A Guide to Good Practice in Clinical Perfusion by Gary Grist RN CCP Emeritus*
On May 27, 2005, there was an incident resulting in the death of a child undergoing heart surgery at the United Bristol Healthcare Trust. The Bristol hospital is a part of the British National Health Service (NHS). The death was caused by a calcium overdose perpetrated by a perfusionist. Another perfusionist-related death had occurred sometime earlier at a different hospital as well. These two incidents prompted a full scale government investigation led by a former NHS director, Mark Gritten. The Gritten Report was not confined to any specific incident, but examined the perfusion profession practices in their entirety throughout Britain and Ireland (1). To my knowledge this is the one and only authoritative and objective review of the perfusion profession ever undertaken. It was not something of which perfusionists could be proud. The report found many deficiencies among the perfusionists in Britain. The ones that stood out for me are these:
- There was no authoritative (governmental) oversite or accountability of the perfusion profession in the way that doctors and nurses are accountable to their licensing boards.
- There were inconsistent protocols in the way that perfusionists performed their work, even within the same hospital.
- There was a lack of effective checklists and double-checking.
- There was poor team communication. Confusion or misunderstanding during a case can be a killer.
- There was no performance management, meaning that even within local programs there were no individual managers responsible to ensure the proper and consistent performance of duties.
- There were no risk assessments, meaning that programs and individuals did not anticipate what problems might occur, how dangerous the problems might be, how they could be avoided or even how they could be mitigated if they could not be avoided.
In addition to the Gritten Report, the case was also investigated by the police and criminal prosecutors as a possible corporate homicide. In 2005, corporate homicide was an informal legal principal in Britain. But by 2007 the Corporate Manslaughter and Corporate Homicide Act (CMCHA) was formalized and took effect in April, 2008 (2). Under this act senior managers and corporations are required to implement good quality health and safety policies and procedures, including risk assessments with regular reviews. By law, guidelines published by existing authorities must be followed. Failure to do so is regarded as reckless disregard for the public (and patient) safety and investigated by law enforcement authorities. (How many American perfusion programs can prove that they adhere to even the rudimentary Practice Guidelines from AmSECT?) Imagine having one or more police detectives investigate you after a harmful perfusion incident; even before giving a deposition to a plaintiff’s attorney which is what usually happens in the USA.
Although senior management personnel do not risk jail time under the CMCHA, their organizations or corporations are subject to public stigma and fines. Fines for NHS entities are paid from public money and may be relatively low. But fines for independent sector providers may be large enough to put them out of business. In addition, individuals can still be prosecuted or sued in civil court for gross negligence and for health and safety offences. The “Publicity Order” after a successful prosecution or confession requires that the entity publicly acknowledge their failings and apologize for them. This would denigrate their reputation and risk the loss of patient trust and possibly end the careers of the senior management personnel involved.
Since there was no authoritative guideline for perfusionists and motivated by the Gritten Report, the Department of Health of the NHS wrote one. It is called “A Guide to Good Practice in Clinical Perfusion” (GGPCP), July 2009 (3). The text may be reproduced without formal permission or charge for personal or in-house use. So, I will let the GGPCP speak for itself:
“The purpose of this document is to provide guidance on systems and processes which need to be in place to assure safe, high quality clinical perfusion services provided by the NHS and independent sector providers to support cardiopulomonary (sic) bypass (CPB) and mechanical circulatory support during cardiac surgical procedures.”
“If adhered to, this guidance and the recommended frameworks should minimise (sic) the risk to Boards and senior executives of being culpable under the Corporate Manslaughter and Homicide Act 2007…”
The GGPCP does not focus on specific details of perfusion practice with one exception; medications. For example, it does not dictate specifics such as which oxygenator should be used or how a pump should be set up. Rather, the focus is on the systems and processes required by the NHS for any of its entities or independent sector providers to perform the safest clinical perfusion practice possible. This is to assure patients and the cardiac surgery team that CPB is being safely practiced by the perfusionist. The NHS estimates that 10% of all hospital patients in Britain experience an unsafe incident. That would imply that CPB patients are at risk as well. Contained within the GGPCP is the method to prevent these errors and accidents. It is called the Quality Management System and Framework.
The GGPCP Quality Management System and Framework focuses on seven key elements:
1 The framework document describes the management system in detail and its sources of authority and responsibility.
2 All standard operating policies, processes and procedures are included.
3 Risk assessments are performed and a registry maintained. Under the GGPCP, each perfusion department needs to have a chief safety perfusionist who is trained in root cause analysis, risk register maintenance and incident investigation. (Does your program have such an individual?) The risk register should include as many conceivable clinical perfusion risks as possible with a prevention or mitigation strategy for each. In the USA a common form of risk registry is called a failure modes and effects analysis (FMEA). The closest thing that American perfusionists have to a risk registry are the FMEAs listed on the AmSECT Safety Page and these are by no means comprehensive.
4 Systematic checking and recording includes effective check lists and a detailed pump record.
5 Medicines management, special clinical perfusion protocols, and patient specific directives (PSDs) are required. PSDs tell the perfusionist how diverse patients are treated differently. For example, what methods are used to pump a routine CABG patient as opposed to a sickle cell patient or a pregnant patient? PSDs can be written in the form of an FMEA.
6 Teamwork and human factors training stresses communication and optimal conditions for practice. For example, if you are too sick to safely pump a case, what back-up options do you have?
7 Peer review is a method of continuously documenting competency that continuing education or case counting cannot assure.
