|Year : 2018 | Volume
| Issue : 2 | Page : 37-42
Developing a consensus on indications, timing and core outcomes for surgical fixation of rib fracture: a Delphi technique protocol
Helen MA Ingoe MRCSEd 1, Catriona McDaid2, William Eardley1, Catherine Hewitt2
1 York Trials Unit, Department of Health Sciences, ARRC Building, University of York, Heslington, York; The James Cook University Hospital, Middlesbrough, UK
2 York Trials Unit, Department of Health Sciences, ARRC Building, University of York, Heslington, York, UK
|Date of Web Publication||18-Jun-2018|
Helen MA Ingoe
York Trials Unit, Department of Health Sciences, ARRC Building, University of York, Heslington, York; The James Cook University Hospital, Middlesbrough
Source of Support: This study was supported by the Educational grant Orthopaedic Research UK. ORUK UK Registered Charity, No. 1111657., Conflict of Interest: None
Background and objectives: Developing a robust surgical trial requires evidenced based indications and selection of relevant outcomes. The aim of this study is to develop a consensus on indications and timing and a core outcome set for surgical fixation of rib fractures following blunt chest trauma.
Design: A systematic review of the literature and identification of a core outcome set using a Delphi survey.
Methods: Systematic review has identified multiple outcomes, indications and timing of surgery following rib fracture fixation. Using the Core Outcomes Measures in Effectiveness Trials (COMET) methodology, three expert panels including clinicians, allied health professionals and patients will be invited to participate in an online questionnaire. Consensus will be achieved after three rounds using a rating scale of 1 (not important) to 9 (critically important). Those responses unable to gain consensus will be dropped through subsequent rounds. All three expert panels will take part in the development of a core outcome set. Clinicians will run a concurrent consensus on timing and indications for surgery.
Discussion: A consensus on the indications, timing and outcomes that should be measured in a surgical rib fracture fixation trial is required to address in ability to compare multiple studies for evidence synthesis.
Ethics and dissemination: Ethics approval has been obtained from the University of York Health Science Research Governance Committee, Health Research Authority (235596) and North West Research Ethics Committee (18/WM/0018). Systematic review is complete, and the consensus rounds will take place between February and June 2018. Protocol version: 2.0. The results of the study will be disseminated in a peer-reviewed journal.
Trial registration: This protocol was registered on the COMET website http://www.comet-initiative.org/studies/details/1104.
Keywords: rib fracture; flail chest; surgical fixation; outcomes; indications; Delphi consensus; protocol
|How to cite this article:|
Ingoe HM, McDaid C, Eardley W, Hewitt C. Developing a consensus on indications, timing and core outcomes for surgical fixation of rib fracture: a Delphi technique protocol. Clin Trials Orthop Disord 2018;3:37-42
|How to cite this URL:|
Ingoe HM, McDaid C, Eardley W, Hewitt C. Developing a consensus on indications, timing and core outcomes for surgical fixation of rib fracture: a Delphi technique protocol. Clin Trials Orthop Disord [serial online] 2018 [cited 2019 Sep 23];3:37-42. Available from: http://www.clinicalto.com/text.asp?2018/3/2/37/233622
| Introduction|| |
Rib fractures are common and can be a consequence of high energy trauma as well as simple falls. Most often patients are treated expectantly with analgesia and physiotherapy but on occasion deteriorating patients could benefit from surgical rib fracture fixation.
Surgical intervention with internal plates or struts restores the mechanical stability of the rib cage and the ability to ventilate effectively. Stabilizing the rib segments is thought to reduce pain and increase the likelihood patients will breathe spontaneously on their own without needing support.
A systematic review of reviews conducted by the research team showed inaccurate meta-analysis data and poorly conducted systematic reviews. A further systematic review of primary research was conducted in response to the low quality data synthesis and found that evidence was lacking due to several reasons: (1) multiple outcomes that could not be compared, (2) multiple indications for surgery that were not comparable and (3) timing of surgery was not consistent or comparable. Both manuscripts are awaiting publication.
