An exploratory descriptive analysis of the 5-year survivorship of total ankle replacements compared with equivalent rates in administrative datasets with further adjusting for co-varieties

This stand-alone non-research study (service evaluation) being undertaken by the Royal Devon and Exeter NHS Foundation Trust is an exploratory descriptive analysis of the survivorship of total ankle replacements and revision ankle replacements comparing with equivalent rates in administrative data sets. The aim of this study is to analyse data from the UK National Joint Registry (NJR) and from Hospital Episodes Statistics data (NHS Digital) to find out how many ankle replacements fail within 5 years of the surgery. This will be the largest study looking at outcomes of these operations. By analysing this data it will give both surgeons and patients more information on the expected outcomes and risks of complications following ankle replacements. The study hopes to improve knowledge of ankle replacements (and complications arising from them) that will improve future patient outcomes, and has support from the National Joint Registry and is being funded by the British Orthopaedic Foot and Ankle Surgery Society (BOFAS) and the Gwen Fish Orthopaedic Trust (neither funder have influence on the outcomes nor suppress any of the findings of the service evaluation).

Failure rate of ankle replacements

A failure is determined as revision, arthrodesis or amputation. This will enable further information to be to be given to patients in the consenting process, and inform surgeons of the likely outcomes for these patients. It will also determine risk factors for ankle replacement failures.

Over 900 total ankle replacements (TAR) are recorded in the NJR each year. There have been varying results reported with these implants, and considerable research is ongoing assessing long term outcomes of these replacements and comparing total ankle replacements with ankle arthrodesis. A previous systematic review on long term outcomes of ankle replacements found that TAR had a positive impact on patients lives, but was unable to make strong conclusions as previous studies have been small and of poor quality. Currently the failure rates of ankle replacements are unknown as those that are revised to an ankle fusion are poorly recorded at present. Therefore it is important to know the true failure rate of ankle replacements for decision about surgical treatment of ankle arthritis.

Patients that have failure of a total ankle replacement can either have a revision procedure, conversion to an arthrodesis or amputation. The Swedish arthroplasty register has demonstrated 188 failures in 1110 primary TARs. Of these 118 were revised with a salvage arthrodesis and 70 with a revision TAR.

The National Joint Registry for England, Wales, Northern Ireland and the Isle of Man started recording total ankle replacements and revision ankle replacements in 2010. Since then over 3500 ankle replacements have been recorded. When a revision is performed an A2 form should be completed. The risk of revision as reported by the NJR in 2019 is 6.9% at 5 years. BOFAS and the NJR believe that this is under-reported and especially those converted to an ankle arthrodesis or amputation are very poorly recorded. The primary outcome of failure is defined as the removal or exchange of any components of the implanted device inserted during ankle replacement surgery.

The aims of this study are:

1. Calculate the failure rate of total ankle replacements and revision ankle replacements

2. Mortality following total ankle replacements

3. Post-operative complications of total ankle replacements and revision ankle replacements

4. Validation of the NJR data set

The study also hoped to identify the incidence of:

5. Post-operative DVT or PE (within 1 year)

6. Post-operative re-admission (within 1 year)

7. Post-operative MI (within 1 year)

8. Post-operative CVA (within 1 year)

9. Post-operative further surgical procedure

10. Mortality

11. Amputation

12. Revision ankle replacement

13. Ankle arthrodesis

The data linkage proposed by this data sharing agreement will ensure the accuracy and completeness of the data recorded by the NJR. At present it is thought that the data recorded by the NJR does not account for all the failures of ankle replacements. The only feasible way to determine the true failure rate is by linking NJR and NHS Digital data.

The data subjects in the NJR cohort group are all patients undergoing either a primary total ankle replacement or revision ankle replacement between 01 January 2010 and 31 December 2018. The cohort group will be sourced from the UK National Joint Registry. The study will perform a deterministic linkage which requires exact match of the fields from both the NJR and Hospital Episodes Statistics (HES) database to confirm they are from the same patients. This linkage will be performed by NHS Digital using NHS Number, Postcode and Date of Birth. A Study ID will be allocated to each NJR patient record so that the data can be pseudonymised by NHS Digital prior to dissemination back to Royal Devon and Exeter NHS Foundation Trust.

