Introduction

The coronoid process of the ulna is one of the main bony structures providing ulnohumeral joint stability, which acts as a buttress to prevent posterior dislocation, and key stabilizer against varus stress in an extended elbow [1]. Fractures of the coronoid process of the ulna generally occur in relatively high energy injuries, and it can be either an isolated finding following elbow dislocation or part of a more complex fracture dislocation [2]. Three types of coronoid fracture were originally described by Regan and Morrey [3]. All were transverse fractures across the coronal plane of the coronoid process, and surgical fixation is recommended for the type III that is a fracture involving more than 50 % of the process [4, 5].

The surgical approach to the elbow should provide an adequate extensile exposure with preservation of the neurovascular structures while permitting early mobilization of a stable joint, which heals without a joint contracture. It is technically demanding because of the high density of neurologic, vascular, and ligamentous elements around the elbow in complex injuries [6]. In addition, the propensity for stiffness or instability to develop after surgery has led many surgeons to avoid complex or extensive approaches to the elbow [6].

Although many articles concerned with treatment of coronoid process fractures have been published, little has been published about the surgical approach related to operative fixation in type III coronoid process fractures [711].

Authors chose anterior approach through the antecubital fossa for better exposure in that case of isolated coronoid process fracture that was not associated ligament injury resulting in joint instability. In spite of several advantages, there has been rare report related to this since Selesnick et al. reported similar approach in 1984 [12, 13].

This paper reports validity of anterior approach for the Regan and Morrey type III, isolated fracture of the coronoid process of the ulna without residual valgus or varus instability.

Materials and methods

Our institute found 32 patients who sustained a displaced coronoid fracture at the base of the coronoid process between January 2007 and January 2010. Our Institutional Review Board approved this retrospective study. The inclusion criteria was the Regan and Morrey type III coronoid process fracture without associated injuries (i.e. radial head or neck fracture, olecranon fracture, distal humerus condyle fracture), which needed additional surgical treatment, and they were included in O’driscoll type III, also. We excluded 18 patients because of concomitant injuries. One patient refused operative treatment for economic status, and another patient was treated conservatively for medical comorbidity. Two patients were lost to follow-up. We evaluated eleven consecutive patients (7 men, 4 women; mean age 44 years; range 25–67 years) who underwent surgical fixation through the anterior approach. The mean follow-up period was 21 months (range 15–34 months).

Eleven consecutive patients, who sustained a displaced coronoid fracture at the base of the coronoid process, underwent surgical fixation between January 2007 and January 2010 (Table 1). Surgery was performed by one surgeon. Seven patients were men, four were women, and their age ranged from 25 to 67 years, with an average age of 44 years old. All the patients were right-hand dominant, and injuries occurred in seven left elbows and in four right elbows.

Table 1 Patient characteristics

All were closed injuries without any open wound, and no neuro-vascular injury was detected. The most frequent mechanism of injury was a fall from a flat ground (3 patients) with walking or from a height (2 patients). The other causes included sports injuries (3 patients) and traffic accidents (3 patients).

All patients were associated with posterior elbow dislocations at their initial visit, and they showed remained posterior instability even after they were initially reduced in the emergency room. The limb was then immobilized in a long arm splint, and the patients underwent operative intervention between 1 and 5 days after the initial injury. They were reviewed and examined after a minimum follow-up of 12 months (range 12–34 months; mean, 19 months). On the initial visit, we evaluated computer tomography (CT) to understand fracture characteristics and to find other concomitant fractures. Preoperative and follow-up roentgenogram was evaluated for assessment of the fracture configuration, and confirmation of fracture union time was defined as the time that the fracture had external bridging of callus across fracture lines in two cortices on the lateral view of the elbow [14, 15]. Follow-up functional analysis included range of motion, Mayo elbow performance score [2], and DASH (Disability of the Arm, Shoulder, and Hand) score [16], and any detectable surgical complications were evaluated.

