Elbow arthroplasty in trauma-current concepts review

Abstract

Introduction

Despite advancements in modern locking plate technology, distal humerus fractures in the elderly remain difficult to treat. A subset of fractures in this osteoporotic bone includes multiple, shallow articular fragments that renders fixation unreliable, precluding early motion and acceptable functional outcomes. Arthroplasty, in the form of either Total Elbow Arthroplasty (TEA) or Distal Humeral Hemiarthroplasty (DHH) are alternative treatment options in this cohort and are being increasingly used.

Methods

This article reviews the use of TEA or DHH for acute distal humerus fracture, including patient selection, pre-operative planning, surgical approach, implant positioning, rehabilitation, outcomes and complications.

Results

Arthroplasties are being increasingly used for acute distal humerus fractures, however they introduce potential complications not seen with fixation. Due care must be employed to correct implant positioning which is a function of implant rotation, implant length and implant sizing. We describe a robust technique for epicondyle repair in DHH and unlinked TEA to avoid instability. Outcomes of DHH and TEA for acute distal humerus fracture are encouraging, however further long-term outcome and comparative data regarding arthroplasty is required.

Conclusions

Short to medium term outcomes demonstrate that both DHH and TEA are valuable options for selected patients, although attention to technique is required to minimise potential complications.

1. Introduction

Distal humerus fractures have an incidence of approximately 5.8 per 100,00 per annum. They occur most frequently in older female patients.¹ Like many fractures in this patient demographic, the incidence is increasing, with one large epidemiologic study demonstrating a sevenfold increase between 1970 and 1995.²

Historically, comminuted distal humerus fractures in the elderly were treated via the non-operative “bag of bones” approach; with early motion encouraged despite articular malunion.³ The development of modern fixed angle locking plates in the 1980s allowed open reduction internal fixation (ORIF) of distal humerus fractures to become the predominant surgical technique.⁴ In the absence of any surgical complications, ORIF has been shown to have improved outcomes when compared to non-operative management.^4,5

Given the patient demographic most commonly sustaining distal humerus fractures, they often occur in osteoporotic bone.⁶ A subset of fractures includes multiple shallow and osteoporotic articular fragments, rendering fixation unreliable despite modern techniques. Good outcomes are reliant on stable fixation allowing early functional motion.

Complication rates for ORIF have been reported as high as 44% in patients aged over 65 years of age.⁷ In particular, poor outcomes are reported in patients with intra-articular comminution and/or severe soft tissue injuries.8, 9, 10, 11 Non-union and metalwork failure is an issue in the elderly and even in those fractures that go on to unite, patients suffer from persistent pain, motion deficits and a risk of post-traumatic arthritis.¹² The loss of elbow function can severely limit patients’ activities of daily living and independence.¹³

Given the challenges and complications associated with ORIF of elderly distal humerus fractures, Total Elbow Arthroplasty (TEA), and more recently Distal Humeral Hemiarthroplasty (DHH) have been increasingly used, in an attempt to more reliably restore function and independence in this patient cohort.

This article aims to review the current use of elbow arthroplasty for the acute treatment of distal humerus fractures, including pre-operative planning, surgical technique and outcomes from published literature and UK National Joint Registry (NJR) data.

2. Current use of arthroplasty for acute distal humerus fractures

2.1. TEA

Since the 1980s, Total Elbow Arthroplasty (TEA) has been used to treat symptomatic non-unions and post-traumatic elbow arthritis as sequalae of distal humerus fractures.¹⁴ TEA for acute fracture in older patients was popularised by Cobb and Morrey in the 1990s,¹⁵ and has now become one of the main indications for primary TEA. In the UK, the data from the NJR shows that this use of TEA has risen from 19.4% of all TEA cases in 2012 to 38.2% in 2020.¹⁶

