Future of surgical registries

The value of accurate healthcare data has never been more apparent than during the past year, and there is increasing recognition of the important role surgical registries play in driving improved patient outcomes, patient safety, and academic research. Successful registries need to be responsive and adapt to the rapidly changing clinical environment.

A limitation of surgical registries has often been their inability to provide data beyond in-hospital or 30-day mortality and complication rates. Longer-term outcomes have not been captured routinely. An exception is the UK National Joint Registry (NJR), which has used revision following joint replacement as an endpoint to identify outlying performance. This measure of long-term quality, combined with the capture of detailed information on hip implants, led the NJR to identify early the problems related to metal-on-metal prostheses¹.

High-profile implantable medical device failures in other surgical specialties have led a number of registries to adopt the NJR model, and there is increasing recognition that linking unique device identifiers to patient-held registry data will improve patient safety by identifying performance issues at the earliest opportunity².

In vascular surgery, the failure of devices used for endovascular aneurysm sealing only became apparent more than 2 years after implantation^3,4. Had registry device capture been established at the time, it is likely to have prevented many patients coming to harm. In response, the UK National Vascular Registry (NVR) added data fields that not only allowed capture of device information for aortic aneurysm repair, at implantation, but also reintervention as a longer-term measure of outcome in 2020. A forthcoming NVR report⁵ will confirm that device and reintervention details have been recorded successfully in the registry, and this information will prove vital for patients and clinicians in the future. The successful adaptations of existing disease-specific registries to capture devices in orthopaedics and vascular surgery are models for other surgical specialties. Furthermore, device outcome data are extremely valuable to industry partners, who contribute to the NJR in return for access to robust anonymized outcome data. This is a potential source of sustainable revenue to support registry maintenance and development.

A central patient-identifiable database that collects key details of the implantation of all devices (collect once, use often), and that is linked to registries for audit and research, should be the ultimate goal². There is no doubt that surgeons of all specialties who implant medical devices in patients should embrace changes to registries to capture this information. Registry data entry should become mandatory, and registries robust enough to allow interrogation and provide assurance of device efficacy and detect harm. It should no longer be acceptable to implant devices without a commitment to monitor device performance during the patient’s lifetime.

Reporting surgeon-level outcomes for specific surgical procedures continues to be controversial and, although transparency remains vital, it is increasingly being recognized that high-quality surgery requires appropriate resources and a functioning interdisciplinary team. Concerns have been voiced that surgeon-level reporting may lead to risk-averse behaviour in patient selection and possible gaming. In cardiac surgery, reporting individual-surgeon outcomes has been established for more than two decades, and these audits have brought transparency, quality assurance, and quality improvement.

Over recent years, surgical autonomy in decision-making has diminished and been replaced by multidisciplinary team-based judgements on the best treatment strategy for individual patients. Furthermore, patient outcomes are influenced significantly by the expertise of intensive care specialists, anaesthetists, and other allied health professionals. For these reasons, the Society of Cardiothoracic Surgery⁶ in Great Britain and Ireland, although continuing to collect surgeon-level data for internal audit and quality assurance, has recently reached agreement to report only unit-level data publicly, better reflecting the move to team-based working and shared decision-making. Robust mechanisms remain to support units where there is significant variation in outcomes between surgeons.

COVID-19 has further highlighted the problems with reporting surgeon-level outcomes. Many units have seen significant rises in postoperative mortality rates as a result of high in-hospital COVID-19 prevalence, and surgeons have justifiable concerns around the reporting of higher mortality rates driven by factors beyond their control. Moreover, there are wider concerns within governance circles that this may lead to a greater reluctance to report poor outcomes. The clear direction of travel for the transparent publication of surgical registry data is away from surgeon-level and towards unit- or hospital-level patient outcomes.

Collaborations between international registry groups are well established and have allowed geographical variations in surgical practice to be studied in large data sets⁷. The COVID-19 pandemic has increased awareness of the importance of population-wide linked electronic patient health records (EHRs). Although these have been developed in Scandinavia in recent years and allowed patterns of disease progression to be studied⁸, COVID-specific legislation in the UK has led to the rapid linkage of EHRs from primary care, hospital episodes, death registry, COVID-19 test results, and dispensing data within a Trusted Research Environment (TRE) at NHS Digital for patients with cardiovascular disease⁹. The data held on cardiovascular disease within the TRE will be further enhanced by the addition of registry data from the National Institute for Cardiovascular Outcomes Research, the Stroke Audit, and the NVR in 2021. These linked data have the potential to enable a very wide range of surgical cardiovascular research.

In recent years, there has also been increasing interest in the development of registry-based RCTs in surgical research. Although recognized as the standard for establishing high-quality evidence, RCTs are often frustrated by delays, high cost, and complexity. Registry-based RCTs focused on clinically relevant outcomes and centred on standard care pathways are potentially more efficient by leveraging existing registry platforms to conduct research¹⁰.

Although such trials are in their infancy in surgery, they have already shown considerable promise for cancer screening and heart disease. In the USA, there are a number of complete or ongoing registry-based RCTs investigating important study questions for hernia surgery¹⁰. In this setting, patient-reported outcome measures are important, and trial designs that incorporate longer-term outcomes are also vital in a similar way to that described earlier for device registries.

Early data suggest that more than 90 per cent of patients have been followed up successfully as a standard of care, with low missing data rates of 2 and 10–15 per cent at 30 days and 1 year respectively. These studies have already provided important information on the clinical and cost effectiveness of laparoscopic versus robotic inguinal hernia repair¹¹. They also underline the importance of case ascertainment, which is the cornerstone all successful surgical registries. High ascertainment rates require not only buy-in from the profession but also resources for registry maintenance, data capture, and analysis.

The importance of registry-based RCTs was also highlighted recently by an interim analysis of a trial¹² of drug-coated devices used to treat patients with peripheral artery disease collected in the Swedish Vascular Registry. This study did not support the findings from an earlier meta-analysis which found that drug-coated devices were associated with an increased risk of death.

At a time when the value of healthcare data has been recognized beyond accepted medical boundaries, it is vital that the surgical profession grasps the exciting opportunities that registries present. By doing so, it will be possible to deliver enhanced patient safety, improved surgical outcomes, and robust cost-effective surgical research.

Disclosure. Member of the Medical Devices Safety Programme Steering Committee (NHSE/I) and Register of Experts, Medicines and Healthcare products Regulatory Agency (MHRA).