Surgical technique

With the patient in the supine position, the operative elbow is extended under tourniquet control. The skin incision commences two finger’s breadth proximal to the elbow flexion crease, a finger breadth lateral to the biceps, and curves across the elbow crease and distally along the ulnar border of forearm about two finger’s breadth distal to the flexion crease (Fig. 1a). During the skin incision, the lateral antebrachial cutaneous nerve should be identified and protected on the lateral side of the biceps tendon [6]. Bicipital aponeurosis is exposed underneath of the subcutaneous tissue and incised perpendicular to the aponeurotic fibers (Fig. 1b). Brachial artery and median nerve, which lie in the medial side, are identified. Biceps tendon is retracted laterally, and pronator teres, median nerve, brachial artery are retracted medially. Brachial muscle insertion is exposed through the space between those retracted structures, and it is longitudinally split (Fig. 1c). Split parts of brachialis tendon is reflected medially and laterally, and the area of coronoid fracture is accessed through this space (Fig. 1d). Joint capsule is usually attached at the tip of the fracture fragment and is separated and elevated from its insertion. After confirming accurate reduction and appropriate provisional fixation of the fragment with special attention to the congruency of the joint by image intensifier (Fig. 2), definitive fixation is performed. The majority of the coronoid fractures are well suited to a buttress type of plating or direct screw fixation. Fracture fragment is large enough to be secured with plate and screws because it is more than 50 % of coronoid process except severely comminuted case. Remained instability is evaluated through direct inspection and image-intensified radiographic examination of the arm while varus, valgus, and posterior stress is being applied.

Fig. 1
figure 1

a A single transverse curved incision is made on cubital area along the flexor crease of the elbow. b Illustration depicting a muscular and neurovascular structure after subcutaneous dissection. The bicipital aponeurosis is incised perpendicular to the aponeurotic fibers. c After medial and lateral retraction of surrounding structures, brachialis muscle insertion is exposed through the space. Longitudinal incision over the brachialis tendon is made. d Incised brachialis tendons are separated and retracted in the opposite direction. The coronoid process fracture fragment can be seen

Fig. 2
figure 2

Intraoperative temporary fixation with two K-wires under the supervision of image intensifier

All of the patients in this series did not show any clinical instability such as redislocation or subluxation after fixation of the fractures. Right-angled posterior long arm splints were applied with well padded dressing for first postoperative 2 weeks. On 15th day of operation, active and active-assisted elbow exercise (flexion, extension, pronation, and supination) was initiated under the supervision of by physical therapist. After a roentgenogram at 3 weeks, the posterior splint was removed, and resistive exercises are initiated 8 weeks after the surgery.

Case presentation

Case 1

A 35-year-old man sustained a fracture of the dominant right elbow after a fall from two stories high building. The plain roentgenogram showed a displaced coronoid process fracture associated with posterior dislocation of the elbow (Fig. 3a, b). Three-dimensional computed tomographic scans detected a fracture of coronoid process involving more than 50 % of height with articular depression. Open reduction and internal fixation were performed through anterior approach, and solid union of fracture was achieved during the follow-up (Fig. 3c, d). He restored 135° of active elbow flexion, 0° of flexion contracture, 80° of supination, and 80° of pronation (Fig. 4). Functional results were also satisfactory with a Mayo Elbow Performance score of 95 points, DASH Score of 3.6 points at the most recent follow-up.

Fig. 3
figure 3

Posterior elbow dislocation with coronoid process fracture. Anteroposterior (a) and lateral (b) image. c, d Plain radiograph of postoperation 1 year shows solid union with maintenance of articular side congruency

Fig. 4
figure 4

Elbow motion of patient in final visit 14 months

Case 2

A 25-year-old man who injured his arm in a fall from skiing had comminuted fracture of the coronoid process and posterior elbow dislocation (Fig. 5). Reconstruction of coronoid process was performed with autogenous bone graft from olecranon tip of the same elbow because severe comminution made it impossible to restore the buttress and articular surface (Fig. 6). Even though it needed additional small incision at the posterior side of the elbow for harvesting olecranon tip as donor bone, main procedure for coronoid fracture was performed through the anterior approach. Sound bony union was achieved postoperative 18 weeks (Fig. 7). He restored flexion of 130° with 80° pronation and 85° supination with a Mayo Elbow Performance Score of 95 and DASH Score of 2.8 points after 14 months of surgery (Fig. 8).

Fig. 5
figure 5

Initial plain radiograph of anteroposterior (a) and lateral (b) image show severe comminuted coronoid process fracture with posterior elbow dislocation

Fig. 6
figure 6

Intraoperative findings of communicated coronoid process fracture (a) and autologous bone graft from the olecranon with plate fixation (b)

Fig. 7
figure 7

Solid union was achieved in plain radiograph during the follow-up

Fig. 8
figure 8

The range of motion of the same patient in final follow-up 14 month

Results

Acceptable reduction and secure fixation of the fracture fragment were accomplished in all cases on the standard postoperative roentgenograms. Fractures were mainly fixed with plate and screws in 8 cases, cannulated screws in 3 cases. Supplemental K-wires were applied in 2 cases among the 8 cases of plate and screws fixation.