When performed for trauma, the trend is to perform a linked TEA.¹⁶ This negates the need for either soft tissue or epicondyle healing for implant stability and no adverse effects of epicondyle excision have been shown regarding longevity or function following linked TEA.¹⁷ Unlinked TEA for trauma can be performed but requires repair of the collateral ligaments or epicondyles around the implant and is less commonly performed. Although the loosening rate of unlinked TEAs should theoretically be lower than linked TEA due to the reduced constraint, there is no clear evidence of superiority in the literature.^18,19

Between 2018 and 2020, 664 linked TEAs were performed in the UK for acute trauma, with over half being the Coonrad Morrey prosthesis (Zimmer-Biomet). Other commonly used prostheses include the Discovery (Lima), the Nexel (Zimmer-Biomet) and the Latitude (Stryker). In comparison, only 8 unlinked TEAs were implanted for acute trauma in the same time period.¹⁶

The outcomes of primary, linked TEA have been demonstrated to be superior to ORIF in osteoporotic, comminuted distal humerus fractures in both retrospective,¹⁵ and prospective comparative studies.¹³ TEA has also been shown to be more cost effective than ORIF in most cases of acute, intra-articular distal humerus fracture in the elderly.²⁰ At medium term follow up, the superior outcomes of TEA over ORIF is sustained with no requirement for late revision at a mean of 12.5 years follow up.²¹ It is however noteworthy that the initial ORIFs in these comparative studies were performed over 15 years ago, prior to the widespread use of current techniques and implants that optimise fixation.

Although mid-term results following TEA are good, TEA does introduce a new spectrum of potential complications, including polyethylene or bushing wear, aseptic loosening, osteolysis and periprosthetic fracture.¹⁵ The results of prior generations of TEA prostheses show that beyond a median of 10-years there are increasing rates of loosening, bushing wear and osteolysis which are potentially difficult and risky to treat in the elderly population.²² Furthermore, the lifelong activity restrictions placed upon linked TEAs to prevent early mechanical failure can be very limiting. This is particularly the case in patients with ‘normal’ elbows prior to their fracture. It can also be potentially disabling for those individuals reliant on their upper limbs for ambulation or transfers.²³

2.2. DHH

In recent years, there has been interest in DHH as an alternative to TEA for distal humeral fractures, as it provides the advantages of TEA whilst avoiding the complications associated with the linked articulation, ulna component and polyethylene bearing surface.²⁴ DHH involves replacement of the capitellum and trochlea with an articular shaped spool attached to a humeral stem. DHH may also be more suited to trauma patients as the required level of activity restrictions are not as strict, although it introduces a new spectrum of potential complications not seen with TEA. As with any hemiarthroplasty, some degree of wear of the opposing chondral surface is inevitable with time and because the articulation is unlinked with reliance on repair of the collateral ligaments or epicondyles, joint instability is more likely.

DHH is therefore relatively contraindicated in patients with pre-existing symptomatic arthritis and in those where primary ulnohumeral stability can't be achieved. Although DDH is indicated for salvage of the sequalae of non-operative management or failed fixation,²⁵ it is less reliable than linked TEA due to contracture and malposition of the collateral ligaments so should be used with caution.

Multiple historic implants have been used as a DHH in the past, however the Latitude EV (Stryker) is currently the only licensed DHH implant available worldwide. It is a modular, cemented implant with a flanged stem and articular “spool”, fixed together by a cannulated pin (Fig. 1). This cannulation through the centre of rotation allows for collateral ligament repair to the implant. If required, the prosthesis can be revised to a TEA without extraction of the stem.^16,25 The spool comes in multiple sizes to optimise matching to the proximal forearm.

Fig. 1.

Latitude elbow DDH implant (Stryker) showing the modular flanged stem and cannulated articular spool.