A clear and measurable consensus of indications for surgery is required prior to undertaking a trial in order to inform eligibility criteria. Suggested indications include intractable pain that precludes adequate secretion clearance as well as acute respiratory distress or chest wall deformity. A radiographic rib fracture score has been developed as a predictor of respiratory failure fixation and predictors of poor outcome (flail chest, more than four rib fractures and the elderly) have been identified in numerous studies.
Timing of intervention
It is also important to identify the most appropriate time to undertake surgical fixation of a rib fracture. It is unknown whether waiting until the failure of mechanical weaning as an indication for surgical intervention exposes patients to increased morbidity related to mechanical ventilation. Uncertainty within the evidence base has highlighted the need for a consensus on the indications and timing of surgery. As standard timings are not used within trials, it is difficult to synthesize primary evidence that is not comparable.
Randomized controlled trials and cohort studies have presented a multitude of outcomes, suggesting a lack of consensus on which outcome is our best marker for treatment effectiveness. This presents difficulties in pooling outcomes in meta-analysis and developing a consensus on outcome measurement, a specific priority in rib fracture research.
To develop a consensus on:
- A Core Outcome Set (COS) for patients undergoing rib fracture fixation following blunt chest trauma;
- Indications and timing for surgical fixation of rib fractures following blunt chest trauma.
| Methods/Design|| |
This protocol is registered on the COMET database of developing outcomes (http://www.comet-initiative.org/studies/details/1104) and is using the methods and reporting standards described in the COMET Handbook.
Determining what to measure?
Identification of indication, timing of surgery and outcomes was gathered by way of systematic review. The review protocol was registered on the PROSPERO database (registration number: 4201605349) and is available from: https://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD4201605349.
The systematic review was completed based on the following PICO (Patient, Intervention, Comparator and Outcome) research question with the last search date completed in March 2017 [Table 1].
|Table 1: PICO (Population, Intervention, Comparator, Outcomes) research questiones|
Click here to view
Two researchers independently screened titles and abstracts for eligibility and were arbitrated by a third researcher.
Data extraction was completed by one researcher and checked 50% by a second researcher.
Timing and indications
The indications were extracted from primary studies and taken directly from the report. A distinct separation has to be made between indications for surgery and inclusion criteria for the study as these are considered two separate entities. For example, authors advocate fixation in all patients who have flail chest and require intubation. However, this study may only include those who also sustained a pulmonary contusion. Although this is the inclusion criteria, pulmonary contusion is not explicitly an indication for surgery. The same reasoning is applied to timing of surgery.
The list of indications from the primary studies was simplified into a concise list on extraction. The simplified list distinguishes primarily by injury definition of flail chest and multiple rib fractures and secondarily with other conditions, examples are, chest deformity, intubation need, or failure to wean.
Timings of surgery were extracted from trials or studies who outlined their rationale for timing of surgery. These were taken verbatim and then transformed into useable statements. GRADE levels of evidence were not ascribed to any statement.
Outcome measures were extracted and grouped into outcome domains. The five outcome domains included adverse events, death, physiological or clinical, life impact and resource and are described by Williamson and Clarke. Extraction of outcomes was done by a single researcher and placed within the sub-domains. Although ideally using verbatim outcomes straight from the primary study followed by allocation of these outcomes to domains would have been preferable, this was not completed. Outcomes were grouped into domains immediately from extraction to make the list of extracted outcomes easier to manage.
Consensus exercise method
Three stakeholder groups will be involved in the Delphi consensus process:
- Medical panel
- Allied health care
- Patient and public panel
Each participant within each panel is considered an expert. All expert panels will form the COS consensus, and a further consensus, but delivered concurrently, will include separate medical panel to encompass indications and timing of surgery ([Figure 1]: Composition of Delphi consensus panels).
Clinicians who undertake the care of patients with rib fractures and are part of the multidisciplinary team who decide on rib fracture fixation will form the clinical expert panel. This may include surgeons who perform rib fracture surgery, general surgeons who often undertake day to day care of rib fracture patients in district general hospitals or intensive care physicians who provide ventilator support and acute pain management. It may also include rehabilitation specialists. This panel will aim to form a consensus on indications for surgery, timing of surgery and a core outcome set and will be drawn from the international community.
