1 INTRODUCTION

Proximal humerus fractures, which constitute 4% of all fractures, are frequently discussed in the orthopedic literature. Although most non-displaced or minimally displaced fractures are treated conservatively, displaced fractures may require surgical treatment.^1,2 Several surgical approaches have been defined to treat unstable displaced humeral fractures, such as the deltopectoral approach, the deltoid split approach, the anterolateral approach, the anteromedial approach, and the transacromial approach.^3–6

The most commonly used approaches are deltoid splitting and the deltopectoral approach.⁷ The deltoid splitting approach has been frequently shown to cause iatrogenic nerve damage, and it has been histologically proven that structural damage may occur.^7–10 Electrophysiological evaluations of this injury have been studied, but no statistical evaluations of clinical significance have been reported.^7,11–16 It has also been demonstrated that the deltopectoral approach provides a better view, especially for tuberculum majus reduction, and allows direct viewing of the plate location.^3,7,8

This study investigated axillary nerve function with clinical correlations in patients treated with the deltoid split approach. We hypothesize that although the Deltoid split approach has a high risk of axillary nerve degeneration, this degeneration does not affect functional results. We also aimed to investigate the effects of this damage on deltoid muscle volume and discuss the effects of volumetric changes and nerve damage on patients' clinical outcomes.

2 Materials and methods

2.2

2.2 Preoperative assessment, surgical approach, and postoperative rehabilitation

All patients were evaluated with a preoperative motor and sensory examination of the axillary nerve as part of the neurovascular examination. All patients underwent anteroposterior and transthoracic shoulder lateral radiographs and computed tomography of the shoulder to assist in classification and preoperative planning. In our clinic, deltoid split approach is preferred in the surgery of Neer III and IV fractures. All patients were operated on by the same surgeon.

The patients underwent preoperative motor and sensory examination of the axillary nerve as part of the neurovascular examination. Sensory examination of the axillary nerve was performed by testing the sensation around the shoulder. The motor examination was performed by the physician's palm, touching on the patient's deltoid muscle and asking the patient to tighten his muscles as abducting his shoulder.

The patients were prepared in the bench-chair position according to the standard orthopedic fashion. The surgical incision site was drawn with a sterile pen and the operation was started. 3.5 mm Proximal Humerus locking plates was used for internal fixation of the fractures. All patients were operated by the same surgical team.

The skin incision was extended between the acromial and spinal parts of the deltoid muscle for the lateral deltoid approach to the proximal humerus (Fig. 1). Care was taken to ensure that the incision did not extend more than 5 cm distal to the acromion to protect the axillary nerve.

Close

Fig. 1 Surgical site marking.

Export to PPT

Muscle fibers were split at the middle of the lateral part of the deltoid. The deltoid muscle split proximally to the acromion to provide the highest point of view, but the muscle protected 5 cm distally to protect the axillary nerve. In order to protect the axillary nerve from damage a stay suture placed at the inferior border. The axillary nerve was palpated on the deep surface of the muscle. After reduction of fracture, proximal humerus Plate placed to bicipital in 2–3 mm posterior to the bicipital groove and 5–7 mm distal to the greater tubercle(Fig. 2). After fixation wound was irrigated, a drain was placed. Finally deltoid muscle, fascia and subcutaneus tissues were closed. The patients underwent neurovascular examinations and wound care on the first postoperative day. All patients had the same rehabilitation protocol after the surgery.

Close

Fig. 2 Intraoperative photo.

Export to PPT

2.4

2.4 Electromyoneurography technique

Patients were evaluated by electromyography at the 45th day, 12th month, and 24th month postoperatively by a board-certified senior physiatrist. Electromyoneurography (EMNG) was performed with a Nihon Kohden 9400 K device. For the EMNG electrode, the recording field strength of a 0.015-mm² bipolar needle electrode with a signal amplitude of 0.1–0.5 mV was used. Results from superficial disc electrodes were recorded. Relative latency delay was more than 1 ms, and amplitude lower than 50% was considered abnormal according to the unaffected side.