The UK NJR data demonstrates a year on year rise in the use of DHH for trauma.¹⁶ Despite an absence of long-term data, there have been good reported functional outcomes in the short and medium term.²⁶ In contrast to TEA, DHHs are predominantly used in the management of trauma, with 89% of UK cases in 2020 being performed for acute trauma.¹⁶

3. Pre-operative planning

3.1. Patient selection

Pre-operative assessment should include a thorough history including previous elbow injury, instability, contracture, previous elbow pathology or infection. History of systemic disease that could preclude surgery should be sought, as should hand dominance, employment status and level of pre-injury function. Examination should include a thorough inspection of the soft tissue envelope, and a detailed neurological examination, with close attention paid to ulna nerve function.

In addition to plain radiographs, multiplanar computerised tomography (CT) is useful for assessing the number, site and size of articular fragments, helping the surgeon decide whether arthroplasty is the most appropriate treatment. It is preferable to avoid an attempted fixation with subsequent intra-operative conversion to arthroplasty due to longer operating and anaesthetic times.

Elbow arthroplasty for the treatment of acute fracture should be considered a surgical option for patients if all the following criteria are met:

• 1.A fracture where stable reconstruction of the articular block is not reliably feasible
• 2.The patient can tolerate a significant surgical procedure
• 3.The patient will be compliant with post-operative rehabilitation regimes
• 4.The patient will benefit from reconstruction i.e. an active or independent individual
• 5.Primary soft tissue coverage and closure can be achieved

A typical fracture pattern suited to arthroplasty is a very distal fracture that primarily involve the articular block, typically with fracture lines in multiple planes creating multiple fragments, alongside poor bone stock (Modified Dubberley type 4) (Fig. 2). Low transcondylar fractures may also be an indication for arthroplasty if the bone stock in the distal fragments is poor (Fig. 3). Fractures that involve the columns tend to have less comminution in the articular segment and are often more amenable to stable fixation, even in older patients.

Fig. 2.

Computerised Tomography 3D Reconstruction of a Modified Dubberley Type 4 distal humerus articular fracture.

Fig. 3.

Pre-operative and intraoperative anteroposterior radiographs of a low transcondylar fracture. Note the significant bone loss once the fracture is disimpacted out to length.

Arthroplasty may also be considered in the setting of a potentially reconstructable fracture in the presence of pre-existing symptomatic arthritis. In addition to acute fracture treatment, arthroplasty can be used for salvage of failed ORIF or failed non-operative management, either non-union or malunion.

Open fracture is not an absolute contra-indication to treatment with an elbow arthroplasty. A multicentre study has shown that patients with a non-contaminated open fracture with a primarily closable wound had no increase in infection or complication rates when treated acutely with primary arthroplasty in comparison to a staged approach with debridement.²⁷

3.2. Implant selection

Once arthroplasty has been deemed the appropriate treatment, a decision regarding DHH or TEA is necessary. There is currently no comparative data to guide clinicians and an individualised decision must be made for each patient. The following however are prerequisites for consideration of DHH over TEA.:

• 1.No pre-existing symptomatic UHJ or RCJ osteoarthritis
• 2.No concurrent coronoid fracture which may result in post-operative instability
• 3.Intact or reconstructable medial and lateral columns
• 4.Intact or reconstructable collateral ligaments²⁸

In general, DHH should be considered in slightly younger patients whose activity levels are likely to lead to aseptic loosening of a TEA, or patients who would be unable to adhere to the restrictions imposed on a TEA. However, in our experience, the best results for both TEA and DHH are in older patients and any arthroplasty should be used with caution in younger patients due to the paucity of long-term data and the lifetime risk of revision surgery.

DHH has been described in the setting of a concurrent olecranon fracture,²⁹ using the fracture to access the distal humerus and via olecranon osteotomy, however outcome may be affected by the damage to the native articular surface.²⁵ TEA can also be performed with an olecranon fracture although it is challenging to reliably fix the olecranon and maintain a stable ulna component. The presence of a concurrent radial head fractures is a relative contraindication for DHH and a TEA may be preferred due to the risk of joint instability.