Allied healthcare panel
This panel is to include allied health professionals that specialize in trauma rehabilitation or chest physiotherapy. This may include physiotherapists, occupational therapist or specialist trauma or intensive care nurses. Each must be registered with a professional body to participate. This panel will participate in the consensus on core outcomes and will be recruited internationally.
Patient and public panel
Patients and carers of patients with knowledge of the condition will be invited to take part. The patient is considered an expert if they have suffered flail chest or multiple rib fractures or have had rib fracture fixation. Carers and family members will also be eligible to participate if they have cared or supported a person with multiple rib fractures, flail chest or had rib fracture fixation. This panel will participate in the consensus on core outcomes and will be recruited from the UK only.
To form a representative expert panel, we seek to include a wide range of allied health professionals and patients from around the UK. The medical panel will seek to include international participants. To take part, the individual must have access to the internet and possess an email address.
Panels will be contacted in the following ways:
- Thoracic Surgery Society, British Orthopaedic Association, Orthopaedic Trauma Society and the Royal College of Anaesthetists will be contacted. We will ask the societies and associations if they have relevant clinicians in their societies who undertake the care of rib fracture patients who may be interested to take part.
- Major Trauma Centres in the UK will be contacted to identify those who are undertaking rib fracture care.
- As part of previous work, a systematic review of literature has been undertaken. Authors’ contact details have been collected from all included studies within the review as possible candidates for an expert panel.
- To increase participation and to keep the sample representative, the medical panel that declined to take part will be asked to suggest a person with similar knowledge or expertise who may take part in their place as well as an allied health professional.
Allied Healthcare Professionals Panel (AHP)
- The Chartered Society for Physiotherapy and the Council for Allied Health Professions Research publish contacts for a number of network research groups. Relevant AHP researches from this directory will be contacted via the website.
- The list of authors derived from recent publications on rib fracture surgery may draw on some allied health professionals who have published their work.
- As stated above, clinicians will be asked to suggest relevant allied health professionals to take part.
Patients and public panel
Patients from relevant public and patient involvement groups and patients identified by clinicians will be invited to take part. To prevent selection bias of participants and make the group representative a one year sample of all admitted chest injuries from James Cook University Hospital will be contacted. Recruitment will also involve posts on trauma support websites (example, After Trauma) and charities (example, ICU Steps) with prior permission from the organization. Contact details of the research contact will be provided and the participants will need to contact the research team if they wish to participate.
All identified participants will be contacted via email shown in Additional file 1 [Additional file 1].
To be representative, we hope to recruit at least ten patients and allied health professionals. Over 200 participants have been identified from the systematic review as relevant corresponding authors for the medical panel; it is hoped half would participate with the consensus exercise. No advice has been given on the size of each panel, the consensus is not bounded by statistical power and attrition could be considerable through the subsequent rounds. Patients and allied health professionals views on outcomes are important to the consensus process and therefore as many as possible will be recruited.
As an incentive to undertake consensus development, medical and allied health professional participants will be offered a certificate of participation if they complete all three rounds. The above conditions are set strictly so that attrition and the subsequent bias is minimized.
A maximum three round Delphi consensus will be undertaken.There will be at least one round of feedback. If consensus is not agreed after the first feedback, a second round of feedback will be undertaken. If consensus is not agreed after the third questionnaire round, no further feedback will be given and the consensus process will end. Any outcomes without a concensus will be elimintaed from the COS.
Invitations describing the study will be sent via email and will include a link to the first consensus round (Additional file 1). A participant information sheet will be attached to the invitation (Additional file 2 [Additional file 2]).
Qualtrics software licensed by the University of York will be used to administer all rounds of the Delphi questionnaire via email (Additional file 3[Additional file 3] ).
The survey will remain open for four weeks to maximize the intake of participants. Three reminder emails will be sent at weekly intervals. An unsubscribe box will be available within the email if people do not wish to receive any further communication. Consent will be taken within the questionnaire in which participants will need to tick a box to acknowledge the conditions of undertaking the study. It is unlikely that the surveys undertaken for the COS will take no more than 15 minutes to fill out. It is expected the medical panel may take up to 25 minutes.
