Avascular necrosis of the capitellum after surgical treatment of a coronal shear fracture of the distal humerus: A case report

Abstract

Introduction and importance

Coronal shear fractures of the distal humerus are rare and are expected to have a high incidence of avascular necrosis (AVN) due to the avascular nature of the capitellar bone fragment and limited soft tissue attachment. However, according to the literature published thus far, AVN is infrequently reported, and some studies suggest that it does not have a significant impact on clinical outcomes.

Case presentation

Two female patients, one aged 72 and the other 70, presented with coronal shear fractures of distal humerus. Both patients were diagnosed with AVN of the capitellum 7 and 10 months after undergoing open reduction and internal fixation. One patient underwent hardware removal, while the other patient declined due to the absence of discomfort. However, at their final follow-up, both patients exhibited good clinical results.

Clinical discussion

The occurrence of AVN may be related to the severity of the initial injury, including posterior comminution. While some studies suggest that AVN of the capitellum may not affect clinical outcomes, hardware removal may be required in cases where there is intra-articular protrusion of the hardware.

Conclusion

Although AVN is a rare occurrence, even when it does occur, it may not significantly affect clinical outcomes. In this study, AVN may be associated with initial injury severity, and surgical treatment may make it possible to develop AVN. Moreover, considering the timing of the occurrence of AVN, it is believed that a close follow-up of more than one year will be required.

Highlights

• •Although AVN of the capitellum is a known complication of coronal shear fracture, its incidence and timing are poorly documented due to its rarity.
• •If AVN does not progress to the trochlea, it may does not have a significant impact on clinical outcomes.
• •Hardware removal surgery may be required in cases where there is intra-articular protrusion of the hardware due to AVN
• •Since the incidence of AVN in this study occurred within one year, a close follow-up of more than one year is necessary

1. Introduction

Coronal shear fractures of the capitellum rarely occur alone but may be accompanied by trochlea, lateral epicondyle, or medial epicondyle fractures [1]. The frequency of avascular necrosis (AVN) is expected to be high in coronal shear fractures because the fragment is an avascular fragment with little soft tissue attachment. However, AVN has been reported to be very rare in several studies [2]. Moreover, the presence of AVN usually does not affect the clinical outcome [3]. However, AVN of capitellum may cause an intra-articular hardware protrusion, so a close follow-up may be required. Here, we report two cases of AVN of the capitellum that occurred after open reduction and internal fixation of the coronal shear fracture but showed good clinical results, along with a literature review. This study was reported in line with the SCARE criteria [4].

2. Presentation of case

2.1. Case 1

A 72-year-old female patient visited the emergency room of our hospital complaining of right elbow pain after a simple fall. She was on medications for hypertension, hyperlipidemia, and Sjogren's syndrome, and there was no history of previous trauma or surgery on the right elbow. Plain radiography revealed coronal shear fracture of the capitellum extending to the trochlea, along with fracture fragments of the medial and lateral epicondyles and diagnosed as Ring classification type 5. Three-dimensional computed tomography(3D-CT) revealed comminuted bony fragments at the posterior aspect of the lateral column and trochlea (Fig. 1). Open reduction was performed using olecranon osteotomy through a posterior approach under general anesthesia, and the double plate was placed in parallel and fixed 4 days after the injury (Fig. 2). AVN of the capitellum was observed 7 months after surgery with plain radiography and there was no crepitus or discomfort. AVN progressed at follow- up 14 months after surgery, and there was crepitus during elbow joint motion. Removal of the metal implant was recommended; however, the patient did not want it because there was no discomfort. At the final follow-up at 21 months after surgery, there was pain of Visual Analogue Scale (VAS) 1; the range of motion was 5° of flexion contracture, 130° of flexion, 90° of pronation, 90° of supination. The result according to the Mayo Elbow Performance Index (MEPI) was good. (Fig. 3).

Fig. 1.

A–D. Preoperative anteroposterior (A) and lateral (B) radiographs showing right distal humerus coronal shear fracture. Preoperative computed tomography (C, D) showing posterior comminution.

Fig. 2.

A–B. Postoperative anteroposterior (A) and lateral (B) radiographs.

Fig. 3.

A–E. Postoperative 7 months anteroposterior (A) and lateral (B) radiographs showing avascular necrosis of capitellum (white arrow). Postoperative 21 months anteroposterior (C) and lateral (D) radiographs showing absorbed bone fragment (white arrow). (E) Photographs of the patient taken at 21 months after surgery.

2.2. Case 2

A 70-year-old female patient visited the emergency room of our hospital complaining of left elbow pain after falling from a 1-m height. There was no history of trauma or surgery on the left elbow. Plain radiography revealed a coronal shear fracture of the capitellum accompanied by a comminuted fracture of the trochlea and a lateral epicondyle fracture and diagnosed as Ring classification type 4. 3D-CT revealed comminuted fracture fragments of the posterior capitellum and the anterior and posterior fragments of the trochlea (Fig. 4). Open reduction was performed using olecranon osteotomy through a posterior approach under general anesthesia. Orthogonal double-plate fixation was performed 8 days after the injury (Fig. 5). A simple radiographic follow-up 10 months after surgery revealed joint space narrowing and intra-articular protrusion of the screw due to AVN of the capitellum, and implant removal was performed (Fig. 6). At the final follow-up, the patient reported a pain score of 1 on the VAS, and the range of motion was 20° of flexion contracture, 130° of flexion, 90° of pronation, and 90° of supination. The result according to the MEPI was good. (Fig. 7).