DHH should be performed using an implant that can be converted to a TEA intra-operatively, in case of intra-operative instability that precludes an unlinked prosthesis. In addition to the usual instrumentation required for elective elbow arthroplasty, plating and wiring systems should be available if required, especially if there are large epicondylar fragments extending into the humeral columns.²⁵

3.3. Site and surgeon selection

In the UK, the “Getting It Right First Time” (GIRFT) programme and the British Elbow and Shoulder Society (BESS) have developed a “hub and spoke” model because elbow arthroplasty is a relatively low volume procedure.^30,31 This patient care model is predicated by evidence of improved outcomes for higher volume surgeons,³² and the relatively small number of elbow arthroplasties performed by surgeons compared to other major joint arthroplasties.^16,33 This model may not be transferable to other health symptoms and countries with differing geography and population density; however it does focus attention on the importance of maintaining volume of arthroplasty and the relationship with outcome, innovation and education.

4. Operative technique

4.1. Peri-operative care and positioning

Pre-operative prophylactic antibiotics are recommended prior to arthroplasty as per local protocols, to reduce the risk of periprosthetic infection. Intravenous tranexamic acid is also recommended in the absence of any contraindications.

The authors prefer a lateral decubitus position with the operating arm positioned over a bar or a padded, unsterile arm holder (Fig. 4) to allow for easy access to both ulna and radial sides of the elbow joint. The supine position is also commonly described. All pressure areas should be well padded and supported, with the contralateral arm readily accessible for anaesthetic support. Fluoroscopy is recommended for DHH to examine for instability and for all arthroplasty where epicondyle reconstruction is required. A sterile tourniquet is used however it is reasonable to perform surgery without tourniquet.

Fig. 4.

Set up for TEA or DHH, with the patient in a lateral decubitus position with the arm supported by an unsterile arm holder. Fluoroscopy can then enter from the contralateral side of the table to the surgeons, AP and lateral views can be obtained through the rotation of the shoulder or c-arm. Placement of the anaesthetic machine at the foot end allows for the positioning of the scrub team in close proximity to the surgical team.

4.2. Surgical approach

Approaches which violate the extensor mechanism are discouraged, whether this be for TEA or DHH. This includes olecranon osteotomy or triceps reflecting approaches. In the setting of DHH, osteotomy fixation may compromise any conversion to TEA either intraoperatively or in the future. It also increases the reoperation rate due to wound complications and metalwork prominence and may accelerate ulna sided ulnohumeral joint wear.^28,34 Furthermore, access to the elbow joint is easier in the setting of acute trauma as removal of the articular fragments provides greater visualisation of both the humerus and ulna, precluding the need to disrupt the extensor mechanism.

Our preferred approach is the lateral para-olecranon approach. A medial para-tricipital window is also required to access the medial joint.^35,36 We also utilise this same approach for the fixation of most intra-articular distal humerus fractures.

A posterior incision avoiding the tip of the olecranon is made with full thickness fasciocutaneous flaps raised. Care is taken to only raise flaps sufficient to access the deep structures. A posteromedial incision may be beneficial to decrease the size of the medial flap which is more vulnerable to flap necrosis.

Our preference is to transpose the ulna nerve anteriorly at this stage, although the nerve can be decompressed along its length and left in situ. The fat flap to secure the subcutaneous transposition is raised and secured prior to wound closure at the end of the procedure. The medial para-tricipital window is created to give access to the medial side of the joint, with release of the posterior capsule and posterior band of the MCL with preservation of the anterior band of the MCL, whether intact or attached to a bony fragment. The lateral para-olecranon window is created by incising the fascia over the anconeus 1 cm lateral to the subcutaneous border of the ulna. This fascial split is extended proximally into the triceps which is split down to the humerus. Distally, the anconeus is subperiosteally elevated off the lateral aspect of the proximal ulna. The posterolateral capsule is incised to visualise the radio-capitellar joint proximal to the insertion of the LCL onto the supinator crest.

