. None of the patients had a defective extension differential.

Discussion: Persistent flexion preoperatively in a patient with a torn anterior cruciate ligament suggests possible presence of a ligamento-fibrous nodule interposing between the femur and tibia at extension. This nodule can go unnoticed at arthroscopy but appears to be unmasked in the Cabott position after partial resection of the subpatellar fat. It is particularly important to look for this nodule when the stump of the torn ACL is not found and the patient has experienced a recent sprain. Since we started looking for this nodule in all cases with resection, we have no longer encountered postoperative cyclope syndrome.

Conclusion: This group of arguments strongly suggests that the conditions necessary for the constitution of cyclope syndrome are probably present before reconstruction of the anterior cruciate ligament.

Purpose: The advent of lapaoscopic disectomy has made it possible to cure discal herniation with minimal trauma and no limitations on indications. We have adopted the technique described by J. Destandau since June 1998. The purpose of this work was to report our early results.

Material and methods: Forty patients were included in a period from June 1998 to August 2000. There were 24 men and 16 women, mean age 43 years (24–78). Eleven patients had an associated stenosis of the spinal canal. Accelerated rehabilitation was employed. Sitting and driving were allowed early.

Results: Mean follow-up was 13 months (2–27 months). Mean operative time was 63 minutes (30–150 min). Mean hospital stay was 3.92 days (2–10). There were 29 patients without stenosis of the lumbar canal. In this subgroup, outcome was excellent in 69%, good in 21% (six patients), fair in 3% (one patient), and poor in 7% (two patients). For the PROLO score, three patients were who were retreated were not included in the analysis. Outcome was excellent in 73% (19 patients), good in 12% (three patients), fair in 12% (three patients, and poor in 4% (one patient). In patients with lumbar canal stenosis, (eleven patients), three were reoperated for wider decompression; there was no haematoma. One patient was reoperated for deep infection. For the other patients the WADDELL score was excellent in five and good; in two the PROLO score was excellent in six and poor in one.

Discussion: The technique favoured a narrow approach. Shorter exposure preserved the anatomy, but for the three patients with an associated stenosis, reoperation was necessary for decompression. For the cases without complications, we noticed that recovery was very rapid, a finding which is exceptional with the conventional technique.

Conclusion: Our early experience with this technique has demonstrated that laparoscopic discectomy is feasible and safe. An associated stenosis is a limitation; we recommend systematic decompression in association with the discectomy.