The data requested is from when the replacement was undertaken until present day to get the most complete data set and accuracy of those ankle replacements that have failed. This includes primary ankle replacements and revision ankle replacements.

The study aims to look at data from England and Wales as these are the geographical areas also covered by the NJR.

The sole Data Controller who also processes the data is Royal Devon and Exeter NHS Foundation Trust.

GDPR LEGAL BASIS FOR PROCESSING

The lawful basis for processing data under GDPR has been reviewed against the guidance provided by IGARD and been assessed as acceptable. Article 6(1)(e) (processing is necessary for the performance of a task in the public interest…) and Article 9(2)(i) (Processing is necessary for reasons of public interest in the area of public health…). Royal Devon and Exeter NHS Foundation Trust are performing a Public Task as per the NHS Act 2006 section 43(5), which describes the functions of authorised NHS Foundation Trusts. The data is required for non-research, service evaluation purposes, for improvement and ensuring high standards and quality/safe care. This project is evaluating the actual failure rate of ankle replacements. This is in the public interest so that it is possible to determine how many ankle replacements will fail. It will also ensure that all the ankle replacements being performed are for the correct patient with highest likelihood of implant survival to prevent patient morbidity and disability.

DATA REQUESTED

Pseudonymised record level HES APC Annual Refresh data extracts from 2010/11 to 2019/20 (10 data periods) with Date of Death derived from Civil Registrations (Deaths) Summary Care Record data set linked via a cohort of approximately 5,000 individuals supplied by the National Joint Registry.

PROPOSED METHOD:

- Northgate Public Services (National Joint Registry) submit a cohort of approximately 5,000 individuals derived from the NJR (permission provided from HQIP) via Secure Electronic File Transfer (SEFT) service to include a Study ID plus the patient identifiers (Gender; Date of Birth; NHS Number; Local Patient Identifier; Hospital ODS code; Post Code).

- NHS Digital will extract HES APC data for the cohort and the date of death for the cohort from the Civil Registrations (Deaths) SCR data set. No identifiable fields have been requested by Royal Devon and Exeter NHS Foundation Trust.

- NHS Digital will remove the patient identifiers, but leave in the Study ID, making the data extract pseudonymised.

- NHS Digital will send the pseudonymised data extracts to Royal Devon and Exeter NHS Foundation Trust Via Secure Electronic File Transfer (SEFT) service.

Record level identifiable health data has been supplied by the National Joint Registry to Royal Devon and Exeter NHS Foundation Trust under a separate data sharing agreement for the purposes of this study. Each record is given a unique Study ID by UK National Joint Registry. All patients have given consent to the National Joint Registry for their data to be used for research purposes. However, as this is a strictly non-research study culminating in a service evaluation, Royal Devon and Exeter NHS Foundation Trust have sought and obtained section 251 approval to send identifiable patient data from NJR to NHS Digital for the purpose of linkage to Hospital Episode Statistics data.

DATA MINIMISATION

All efforts have been used to minimise the data requested. All data related to the surgical procedure of ankle replacement is collected by the National Joint Registry, including implants, surgical technique and patient demographics.

HES APC will be limited to the patient cohort and operation codes provided by the customer. Civil Registrations (Deaths) - Summary Care Record will also be limited to the patient cohort. Only specific codes for further surgery following ankle replacements have been included.

The data processor does not have access to unique patient records or any identifiable information. All variables such as gender, age, centre performing the procedure and others will be analysed and anonymised prior to publication. The appropriate HES suppression and disclosure rules will be applied prior to publication. Royal Devon and Exeter NHS Foundation Trust will make no attempt to re-identify individuals.

Data processing only carried out by substantive employees of Royal Devon and Exeter NHS Foundation Trust. Substantive employees of Royal Devon and Exeter NHS Foundation Trust are appropriately trained in data protection and confidentiality rules and regulations.