The articular fragments are excised and kept for use as bone graft behind the anterior flange and to augment condylar fixation if necessary. The anterior band of the MCL and the LCL should be preserved either to their intact distal humeral attachments or their epicondylar fracture fragments for later repair if an unlinked TER or DHH is being performed. The anterior capsule is carefully released off the humerus and the distal humerus is delivered through the triceps split. (Fig. 5).

Fig. 5.

Intra-operative photograph showing the lateral para-olecranon window with trial DHH in situ.

4.3. Implant positioning

Anatomical implant positioning is required to ensure joint stability and to reduce wear. Optimal implant positioning is largely a function of implant rotation, implant length and implant sizing. The roof of the olecranon fossa is the landmark for safe entry into the humeral canal. Reamers and broaches should be directed anteriorly from this starting point to ensure satisfactory sagittal plane alignment.

4.3.1. Implant rotation

Anatomic implant rotation is key to avoid asymmetric forces across the implant which can lead to instability, premature wear or loosening. In the elective setting the native centre of rotation is established using the epicondylar axis. This reference is lost in the setting of acute fracture. The posterior humeral line is the axis of the flat bony surface proximal to the olecranon fossa, is intact in most fractures and can be used as a reference. The axis of rotation is on average 14° internally rotated to this line (Fig. 6).³⁷ Humeral rasps and implant should be implanted with attention to this rotation.

Fig. 6.

The axis of rotation is on average 14° internally rotated to the posterior humeral line proximal to the olecranon fossa.

In TEA, care must also be taken to implant the ulna component in the correct rotation to avoid the same complications as with the humerus. In TEAs without a radial head component, the flat spot of the ulna has been shown to be a useful anatomic landmark to guide ulna component rotation (Fig. 7), as it is nearly perpendicular to the guiding ridge of the sigmoid notch.³⁸

Fig. 7.

The use of the ulna flat spot to guide ulna component rotation.

In TEAs with a radial head component, optimising radial head tracking must also be considered to avoid point loading of the radial head if it is retained or replaced. The axis between proximal ulna and proximal radius can therefore be used to set the rotation of the ulna component. It is noteworthy that there is slight discrepancy between these 2 axes of rotation for the ulna component (Fig. 8), although there is no current comparative data.

Fig. 8.

The discrepancy between rotational axes between the flat spot axis (pink) and the proximal radioulnar axis (green).

4.3.2. Implant length

Correct implant height is crucial to restore range of motion, reduce contact forces in DDH and reduce early loosening in TEA. As with implant rotation, loss of the native epicondylar axis in the fracture setting makes referencing implant length difficult. The origin of the MCL can be used to check appropriate implant length, either by comparison to intact origin or by reducing the medial epicondylar fragment to the intact humerus/implant (Fig. 9). This can be done with both the trial and definitive implant and should be performed as a combination of direct visualisation and intra-operative fluoroscopy. With the Latitude implant, the position of the spool cannulation corresponds to the implant axis of rotation.

Fig. 9.

Radiograph showing a with the correct length referenced from the medial epicondyle. Despite some bone resorption at the epicondyle fixation sites the patient had full pain free range of motion.

In addition to the use of the MCL origin, intra-operative trialling will help guide the surgeon as to appropriate implant length, as can the position of the implant relative to the roof of the olecranon fossa, if intact.

