Among the observations and conclusions of the Channel 4/BMJ commentary were:

1. Surgical Devices are not subjected to the same degree of robust evaluation and trialling as are drugs and medicines. Many implantable devices are introduced to the market without formal clinical trials, or because their function is similar to devices already permitted and in use.
2. Regulation and evaluation of Surgical Devices is weaker in Europe than in the United States [4, 5].
3. Reporting systems are weak, and there can be considerable delays before problems with particular devices emerge. “Commercial Confidentiality” protects the industry rather than patients in the short term. It hinders disclosure of key information, and “there is a lack of transparency in publishing research findings, device related complications, and competing interests”, as pointed out by Peter Wilmshurst [4].
4. Manufacturers can potentially have an unhealthy financial relationship with the clinicians who use and promote the use of particular devices.
5. Better registries are urgently needed, along with much greater engagement of clinicians with the reporting process.
6. It is a challenge to strike a balance between regulation, innovation and patient protection. Clinical devices have produced dramatic improvements in health measures and technical outcomes, as for example following the introduction of cardiac valves, artificial joints and vascular stents, and sometimes when the devices have been used outside their original design specifications [10].

Surgical Devices and their classification

Mechanical and electromechanical devices are now ubiquitous in surgical practice. Over the past 50 years or so, there has been an explosion in the products and technologies which advance our trades and capabilities. Surgical devices can be classified in various ways, as by function, intended use and user group, by complexity, or by risk to the patient.

a) Classification of surgical devices by function

This broad means of classification includes:

• Those devices which are used for surgical diagnostics, including imaging technologies and endoscopy.
• Those devices which advance our technical operating capabilities, as for example laparoscopic instruments, and
  ...technical operating capabilities, as for example laparoscopic instruments, and electromechanical equipment such as diathermy machines or bypass pumps.
• Those devices which are implanted into the body to secure therapeutic gain, including surgical meshes, stents and grafts, and orthopaedic ‘metalwork’.
• Those devices which are not in the immediate ‘hand to patient’ eye-line, but which are central to the way that we deliver surgical product, including anaesthetic and ITU equipment, and patient handling devices.

We may also include computers and software systems in this definition, as these become more intrusive in the surgical care pathway, in the operating theatre, and in the recording and appraisal of our outputs. Surgeons now work in information - technology directed environments, where sudden technical failure or disruption (as of digital image transmission or retrieval) can have a significant bearing on the safe conduct of surgery. A parallel can be found in the world of flight, where much innovative work on risk mitigation has originated, and in the move from analogue and manual pilot controls to digital ‘fly by wire’.

b) Classification of surgical devices by intended use and user group

We can classify surgical devices by their intended uses, and by their user groups, which fall broadly into the sub-speciality divisions of the profession. While our focus at the ASGBI is primarily upon the work of surgeons in the directly affiliated specialties, we intend that our work in progress will benefit all surgical disciplines and other medical professional groups.

c) Classification by complexity

We may also classify surgical devices by their complexity, which has a bearing on fault recognition and fault fixing.

• Basic mechanical devices, including most surgical instruments, are characterised by robustness and simplicity of design. Direct inspection will allow a full appraisal of faults.
• Electromechanical and digital electronic devices can be difficult to evaluate and repair. Complexity adds risk, and inopportune equipment failure can compromise safe practice in a variety of ways.

c) Classification of surgical devices by inherent risk to the patient

By convention among the regulatory authorities, devices are classed as low risk (Class I), medium risk (Class II) and high risk (Class III).

• Class I devices include patient handling equipment and demonstrably safe devices (eg. wheelchairs, operating tables, contact lenses).
• Class II devices include those with an invasive procedural use, or where the consequences of technical failure would not usually be injurious to health (eg. endoscopes, ultrasound devices).
• Class III devices include implantable devices, such as artificial joints, stents, defibrillators, biological meshes and hearing aids, where there is a significant impact of short, medium or long term failure and of adverse clinical events.

The US Food and Drug Administration characterises Class III devices “as those that support and sustain human life, are of substantial importance in preventing impairment of human health, or which present a potential, unreasonable risk of illness or injury”. The existential problem with this definition is that these risks may not become fully apparent until the device has been in widespread and long term use. For this reason, robust long term reporting and surveillance is now essential, and why a robust programme of clinician engagement in the reporting system is now apparent. Implanted devices may have to survive intact and functional for many decades, while doing no harm either to local tissues or systemically. In the biologically reactive and corrosive environment of the human body, this is a ‘big ask’ for any engineered device.

