Purpose of the study: The purpose of this study was to assess at more than twenty years follow-up, the results achieved with the Bankart operation, focusing on functional and radiographic outcome of the glenohumeral joint in patients operated on when they were young.

Materials and Methods: This retrospective analysis reviewed clinically and radiographically patients who underwent shoulder surgery for instability between 1971 and 1986. The Bankart operation was performed in all patients followed systematically by immobilisation. Self-controlled rehabilitation was the rule. The clinical assessment used thed Duplay and Rowe scores. The Prieto and Samilson radiographic score (four stages) was noted.

Results: Mean follow-up was 26 years for 49 patients (50 shoulders). There were three women and 46 men (sex ratio = 15.3), mean age 25 years at surgery. Eighty percent practiced competition sports (a contact sport for 72%). The time from the first dislocation to surgery was four years on average. The rate of recurrence was 16%. All recurrences were provoked by a violent accident. 94% of patients returned to sports activities (on average 4.6 months after surgery), 80% at the same level. 86% of patients were satisfied. Mean Duplay and Rowe scores were 81.3 and 82.2 respectively. Average deficiency of external rotation was 9° compared with the contralateral side. Normal radiographs were noted for 13 shoulders (26%). The Prieto and Samilson classification was: stage I (n=18), stage II (n=5), stage III (n=5) and stage IV (n=1).

Discussion: In this cohort with 80% competition sports athletes and 94% return to sports activities after surgery, the Bankart operation demonstrated its efficacy for contact sports. The rate of osteoarthritis after this operation is comparable with that observed with other types of bone blocks, but the follow-up here was twice as long. The deficit in external rotation was not greater than with non-anatomic operations.

Purpose of the study: Basal joint osteoarthritis remains one of the most invalidating degenerative diseases of the upper limb. Most patients are women and the main symptom is pain. Several surgical approaches can be used: total trapeziectomy with or without a ligament reconstruction for intermetacarpal stabilization associated or not with tendon interposition (possibly with a synthetic insert); cemented or non-cemented trapezio-metacarpal prosthesis and arthrodesis. The purpose of this work was to report our experience with the modified Eaton-Littler technique and determine the current contribution of total trapeziectomy – ligament reconstruction – tendon interposition.

Material and methods: We report a series of 26 patients who underwent surgery between 1994 and 2002 for trapezium resection associated with intercarpal ligamentoplasty and «anchovy» interposition using a flexor carpi radialis hemi-tendon. The series involved 21 patients, 16 women and 5 men, with invalidating basal joint osteoarthritis unresponsive to medical treatment. The Dell classification was stage II and III. Mean age was 58.6 years. Five patients underwent surgery on both wrists. One patient had had a prior operation on one side and a Swanson implant on the other. The mean follow-up was 66.1 months (range 25–131 months).

Results: For these 22 cases, outcome at last follow-up was satisfactory with complete resolution of pain, excellent joint mobility with a mean Kapandji score of 9.5/10 (range 8–10). Half of the wrists exhibited deficient pinch and grasp force. Mean postoperative force was 20 kg (range 8–28 kg). Radiographically, scaphometacarpal collapse was nearly complete in all cases. For four cases, outcome was considered poor because of nearly constant pain with reduced mobility and a mean Kapandji score of 5/10. Three patients developed a reflux dystrophy: one of these patients achieved a final satisfactory result. There had been no revisions at last follow-up.

Discussion: All the proposed techniques appear to give good results in terms of pain relief. Trapiezectomy with ligamentoplasty appears to provide good results at mean six months follow-up, results which are reliable and sustained over time. Trapezio-metacarpal prostheses give good immediate results in selected patients but the rate of failure at mid-term remains high, mainly due to loosening. Second generation cemented prostheses are promising. Trapezial implants have specific complications. Trapeiometacarpal arthrodesis gives good stability of the thumb column and is particularly indicated for young active patients with less consideration on the esthetic aspect of their hand, keeping in mind the significant rate of nonunions.

Conclusion: Despite the significant progress achieved with trapeziometacarpal prosthetic arthroplaasty, we still believe that complete resection of the trapezium with ligament reconstruction associated with tendon interposition remains the gold standard in terms of patient satisfaction and reliable results for the surgical treatment of basal joint osteoarthritis.

