The humeral shaft fracture accounts for about 1% to 3% of the total body fractures. Although most of these fractures can be treated by conservative treatment, the limbs need to be immobilized for a long time. About 40% of patients with humeral shaft fractures have been reported in the literature to remain different degrees of shoulder and elbow function loss after conservative treatment. Therefore, the surgical treatments of displaced humeral shaft fractures have become a consensus in the field. In the choice of surgical methods, most doctors still advocate open reduction and internal fixation, but the traditional surgery requires extensive dissection of soft tissue, and the big scar remains after surgery. In recent years, the interlocking intramedullary nails have been increasingly used in the treatment of humeral shaft fractures, but it is difficult to use for distal third of the humeral shaft because of flattened and narrowed distal cannel. In contrast, minimally invasive plate osteosynthesis (MIPO) is increasingly recognized and popularized for its biological advantages and satisfactory results.

1.General Information:From May 2014 to May 2017, our department applied MIPO technology combined with locking compression plate to treat humeral shaft fractures and obtained 26 cases of follow-up, all of which were closed humeral shaft fractures, including 18 males and 8 females; average age was 42.6 years old (19 to 72 years old) ; 19 cases were on the right side and 7 cases were on the left side. Causes of injury: 11 cases were traffic injuries, 8 cases were flat falls, 4 cases were falls from height, and 3 cases were sprain injuries. According to the AO classification, 6 cases were type A, 15 cases were type B, and 5 cases were type C. One patient had radial nerve paralysis, one patient had ipsilateral ulnar and radial shaft fractures, one had ipsilateral distal radial fracture, one had contralateral humeral intercondylar fracture, and one had ipsilateral intertrochanteric fracture. The time from injury to surgery was 3 to 12 days (average 7.2 days) . 2.Surgical Methods: (1) The anterior lateral approach:In this group, 20 cases used the anterior lateral approach. The operation was performed under general anesthesia. The patient was in supine position with upper arm abduction and forearm supination. The proximal incision was located between the medial edge of the deltoid muscle and the lateral edge of the biceps muscle and was 3 to 5 cm long. If the fracture was on the upper segment, the incision could be moved up to the deltoid pectoralis major space. The distal incision had the same length and was from the proximal end of the elbow transverse line, close to the lateral side of the biceps muscle. The biceps muscle was retracted to the medial side, the brachialis muscle was cut open lengthwise to reveal the lower part of the humerus cortex. The channel for the plate was prepared using a periosteal elevator. The reduction was achieved by traction in the posture of elbow flexion. The corrections of shortening, angulation and rotational deformity were confirmed under X-ray. For most A, B type fractures, percutaneous Kirschner wire fixation could be used to temporarily maintain the reduction. A narrow 4.5 mm locking compression plate (LCP) of sufficient length was placed through the proximal incision, and two locking screws were drilled into each of the proximal and distal bone blocks. For the older patients with significant osteoporosis, three locking screws could be used at each bone block. (2) The posterior lateral approach:In this group of patients, 4 cases were B-type fractures of the distal 1/3 of the humerus, and the wedge-shaped bone block was located on the anterior side, resulting in a lack of sufficient space on the anterior side of the distal bone to screw in 2 screws. Two cases had the skin abrasion of the anterior side of the arm. So the posterior lateral approach was selected in this group. The proximal incision started from 6 to 7 cm below the acromion. The deltoid muscle was cut open lengthwise to avoid damage to the axillary nerve. The distal incision was located at the lateral edge of the triceps muscle. The plate was placed on the posterolateral side of the humerus. When preparing the channel for the plate in this approach, it should be noted that the process should be gentle and close to the cortex to avoid damage to the radial nerve. (3) Postoperative treatment:Except for the radial nerve injury patients, the other patients did not need the external fixation. On the first day after surgery, the patient was instructed to perform shoulder and elbow function exercises without stress, and the suture was removed 2 weeks later. The follow up examination was performed every 2 weeks after surgery, and the functions of the shoulder and elbow were recorded. AP and LAT X-ray of the humurus were taken every 4 weeks to observe the fracture healing process.3.Observation indicators:The healing of the incision was observed and recorded. The shoulder and elbow function scores were recorded every 2 weeks postoperatively. The Ellman method of the UCLA shoulder scoring system was used to evaluate the recovery of shoulder function (A total of 35 points included 10 points of pain, 10 points of function. 5 points of active flexion motion, 5 points of flexion strength test, and 5 points of satisfaction; 34 to 35 was excellent, 28 to 33 was good, and 21 to 27 was fair, less than 21 was poor) . Elbow function was evaluated using Mayo elbow function score (A total score of 100 points included 45 points of pain, 20 points of motor function, 10 points of stability and 25 points of daily activities; 90 points or more was excellent, 75 to 89 points was good, 60 to 74 points was fair, less than 60 was poor) . For patients with radial nerve paralysis, the recovery of the nerve function was recorded. X-ray films were taken every 4 weeks to record fracture healing process and healing time.

The incisions of all patients were primary healed. The average follow-up time was 16.4 months (12-35 months) , and bony union was achieved in all fractures. The average healing time was 14.8 weeks (10-22 weeks) . There was no looseness or breakage of internal fixation. At the last follow-up, the UCLA scores of the shoulder were 24-35 points, of which 19 were excellent, 6 were good, and 1 was fair. The excellent and good rate was 96.1%. The Mayo elbow scored 91 to 100 points, and all 26 patients were excellent. The only one patient with fair shoulder function was a 72-year-old woman with a C3 type fracture of the middle and upper humerus caused by a car accident. The fracture line involved the head and neck of the humerus. The anterior lateral approach was performed with an extended PHILOS fixation. The patient also had an ipsilateral femoral intertrochanteric fracture. It was difficult for her to cooperate with functional exercise due to fear of pain. At 6 months postoperatively, the shoulder UCLA score was 24 points.Among the patients in this group, 1 patient had radial nerve paralysis after operation, and after oral administration of neurotrophic drugs, the functions of radial nerve was completely restored after 3 months; the other patient was complicated with radial nerve injury before operation. The radial nerve was not exposed in the surgery, and the neurological function recovered 2 months after surgery.

The minimally invasive percutaneous plate fixation treating complex humeral shaft fractures had the advantage of a smaller surgical trauma, a safer and easier operation, a higher fracture union rate, and a higher satisfaction rate.