Fracture union

The fractures had united in all cases with no evidence of post-traumatic arthritis at the last follow-up. The average time to radiologic union was 15.2 weeks, ranging from 11 to 20 weeks (Table 2).

Table 2 Patient results

Range of motion

Mean flexion contracture was 3.6° (range 0°–10°), and further flexion was 130.9° (range 5°–140°) at the last follow-up. The average pronation and supination at the last follow-up were 73.2° (range 60°–80°) and 75.9° (range 65°–85°). All patients achieved a full functional range of elbow motion for daily activities.

Instability

Concentric reduction of the dislocated joint was achieved in all cases. Examination after fracture fixation in surgery revealed that none of the cases had instability that needed ligament repair. All patients maintained same articular spaces compared to uninjured elbow on plain radiographs, and they did not have any subjective or objective elbow joint instability at their final follow-up.

Functional score

All patients returned to their former work position, and their average Mayo Elbow Performance Score was 92.3 points ranging from 80 to 100. There were 8 excellent results, 3 good, and no patients showed fair or poor range of the score. In the analysis of DASH score, mean score was 5.7 points (range 1.6–8.3 points).

Complications

None of the patients showed any neurovascular complications with this anterior approach. There was one case of heterotropic ossification, which was small size, 3 mm diameter on lateral plain radiograph, anterior to fracture fixation. But it did not intervene in the elbow motion, and there was no subjective symptom.

Discussion

Coronoid fractures are relatively uncommon injuries occurring in approximately 2–10 % of patients with dislocation and occasionally found as isolated injuries [2, 12, 17]. Large coronoid fractures have a high incidence of elbow instability, and the current consensus of treatment is to fix virtually all coronoid fractures associated with elbow instability [18, 19]. The mechanism of coronoid process fracture is considered to be mainly axial-loading, and it is especially vulnerable to fracture when the elbow is flexed at 80° [20].

Two classifications are most commonly applied for clarifying the severity of the coronoid fractures. The classification of Regan and Morrey is based on size of fracture fragment in the plain lateral roentgenograms and is considered to be the clinically relevant classification system with biomechanical evidence [3]. Recent study of 103 acute fractures of the coronoid process, 51 were type I (49.5 %), 29 were type II (28.2 %), 23 were type III (22.3 %) [7], reported that large fragment fractures (type III) have a poor outcome when treated conservatively. Cabanella and Morrey [4] recommended early motion for stable type I and type II injuries and reduction and fixation for type III fractures. According to the biomechanical study based on the type of coronoid fracture by Closkey et al. [20], in response to axial load, elbows with a fracture involving more than 50 % of the coronoid process displace more readily especially when the elbow is flexed 60° and beyond.

This Regan and Morrey classification was subsequently modified by O’Driscoll into type I (transverse fracture of the tip of the coronoid process), type II (fracture of the anteromedial facet of the coronoid process), and type III (fracture of the base of the coronoid process) fractures [21]. In our study, we only included base fracture of coronoid process, which belongs to Regan and Morrey type III fracture, and it is the same category of type III in O’Driscoll classification.

The coronoid process has three soft tissue insertions such as the anterior joint capsule of the elbow, the brachialis muscle, and the medial ulnar acollateral ligament (MUCL). The anterior capsule inserts very close to the tip [22], the brachialis inserts anteriorlly, distal to the capsule, and the anterior bundle of the MUCL inserts distally and medially on the sublime tubercle of the distal ulna.

Cadaveric dissections performed by Ablove et al. [17] found that the average coronoid tip to capsule insertion distance was 2.36 mm, and the average coronoid tip to brachialis insertion was 10.13 mm with the average height of the coronoid process measured at 16.98 mm. It is, therefore, likely that the brachialis muscle insertion will only be involved in Regan and Morrey type 3 and O’Driscoll type 3 coronoid fractures. From that reason, some part of the brachialis muscle insertion should be separated and elevated for manipulating and reducing the fracture fragment in author’s anterior approach.

Repair of coronoid fractures has been described from lateral [2325], medial [5, 9, 21], posterior [5, 21, 26], and anterior approaches [12, 13], as well as a combination of these. Multiple means of fixation of coronoid fractures have been proposed depending on the nature of the fractures, and various techniques have been also described for osteosynthesis of larger fractures.