4.3.3. Implant sizing

In TEA, implant sizing is guided by the bony rasps, ensuring an appropriate size of implant and cement mantle relative to the native bone. In DHH, sizing is determined by matching the anatomic spool to the native proximal radius and ulna. This is both in terms of the radius of curvature of the greater sigmoid notch and also the width of the implant. Care should be taken to avoid radiocapitellar overhang in particular, which can lead to point loading, impingement or failure of the lateral epicondyle repair. Intra-operative trialling will ensure the appropriate choice of implant. It is suggested in cadaveric data that if between sizes for DHH one should chose the larger size to reduce point loading of the native joint.³⁹

4.4. Epicondyle repair

As DHH is an unlinked articulation, repair of the collateral ligaments is essential to avoid instability. In some fractures, the anterior band of the MCL and medial epicondyle may be intact. If possible, this should be preserved throughout the procedure, and if released intra-operatively, should be repaired after prosthesis implantation. In the setting of distal humeral fractures requiring arthroplasty, there are often epicondyle avulsion fractures. Care should be taken to preserve these for later repair, particularly when excising the articular fragments. Repair of the ligaments/epicondyles is not essential in a linked TEA, although it may be beneficial for proprioception.

Our technique involves multiple high strength sutures for each epicondyle (Fig. 10). The epicondyle fragments with their attached ligaments are carefully dissected from overlying soft tissues. Whip stitches are passed along both collateral ligaments, starting and ending at the humeral origins. These sutures are then passed through the cannulated spool, and then the opposite epicondyle or soft tissues as a mattress suture (Fig. 10 – A/B/C). The epicondyles are then reduced to the implant using reduction forceps and the sutures are tied (Fig. 10- D). Each epicondyle fixation is then augmented with a tension band suture through small drill holes in the humeral columns (Fig. 10 – E) and cerclage sutures through the central aperture of the humeral stem (Fig. 10 – F).

Fig. 10.

Illustration of epicondyle repair to a DHH.

Sometimes there are larger epicondyle/metaphysis fragments extending into the columns of the humerus. These fragments can be fixed with Kirschner wires into the cement mantle (Fig. 11) or with plates, particularly if they are large fragments involving the meta-diaphyseal cortex.

Fig. 11.

Radiograph of DHH with medial epicondyle fixed using Kirschner wire into posterior cement mantle.

4.5. Wound closure and post-operative care

If used, the tourniquet is deflated to ensure haemostasis and avoid post-operative haematoma and wound breakdown. Closure in layers is performed to create a watertight seal around the implant. Prior to the closure of the lateral para-olecranon window, the senior author's current practice is to place 1g of Vancomycin powder beneath the triceps. Use of intra-wound antibiotics still requires further investigation and should be used according to local protocols. Following skin closure the elbow may be splinted in extension for 7–10 days if the soft tissue quality is poor, however in most instances a simple wool and crepe bandage is used. Elevation for the first few days is recommended, with encouragement of hand exercises to manage swelling. It may be preferable to avoid extension splinting for DHH due to the risk of instability. A polysling is given for comfort. Routine prophylaxis for heterotopic ossification is not used.

5. Rehabilitation

Once the wound has healed, early rehabilitation of DHH focuses on maintaining movement whilst protecting the ligament repairs. A supine overhead rehabilitation protocol is used which increases joint compressive forces and dynamic elbow stability, allowing early active motion whilst avoiding varus or valgus torque on the joint.⁴⁰ We recommend avoidance of shoulder abduction of more than 45° for 6 weeks to reduce varus torque, alongside concurrent shoulder exercises to prevent capsulitis. For linked TER unrestricted motion can be commenced once the wound has healed, particularly if the triceps insertion has been preserved in the surgical approach. No splinting or braces are used routinely.

After DHH, advice regarding excessive activity and loading of the joint is given to patients in order for them to understand the importance of maximising longevity of the prosthesis, however specific weight restrictions are not given in our practice.⁴¹ Following linked TEA, some authors recommend avoiding single event lifting over 5 kg, and repetitive lifting of any object over 1 kg. Our practice is to educate patients on the rationale for avoidance of heavy activity rather than place stringent weight limits.