The international regulation of surgical device associated clinical risk

The regulatory environment for surgical devices is both more complex and less demanding than for drugs and medicines [5 to 8]. Europe and the US generally provide worldwide leadership in such matters. However, Peter Wilmshurst states that “a medical device may obtain a CE (European Conformity) mark from one of dozens of notified bodies”. The MHRA confirms that the responsibility for checking the CE marking applied by the manufacturer for all except the simplest devices lies with a third party Auditing and Accreditation Body known as a Notified Body, of which there are approximately 80 across Europe.

The European Committee for Standardisation (CEN) is a business facilitator in Europe. CEN produces European Standards to meet European Union directives for medical devices (93/42/EEC), active implantable medical devices (90/385/EEC) and in vitro diagnostic devices (98/79/EC).

While the US regulatory environment for devices is more rigorous, it still has loopholes which allow individual devices to evade rigorous testing.

The International Organisation for Standardisation (ISO) promulgates worldwide proprietary industrial and commercial standards. ISO 13485 is a Quality Management Standard for Medical Devices.

Patient Safety

In respect of Class III implantable devices in particular, there are a number of stages where regulation can help mitigate clinical risk, including:

• Design and manufacture.
• Materials selection and engineering.
• Appropriate in vitro and in vivo testing and clinical trials.
• Education and training for clinical users.
• Long term surveillance systems with ease of reporting for clinical users and patients.
• An early warning system for perceived flaws, with appropriate investigation, testing and recall.

Manufacturing standards and the quality control of materials and components for such devices are under seemingly strict and demanding regulatory and commercial disciplines. However, given the size and profitability of the device market, the possibility of counterfeit manufacture of flawed devices must also be considered.

Device regulation in the UK

Within the UK, there are many official bodies and agencies which have a regulatory and supervisory interest in device safety alongside the MHRA. The documentation on the Interventional Procedures Programme Process Guide of the National Institute for Health and Clinical Excellence (NICE) cites more than 20 such organisations, including:

• Individual NHS Trusts, their medical directors, and the Clinical Negligence Scheme for Trusts (CNST).
• NICE.
• Healthcare (Care Quality) Commission.
• Department of Health.
• Welsh Assembly Government.
• Northern Ireland DHSS.
• National Institute for HTA.
• National Horizon Scanning Centre.
• National Patient Safety Agency.
• NHS Connecting for Health.
• NHS Litigation Authority.
• Welsh Risk Pool.
• NHS Quality Improvement Scotland.
• NHS Research & Development.
• NHS National Technology Adoption Hub.
• Association of British Healthcare Industries (ABHI).
• Medical Devices in Scotland.
• Device Manufacturers.
• Patient organisations.
• Professional bodies, including the Royal Colleges and ASGBI.

This is clearly a complex field for any clinician to understand and work within. Streamlining and simplification is urgently needed.

The scope of medical device related problems in the UK

The MHRA has well established reporting procedures for Medical Device Adverse Events, which are set out on its website. In practice, very few surgeons will as yet be familiar with the system, will have visited the MHRA website, or will have used it. The challenge for the ASGBI, for the MHRA and for surgeons in general lies in understanding:

• What is the true incidence of clinically significant device related failures?
• What can be done to facilitate reporting to the appropriate regulatory authorities?
• Where/If there is a significant number of such events to justify concern, what should be done to ensure that information is reported and disseminated to other surgical users?

The MHRA quotes the following recent (2010) data for device related problems:

There are over 90,000 medical devices on the market and 1 in 25 of hospital admissions involves an implant. The cost to the NHS is approximately £11 billion in addition to a follow-up or maintenance cost of around £200 million.

Eleven percent of patients in hospital suffer adverse events costing £2bn per year in additional hospital stays; 20 percent are due to medical devices and 50 percent are preventable. The medical negligence bill for device related adverse events is in the order of £460 million.

The challenges facing the use of medical devices relate to the increasing sophistication of the devices, the development of the necessary technical skills for implantation, the use of implants in younger patients with longer life expectancy and "local interference" with design and deployment (non approved modification).