Additional open questions
The first part of the survey will ask participants an open question. This allows them to list what outcomes they feel should be measured. If identified without prompting, it is more likely to be important to that individual and may prevent researcher bias in the study.
The structure of the items in the COS consensus exercise will be ordered as per the framework by Williamson and Clarke., The outcomes listed under each heading will be listed in alphabetical order to prevent imbalance of questions or bias in outcome list order. Each outcome will be associated with an explanation of the outcome and will be written in lay terms at a reading age of 8 years old.
A further free comment box will be added at the end of the survey to allow participants to recommend any further additional outcomes/indications that were not included within the initial list.
The second part of the survey will comprise of a list of statements on outcomes and for the medical panel only a list of statements on indications and timing of surgery are listed.
Participants will be asked to rate the relevance of each statement to be included within the minimum dataset or final consensus statement. The rating scale will be 1–9, 1 being not important and 9 being critically important.
Retaining or dropping items between rounds
Outcomes and statements will be retained and entered into the next round if the groups average of all three panels’ scores 50% or more between 7–9 and less than 15% score 1–3. Outcomes and statements not meeting these criteria will be dropped from the consensus process. The three panels score will be grouped together for ease of analysis.
Feedback will be delivered as a percentage of respondents from each stakeholder panel to each rating. All ratings 1 to 9 will be associated with a percentage score from each of the stakeholder panels. All stakeholder panels will see the percentage rating for all other panels. Personal feedback of their previous ratings will not be provided.
Round two and three
A second round will be undertaken with only the respondents from the first round. Suggestions from the first round of additional outcomes will be included within this second survey, but not duplicated, if felt this outcome or indication has been covered in the first round. If any comments pertain to combining or expanding outcomes or indications due to similarity or narrowness, this will also considered if there are significant comments to support this (Additional file 4 [Additional file 4]).
New statement scoring
The new statements identified from the first round will then be submitted into the second round for rating of 1–9. Retaining and dropping of outcomes will be reached as per the same guidance in round one with the new outcome first being scored 50%/15% in the second round and then 70%/15% in the third round.
- A consensus will be reached if more than 70% of the whole participant group consider a statement highly relevant (7–9) AND less than 15% scored not relevant (1–3).
- A consensus will also be reached if more than 70% of the whole participant group say that the statement is not relevant (1–3) AND less than 15% say the statement is highly relevant.
All statements that do not satisfy these criteria are considered not to have consensus and therefore will be dropped from the COS.
The consensus process will cease at this point. No further consensus conference, stakeholder meeting or face to face interviews will take place. The full overall processes of the consensus rounds are shown in [Figure 2].
The level of consensus
A brief assessment will be made on the degree of consensus in each round by looking at the spread of group scoring (with histograms) and the change in group scoring between rounds.
Emphasis will be made during communications with participants that the reliability of the results may be compromised if participants drop out and even if participant’s opinions are in the minority, they should continue to take part as the consensus could be overestimated if they drop out.
Ethics approval has been obtained from the University of York Health Science Research Governance Committee. Health Research Authority (235596) and North West Research Ethics Committee (18/WM/0018) granted approvals on 11 January 2018. Further R&D approvals were granted from James Cook University Hospitals for access to patient addresses on 26 January 2018.
All data will be anonymized and stored in accordance with guidelines of the York Trials Unit.
Any identifiable data we collected about an NHS patient (home address) will be stored securely at James Cook University Hospital and only accessed by the chief investigator. All other participant information including email address supplied to the study team will be kept at the University of York. Only members of the regulatory authorities, research team (University of York) will have access to this data. Participants’ personal details such as email address will not be provided to anyone else, or used for any other purpose.
