Midterm Results of Minimally Invasive Plating Osteosynthesis Versus Open Reduction and Internal Fixation for the Treatment of Mid-distal Humeral Shaft Fractures: A Case-match Controlled Comparison Study


 Background: Surgical treatment of humeral mid-distal shaft fractures is controversial. The purpose of this case-match controlled comparison study was to determine the safety and effectiveness of applying the MIPO technique compared with conventional ORIF for treating humeral mid-distal shaft fractures.Methods: This study was conducted from January 2012 to December 2016, and patients of mid-distal humeral shaft fractures were eligible for this study. The patients were followed up for a minimum period of 1 year. The clinical and radiographic outcomes were evaluated. The inclusion criteria were age between 18 to 60 years and acute displaced mid-distal humeral shaft fracture. The exclusion criteria were intra-articular fractures of the elbow, vascular insufficiency of the upper limb, pathological fracture and multiple or open fractures. The medical records and radiographs of all eligible patients during hospitalization and follow-up after discharge were reviewed. We use a 1:2 (MIPO/ORIF) case-match based on gender and age. All patients had at least 3 years of postoperative follow-up.Results: In total, 216 patients with mid-distal humeral shaft fracture underwent surgery at the departments of orthopedics of the Seventh Medical Center of PLA General Hospital and Beijing Chaoyang Hospital. Of them, 28 underwent MIPO and 56 case-matched controls underwent ORIF; all of them had complete 3-year follow-up data. No significant differences were observed in baseline characteristics between both groups. UCLA scores and MEPS were significantly higher in the MIPO group than in the ORIF group. Furthermore, UCLA score and MEPS grades in the MIPO group were significantly superior to those in the ORIF group. There was no statistically significant difference in major complication rates between both groups; however, the total major complication rate was significantly different between both groups.Conclusion: This study demonstrates that MIPO has a statistically significant clinical benefit over ORIF, including better shoulder and elbow joint function, with few overall major complications after at least 3 years of postoperative follow-up. MIPO is a safe and effective method for treating mid-distal humeral shaft fractures when surgery is indicated.


Introduction
MIPO is currently applied to treat long bone fractures of limbs. 1 The operation has the advantages of small incision, less soft tissue dissection, preservation of hematoma around the fracture and little in uence on blood supply of fracture block at the fracture end. Therefore, the fracture healing rate is high, the risk of infection is reduced, and bone grafting is basically unnecessary.
The surgical treatment of mid-distal humeral shaft fracture is controversial. Diverse opinions exist in the literature between the values of open reduction and internal xation (ORIF) with plate 2 and MIPO for treating mid-distal humeral shaft fracture. ORIF has been regarded as the gold standard in the treatment of humeral shaft fractures. The advantages of ORIF are anatomical reduction, strong xation, and little impact on elbow and shoulder joint function. 3 Nevertheless the disadvantages of this technique are also obvious, including large incision, more soft tissue dissection, serious blood supply damage to the fracture end, resulting in fracture healing di culty, prone to bone nonunion, and high risk of iatrogenic radial nerve injury. 4 MIPO for the treatment of mid-distal humeral shaft fracture has the advantages of small soft tissue dissection, avoidance of exposuring radial nerve, and low risk of iatrogenic radial nerve palsy. 5,6 This case-match controlled comparison study aimed to discover the safety and effectiveness of MIPO compared with conventional ORIF for the treatment of mid-distal humeral shaft fractures.

Study subjects and criteria
This retrospective case-controlled study was implemented at the Seventh Medical Center of PLA General Hospital and Beijing Chaoyang Hospital. All cases with mid-distal humeral shaft fracture between January 2012 and December 2016 were eligible for this study. The MIPO group comprised patients who underwent closed reduction and MIPO. The ORIF group comprised patients who underwent ORIF with plate and was considered the control group. The inclusion criteria were patients aged between 18 and 60 years and those with acute displaced mid-distal humeral shaft fracture. The exclusion criteria were cases involving intra-articular fractures of the elbow, vascular insu ciency, pathological fracture, and multiple or open fractures. The medical records and radiographs of all eligible patients during hospitalization and follow-up after discharge were reviewed.
We used a 1:2 (MIPO/ORIF) case-match based on sex and age (± 3 years). All patients had at least 3 years of postoperative follow-up. The research was approved by the research ethics committees of the two hospitals.