6. Outcomes and complications

6.1. TEA

The UK NJR data shows that TEA for acute trauma has a 6% 8-year revision rate, with revision procedures almost exclusively due to aseptic loosening or infection.¹⁶ Data from the Swedish Joint Registry reports 90% survival at 10-years.⁴² Meta-analysis of case series calculates a revision rate of 7% at an average follow up of 40 months.⁴³

TEA for acute trauma has demonstrated good short and long-term outcomes, with the literature describing good to excellent results in 85–90% of patients.^13,21,44 Weighted mean Mayo Elbow Performance Score (MEPS) after TEA has been reported as 88 in recent meta-analysis.⁴³ A score of 75–89 is interpreted as ‘good’ elbow function.

Residual extension deficits have been reported at 10–33° after TEA for acute fracture.⁴⁵ A mean flexion arc of 102° and prosupination arc of 145° were reported in recent review.⁴⁴

The complication rate after TEA for acute fracture is 22%, including a 4.7% rate of ulna nerve symptoms, a 4.7% rate of periprosthetic fracture, a 2.7% rate of heterotopic ossification and 2.3% rate of loosening, at an average follow up of 49 months.⁴⁴

6.2. DHH

There is less long-term data regarding DHH, with largely small series with less than 20 patients. The largest series in the literature includes 42 patients with a mean follow up of 34 months.⁴⁶ Recent meta-analysis of published studies has found a revision rate of 4% (range 0–15%), with a mean follow up of 49 months.⁴⁷ All revisions were to TEA, for either loosening (3.6%) or periprosthetic fracture (0.4%). UK NJR data reports an 8% revision rate for DHH at 2 years, with revision predominantly attributed to instability or infection.¹⁶

The data for patient recorded outcome measures (PROMs) also comes from small series, also most commonly in the form of the MEPS. The mean MEPS reported in the literature is 83.6, with a range from 50 to 100, interpreted as ‘good’ elbow function.⁴⁷

A mean flexion/extension arc of 107° and prosupination arc of 153 is reported after DHH in recent meta-analysis²⁶ and is within the range required for functional daily activities.⁴⁸

DHH has a complication rate reported as high as 63%.⁴⁷ However the majority of these were minor complications not requiring intervention, including asymptomatic heterotopic ossification, radiographic ulnohumeral wear and transient ulnar nerve irritation. When solely radiological complications are excluded, the complication rate is around 22%.⁴³ The incidence of major complications, including those requiring surgical intervention is 11%,⁴⁷ with the most common being symptomatic loosening requiring revision, intraoperative fracture and permanent ulna nerve palsy.

A recent systematic review reported that PROMs and range of motion were superior following DHH compared to TEA for acute fracture with similar complication rates. This was however deemed a low level of evidence due to the nature of the studies included and the patient heterogeneity.⁴³ Rates of aseptic loosening of TEAs are found to be higher in meta-analysis (3.8% vs 1.4%) which may be indicative of the reduced constraint of DHH but also the longer term follow up available for TEA. The preliminary results of a randomised controlled trial carried out in Sweden⁴⁹ has shown no statistically significant difference in PROMS or range of motion at 2 years between DHH and TEA.⁵⁰

7. Conclusion

Intra-articular distal humeral fractures in the elderly remain difficult injuries to treat. Although ORIF with modern locking plate fixation remains the gold standard, the patient demographic means some fractures are beyond the scope of reliable fixation methods. Further long-term outcome and comparative data regarding arthroplasty is required, but short to medium term outcomes demonstrate that both DHH and TEA are valuable options for select patients, although attention to technique is required to minimise potential complications.

Author statement

Andrew Stone: Methodology, Writing – Review & Editing, Visualisation, Project Administration Chan G: Writing – Original Draft Sinclair L: Investigation Phadnis J: Conceptualisation, Writing – Review and Editing, Supervision.

Informed consent (Patient/Guardian)

N/a (review article).

Institutional ethical committee approval

N/a (review article).

Funding/sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Declaration of competing interest

None.