Most device related adverse events are due to several identifiable factors. These include shortcomings in the device itself; modifications and adjustments; inadequate instructions for use (or not read); failure to perform a pre-use check; inadequate servicing and maintenance; user error, including inadequate training and inappropriate use (eg. single use); the environment in which it is used or stored; inappropriate choice of device; incompatible device combination; and the failure to report adverse events.

The MHRA also states that:

If an adverse device related event is recognised it is important to investigate and establish the root cause for this "failure". The Consultant should discuss the case with the involved healthcare workers and with manufacturer, and similar devices should be tested. Appropriate advice should be issued to the health service and relevant information disseminated, utilising safety notices websites and email as appropriate. If the adverse effect is serious a "medical device alert" should be issued and the device recalled pending further investigation.

All adverse device-related events should be reported to the Medicines and Healthcare products Regulatory Agency (MHRA). This should include individual incidents, trends, user problems...

The Information Environment and Surgical Device Safety

The information provided to ensure the safe use of implants and devices often appears to be written more from a defensive liability perspective than to educate users through simplicity and clarity. For example, the Instructions for Use of vascular stent grafts in some cases run to several hundred pages. There is a need for better documentation on the use of implants and devices, written for the ease of use of the individual surgeon and of the team using the equipment.

Factors within the Surgeon’s Control which May Lead to Device Failure

The misuse or inappropriate use of devices and implants may cause clinical problems. This can occur both within the technical design brief and applications envelope of the device, and in novel clinical applications, in suboptimal clinical circumstances; or where there are unforeseen co-morbidities which affect device use. There is also concern that devices may be used 'off label', which means outside the intended design envelope. While novel uses may produce significant clinical gains, it is important that there is a continuous flow of information to the regulatory agencies so that the safety, or otherwise, of any device can be continuously appraised.

The Training and Simulation Environment

'In vitro' training and simulation can be powerful tools for testing and rehearsing the applications of devices, both in the design and development phase, and subsequently in the training of individual theatre teams in the use of the devices. Simulation and team training facilities have been a longstanding deficiency in surgical practice. The Armed Services have been very innovative in the development and work of surgical team training facilities, both at the National Field Hospital Trainer in York, and in the use of facilities at the Royal College of Surgeons of England.

Inculcating a Safety Critical Attitude

The aviation and marine industries have been very vigorous in promoting a safety critical attitude at all levels to ensure maximum physical and governance protection for passengers (patients) and for pilots (clinicians), while supporting innovation. The Confidential Human Factors Incident Reporting Programme (CHIRP) promotes feedback and dissemination of experience on adverse events and near misses. The Safety Critical mindset and procedures are not yet embedded in surgical practice.

Reporting Systems for Surgical Device Safety

At present, very few reports feed back to the MHRA from surgeons. The present data covering device usage and problems is incomplete and anecdotal. A formal reporting system which recorded the use of each and every specified device over its lifespan and which made the data publicly available would be a major step forward in patient protection.

A voluntary reporting system, universally and willingly adopted by manufacturers and by the profession, would be preferable to an imposed regulatory model. However, it may well be that compulsory reporting will become the norm if the profession does not act swiftly to develop robust reporting systems for surgical device safety.

Fortunately, web-based reporting technologies are becoming increasingly familiar in the surgical workplace, as in training, appraisal and national audits. The disciplines of recording and tracking devices and physical items in hospitals and operating theaters and clinical supply chains are already well established, and it should not be unduly difficult to integrate a web-based devices reporting system, using electronic tagging (through bar codes or RFIDs) into the clinical work space.

The on-line reporting infrastructure already exists through the MHRA website, so implementation of a more robust reporting system has a low cost. The adaptation of various Surgical Association and College websites to encourage reporting and to include links to the Adverse Incident Reporting System on the MHRA website is also straightforward.

Long Term Device Surveillance

By European law, manufacturers must have a post-market surveillance system in place, but there are many unacknowledged problems. The post-market surveillance system is often inadequate for the risk of any particular device and is not always followed up by the Notified Body, and that high risk devices are falling under systems of customer complaint without systematic assessments in place. Concerns raised about long term surveillance include:

• The high cost per patient, particularly with the high risk devices where there may be relatively few devices used.
• The decline in physician interest over time.
• Difficulties in interpretation as to whether late adverse events are device related or not.
  In setting and monitoring standards in any new regulatory framework, we will need to establish:
• Criteria for the success or otherwise of a new regulatory framework, combining voluntary and compulsory elements.
• Criteria for the risk/benefit assessments and classification of individual devices.
• Performance and cooperation monitoring and its frequency.
• The reliability of reporting systems.

