Humeral shaft fracture is common in clinic, and it accounts for 1%-1.5% of all fractures in whole body. The fracture is commonly located in the middle and distal parts of humerus, which is usually unstable. The conventional treatment method was internal fixation, and the strategies for treating the fracture of middle-lower humerus include anterolateral approach, posterior approach and anterior approach. The anterolateral approach was the most commonly used previously, which was also adopted by the author in early years. With the development of internal fixation device, the posterior approach was used by more and more surgeons for the treatment of middle-lower humeral fracture due to clear operation field, adequate radial nerve exposure, flat surface of posterior humeral shaft, and easy plate placement without shaping. For posterior approach, the posterior humeral shaft was exposed via triceps-splitting approach or posterior incision and lateral intermuscular septum approach. Using posterior incision and lateral intermuscular septum approach, the author found that this strategy had the advantages of clear anatomic structure, adequate exposure and easy internal fixation. Therefore, this study retrospectively analyzed the clinical effects of posterior incision and lateral intermuscular septum approach and anterolateral approach for the treatment of middle-lower humeral fracture.

(1) General information. From May 2013 to May 2015, a total of 44 patients (25 males and 19 females) with middle-lower humeral fracture were treated in the Number One Orthopaedics Hospital of Chengdu. The age ranged from 24 to 58 years with an average of 36.81 years. All cases were fresh closed fracture caused by trauma, including 12 cases of fall damage, 4 cases of wrist injury, 15 cases of high fall and 13 cases of traffic accident. Among them, 19 cases had the left side affected, and 25 cases had the right side affected. Furthermore, there were 4 cases of transverse fracture, 3 cases of multiple fracture, 24 cases of oblique fracture and 13 cases of spiral fracture (including 24 cases of comminuted fracture that accounted for 54.55% of all fractures) . According to AO classification, there were 20 cases of type A fracture, 20 cases of type B fracture and 4 cases of type C fracture. The posterior humeral approach and lateral intermuscular septum approach were used in 25 patients, and the anterior lateral approach was used in 19 patients. There was no statistically significant difference between two groups in age, gender, body mass index and ASA classification, and a P value >0.05 was regarded as comparable. (2) Inclusive and exclusive criteria. Inclusive criteria:① middle-lower humeral fracture in accordance with diagnostic criteria;② fresh closed fracture without combined injury, no age and gender limitations;③ patient with open reduction and plate and screw fixation;④ Grade I-III of ASA classification without renal dysfunction, blood coagulation dysfunction or severe basic diseases.Exclusive criteria:① open fractures or combination of multiple injuries;② combination of radial nerve fractures;③ combination of psychological illness. (3) Surgical procedures. Posterior incisional and lateral intermuscular septum approach group: after general anesthesia, the patient was in lateral position of the healthy side. The affected elbow was placed in hand cradle. The incision was made from the posterior olecranon vertex to the acromion extension line or lateral side upwardly, which was normally located below the deltoid insertion. The incision could be properly extended if the fracture line is long. The skin, subcutaneous tissue and deep fossa were cut open to laterally isolate flap. With the traction of flap, the lateral intermuscular septum was carefully dissected. Firstly, the lateral cutaneous nerve of upper arm was found 11-14 cm proximal to the upper condyle. Then, the radial nerve was found at the provenance of lateral septal perforation via proximal tracking. The radial nerve was pulled to the sulcus up and down for tension free and then protected by lateral traction with children urine tube. The fracture end was exposed for a length equivalent to the distance between at least 3 screws on plate. If the fracture line was overlong, part of the deltoid insertion should be peeled off. The fracture was reduced with reset forceps and temporarily fixed with Kirschner wire for the placement of humeral lateral locking plate. The locking plate with proximal end of 5.0 mm and distal end of 3.5 mm should be selected prior to the one with proximal end of 3.5 mm and distal end of 2.7 mm. If the fracture line is close to the distal humerus, 3.5 mm humeral distal anatomical locking plate was selected to ensure the distal fixation with enough screws. In this study, double plate fixation was not used. The distal end of metaphyseal locking plate was slightly shaped forward, and the plate was placed on the proximal end or lateral side of olecranon fossa. The transverse fracture was conventionally fixed by fracture end compression. The periosteum should be protected for oblique and comminuted fractures without the pursuits of anatomical reduction and compression fixation. After satisfactory reduction, the plate was locked from distal to proximal with at least 4 screws on the distal end and 3 screws on the proximal end. Conventional intraoperative fluoroscopy was performed to check the alignment of fracture, the position of internal fixator and the length of screw. The drainage tube was placed as usual. The external fixation was unnecessary for 5.0 mm plate system. For 3.5 mm plate system, brace should be used for dynamic immobilization for 4 weeks. As young adult with strenuous exercise represented the majority of patients (the strength of 3.5 mm plate was less than that of 5.0 mm plate) , the dynamic immobilization was adopted to avoid internal fixation failure caused by excessive activity. More compliance was required if suspension immobilization was taken, and we believed that the brace was more reliable without affecting elbow joint function. Most of the middle-lower humeral fractures were oblique or comminuted, and only 1 case of transvers fracture was in the posterior approach group. The bridging fixation was considered as being able to avoid stress concentration, and thus we preferred locking plate. As no obvious biomechanical advantage was found in the treatment of transverse fracture with locking plate or common plate, we treated one of the transverse fractures with locking plate out of habit. However, the fracture end was compressed for rigid fixation. Anterolateral approach group: the patient was in supine position with the affected arm abducted and the ipsilateral shoulder elevated. A longitudinal incision was made 1 cm medial to the lateral margin of biceps, and the incision was dissected deeply along the lateral margin of biceps to expose and internally pull the biceps and part of brachialis; When the dissection was continued 2.5 cm above cubital crease along the lateral margin of biceps, the elbow joint was flexed in 70° with the forearm supinated to avoid the damage of forearm lateral cutaneous nerve through deep fascia; the radial nerve was isolated and explored between brachialis and brachioradialis. The deep dissection of brachialis should be carried out vertically and bluntly from the middle and lateral 1/3 of brachialis to avoid the damage of radial nerve located laterally to brachialis and anterior musculocutaneous nerve; The fracture end was reduced after the debridement of soft tissue impacted inside, and then the LC-DCP or DCP plate was shaped and placed on the flat surface of anterolateral humerus. The fixation was conducted based on AO operation technique; to check for damage, the radial nerve was explored between brachialis and brachioradialis and repaired with neuroectomyolysis or epineurial neurorrhaphy. The radial nerve should be sutured in brachioradialis to prevent its contact and friction with plate; the long strip of fracture fragment should be fixed with Kirschner wire firstly, which was later locked by plate with at least 3 screws at each end respectively. The other procedures were the same as those in the posterior incisional and lateral intermuscular septum approach group. (4) Postoeprative management. Regular rehydration, infection prevention and support therapy were adopted after operation. The affected arm was immobilized with forearm sling, and the drainage tube was removed from inside incision within 48 hours. Active extension and flexion of elbow joint was imitated on the 2nd day after operation. The affected arm was conventionally suspended for 6 weeks, and intense activities should be avoided within half a year. The regulating rechecks of X-ray were performed on the postoperative 1st, 3rd, 6th and 12th months. (5) Follow ups.This study was focused on outpatient follow ups. The life quality of patients such as fracture healing and physical activities of limb was collected in the subsequent visit and the postoperative outpatient follow-up of the 1st, 3rd, 6th and 12th months. (6) Statistical analysis. SPSS 20.0 software was used for statistical analysis. The measurement data of two groups were presented as±s , which were tested for normality. The independent sample t test was used if the they were in accordance with normal distribution, and 2-independent samples nonparametric tests were used if the measurement data of one or two groups did not comply with the normal distribution. The enumeration data were compared with chi-square test. A value P<0.05 was regarded as statistically significant.