MIPO
To perform MIPO, patients were placed in the supine position under brachial plexus or general anesthesia.
The shoulder was abducted to 90° on a radiolucent table to allow fracture reduction and intraoperative radiography. Two mini-incisions were made. One is a proximal incision, located between biceps and deltoid, about 4 cm long, whereas the other was a distal incision approximately 4 cm long along the lateral spine of the humerus in the gap between the triceps and brachioradialis. Through these two incisions, a tunnel was made under the muscle across the fracture fragments and the locking compression plate (LCP) was placed into the incision. After the indirect reduction quality was checked, the proximal side was xed with three 3.5 mm diameter locking screws and the distal side was xed with four 2.5 mm diameter locking screws (Fig. 1).

ORIF
To perform ORIF, the patient was lying on the healthy side with the affected side on the upper side. The shoulder joint was abducted and extended 90°. After a median incision was created, the triceps was split and the proximal periosteum of the fracture was peeled to avoid the radial nerve. Radial nerve exposure is not necessary. The fracture ends were anatomically reduced and rmly xed with dynamic compression plate (DCP) under direct vision. The fracture ends of type A were compressed (Fig. 2).

Follow-up and outcomes
All cases were evaluated at 1, 3, 6, and 12 months and 2 and 5 years postoperatively. The primary outcomes compared were shoulder and elbow joint function, including the University of California at Los Angeles (UCLA) scoring system and Mayo Elbow Performance Score (MEPS). Shoulder function was graded excellent (≥34 points), good (33-29 points), or poor (<29 points) based on the UCLA scoring system, whereas elbow function was graded excellent (≥90 points), good (75-89 points), fair (60-74 points), or poor (<60 points) based on MEPS. Secondary outcomes were operation time, intraoperative blood loss, bone union time, and major complications, such as iatrogenic radial nerve palsy, infection, myositis ossi cans, and nonunion. Radiographic assessment was assessed by anteroposterior and lateral radiographs of the humerus. Bone union criteria were de ned as complete union of at least three of the four cortices in the radiographic assessment, and no clinical symptoms of pain.
Statistics SPSS version 23.0 (IBM, Armonk, NY, USA) and GraphPad Prism version 8.3.1 (GraphPad Software, USA) were used for statistical analysis. Continuous variable was described as mean ± standard deviation and compared using the independent samples t-test. Categorical variable was described as rate and compared using the Pearson chi-square test. Additionally, all patients included in the case-match controlled comparison study were considered as a cohort, and survival analysis was performed using the Kaplan-Meier method, in which total major complications and time interval between operation day and complication time were considered as the outcome variable and "time to event," respectively. The log-rank test was performed to compare survival curves of the two groups, and the Mantel-Haenszel test was used to estimate hazard ratio (HR) of total major complications. P <0.05 was statistically signi cant, and P < 0.01 was highly signi cant. There were 18 men (64.3%) and 10 women (35.7%) in the MIPO group, with a mean age of 36.0 ± 12.3 (range 18-59)). The mechanisms of injury were wrestling in 10 (35.7%) patients, throwing in 9 (32.1%), fall in 6 (21.4%), and car accidents in 3 (10.7%). The classi cations were A1 in 13 (46.4%) patients, B1 in 12 (42.9%), A2 in 1 (3.6%), B3 in 1 (3.6%), and C1 in 1 (3.6%) depending on the Orthopaedic Trauma Association (OTA) classi cation. In the ORIF group, there were 36 males (67.9%) and 18 females (32.1%), with a mean age of 36.8 ± 12.8 (range 18-58) years. The mechanisms of injury were wrestling in 23 (41.1%) patients, throwing in 17 (30.3%), fall in 9 (16.1%), and car accidents in 7 (12.5%). The OTA classi cations were A1 in 29 (51.8%) patients, B1 in 18 (32.1%), A2 in 3 (5.4%), C1 in 3 (5.4%), A3 in 2 (3.6%), and B2 in 1 (1.8%). Moreover, four (14.3%) cases in the MIPO group and 7 (12.5%) in the other group had preoperative radial nerve injury. There was no signi cant difference in the baseline characteristics ( Table 1).