COHORT SUBMITTER

Northgate Public Services are the contracted IT services supplier for the National Joint Registry and manage collection and management of data and technology. The NJR data systems are hosted by Northgate Public Services.​ Northgate Public Services will submit the cohort. They will not receive any data from NHS Digital and are therefore not listed as a Data Processor.

Data processing is on Royal Devon and Exeter NHS Foundation Trust computers which is a secure server, backed up overnight and accessible only to authorised individuals via user name and password. A technical solution is in place to monitor if data is downloaded from the server to an individual machine, and all staff involved in the processing of data are instructed not to download the data to another geographically distant location. Therefore data processing can be undertaken on the sever with a geographically external device via a secure VPN connection to the server, but data is not permitted to be stored on a geographically external device. There are no geographical external back-up storage facilities and no Cloud Platform services being used.

HES and ECDS DISCLOSURE CONTROL / SMALL NUMBER SUPPRESSION

In order to protect patient confidentiality, when presenting results calculated from HES record level data, outputs will contain only aggregate level data with small numbers suppressed in line with HES Analysis Guide. When publishing HES data, you must make sure that:

· cell values from 1 to 7 are suppressed at a local level to prevent possible identification of individuals from small counts within the table.

· Zeros (0) do not need to be suppressed.

· All other counts will be rounded to the nearest 5.

Data will not be made available to any third parties other than those specified except in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

As a result of analysis of the data:

- A report will be made to the National Joint Registry to determine the true failure rates of ankle and revision ankle replacements, and a report will be included in a future NJR annual report.

- Presentations at the British Orthopaedic Foot and Ankle Society Conference and other Orthopaedic conferences.

- A submission to Bone and Joint Journal and other commonly cited Orthopaedic Journals.

Royal Devon and Exeter NHS Foundation Trust has undertaken informal discussions with patients. The main output by undertaking this data analysis is that surgeons will be able to give more informed information to patients that are considering an ankle replacement, as currently it is not known how many of these fail.

The aim is for data analysis and initial reports to be written within 1 year, and published in the National Joint Registry report, which is published annually and is freely available to view from their website.

The NJR have an Implant Outlier Process in place. The NJR Implant Outlier Process identifies implants which fall outside designated patient time incidence rate (PTIR) confidence intervals and is a defined mechanism for contacting the implant manufacturer and regulators in order to provide early warning of issues relating to patient safety, so appropriate action can be taken. Further information about the outlier process is available on the NJR website - https://www.njrcentre.org.uk/njrcentre/NJR-Accountability-and-Transparency-Model/Implant-Outlier-Process.

All outputs will be aggregated with small number suppression applied according to the HES Analysis Guide to avoid potential patient re-identification.

This evaluation aims to determine the true 5 year failure rates of ankle replacements, and which patients are at a high risk of failure. This will enable more informed decisions to be made by clinicians and patients regarding the management of ankle arthritis in individual patients. This study will provide a service evaluation which will improve the current knowledge of ankle replacements potentially improving future patient outcomes.

With increasing numbers of ankle replacements being performed it is vital that the National Joint Register produces accurate data on the outcomes of these prosthesis. This will evaluate the results of all prosthesis being currently implanted in a similar manner to other arthroplasty procedures. This will ensure that patients are receiving accurate information in the consenting process with regards the likely requirement for further surgery due to ankle replacement failure.

At present the only publication which can inform patients of the risks of ankle replacements is a patient information booklet available from the National Joint Registry. This, though, only has the failure rate as those converted to a revision ankle replacement, rather than all those that fail and require further surgery. This document is due for review, and therefore the findings and evaluation from this study will be able to be used when this document is rewritten.

As this study is being undertaken in partnership with the National Joint Register its dissemination will be widespread and will enable guidance on future management to be produced.

There are currently over 900 ankle replacements performed per year in the UK and this number is increasing. By increasing our knowledge it will ensure that the correct patients who are most likely to benefit from ankle replacements are undergoing this procedure.

All outputs will be aggregated with small number suppression applied according to the HES Analysis Guide.

Yielded Benefits is not a requirement for new applications.