In this study we compared the clinical outcomes of patients in three different groups. Based on the results, there were 3 cases of transient radial nerve palsies patients in the non- ultrasound group. But No nerve complications occurred in patients who accepted ultrasound guidance before or during the operation at the follow-up evaluation. In addition, the function scores including VAS, MEPS, and DASH scores, and ROM of the elbow had similar outcomes among the non-, pre- and intra-operative ultrasound groups at 6 and 12 months postoperatively.
The use of elbow arthroscopy for related elbow joint procedures has continued to increase over time, necessitating a greater assessment of the associated complications after elbow arthroscopy [21,22,23]. Jinnah et al.  reported that 222 nerve injuries occurred in 372 respondents; ulnar, radial, and posterior interosseous nerve injuries occurred in 38%, 22%, and 19% of the patients, respectively. These findings indicate that major nerve injuries after elbow arthroscopy are not rare occurrences. To the observations of general elbow arthroscopy, we also focused on the outcomes of patients accepting different forms of ultrasound guidance. Because of the high probability of radial nerve injuries, we mainly observed radial injuries after elbow arthroscopy. The nerve injuries in the non-ultrasound was the radial nerve in 15% of the patients. The probability of nerve injuries in our study was lower and the main injuries involved the radial nerve.
Previous studies [6, 7, 24] have demonstrated that the radial nerve is prone to injury during the anterolateral approach in elbow arthroscopy. The radial nerve is directly anterior to the radial head, and the anterolateral portal is located anterior to the articulation of the humeroradial joint . Thaveepunsan et al.  reported a needle-and-knife technique for anterolateral portal placement during elbow arthroscopy. The anterolateral portal placement was also the focus in our study, and we suggested a safe and effective technique with ultrasound during surgery. We could directly see the radial nerve under the ultrasound image, and there were no complications using this technique. Using an ultrasound-assisted technique provides precise location of the nerve in the case of nerve variation.
Powell et al.  reported that preoperative mapping facilitates planning of surgical access. They preformed preoperative sonographic ulnar nerve mapping following various elbow operations. It was concluded that sonographic mapping of the ulnar nerve mitigates the potential inaccuracy of nerve palpation in a complicated postoperative elbow joint, and this technique may reduce the risk of ulnar nerve injury. Preoperative mapping under ultrasound technique was our preference, but we focused on the radial nerve. We drew the path of the radial nerve from the distal humerus to the mid-forearm as a reference during surgery. There was no patient with radial nerve palsy in the preoperative ultrasound group. Our study was a further investigation for the application of preoperative mapping under ultrasound, and we concluded that preoperative radial nerve mapping under ultrasound is safe and effective.
Ohuchi et al.  presented a technique applicable to portal placement of the elbow using an ultrasound-assisted method. However, they did not perform this technique in a large number of cases. Although we used the ultrasound-assisted technique during elbow arthroscopy, we focused on the anterolateral portal placement of the elbow joint. Portal placement with an ultrasound-assisted technique in elbow arthroscopy is a safe surgical method. Although the application of ultrasound in elbow arthroscopy may add surgical time and preoperative preparation, this method is a good way to avoid nerve injuries.
Hackl et al.  pointed out that the radial nerve shifts due to flexion and joint insufflation. The distance of the median and radial nerves to osseous landmarks doubles from extension to 90°flexion and triples after joint insufflation. In our study, preoperative markers of the radial nerve could shift according to Hackl et al. But in the results, there was no nerve injury in preoperative ultrasound group. We thought that although the radial nerve shifts due to flexion and joint insufflation, drawing path of the radial nerve preoperatively could also provide a valuable reference.
The VAS score has been used to evaluate pain, and elbow function can be assessed by MEPS and DASH scores, subjectively and objectively [27,28,29]. The ROM of the elbow in the neutral position reveals significant values to evaluate elbow function. Previous studies have demonstrated that arthroscopic elbow surgery leads to improved ROM and health-related quality of life in elbow disorders [30,31,32]. In our study, there were no significant differences in VAS and functional scores among the three groups at the 6 and 12-month follow-up evaluation.
The surgery was performed without ultrasound guidance at initial. then we gradually performed elbow arthroscopy with ultrasound assistance. All surgical procedures were performed by a senior surgeon with 17 years of experience in joint surgery. The surgeon with extensive knowledge of elbow anatomy might reduce the influence of the surgeon’s experience on the operation complication.
Although evaluation of patient outcomes in our study was more comprehensive, this study had several limitations. Because of the strict inclusion criteria, the study population was small. In addition, this was a retrospective study with all the associated biases, and future prospective randomized controlled studies with long-term follow-up are warranted to confirm our findings.