Purpose of the study: Unlike thoracic and abdominal stab wounds, little has been reported about blade wounds to the forearm. We report a serie of machete wounds to the forearm treated in the Caribbean island Martinique, between 1997 and 2004.

Material and methods: This study included 14 open fractures of the forearm caused by machete wounds. This retrospective analysis was based on the patient files. We studied the mechanism of the fracture, the type and level of the fractures, the associated lesions, the type of treatment given, and complications observed.

Results: Mean follow-up was seven months. Among the 14 patients studied, 14 presented an ulnar fracture, and five a radial fracture. Five patients suffered a complete amputation of the hand. Ten patients (71%) also had associated tendon injuries, all on the ulnar side. Three associated vascular injuries were noted (21%), two on the ulnar side. There were four nerve lesions (29%) involving the ulnar nerve alone (n=2),the ulnar and medial nerves (n=1) or all of the nerve trunks (n=1). The fractures involved the distal third of the forearm in nine patients (64%) and were comminutive for ten (71%). A complete fracture was noted in twelve patients (86%) with a partial fracture in two. Osteosynthesis was performed in all cases. There were nine complications: early infection (n=2, due to late referral), stiff joints (n=6, 43% including tendon retraction in five), nonunion (n=2, one repeated case) and one late healing at one year. Motor and sensorial sequelae were observed at last follow-up in all patients with an initial nervous lesion.

Discussion: The mean follow-up in our patients was short because of the specific context (homelessness, drug addiction). Most of our patients refused medical follow-up. The strong predominance of bony or soft tissue injuries observed on the ulnar side of the forearm corresponds to the mechanism of defense used by the victims. Despite the fact that the wounds were soiled and that the patients failed to comply with medical advice, the rate of early and secondary infection was low. Stiff joints due to tendon retractions and motor deficits were however frequent and compromised the functional outcome.

Conclusion: Fractures of the forearm by machete wounds generally occur in a typical situation of self defense. The characteristic injury to the ulnar side of the forearm results from this mechanism. Tendon and nervous complications are common and cause invalidating sequelae. Secondary infection is exceptional. Prolonged regular follow-up could probably improve the functional outcome of these particular injuries.

Purpose of the study: Three-bone arthrodesis, described in 1997, is designed for radiocarpal osteoarthritis with mediocarpal extension. The procedure consists in a capitolunohamate fusion after resection of the scaphoid and the triquetrum. It is associated with carpal shortening proportional to the degree of preoperative wrist stiffness. The objective is to achieve less stiffness than with four-bone fusion.

Material and methods: To verify our hypothesis, the first 24 patients (25 wrists) were reviewed with mean 5.2 years (2–8.5 years) follow-up. All wrists were painful and stiff, and presented radiocarpal and mediocarpal osteoarthritis. There were twelve SLAC III, nine SNAC III and four SCAC III. Mean age was 59 years (range 37–79 years). Mean preoperative range of motion was 50.5° flexion-extension (range 10–105°), mean force was 17 kg (range 10–35 kg). Radiological assessment was performed preoperatively and at last follow-up to determine the Youm index (carpal height) and the Bouman index (carpal translation) and to study the radiolunate joint space.

Results: At last follow-up, all patients had improved but one. Ten wrists were pain free twelve caused some pain at forced wrist movements, and two caused pain daily but at a level below the preoperative level. One patient still suffered from severe pain and required revision for total radiocarpal arthrodesis. The final mean flexion-extension range of motion was 67.8°, for a 13.3° gain in extension and a 3.8° gain in flexion. Ulnar inclination was improved 14° on average. Mean force was 24 kg (73% of healthy side), for a 40% improvement over the pre-operative force. RAdiographically, there was one case of capitolunate nonunion. The radiolunate space remained unchanged. Carpal height decreased 15% on average and the Bouman index increased from 0.90 to 0.93 with no significant ulnar misalignment on the carpus.