The most specific guideline in the GGPCP (#5 above) deals with the 28 solutions and medications that a perfusionist may use. The perfusionist must be aware of his/her legal status in administering the medications, the form, the strength, the indications, the contra-indications, the dose range, the maximum dose and the acute side effects. (I would have added knowledge of any allergies to this list as well, but that’s just me! Perhaps the NHS will correct that omission when the GGPCP is revised.)
The GGPCP is compulsory in Britain. Without it, senior managers and the hospitals where they work could find themselves to be in serious trouble. Perfusionists don’t have anything like it on this side of the pond. The AmSECT Practice Guidelines are good, but they are deficient in several ways. 1. They are voluntary. 2. They lack a quality management system structure. 3. They lack accountability in the form of a revocable certification. 4. They lack risk assessment. 5. They lack peer review as a means to continuously document competency of individuals or an entire perfusion department. 6. They lack PSDs. 7. They lack teamwork and human factors training.
In 2001 (well before the GGPCP), the Joint Commission (JC) Leadership Standard LD 5.2, “Support of Patient Safety and Medical/Health Care Error Reduction”, was introduced. The goal of LD 5.2 is to reduce sentinel events and significant errors. Healthcare providers are required to prevent adverse events and errors, rather than just react to them, by conducting proactive risk assessments. A sentinel event root cause analysis is reactive and does not, on its own, meet compliance to LD 5.2. Under this broad-based standard, American healthcare providers must develop a “failure mode analysis” for proactive process and risk review (4). This comes as close as anything to the GGPCP. But it is generally ignored by American perfusionists because the JC has not focused on them the way that the NHS focused on the British perfusionists.
The JC does not have any CPB guidelines itself, but refers to the “Society of Thoracic Surgeons, The Society of Cardiovascular Anesthesiologists, and the American Society of ExtraCorporeal Technology: Clinical Practice Guidelines”. Currently these are limited to anticoagulation, temperature control, inflammatory response and blood conservation. These are “how to do it”, evidence-based instructions. They do not address the safety issue directly, implying that if the guidelines are followed then the best outcome will result; a very admirable goal. But if the best outcome is obtained does that, in and of itself, mean that the perfusion program is safe? There is no real mention in these guidelines of risk assessment or risk management in the day-to-day, patient-to-patient and perfusionist-to-perfusionist practice of CPB perfusion. The guidelines do refer to a risk>benefit and benefit>risk type of categorization, but “risk” is not well defined and there is no risk registry. In other words, they are not LD 5.2 compliant and I don’t think they are meant to be. A perfusion program that strictly follows all of these guidelines may not be any safer in terms of errors or accidents than programs that do not strictly follow the guidelines. A program’s record of good operative outcomes does not mean much to the family of a patient killed by an unmitigated perfusion accident.
So what can be done in the USA? An AmSECT sponsored “Center of Excellence” (CoE) in perfusion has been suggested. If such a plan is adopted, a quality management system like the GGPCP should be its backbone, IMHO. In fact, there is no reason that the GGPCP could not be immediately adopted, as is, by AmSECT. It would save a lot of work. Maybe any perfusion program that whole heartedly adheres to the GGPCP and the AmSECT guidelines could be designated as a Perfusion Safety CoE by AmSECT (for a fee, of course).
An AmSECT CoE should not be based on outcome but on safety, i.e. how infrequently any unsafe incident occurs. Unsafe incidents could be ranked on a scale from a minor to a major risk. Is there a liability threat to AmSECT in doing this? I don’t think so! When was the last time that the JC was sued because a hospital adhered to one of its requirements? In fact, liability insurers often want their clients to have a JC certification.
A CoE can be sponsored by a variety of entities including a professional society like AmSECT or a government supported organization like the National Cancer Institute or a private group seeking the treatment or cure for a specific disease. For example, the Extracorporeal Life Support Organization (ELSO) awards a voluntary “Excellence in Life Support” recognition (for a fee), the purpose of which is to recognize and honor ECMO programs with the highest level of performance, innovation, satisfaction and quality. There is no outcome or safety requirement that I can see. And the ELSO award still lacks many of the important aspects of the GGPCP. ECMO is a risky business. Perhaps ELSO thinks that if an ECMO program does a good job, then they don’t need to worry about risk assessment and management. By comparison, a high wire acrobat may be good at his job, but he is still unsafe unless he consciously wears a safety tether. Remember, in perfusion an unforeseen error or an unmitigated accident can ruin two lives; the patient’s and the perfusionist’s (or the ECMO nurse or therapist).
The American Association of Blood Banks’ Standards for Perioperative Autologous Blood Collection and Administration use a quality management system similar to the GGPCP. But there is no requirement for a focused risk assessment that I remember (but then again I have been out of the clinical end for four years).
Many cardiac surgery CoE’s are self-designated (5,6). In the past, there was a Medicare CoE for cardiac surgery. It originated in 1991 (7). However I am unable to verify if that designation currently exists. (My contact at Medicare could not find a current reference to it.) This noncompulsory CoE was awarded to high volume, high quality cardiovascular programs that offered lower prices to Medicare. In other words, the Medicare CoE was bought simply by lowering costs to Medicare. Nonetheless, applicants were required to complete an in-depth application with a clinical and organizational quality review process prior to receiving the Medicare CoE designation (8). I don’t know if the application required continuous updating or was a onetime requirement. I am not aware of any perfusion-specific requirements in that application.
But someday (perhaps in a push for healthcare accountability like the Brits), Medicare, other health insurance providers or even the JC may require something akin to the GGPCP to be mandatory in the USA. Will American perfusion programs be ready?
*Disclaimer: The views expressed in this article are my own and do not necessarily reflect the views of members of the AmSECT Safety Committee or any of AmSECT’s leadership personnel.