Purpose of the study: Benign tumors of peripheral nerves are exceptional. Schwannomas predominate. Most tumors are revealed by tumefaction or pain over a nerve trajectory. The risk of degeneration is very low. Magnetic resonance imaging is the exploration of choice. The risk of sequelae or recurrence must nevertheless be determined with precision. We reviewed our experience with 93 benign tumors of peripheral nerves to search for factors predictive of prognosis.

Material and methods: This retrospective analysis included patients seen between 1979 and 2004. We collected a series of 89 patients, 41 women and 48 men, mean age 48 years, age range 18–80, with 93 benign tumors. Mean time from symptom onset (pain) to diagnosis was 20 months. The patients consulted for pain (n=78), presence of a mass (n=79) or both (n=66). Percussion produced paresthesia in 54 patients. Pre-operative magnetic resonance imaging was available for 45 patients. The same surgeon performed nerve microsurgery in all patients. A prior procedure had been performed in another institution for 23 patients. The tumors were: schwannoma (n=74), neurofibroma (n=14), plexiform neurofibroma (n=3), angiolipoma (n=1) and intranervous lipoma (n=1). Mean tumor size was 31 mm (range 7–120 mm). Tumors were located in the brachial plexus (n=13), the upper limb (n=29), the trunk (n=1) and the lower limb (n=50). Complete resection was achieved in 83 cases, with removal of a non-stimulatable fascicle in 50 cases and a motor fascicle in

4. Nerve repair was required for 11 cases: 5 by direct suture and 6 with grafts. Resection was impossible for 4 tumors treated by neurolysis, decompressive epineu-rotomy, biopsy and interfascicular dissection.

Results: Mean follow-up was 96 months (range 3–300). Outcome was very good for 42, good for 25, fair for 8 and poor for 5 (all seen secondarily). Nine patients were lost to follow-up. There were no cases of recurrence.

Discussion: Microsurgical procedures are necessary for resection of nerve tumors in order to preserve the fascicles and thus function. Unresectable tumors and secondary grafts yield les satisfactory results, in our series and in the literature. Similarly, the duration of the symptoms and the size of the tumor increase the risk of operative difficulty and sequelae. Despite high-performance imaging techniques, surgery is the only sure way to establish certain diagnosis.

Purpose of the study: The osteoconductive properties of hydroxyapatite surfacing improves the biointegration of orthopedic implants. Current high- and low-temperature resurfacing techniques have several drawbacks, particularly concerning the control of phases. The «low-temperature nanocrystalline apatite resurfacing technique using amorphous phosphate» was developed to avoid this type of inconvenience. The purpose of this study was to examine the biocompatibility of resurfacings produced with this patented technique and to compare biological efficacy with that of the reference technique of plasma torch resurfacing.

Material and methods: The cytocompatibility tests included cell proliferation and attachment tests using human osteoprogenesis cells, and phenotypic characterization of phosphatase alkaline (PAL) and pro-collagen (type I) activity. Biocompatibility studies were performed. Cylinders of natural titanium or titanium resurfaced with the plasma method and the low-temperature method (single layer, bilayer) were implanted in 16 rabbits in condylar and tibial sites. Histological examinations without decalcification were performed one and three months after implantation (n=8 for each time and condition). The implant-quantity of bone in contact ratio was determined by histomorphometry. Scan electron microscopy was used to ascertain the persistence of the resurfacing.

Results: The cell attachment rate of 30–40% confirmed earlier results. The cells grew, and preserved and maintained their differentiation properties (PAL activity at 7, 14 and 21 days). The histological results revealed that all types of resurfacing were well tolerated. HIstomorphometry confirmed the influence of the implantation site on the tissue reaction. One month after implantation, the low-temperature amorphous resurfacing appeared to produce a better result with an optimal ratio for the bilayer in the tibial site and an optimal ratio for the monolayer in the condylar site. The trend was the same three months after implantation, but was less pronounced compared with the plasma torch resurfacing. Paradoxically, the absence of treatment produced a very satisfactory ratio at the condylar level. Scan electron microscopy demonstrated rapid resorption of amorphous resurfacing unlike plasma torch resurfacing with was detectable three months after implantation.

Discussion and conclusion: The different performance levels of bilayer and single-layer resurfacings depending on the implantation site might be explained by the cortical or cancellous nature of the neighboring bone. Low-temperature resurfacing would be more appropriate for implants inserted into cortical bone. In vivo, this resurfacing is resorbed but appears to enable, like the plasma process, the formation of peri-implant bone formation. It offers the advantage of enabling incorporation of compounds of interest (antibiotics, growth factors).