Fig. 4.

A–D. Preoperative anteroposterior (A) and lateral (B) radiographs showing left distal humerus coronal shear fracture. (C,D) Preoperative computed tomography showing posterior comminution.

Fig. 5.

A–B. Postoperative anteroposterior (A) and lateral (B) radiographs.

Fig. 6.

A–D. Postoperative 10 months anteroposterior radiograph (A) showing a narrowing of radiocapitellar joint (white arrow) and lateral radiograph (B) showing penetrated screw into the joint (white arrow). (C) Postoperative 2 weeks computed tomography and (D) Postoperative 10 months computed tomography showing avascular necrosis of capitellum.

Fig. 7.

A–C. (A,B) Radiograph after removal of implants. (C) Photographs of the patient taken at 1 year after surgery.

3. Discussion

In the past, capitellum fractures were considered and classified separately from distal humeral fractures [5]. Later, Mckee et al. reported a coronal shear fracture containing the trochlea and the capitellum, emphasizing the importance of the trochlea in capitellar fractures [6]. Ring et al. classified articular distal humerus fractures into five types according to their location, including the capitellum, trochlea, lateral epicondyle, and medial epicondyle [1]. In Ring classification type 3 or higher, which involves the posterior part of the capitellum or trochlea, impaction or comminution of the posterior fragment is often observed. Shortening due to the impaction of the posterior fragment makes it difficult to achieve anatomical reduction and increases the probability of nonunion. The comminution of the posterior fragment makes it difficult to fix the anterior fragment to the posterior with screws alone. Dubberley et al. suggested a new classification based on the presence of posterior comminution and emphasized its importance. They suggested that bone grafting or additional plate fixation was necessary when posterior comminution was present [7]. Furthermore, Brouwer et al. reported that nonunion occurred in 8 of 18 Dubberley classification type 3 B fractures and that stable fixation could not be obtained with posterior comminution [8].

Posterior comminution affects rigid fixation and the blood supply to the anterior bone fragments. According to a cadaveric study by Kimball et al., which analyzed the blood supply in the distal humerus, the medial columns were supplied with blood from the anterior and posterior sides, whereas the lateral columns were supplied mainly from the posterior side [9], [10], [11], [12]. Thus, the blood vessels entering from the medial and lateral sides do not connect within the bone, forming a watershed area in the trochlear region. When there is posterior comminution, damage to the posterior blood vessels often occurs, and additional damage easily occurs during the stripping of soft tissue while applying the posterolateral plate. For these reasons, blood flow to the capitellum is blocked, which affects the development of AVN [9].

The prevalence of AVN may be related to the severity of the initial injury, including the posterior comminution. In many studies, the majority of AVN cases were observed in Ring classification type 3 or higher [3], [7], [13], and the patients in this study were also classified as Ring types 4 and 5. However, studies on the relationship between injury severity and AVN incidence have not yet been conducted, and further research is required to investigate the incidence of AVN according to the severity of the initial injury. Despite being an avascular fragment with little soft tissue attachment, AVN rarely occurs in the capitellum. Carroll et al. reported AVN in only 6 out of 256 cases from 23 studies in a meta-analysis of capitellum and trochlear fractures [2].

Ring et al. reported that osteonecrosis of the capitellum does not seem to be a clinical problem [1]. Mighell et al. reported obtaining excellent or higher clinical results in two out of three cases with AVN and found no statistical correlation between poor results and AVN [3]. In this study, both patients showed good clinical results. However, some studies have reported poor clinical results, and arthroplasty was necessary in some cases [13]. The contradictory results seen in different studies may be due to differences in the extent and progression of necrosis. Sabo et al. conducted a biomechanical study and found that excision of an irrepairable capitellum fracture did not contribute to elbow stability, while excision of the trochlea led to multiplanar instability of the ulnohumeral and radiocapitellar joints [14]. These findings suggest that AVN confined to the capitellum may not have a significant impact on elbow stability and could be a factor contributing to relatively good clinical results.

AVN of the femoral head after a femoral neck fracture is known to occur within 2 years in 80 % of cases, but there are no reports on AVN of the capitellum [15]. Although AVN may have a minor impact on clinical outcomes, it is still important to monitor its occurrence as it may necessitate hardware removal. In this study, AVN occurred at 7 and 10 months after surgery, highlighting the need for careful follow-up of at least 1 year to monitor the possibility of AVN in cases of coronal shear fracture of the capitellum.

4. Conclusion

A coronal shear fracture is a rare fracture, and the incidence of AVN of the capitellum is even lower. The cases presented in this study are valuable in assessing the correlation between the injury severity and the occurrence of AVN, as well as determining the timing of AVN. In this study, AVN was found in cases with Ring classification 4 and 5, which may indicate a connection between the injury severity and the occurrence of AVN. Since AVN was detected at 7 and 10 months after surgery in this study, it will be necessary to have a follow-up period of over a year, with careful attention given to the possibility of AVN which may require removal of hardware.

Sources of funding

N/A.

Ethical approval

Ethical Approval was provided by the authors' institution.

Patients consent

Written informed consent was obtained from the patients for publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Research registration (for case reports detailing a new surgical technique or new equipment/technology)

N/A.

Declaration of competing interest

N/A.