(1) Operation time: through the references of operation records, the operation time of posterior incision and lateral intermuscular septum approach was obviously shorter than that in the anterolateral approach group. The operation time of both two groups were compared and analyzed and the difference was statistically significant P<0.05) . (2) Intraoperative blood loss: through the references of course records, the intraoperative blood loss of posterior incision and lateral intermuscular septum approach was obviously less than that in the anterolateral approach group. The operation time of both two groups were compared and analyzed and the difference was statistically significant (P<0.05) . (3) Radial nerve injuries: through the references of course records, 1 case of radial nerve injury was found in the anterolateral group, but no case was found in the posterior humeral approach and lateral intermuscular septum approach. There was no statistical difference between two groups (P>0.05) . (4) Fracture healing time: through the references of follow-up records and imaging data, all patients of both two groups obtained osseous healing within 1 year. There was no statistical difference between two groups (P>0.05) . (5) Broberg & Morrey scores of elbows joint after 1 year: through the references of follow-up records and imaging data, all patients of both two groups obtained good elbow function within 1 year. There was no statistical difference between two groups (P>0.05) .

The posterior incision and lateral intermuscular septum approach combined with posterior locking compression plate for the treatment of middle-lower humeral fractures has the advantages of short operation time, easy operation, reliable fixation and high healing rate, which allows early postoperative rehabilitation with satisfactory outcome. With proper rehabilitation guidance, no patient has elbow joint dysfunction. This approach can be used as a routine approach for the treatment of middle-lower humeral fracture.