Motor unit potential analyses were performed by evaluating axillary nerve functions from the anterior and posterior fibers of the deltoid muscle by needle electromyography (EMNG) technique (Fig. 3).

Close

Fig. 3 EMNG technique.

Export to PPT

2.5

2.5 Volume measurement technique with magnetic resonance

Deltoid muscle volumes of patients were investigated, and atrophy caused by axillary nerve damage was revealed. Bilateral shoulder MRI was performed with a General Electric Medical Systems SIGNA Explorer GEM Suite 1.5 T MR spectrometer using HD16-channel large flex coil axial T1W images (slice thickness, 4.0 mm spacing, 1.0 mm; field of view 18.0; matrix size 224 × 288) obtained from the supraclavicular region of both shoulders, including the adherence site of the caudal deltoid muscle to the humerus on 24th month. A manual drawing was made from the outer edge of the muscle in each section by a board-certified radiologist, and the muscle volume was calculated automatically by configuring with the OsiriX MD program (Fig. 4).

Close

Fig. 4 3D modeling obtained during deltoid volume measurements.

Export to PPT

3 Results

Twenty-five patients operated on with open reduction internal fixation were prospectively observed. Patients' characteristics are given in Table 1. None of the patients had complications such as superficial infection, avascular necrosis, nonunion, implant failure, or subacromial impingement. Neurological deficits were not detected in any patients according to the axillary nerve examination performed at the first presentation of the patients clinically.

In the EMNG measurements of the patients on the 45th postoperative day, partial degeneration was observed in the anterior part of the axillary nerve in all cases (100%). In the control EMNG measurements performed at the 12th month, normal values were obtained for 15 (60%) of the patients, while findings of ongoing regeneration were detected for 10 (40%) of the patients. The final EMNG results in the 24th month we obtain normal results from all patients.

Patients' CM scores DASH scores are shown in Table 2. A high-level negative correlation with 99% reliability was found between the DASH and CM scores. As CM scores increase, DASH scores decrease (p = 0.000). The difference was not statistically significant in any period between the normal and abnormal EMNG groups (p = 0.281) (Table 2).

The deltoid volumes of patients in the 24th month are shown in Fig. 5. The difference between the groups with normal and abnormal 12th-month EMNG results was statistically significant (p = 0.007). Clinical deltoid volume results in patients with normal and abnormal EMNG values were analyzed on the 24th month. The group with normal EMNG results and the group with changes in EMNG, CM scores in the third, six^th, 12th^, and 24th months did not correlate with the deltoid volume decrease. The deltoid muscle thicknesses of the patients were also measured at the mid-glenoid level and compared with the contralateral thicknesses. There was a statistically significant difference between normal EMNG results and the group of patients with abnormal EMNG in anterior, posterior, and lateral muscle thicknesses. Anterior muscle thickness was statistically higher in the normal patient group than those with abnormal EMNG results (p = 0.000). Lateral and posterior muscle thicknesses were not statistically different (p = 0.246; p = 0.510).

Close

Fig. 5 Deltoid volumes according to sixth-month EMNG results (mm3); asterisks signify a statistically significant difference (p = 0.0073).

Export to PPT

4 Discussion

We electrophysiologically, clinically, and radiologically evaluated patients with proximal humerus fractures operated on with a deltoid split approach. The most important result of this study is that although the deltoid volumes are affected, the patients' clinical scores are not affected by electrophysiological changes.