Results
The UCLA scores in the MIPO group were great higher than the ORIF group (34.4±1.7 vs. 31.2±3.9, p=0.000). In the MIPO group, 27 (96.4%) patients showed excellent results and 1 patient (3.6%) had a poor result. In the ORIF group, there were 35 (62.5%) excellent cases, 14 (25%) good cases, and 7 (12.5%) poor cases. The MIPO group was highly signi cantly superior to the ORIF group in terms of UCLA scores and grades (p=0.003). The patient with poor shoulder function in the MIPO group was obese. Due to poor compliance, the shoulder joint did not undergo timely functional exercise, resulting in adhesion around the shoulder joint. In the patients with poor shoulder function in the ORIF group, 3 had nonunion, 2 had myositis ossi cans, and 2 had iatrogenic radial nerve injury.
Similar results were observed for MEPS. MEPS in the MIPO group was great higher than the ORIF group (97.9±4.2 vs. 86.7±13.6, p=0.000). In the MIPO group, 27 (96.4%) patients showed excellent results and 1 (3.6%) patient had a good result. In the ORIF group, there were 37 (66.1%) excellent cases, 9 (16.1%) good cases, 8 (14.3%) fair cases, and 2 (3.6%) poor cases. The MEPS grades of the MIPO group were statistically superior to those in the ORIF group (p=0.02). One patient with myositis ossi cans and one patient with intraoperative radial nerve injury had poor elbow function in the ORIF group ( Table 2).
Iatrogenic nerve injury was not observed in the MIPO group and 4 patients with iatrogenic radial nerve injury (7.1%) were postoperatively identi ed in the ORIF group with no statistical difference (X 2 =2.10, p=0.147). Of the patients with iatrogenic nerve injury, two recovered within 4 weeks postoperatively without intervention and two did not recover 6 months postoperatively; these patients underwent radial nerve exploration, revealing that the radial nerve was cut off in one patient and pressed under the plate in another patient.
No deep infections were observed in the MIPO group, whereas 3 cases (5.6%) in the ORIF group developed deep infection; however, all infections were treated with intravenous antibiotic therapy. No statistical difference was found between the two groups (X 2 =1.56, p=0.21).
All cases in the MIPO group met the standard of bone union within 9 months postoperatively. Three patients (5.6%) in the ORIF group had nonunion, and each nonunion was subsequently managed by implant removal and ORIF with bone graft. There was no statistical difference in the incidence of nonunion between the two groups (X 2 =1.56, p=0.21).
There were 4 patients (7.1%) with myositis ossi cans in the ORIF group; however, this complication was not observed in the MIPO group with no statistical difference between the two groups (X 2 =2.10, p=0.15).
Only one myositis ossi cans needed elbow joint release to improve elbow function.
Internal xation removal was performed in 20 cases in MIPO group without complication and in 14 cases in ORIF group with 2 complications. One patient had radial nerve injury during the surgery, whereas one patient had refracture after surgery in the ORIF group. No statistical difference was observed between two groups (X 2 =3.04, p=0.08) ( Table 2).
The overall major complication rates in midterm follow-up were 0 and 16 (0% vs. 28.6%) in the MIPO and ORIF groups, respectively, which signi cantly differed between the two groups (X 2 =9.88, p=0.002). Survival analysis and cumulative complication incidence curves showed that the total major complication rate was highly signi cantly lower in MIPO group than in ORIF group (p=0.000) (Fig. 3). The Mantel-Haenszel test showed that HR (MIPO/ORIF) was 0.20 with 95% CI of 0.07-0.56, indicating that every ve patients who underwent MIPO can avoid one major complication compared with those who underwent ORIF.