The completed consensus work will contribute towards an M.D. thesis. It is intended that the consensus work will be written as a report and disseminate to those who undertook the consensus but also a journal article that will be published in a peer reviewed trauma specialty journal. The report will be structured as recommended by the guidance by the COMET group for the Delphi method with the checklist developed by Sinha et al. and Core Outcome Score with the Core Outcome Score -STAndards for Reporting (COS-STAR) checklist.
| Discussion|| |
A consensus on the indications, timing and outcomes that should be measured in a surgical rib fracture fixation trial is required to address in ability to compare multiple studies for evidence synthesis. These include (1) multiple outcomes measures, (2) multiple indications for surgery and (3) timing of surgery. In preparation for the consensus rounds information was gathered by systematic review. Limitations of this review include the potential for a small sample size and attrition after each round. Invited are 230 clinicians, 60 patients and 18 allied health professionals.
| Trial Status|| |
The Delphi study is currently ongoing. The first Delphi questionnaire has been delivered; it is anticipated that the second Delphi questionnaire will be sent to all panelists who agreed to participate in the second round in March 2018. A published guideline on the methods for selecting outcome measurement instruments to be included in a COS is expected by the end of August 2018.
Additional file 1: Patient recruitment leaflet and invitation email.
Additional file 2: Participant information sheet.
Additional file 3: Model consent and questionnaire.
Additional file 4: Second round invitation email.
HI is the principle investigator. HI and CMcD are responsible for the search of the literature review. CH and WE have contributed to the design of the study and the content of the study protocol with important intellectual revisions. HI is responsible for drafting the protocol manuscript. All authors have read and approved the final protocol manuscript.
Conflicts of interest
CMcD and CH have received funding from the British Orthopaedic Association outside the scope of this project. CMcD is a member of the NIHR HTA & EME Editorial Board. None of above influences the work completed in this study.
Educational grant Orthopaedic Research UK. ORUK UK Registered Charity, No. 1111657.
Institutional review board statement
Ethics approval has been obtained from the University of York Health Science Research Governance Committee, Health Research Authority (235596) and North West Research Ethics Committee (18/WM/0018).
Declaration of participant consent
The authors certify that they have obtained all appropriate participant consent forms. In the form, the participants have given their consent for their images and other clinical information to be reported in the journal. The participants understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
This study follows the Consolidated Standards of Reporting Trials (CONSORT) statement and Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidance for protocol reporting.
Copyright transfer agreement
The Copyright License Agreement has been signed by all authors before publication.
Data sharing statement
Individual participant data that underlie the results reported in this article, after deidentification will be available beginning immediately and ending 36 months following article publication to researchers who provide a methodologically sound proposal. Proposals should be directed to email@example.com.
Checked twice by iThenticate.
Externally peer reviewed.
Funding: This study was supported by the Educational grant Orthopaedic Research UK. ORUK UK Registered Charity, No. 1111657.
| References|| |
Cataneo AJ, Cataneo DC, de Oliveira FH, Arruda KA, El Dib R, de Oliveira Carvalho PE. Surgical versus nonsurgical interventions for flail chest. Cochrane Database Syst Rev
Caragounis E-C, Fagevik Olsén M, Pazooki D, Granhed H. Surgical treatment of multiple rib fractures and flail chest in trauma: a one-year follow-up study. World J Emerg Surg
Granetzny A, Abd El-Aal M, Emam E, Shalaby A, Boseila A. Surgical versus conservative treatment of flail chest. Evaluation of the pulmonary status. Interact Cardiovasc Thorac Surg
Chapman BC, Herbert B, Rodil M, et al. RibScore: A novel radiographic score based on fracture pattern that predicts pneumonia, respiratory failure, and tracheostomy. J Trauma Acute Care Surg
Williamson PR, Altman DG, Bagley H, et al. The COMET Handbook: version 1.0. Trials
. 2017;18(Suppl 3):280.
Dodd S. Classification of COS outcomes. Paper presented at: 6th
COMET Meeting 2016; Amsterdam, The Netherlands.
Bardecki MJ. Participants’ response to the Delphi method: An attitudinal perspective. Technol Forecast Soc Change
Sinha IP, Smyth RL,Williamson PR. Using the Delphi Technique to Determine Which Outcomes to Measure in Clinical Trials: Recommendations for the Future Based on a Systematic Review of Existing Studies. PLoS
Kirkham JJ, Gorst S, Altman DG, et al. Core Outcome Set–STAndards for Reporting: The COS-STAR Statement. PLoS Med
[Figure 1], [Figure 2]