Several studies in the literature evaluate the electrophysiological changes of patients operated on with a deltoid split approach.^7,13–15,17 In the study of Westphal et al., temporary electrophysiological changes were reported in 3 (7.5%) patients and permanent changes in 3 (7.5%) patients. The clinical scores of the patients with permanent nerve damage were reported to be worse, and the clinical scores of patients with temporary injury were reported to be close to those, but it was not explained which statistical method was used.⁷ In the study in which Röderer et al. evaluated 10 out of 23 patients treated with the minimally invasive deltoid split method for proximal humerus fractures, EMNG changes were reported in all patients (100%) at six weeks and in 7 (70%) patients at seven months. According to the severity of the electrophysiological changes, it was stated that CMS values decreased, but no statistical evaluation was performed.¹⁶ In the study in which Wu et al. compared the deltopectoral and deltoid split approaches, the EMNG evaluations performed in the 3rd month among 28 patients operated on with the deltoid split approach found EMNG changes in 7 (25%) patients. They further reported not finding any clinical abnormalities in any patients.¹⁴ Khan et al. prospectively analyzed 14 patients operated on with a lateral deltoid split approach due to proximal humeral fractures. They reported mild changes in the anterior branch of the axillary nerve in only one of the patients upon examination of axillary nerve function by EMNG between 6 and 12 months.¹³ However, the details of the electrophysiological test technique were not explained, and EMNG times were not homogeneous. In the same study, the mean CM score of the patients without EMNG change was 81.6 while the mean DASH score was reported as 15.78, whereas in the patient with EMNG change, the CM score was reported as 88, and the DASH score as 9. Since only one patient had EMNG change, it was impossible to group and compare patients statistically. However, the authors reported that the clinical results of the patient whose axillary nerve was injured were the same as those of the other patients.¹³ We found damage in the anterior motor branch of the axillary nerve at 100% in the sixth week and 40% in the 12th month, and all nerve damages healed in the 24th month in our patients who had been treated with the deltoid split approach. That damage was found to cause muscle atrophy, which was especially evident in the anterior part of the deltoid, but this atrophy and nerve injury did not affect the clinical scores.

The deltoid volumes of all patients decreased compared to the non-operated sides. According to the results of 12th-month EMNG, deltoid muscle volume reduction was found to be statistically significant in the group in which reinnervation continued. Considering the results, the decrease in deltoid volume did not affect the clinical results of patients in the 24th month. However, the mean CM score in the sixth week in the group in which EMNG changes were observed was lower, although this did not represent a statistically significant difference. This suggests that patients experiencing EMNG changes in the 12th month may have clinically worse results in the early period. The clinical results of these patients improved rapidly until the sixth month, and the difference between the patients disappeared after the sixth month.

Our study had several limitations. First of all, our study was based on the data obtained through 24 months of follow-up among a group of 25 patients, which prevented us from comparing long-term results. Studies with larger samples and more extended follow-up periods are needed. Another possible limitation is that it cannot be demonstrated whether the axillary nerve injury of patients occurred at the time of trauma or during surgery. The same issue was discussed in the editorial correspondence between Westphal et al. and Visser et al.^18,19 According to current information, the electrophysiological evaluation that can be performed 14 days after trauma to the nerve is not recommended, and the surgery of the patients is performed within this period. Although motor examination of the axillary nerve is performed in patients, the contractions of the deltoid muscle cannot be clearly evaluated due to the patient's pain. Thus, the only possible evaluation of the axillary nerve in the case of a proximal humeral fracture is the two-point discrimination of the sensory branch. This evaluation has no reference values in the literature. Based on the information in the literature, there is still a need for a method that will reveal whether nerve injury occurs during trauma or surgery. We think this is not currently possible in light of the available information.

5 Conclusions

In proximal humeral fractures treated with the deltoid split approach, there may be iatrogenic damage of the anterior branch of the axillary nerve. This damage causes atrophy only in the anterior part of the deltoid. However, this atrophy and axillary nerve damage do not affect the patients' clinical scores in the early and middle terms. Routine electrophysiological examination in the follow-up of these patients is not considered necessary because the possible injury does not affect the clinical results of the patients. The deltoid split approach can be used safely to treat proximal humerus fractures.

Conflict of interest

None.

Funding/sponsorship

None.

Institutional review board approval or patient consent

Our study was approved by the local ethics committee with the number E−2369 and informed consent was obtained from the patients whose data and images were used.

Author's contribution

UG, EC, IAY, FGY, EC, FS, KE performed data collection, statistical analysis and writing of manuscript. AOY, KE, FS contributed to data collection and manuscript review. UG, KE, AOY manuscript editing. All authors contributed to study conception and planning, identification of patients, manuscript editing.