Work done in the UK needs to translate into an internationally standardised system which crosses jurisdictions. Conversely, many devices are designed, manufactured and used outside UK jurisdiction, so surveillance and reporting needs to be international.

We will also need to consider matters of non-compliance with any future reporting system for surgical device safety, such as:

• Where and how will it be detected?
• Who will enforce and what action/sanctions will be in place?
• What will be the consequences of non-compliance for manufacturers, for individual surgeons, for clinical teams and for the regulatory agents/agencies themselves?
• What will be the costs of the reporting systems, and how will they be met?
• How will findings be communicated and disseminated?

There may be lessons to be learned from other safety critical industries and operating organisations, such as the civil aviation and nuclear power industries.

The concerns of surgeons about mandatory reporting systems

There are a number of legitimate concerns which individual surgeons will feel about the voluntary reporting of device associated problems, in respect of confidentiality; the apportionment of responsibility and “blame”; the costs in time and effort in undertaking reporting; and medico-legal considerations and allegations of negligence. These concerns must be addressed, but they must also be balanced against the primary needs of the patient to be protected from harm.

Summary and recommendations

1. The issue of surgical device safety; the limitations of present regulatory and reporting systems; and the limits on professional engagement in reporting are now a matter of public interest and debate, which must be addressed with urgency.
2. At present, there is no systematic approach to the collation or reporting of data on the use, misuse and complications arising from invasive and complex surgical devices and implants. There is a voluntary reporting process which is almost certainly incomplete.
3. The wellbeing and protection of patients must remain the primary driver of change.
4. The MHRA is the statutory government agency charged with responsibility for medical devices, and the established working partner for the ASGBI in taking the professional advisory lead on behalf of surgeons. ASGBI will aspire to work with a wide range of professional representative bodies, including the Royal Colleges, to secure sign up to a robust reporting system.
5. The existing MHRA reporting system provides the technical and administrative framework for an enhanced surgical device reporting strategy.
6. There needs to be whole-life device registration and tracking system, such that medium and long term problems with individual devices will be picked up and addressed.
7. A culture of professionally led, cooperative, no-fault reporting (as with civil aviation fault reporting systems) and voluntary compliance is preferable to a compulsory regulatory system.
8. Transparency of reporting will bring significant professional benefits and establish greater confidence among patients and their support groups.
9. Open and trusted reporting processes will minimise commercial risk and long term cost arising from product fault or misuse, not least through early identification of problems and early recall before settlement costs escalate.

Acknowledgements

I am grateful for help in drafting this paper to Dr Susanne Ludgate, Clinical Director of Devices at the Medicines and Healthcare products Regulatory Agency (MHRA); to Professor Michael Horrocks, Past President of the Association of Surgeons of Great Britain and Ireland; and to Dr Julie Cullen, recently Consultant Nurse Advisor on Clinical Standards and Patient Safety, Portsmouth City Teaching Primary Care Trust.

References

[1] Patient Safety: A Consensus Statement
ASGBI, April 2010
The truth about medical devices: a range of articles in the
British Medical Journal, Volume 342, 21st May 2011,
including:

[2] Out of joint Cohen D. BMJ 2011 342 1116-1122

[3] Commentary: Metal on Metal Hips Skinner J and Kay P
BMJ 2011 342 1123

[4] The trouble with medical devices: Editorial. Godlee F. BMJ
2011 342 p1091

[5] The Regulation of Medical Devices: Wilmshurst P. BMJ
2011 342 1093-1094

[6] Evaluating and regulating device therapy. Freemantle N.
BMJ 2011 342 1129

[7] Standards need international collaboration. Fraser A G,
Krucoff M W, Brindis R G, Komajda M, Smith S C Jr. BMJ
2011 342 1130

[8] Medical device recalls and transparency in the UK.
Thompson M et al. BMJ 2011 342 1131-32

[9] Europeans are left to their own devices. Cohen D and
Billingsley M. BMJ 2011 342 1124-127

[10] Commentary: The risk of over-regulation. Di Mario C,
James S, Dudek D, Sabate M and Degreerik M. BMJ
2011 342 1128