Discussion
As far as we know, this study was the rst to report midterm results of MIPO versus ORIF treating middistal humeral shaft fractures. After at least 3 years of postoperative follow-up, shoulder and elbow joint function in MIPO group was statistical superior to that in ORIF group. There was no statistically signi cant difference in the incidence of each major complication between the two groups; however, the overall major complications rate between the MIPO and ORIF groups had highly signi cant difference, with HR (MIPO/ORIF) of 0.20. This indicates that every ve patients who undergo MIPO can avoid one major complication compared with those who undergo ORIF.
Humeral shaft fracture is a common fracture and can be the result of low-or high-energy trauma, accounting for 3%-5% of all fractures. 7,8 Although the incidence of humeral shaft fractures is high, the incidence of the mid-distal shaft fractures is low, accounting for about 20% -30% of all humeral shaft fractures. 9 Zogbi et al. reported 15 patients with this type of fracture in 6 years 10 . In the present study, two hospitals treated 84 patients with MIPO and ORIF in 5 years. Most of the injuries were caused by wrestling and throwing, and the most common types of fractures were types A1 and B1 of the OTA classi cation.
The optimal clinical management of a mid-distal humeral shaft fracture is controversial. Intramedullary nailing (IMN) and ORIF of humeral shaft fractures have been reported to obtain similar good clinical results. 11,12 However, for some distal fractures, it is inappropriate to use an intramedullary implant; there is obliteration of the intramedullary canal proximal to the supracondylar region, and this site can easily be affected by a fracture. 13 In the present study, all patients were not candidates for treatment with IMN.
Although MIPO has been successfully applied for femoral and tibial fractures for nearly 20 years, 14,15 its use in humeral fracture was not reported until 2005 owing to the fear of injury to the radial nerve. At that time, Apivatthakakul et al. initially performed an anatomic cadaveric study to prove the safety and feasibility of MIPO for humeral shaft fracture. 16 Thereafter, the effectiveness and safety of MIPO for the treatment of distal humeral fracture were further con rmed in the short-term clinical results of several case series. 5,6,10,[17][18][19][20][21][22][23] Some comparative studies have also determined the advantages of MIPO for the treatment of humeral shaft fracture in short-term follow-up. Matsunaga et al. performed a clinical trial to verify clinical and radiographic outcomes between patients who had been treated with MIPO and nonoperation with a functional brace and found that MIPO had a statistically signi cant advantage with respect to joint function score. 24 An et al. retrospectively analyzed 33 cases with mid-distal humeral shaft fractures treated with MIPO and ORIF. Their results showed that MIPO can reduce the incidence of iatrogenic radial nerve injury (0% vs. 31.3%). 25 Oh et al. compared the clinical e cacy of MIPO (n = 29) and ORIF (n = 30) in the treatment of humeral shaft fractures, and found that the average operation time in MIPO group was shorter than that in ORIF group (110 min vs. 169 min, P < 0.05). They believe that MIPO potentially reduces the complications with a shortened operation time. 26 Our study yielded similar clinical results.
Malunion is a more common complication reported in treating humeral shaft fracture. 5,19,27 Li et al. compared the degree of poor rotation and shoulder joint function after treatment of humeral shaft fracture with ORIF and IMN, and found that there was no poor rotation in patients receiving ORIF treatment, but the rate of poor rotation in IMN group was 27.2%, which led to the reduction of shoulder joint range of motion. 28 Wang et al. Prospectively divided 53 patients with humeral shaft fractures into MIPO group or ORIF group. The incidence of postoperative malrotation in MIPO group was signi cantly increased (40.9% vs. 0%; P < 0.01), which was related to shoulder joint degeneration. 29 In this study, no patient in the MIPO group showed malunion. This may be because the locking compression plate (LCP) was contoured to conform to the anterolateral surface of the mid-distal humerus, possibly reducing the incidence of malunion and making fracture reduction more convenient.
Our study proved that the MIPO group had statistically signi cantly better functional outcomes than the ORIF group in midterm follow-up. We believe that the advantages of MIPO in the optimal safety pro le and gaining strong xation without damaging the surrounding soft tissue were the foundation of obtaining good joint function. Mahajan et al. evaluated the e cacy of MIPO in the treatment of middle humeral fractures in patients mainly engaged in overhead shoulder movement (athletes and manual workers), and found that most patients had good functional results, which was similar to our clinical results. 30 Our study had three limitations. First, the LCP in the MIPO group was contoured to conform to the anterolateral surface of the mid-distal humerus, which can reduce the strength of the plate, although we did not nd any cases of failure of internal xation. Therefore, a new anatomical plate designed for this area is needed. Second, the number of patients included is relatively small, because it is di cult to recruit hundreds of patients with such fractures in clinical practice. We designed the case-match controlled study to reduce type II error. Second, the number of patients was relatively small because it was di cult to enroll hundreds of patients in clinical practice. Therefore, we designed a case-match controlled study to reduce type II error. Lastly, this study was carried out at two centers and there might be a bias on the surgical technique and functional assessment. In the research preparation stage, we developed standard surgical procedures and all surgeons had vast experience to minimize potential bias.

Conclusions
This study demonstrates that MIPO has a statistically signi cant clinical bene t over ORIF, including better shoulder and elbow joint function, with few overall major complications after at least 3 years of postoperative follow-up. MIPO is a safe and effective technique for treating mid-distal humeral shaft fractures.

Declarations
Ethics approval and consent to participate Consent for publication Signed informed consent for publication was obtained from all authors.

Availability of data and materials
The data and materials used and/or analyzed during the current study are not publicly available but available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.