Discussion: For pain and force, these results are similar to those achieved with four-bone fusion. The overall results for range of motion are however better for flexion-extension and unlar inclination. In our practice, we have decided to replace the four-bone technique by three-bone fusion because the outcome is a less stiff wrist with a simpler surgical technique. Better results are obtained for stiffer wrists which achieve a significant improvement in motion due to carpal shortening.

Purpose of the study: Analyze failures (recurrent anterior instability) and other complications (pain, stiffness) compromising the overall outcome after arthroscopic anterior shoulder stabilization.

Material and methods: We studied failures and complications in a consecutive inaugural series of 43 patients undergoing an arthroscopic Bankart procedure for chronic anterior shoulder instability. The procedure was performed with knitted resorbable threads on metallic anchors. Outcome was reviewed at mean 26 months (range 6–63 months). There were 19 recurrent dislocations, 12 recurrent subluxations, 4 cases of recurrent subluxation and dislocation and 6 cases of painful unstable shoulder. Mean patient age was 35.6 years (range 19–59 years). Thirty-two patients practiced sports, including 21 who practiced high-risk sports.

Results: One patient, a competition basketball player, presented recurrent traumatic dislocation due to a violent shock after premature resumption of sports activities five months after surgery. There were no recurrent dislocations among the subluxation cases. Pain persisted in three of the six painful unstable shoulders. Sixteen patients presented persistent apprehension but none complained of instability. Nine patients had a positive relocation test. Limited external rotation of less than 30% as observed in five patients and of 30–50% in two. Residual pain was observed in 14 patients (33%) (when carrying a heavy load with the arm hanging along the body, with fatigue, and for forced movements without warm-up in the morning). For four patients, pain occurred in the armed position. Seventeen patients (43%) interrupted their sports activity. The Duplay score showed 13 (30%) fair and poor objective overall results. Subjectively, only seven patients (15%) were only partially satisfied or dissatisfied.

Discussion: Analysis of failures and complications disclosed a discordance between the low rate of failure using the classical definition (recurrent dislocation or subluxation) and the high rate of fair or poor overall outcomes. Residual pain and non-resumption of sports activities appeared to be the major problems. These two factors were analyzed in detail to compare this series with data in the literature. It was found that non-resumption of sports activities is not always related to shoulder instability or apprehension and that pain is often related to associated injury (SLAP, cuff). Conversely, pain associated with a positive relocation test should be considered as a true recurrence, especially in a subject who was unable to resume sports activities.

Purpose: The bifoliated vascularised fibula graft is an attractive alternative for reconstruction of large bone segments. The purpose of this work was to evaluate mid-term results and the usefulness of two surgical techniques: skin island flap monitoring and the arterio-venous loop.

Material and methods: This retrospective analysis included fourteen patients (eleven men and three women) treated between 1992 and 2002. Mean age was 30 years (10–54). Indications were complications of open fractures in nine patients, major bone loss in two, septic nonunion in four, and aseptic nonunion in three. Reconstruction was performed after bone tumour resection in five patients involving immediate reconstruction after failure of an infected massive allograft in four of them. Localisations were: tibia (n=6), femur (n=5), humerus (n=2), and pelvis (n=1). Average bone loss was 10 cm (7–15 cm). Minimal pinning, cerclage or screwing was used to stabilise the flap completed by internal fixation in four patients and external fixation in ten. A monitoring skin island was used for twelve patients (the island was technically impossible in two patients). Vascular anastomoses were performed in seven patients using an arteriovenous loop, performed as a preliminary measure in six.

Results: Mean follow-up was 35 months. One patient died early from tumour progression. Among the seven patients who had an arteriovenous loop, one required revision for a vascular complication. For the seven “classical” bypasses, there were three intraoperative or early complications requiring revision of the anastomoses. Nonunion developed despite early revision in the four patients whose monitoring skin island suffered. Consolidation was achieved without revision in all patients who skin island did not suffer; time to bone healing was eleven months for seven of them.

Conclusion: Bone healing was related to the quality of graft vacularisation. Clinical observation of the monitoring island was the best way to identify vascular complications early and initiate treatment. Use of a preliminary arteriovenous loop decreased the risk of vascular insufficiency inherent with long bypasses and shortened operative time.