Lateral approach to the elbow has become a standard means to gain access for contracture release and to manage for fractures on the lateral side of the elbow, such as the radial head and the capitellum [27]. When the coronoid process fracture is associated with radial head fracture, this approach could be one of best options for fracture fixation. Also lateral approach is performed, taking care to utilize any pre-existing tears or defects in the common extensor origin, lateral capsule, or lateral ulnar collateral ligament [24].

Approach from the medial side is generally recommended for the isolated coronoid fracture [27]. Especially in patients with varus posteromedial rotational injury, the anteromedial facet of the coronoid is fractured, and fixation with plate by the medial approach is the best treatment [9, 21]. But in the medial approach, it is hard to exposure whole anterior part of the coronoid process, and it needs relatively long incision and more soft tissue dissection such as flexor carpi ulnaris and flexor-pronator mass [21]. In addition, it is technically difficult to provide strong vertical compression of the fracture fragment against the intact base of coronoid for stronger stability with simple instruments.

The advantage of the anterior approach is the wide exposure of the joint, which allows the surgeon to reduce the intra-articular fragment and inspects the articular lesions inside the elbow joint with the joint slightly flexed. In addition, this approach avoids damaging the normal anatomic structure of the elbow joint such as the medial collateral ligament, flexor-pronator muscle mass, and the ulnar nerve.

Because we believed that smaller volar side fragment could be fixed to the main dorsal stock in an easier and firmer way using this approach, the required instrumentation was less complicated, which include small cannulated or headless screw, plate, and screws.

Large fracture fragment can be secured with a buttress plate pushing the fragment against intact coronoid and helping to resist compressive forces across the joint, with or without additional screw fixation [28]. In case of severely comminuted fracture that could not be realigned, the coronoid process can be reconstructed by autogenous bone graft. One case of our series that was not amenable to fix the fragment due to severe comminution was reconstructed by bone grafting from the olecranon tip.

Authors did not use an additional separate medial or lateral approach because no patients associated with instability that needed ligament repair. This exposure brings the surgeon directly onto a tip fragment of the coronoid, and an anteromedial fragment superior to the sublime tubercle and direct anterior fixation can get stable fixation of fracture. In cases of significant comminuted fracture, supplemental fixation may be obtained using additional screws, K-wires, a second plate, suture, threaded wires, or suture anchors.

Even though the medial and lateral collateral ligaments are torn in most uncomplicated elbow dislocation [2931], the ligament and capsule structure repair is usually unnecessary due to the good bony support in the humeroulnar joint [5, 19, 32]. In more severe case like elbow fracture dislocation, there is higher tendency to show instability requiring ligament repair. But authors did not repair collateral ligament because secure bony fixation converted the fracture dislocation into simple dislocation, and there were no instability in all cases at the last follow-up. Jeon et al. [8] also reported functional outcome of eight coronoid type III fracture patients treated with early open reduction and stable internal fixation without LCL reconstruction. Among the patients, only one with a ligament injury who had had a primary ligament repair showed medial instability at follow-up, which was asymptomatic.

With regard to complications, no patient showed any neurovascular injuries despite nearby neurovascular structures including median nerve, brachial artery, and their branches. Heterotrophic ossification around the elbow joint has variable incidence. The most common symptoms are pain and limitation of motion, although erythema, edema, and warmth may be seen as well [33]. Surgical excision of heterotrophic ossification is critical in the management of traumatic injury to the elbow for patients suffering from nerve compromise, pain, or impaired range of motion affecting activities of daily living [34]. One of our cases showed heterotrophic ossification on follow-up radiographs, but we just did close observation and checked plain radiograph by periods because this patient did not show any subjective and objective symptoms.

Regan and Morrey published their experience of surgical treatment for coronoid process fractures [3]. Four of the five type III fractures had poor results secondary to stiffness, pain, and recurrent elbow instability in their study.

Recent study by Jeon et al. [8] showed average active elbow joint motion at the most recent follow-up was 105°, and the average Mayo Elbow Performance Score was 76.9 (range 50–95).

Comparing our results with other reports is inappropriate because the patients in our study had neither any other combined fractures that required internal fixation nor massive ligaments or soft tissue injuries such as open fracture. This study has limitation that there is no control group to prove the merits. However, authors believe that Regan and Morrey type III isolated coronoid process fractures could be managed through this anterior approach with the advantages of direct vision and minimal soft tissue dissection without damaging surrounding tissue.

Conclusion

Anterior approach through the antecubital fossa could be an alternative surgical approach for fixation of Regan and Morrey type III coronoid process fractures, which can give better exposure of fracture fragment with less soft tissue damage.