Purpose: We report a retrospective consecutive series of 57 hemiplegic patients (32 men and 25 women) who underwent surgery between 1995 and 2000 for spastic talipes equinovarus associating fascicular neurotomy of the tibial nerve and tendon release in order to recover sole to floor walking capacity.

Material and methods: Mean patient age was 47 years (16–75). The hemiplegia resulted from stroke (n=41), trauma (n=8), and other causes (n=8). All patients had spastic talipes equinovarus and 46 required a walking aid. Triceps force and spasticity were scored 2.1 (MRC) and 3.66 (Ashworth) respectively. Pedial hypoaesthesia was present in 23 patients. The mean functional ambulation classification (FAC) score was 3.3, with severe disability (FAC 1 or 2) in 13 patients. Surgery was performed three years (average) after the causal event. After identification by electrical stimulation, we performed microsurgical section of 4/5 nerve fibres of the terminal branches of the tibial nerve in 55 patients. After physical exploration of musculotendon retraction (triceps and toes flexors) and dorsiflesion palsy, we released tendons as needed. The Achilles tendon was lengthened percutaneously in 13 patients, Bardot tenodesis or transfer of the anterior hemi-tibial anterior tendon was performed in 29, and tenotomy of the toe flexors in 12.

Results: Mean follow-up was three years (1–6). Triceps force and foot sensitivity were not modified by the surgery. Spasticity was scored 1.08 and 1.19 (Ashworth) postoperatively and at last follow-up respectively. The FAC walking score was 4.13 and 4.15 postoperatively and at last follow-up respectively. The walking aid was no longer necessary or was improved in 52 patients. Tibiotalal arthrodesis was necessary for recurrent spasticity in three patients.

Discussion: The preoperative assessment of foot deformity and gait is an essential element. Correct preoperative assessment enables microsurgery for fascicular neurotomy of the posterior tibial nerve and tendon lengthening or tenodesis during the same operation. These procedures are indicated for severe deformity and should be used as the first intention treatment before arthrodesis which involves greater morbidity.

Purpose: Percutaneous suture of the Achilles tendon is an excellent alternative to open surgery with the known risk of cutaneous and septic complications and also to orthopaedic surgery with the risk of recurrence. We propose our technique of percutaneous suture applied in a series of 76 patients.

Material and methods: Achilles tendon sutures were performed from 1998 to 2002 in 76 patients (17 women), mean age 41 years (22–66). The procedure was conducted under local anaesthesia in the outpatient setting. A resorbable knitted thread mounted on a needle was used to make the two-point frame percutaneous suture. The patients were immobilised for three weeks (average). The Thermann scores, modified by McComis, were determined and the objective Cybex test at 30 and 60/s was used to measure peak torque, power, and maximal power followed by 30 cycles at 120/s to evaluate muscle fatigability.

Results: Mean follow-up was 35 months (10–66). Outcome scores (maximum 100 points) showed excellent or good results in 73 patients and poor results in three. Mean operative time was 15 minutes with no immediate complications. Weight bearing was allowed as early as day 1 (range d1-d20) and patients resumed their occupational activities on day 40 (d8-d100). Sports activities were resumed at six months (4–8 months). Ankle motion was symmetrical and pain free in 73 patients. Calf circumference was 2 cm less than on the contralateral side (0.5-3.0). There were two cases of recurrent tears in patients with poor outcome. Decreased muscle force involving the triceps was never greater than 35% compared with the healthy side (Cybex).

Discussion: The objective Cybex measurements were tightly correlated with the McComis scores. This percutaneous technique does not require removal of the material and is perfectly reproducible in outpatients under local anaesthesia. It is an inexpensive method but remains contraindicated for tears seen late, recurrent tears, or very distal tears.

Conclusion: The very good results obtained in this series suggest we should continue with this low-cost percutaneous method.

Purpose: Progress in medical and surgical management has reduced the incidence of osteitis. Nevertheless, this type of complication remains a difficult therapeutic challenge for frail polyoperted patients exposed to infection for several months. Based on a technique developed by A.C. Masquelet and cooperative work with the infectiology unit, we propose an alternative to eminent amputation.

