Management of medial humeral epicondyle fractures in children: a structured review protocol for a systematic review of the literature and identification of a core outcome set using a Delphi survey

Medial humeral epicondyle fractures are one of the most common injuries in childhood and often require surgery [1, 2]. Long-term sequelae may include non-union, mal-union, instability, nerve injury, growth arrest, deformity and loss of function [1, 3]. There are currently no standardised outcome measures to assess progress after an elbow injury in a child. Wide variation in currently reported outcomes (radiographic-, surgeon- and patient-reported) makes comparison of the treatment of elbow fractures in children difficult.

In particular, the management of medial epicondyle fractures in children is controversial [1]. National Institute for Health Research (NIHR) Health Technology Assessment (HTA) have made children’s orthopaedic surgery a priority area and are considering a clinical trial in the management of medial epicondyle fractures in children [4]. A major barrier to these trials is the lack of a clear outcome, i.e. pain, function, fracture union, cosmesis, quality of life.

An emerging strategy to overcome these issues has been the development of core outcome sets (COSs). COSs represent the minimum required dataset for all randomised control trials (RCT) of a certain condition [5‐14].

The use of COSs aim to reduce heterogeneity in research allowing easy comparison of studies, improve accuracy of data interpretation and reduce outcome reporting bias. Standardisation of outcomes will also lead to a reduction in omissions and increased statistical power for meta-analysis [15].

In recent years a small number of COSs have been developed for use in adult orthopaedics; however, these are limited to specific conditions (hip fracture [16], back pain [17] and total joint replacement [18]). There are currently no COSs available for use in clinical trials across the whole of paediatric orthopaedics.

A list of all potential outcomes will be identified from the systematic review as described above. Outcomes will be listed both individually and by domain to aid interpretation. All outcome domains and included outcomes will be reviewed by the Study Steering Group (SSG) to assess suitability of domain name and grouping. The SSG will consist of authors SD, JJK and DP.

To identify the outcomes of importance to all stakeholder groups a Delphi approach will be used. An overview of the COS developmental process is shown in Fig. 1. This will enable participants to provide anonymous opinions with equal influence given to all participants. This method also avoids individual participants’ responses being influenced by the opinion of their peers. Key stakeholder groups will consist of children, parents (Delphi stream 1) and clinicians (Delphi stream 2).

It is essential that consideration is given to the opinions of parents and children regarding the treatment of medial humeral epicondyle fractures. They should be given the opportunity to identify the most important outcomes and domains based own their own experiences and beliefs.

The opinions of children are key as they represent the group upon which the short- and long-term benefits and adverse effects of treatment will have the largest impact.

Medial epicondyle injuries typically affect older children and, as such, it is unlikely that our cohort will include any young children due to the nature of the injury. A recent national audit of medial epicondyle injuries indicated that 7-year-olds were the youngest children presenting with this injury pattern. Children over the age of 7 years with a previous medial humeral epicondyle fracture will be identified from a UK tertiary paediatric centre and invited to interview. A minimum of 10 children will be recruited with no upper limit.

Eligibility will be based on history of an acute medial humeral epicondyle fracture treated non-operatively or operatively and the ability to complete the interview in English without the need for a translator. Eligible children will be invited to attend a one-off interview structured by the use of a questionnaire. The authors feel this will improve compliance and accurate completion of the questionnaire. This will not be audio-recorded. It will consist of a researcher reading out the questions and recording the answers. The survey questionnaire (Delphi stream 1) has been designed to meet the developmental needs of a broad array of children using a traffic-light system for grading of outcome importance (Green ‘not important’, Amber ‘important but not critical’ and Red ‘critical importance’. The traffic-light scoring system has been approved by a Young Person Advisory Group (YPAG).

Children will also have the opportunity to add any additional outcome they feel may have been missed. There will be no formal qualitative interview or qualitative analysis.

A separate sample of 20 parents will be identified from an existing trauma database at a UK tertiary paediatric centre. Parents will be invited to complete an online Delphi questionnaire (Delphi stream 1). Parents will complete all rounds of the Delphi process.

A plain English explanation for all outcome measures listed will be included in the stream-1 questionnaire for parents and children. Language will be approved by the YPAG.

Consultation with the HRA deemed this study a service evaluation project with no requirement for ethical approval (reference 60/89/81). Informed consent will be assumed if participants agree to fill in the surveys.

There is currently no published COS for medial humeral epicondyle fracture in children. The development of COSs in this clinical area will improve the quality of future studies reducing bias, allowing easier comparison and enhance opportunities for a larger meta-analysis. It is anticipated that this COS will form part of the feasibility to an NIHR HTA-funded trial concerning the management of medial humeral epicondyle fractures in children.