Domestic and foreign data have confirmed that clavicular fractures account for approximately 2%-3% of all adult fractures, and midshaft fractures are much more common with obvious displacement or comminution. The accompanied butterfly fragments increase the difficulty of maintaining the stability after fracture reduction. In recent years, literatures have reported that surgical treatment superior to conservative treatment. In 1997, foreign scholar Hill et al. reported that the nonunion rate of non-surgical treatment was as high as 15%. It was also reported that 30% of patients had poor shoulder joint function. For the clavicular fractures with obviously displaced or comminuted midshaft, surgical treatment has become the mainstream, which can significantly reduce the incidence of complications such as nonunion and malunion and has the merits of good and fast recovery of shoulder joint function.

1．General information: From January 2014 to February 2017, 28 adult patients (19 males and 9 females) with comminuted midshaft clavicular fractures were treated with Nice knot combined with anatomic locking plate in our department. The age ranged from 22 to 73 years old with an average of (46.93±14.16) years old. Causes of injury: 8 cases of automobile accident, 10 cases of electric bicycle accident, 5 cases of bicycle fall and 5 cases of sports injury. According to Allman classification, all patients were type I fractures and were operated with open reduction and internal fixation within 3 to 5 days after injury. Inclusive criteria: (1) Patients over 18 years old with comminuted midshaft clavicular fractures of obvious displacement and shortening displacement of over 2 cm; (2) Fractures with obvious angular deformity and the risk of sink puncture; (3) Informed consent of the patient and family; (4) Patients without severe internal medicine disease and who can tolerate brachial plexus block or general anesthesia. Exclusive criteria: (1) Fractures with displacement of less than 2 cm; (2) Comminuted fractures without the risk of skin puncture; (3) Combination of multiple fractures; (4) Pathological clavicular fractures. 2. Surgical treatment: After brachial plexus block anesthesia, the patient was placed in beach chair position with conventional disinfection and draping. A straight or curved incision of approximately 6 to 10 cm with the fracture ends as center was made along clavicle. The tissue was cut open layer by layer, and the supraclavicular nerve was protected to the greatest extent. The fracture ends were stripped limitedly, and the blood clots were removed to secure fracture reduction. Attention should be paid to protect blood supply. During the operation, different reduction methods were adopted based on the size of butterfly fragment. For the fracture fragment of over 1 cm, cloth clamp was used for reduction. After anatomic reduction, No. 0 absorbable suture was used for fixation with 2 Nice knots. The reduction was beneficial for maintaining the stability of fracture and the length of clavicle. As the distal and proximal clavicular fractures were reduced, the anatomic plate of appropriate length was placed and fixed with drilling on both the distal and proximal ends. For the fracture fragment of 0.5 to 0.8 mm, the fracture was reduced at the distal and proximal sites, and an anatomic plate was placed to maintain the length of clavicle for bridge fixation. Then, the small butterfly fragment was reduced and fixed with 2 Nice knots of No. 0 suture. During the operation, attention should be paid to protect the blood supply of fracture end and butterfly fragment. At least 3 sextuple-layer cortex screws were used on each side of the plate for fixation. At least 2 Nice knots should be used for binding based on the size of butterfly fragment. The length of plate was selected in accordance with the comminution of fracture, and the plate with 8 to10 holes was most commonly used. The reduction, plate position and screw length were monitored under fluoroscopy with C-arm. After satisfactory reduction and fixation, the cavity was irrigated, and the wound was closed layer by layer.3.Postoeprative management: To prevent postoperative incision infection, 1.0 g of cefotiam dissolved in 100 ml of regular saline was conventionally used via intravenous infusion twice per day for 24 to 48 hours. For those who were allergic to cephalosporin and penicillin, 0.6 g of clindamycin dissolved in 100 ml of regular saline was used via intravenous infusion twice per day. The affected limb was immobilized with forearm sling for 4 to 6 weeks. The patients were instructed to conduct pendulum exercise 1 week after the operation. The active functional training of shoulder joint was started 3 weeks after the operation, and weight bearing should be avoided within 6 weeks. 4. Efficacy evaluation criteria: (1) Records of operation time and intraoperative blood loss for all patients; (2) Patients’ satisfaction with treatment; (3) Complications, including wound infection, neurovascular injury, internal fixation loosening, breaking, skin irritation, etc; (4) Shoulder function evaluated according to Constant-Murley score and therapeutic evaluation judged according to Lazzcano criteria. Patients' satisfaction, complication and shoulder function were evaluated at the last follow up.

28 patients were followed up for 6 to 16 months with an average of (10.27±3.22) months. The operation time was 55 to 90 minutes with an average of (63.33±21.27) minutes. The intraoperative blood loss was 40 to 100 ml with an average of (62.67±19.07) ml. All the surgical incisions were healed during the first stage without wound infection or neurovascular injury. Two slim patients were suffered from the skin irritation respectively caused by the cocked proximal and distal ends of plate. The symptom disappeared after the removal of internal fixator. During the last follow-up, the Constant-Murley score ranged from 80 to 100 points with an average of (90.00±5.98) points. The therapeutic evaluation was conducted based on Lazzcano criteria. Among the patients, there were 20 cases of excellence, 6 cases of good and 2 cases of moderate, and the good and excellent rate was 92.86%. Complications: 1 case of proximal skin irritation and 1 case of distal skin irritation.

As a new choice for the treatment of comminuted midshaft-clavicular fracture, Nice knot combined with anatomic locking plate can achieve good anatomic reduction, stable internal fixation, high healing rate and less complication. Since the binding of Nice knot plays an important role in maintaining the length of clavicle, satisfactory reduction can improve the trust of patient and family for doctor and reduce the contradiction between doctor and patient. Small suture knot can solve major clinical problem. With its good application value, Nice knot combined with anatomic locking plate is worth of applying clinically in the fixation of clavicle fracture with butterfly fragment.