Material and methods: The series included eighteen surgery patients (fifteen men and three women, mean age 37 years) Mean follow-up was fifteen months. Fourteen patients had undergone one or several operations. Bone loss varied from 5 to 17 cm with shaft loss in ten patients and metaphyseal loss in eight. Fifteen reconstructions involved the lower limb: three femurs, nine tibias and three tarsal bones. Three cases involved the upper limbs: elbow, radius, and radiocarpal bones, one each. Meti-R bacilli were identified in eleven patients. Reconstruction was based on the Masquelet spacer technique to induce membranes in all cases. External fixation was used in sixteen cases and pinning in two. A second operation was performed after normalisation of biological parameters and wound healing.

Results: There were no early complications. Late wound healing required a secondary flap in three patients. There was one graft failure after early reinfection. The spacer was in place for a mean four months. Healing was achieved at six months on the average. A second graft was required in two patients.

Discussion: This series demonstrates the usefulness of this bone reconstruction method irrespective of the soft tissue trophicity and the degree of bone loss. It allows maximal debridement of infected tissue, the only method allowing effective eradication of infection. The close cooperation with the infectiology unit was particularly helpful for the management of the more difficult cases allowing bone grafting beyond usual limits.

Purpose: Meralgia paraesthetica is usually caused by entrapment of the lateral femoral cutaneous nerve (LFCN) at the inguinal ligament. We present our experience with 114 patients who underwent surgical management for meralgia paraesthetica.

Material: We reviewed 114 patients (48 men, 66 women, five bilateral cases) who underwent surgery for meralgia paraesthetica between 1987 and 1999; local anaesthesia was used for neurolysis in most cases. We identified five aetiologies: idiopathic (n=69, three bilateral), abdominal surgery (n=19), iliac graft harvesting (n=12, one bilateral), hip surgery (n=7), trauma (n=7, one bilateral).

Methods: We analysed outcome at more than two years follow-up for the entire series and by aetiology using a standard 12-point evaluation scale accounting for residual pain, sensorial disorders, and patient satisfaction.

Results: The overall results were good, mean score 9/12 (range 1–12). Ninety-two patients were very satisfied or satisfied. Among the 27 patients who were not satisfied, five developed recurrence. Mean time to full pain relief was 70 days (range 1 – 364 days). Recovery of thigh sensitivity was noted at 128 days (range 1 – 364).

Discussion: The essential criteria of poor prognosis were duration of the meralgia before surgery and its aetiology. Neurolysis of an LFCN injured by trauma or iliac graft harvesting provided less satisfactory results (scores 7 and 6 respectively) than for idopathic meralgia paraesthetica or abdominal-surgery injury (scores 9 and 10 respectively). Eight of the neurolysis procedures in this series did not provide satisfactory results (score 5).

Conclusion: Neurolysis appears to be the surgical treatment of choice for mearlgia paraesthetica. In skilled hands, neurolysis can be performed under local anaesthesia, although certain difficulties can be encountered: obesity, modified anatomy due to prior operations, nerve variability (frequent). Knowledge of these different elements is essential not only to achieve neurolysis but also prevent iatrogenic injury.

Purpose: Mucoid pseudocysts are infrequent benign tumours affecting all peripheral nerves near joints. Reported for the first time in 1891, these cysts raise an unresolved question as to their nature. Identifying the origin of these cysts could have an impact on management and the risk of recurrence.

Material and methods: Twenty-three patients (21 men two women), mean age 38 years, were followed for a mean six years. The mucoid cyst was located at the level of the common fibular at the neck of the fibula in 16 cases, the tibial nerve at the knee in one, the medial nerve in one, the ulnar nerve in three and the supra-scapular nerve in two. Pain was local in 18 patients, irradiated to the concerned territory in 20. In 17 patients, the nerve lesion was discovered due to development of motor deficit. An EMG was performed in all cases, an ultrasound exploration in 15, a CT scan in seven and an MRI in ten. All patients underwent surgery with neurolysis under microscopy for intraneural cysts. A pedicle communicating with the joint was search for during each intervention.

Results: An articular communication was found in 17 cases. Mean delay to recovery of motor force was five months with recovery of normal sensitivity in 16 cases at seven months. Total lack of recovery was noted in one patient. There was one recurrence requiring tibiofibular arthrodesis.

Discussion: Three theories have been put forward. Besides the cystic degeneration of certain schwannomas, degeneration of the nerve sheath connective tissue, the joint theory appears to be the most probable. The presence of a pedicle linking the cyst to the joint in more than 60% of the cases, the periarticular situation of the nerves involved and sometimes the migration along an articular nerve as well as the mucoid content of the cysts is in favour of an articular origin. The notion of recurrrence after complete minute resection is also in favour of a joint disease.

Conclusion: A mucoid cyst is a likely diagnosis for neruological lesions lying near a joint. Search for an articular communication both before and during the surgical procedure is important to limit the risk of recurrence.

Purpose: Radial paralysis is a major complication of humeral shaft fractures. In most cases, the paralysis is regressive but in certain patients surgical repair is required to achieve full neurological recovery. We reviewed retrospectively our patients to determine the causes of non-recovery and evaluate the efficacy of different treatments.

Material and methods: Thirty patients were operated between 1990 and 1997 for radial nerve paralysis that was observed immediately after trauma or developed secondarily. Mean follow-up after surgery was 6.3 years. There were 22 men and 8 women, 16 right side and 14 left side. Mean delay from injury to surgery was four months (0–730 days). Elements that could be involved in radial paralysis were noted: type of fracture, level of the fracture, treatment, approach, material used. There were ten cases with non-union. Neurological recovery at three years was assessed with muscle tests and with the Alnot criteria. An electrical recording was also made in certain patients. Surgery involved neurolysis in 23 cases, nerve grafts in five and tendon transfers in two.

Results: Outcome was very good and good in 22 patients, good in one and could not be evaluated in one (tendon transfer). There were three failures (two neurolysis and one graft) and two patients were lost to follow-up. After neurolysis, mean delay to recovery was seven months; it was 15 months after nerve grafts. Recovery always occurred proximally to distally.

Discussion: Radial paralysis after femoral shaft fracture regresses spontaneously in 76% to 89% of the cases, depending on the series. There is a predominance in the 20 to 30 year age range. Several factors could be involved in radial paralysis (fracture of the distal third of the humerus, spiral fracture, plate fixation, nonunion). The anterolateral approach allows a better exposure of the nerve. Unlike other authors, we do no advocate exploration of the injured nerve during surgical treatment of the fracture because it is most difficult to determine the potential for recovery of a continuous nerve.

Conclusion: The risk of radial nerve paralysis is greatest for spiral fracture of the distal third of the humerus. In such cases, it may be useful to explore the nerve during the primary procedure and insert a plate. For other cases, we prefer to wait for spontaneous nerve recovery. If reinnervation is not observed at 100 days, we undertake exploration.

Purpose: Using the transfemoral approach for a locked resurfaced implant without cement appears to be an interesting solution for reconstruction after loss of bone stock.

Material and methods: We report a multicentric consecutive series of a homogeneous group of 94 revision femoral implants in 26 women and 68 men, mean age 67 years (38–88). Mean follow-up was 19 months (12–38 months). All the patients were operated via the transfemoral approach which allowed removal of the implant, insertion of a new locked stem without cement and stabilisation of the femoral fragment in close contact with the implant. The cause of the revision was loosening of the femoral component in 83% of the cases in association with important loss of femoral bone. In addition, this technique was applied for stem fracture (6.5%), prosthesis fracture (5.5%), or another cause (3.5%). The loosened stem had been cemented in 87% of the cases; a first stem in 68% and a second one in 27%.

Clinically, the patients were assessed with the Postel Merle d’Aubigné (PMA) and Harris scores using the SOFCOT and Picault-Vives scores. We also assessed healing of the femoral fragment, the stability of the implant, and the bone response around the implant. Preoperative PMA and Harris scores were 8.5 (1–17) and 38 (5–86) respectively. In 91% of the cases, bone construction was important radiographically. The locked stem without cement was short in 32% of the cases, and long in 68%.

Results: At last follow-up the PMA and Harris scores were 15 (9–18) and 78 (37–99) respectively. The femoral fragment healed in 93% of the cases at one year. In 72%, bone reconstruction was significant. Two patients died. Four others required revision of the femoral implant.

Discussion: The advantages of this technique compared with other solutions is discussed. The main points of the operation are recalled. The causes of failure are analysed.

Conclusion: The transfemoral approach allows reconstruction of the bone defect. The primary stability allowed by locking gives time for osteointegration of the implant. These preliminary results confirm that secondary stability of the implant is achieved. In case osteointegration does not occur, a less aggressive surgical solution could be discussed.