Recurrent wrong‐route drug error – a professional shame

‘To err is human, to cover up is unforgivable, and to fail to learn is inexcusable.’ (Sir Liam Donaldson [Chief Medical Officer]– launch of the World Alliance for Patient Safety – Washington DC – 27 October 2004) Human error, particularly when fatal, captures a reader's interest, if only as a diversion from the more demanding scientific papers, and the cluster of drug administration errors reported in this month's journal should achieve that goal [1-3]. The depressingly recurrent theme of wrong-route error [4, 5] demonstrates, however, the challenge in converting that professional interest into the definitive action demanded above by the CMO. Root cause analysis [6] has provided us with an understanding of both human error and system or design weaknesses as either causal or contributory for critical incidents [7], and an opportunity therefore to develop logical and robust solutions. For the same class of error to recur, there are clearly impediments to the endorsement, dissemination or implementation of such solutions, suggesting that analysis of the professional response to error is at least as important as analysis of the error itself. When approaching the described cases on this basis, the preliminary question would be whether the specialty should be engaged by incidents which occur on the ward rather than in theatre, and at the hands of nursing staff rather than practising anaesthetists. The answer has to be ‘yes’ as once an anaesthetist inserts an epidural, central venous catheter or arterial line for presumed patient benefits they assume certain responsibilities for the future safe management of that device. Furthermore, the same class of wrong-route error does occur within anaesthesia with fatal consequence [8, 9] and, applying the time-honoured Heinrich pyramid principle [10], it is very likely that a significant number of incidents with lesser consequence go unreported. Whether this fulfills the CMO's criteria for a ‘cover up’ is debatable, given the current lack within the UK of a specialty-specific and responsive reporting system. Furthermore, as the CMO's message is arguably more critical of those individuals and bodies capable of considering and implementing change, it is indeed timely to consider the role and progress of the professional and regulatory bodies on the issues raised by these incidents. In analysing the nature of error in the reports, it is important that the case of induced hypoglycaemia [2] does not dominate the collective learning process by encouraging nihilism towards technical solutions. Successive failures to consider potential sources of error and to place an idiosyncratic high glucose result in the context of the patient's clinical condition, the escalation of symptomatic treatment without formal laboratory confirmation, and ignoring a result that conflicted with a primary erroneous conclusion, collectively illustrate the unlimited permutations of human error and the elusiveness of its elimination, even with more explicit procedures and technical solutions. However, the other two reports simply illustrate new variants of a persistent and much more frequent theme, that of wrong-route drug administration error [1, 3]. It is possible to identify certain procedural factors such as checking failure prior to administration, and the absence of warning mechanisms such as unique colouring or explicit labelling for different invasive routes into the body, but the ubiquitous nature of the Luer connector as a fundamental contributory factor has to be recognised and addressed. The Luer system, in the form of a glass syringe with a male 6% tapered nozzle, was patented in 1898 and, apart from the addition of the threaded collar to create the Luer-Lok in 1925, has remained unchanged since that date [11]. The combination of ease of use and simplicity in manufacture from a range of materials, glass, metal and plastic, has resulted in its adoption as the virtually universal connector for a vast range of medical devices from delivery, monitoring or sampling by the enteral, arterial, venous, spinal and epidural routes, to the insufflation of gas into laparoscopic or bronchoscopic equipment, intravascular balloon devices, tracheal tube cuffs and sphygmomanometers. Such commonality permits fatal wrong route errors when only one Luer system is active [12]. The multiplicity of devices attached to the acutely ill, as the case reports illustrate, compounds this risk. Even with just one practitioner taking responsibility for all aspects of care and, indeed, placement of all such devices, as in the anaesthetised patient, such multiplicity creates an inherent risk of fatal error, usually when substances such as bupivacaine or air are administered intravenously [8, 9]. Although insulin has not previously been reported, most anaesthetic agents and a range of other drugs have been administered via the epidural route [13-15], and those intended epidurally, given intravenously [16]. Logic dictates that such serious outcomes are significantly outnumbered by actual error rate, borne out by published series of critical incident reports, in which drug administration error dominates other causes [17, 18]. Given that such events have been reported for a significant period of time and are clearly recurrent both in the operating theatre and on the wards, it is reasonable to take an overview as to the status of current strategies to modify such risk and where responsibility lies for this task. On a European basis, the Committee for Standardisation (Comité Européen de Normalisation, CEN) recognised the misconnection hazards associated with the proliferation of the Luer connector in the early 1990s. The committee established a task group in 1997 and produced a consultation document in July 2006 entitled Small bore connectors for liquids and gases in healthcare applications[19]. The report reiterated earlier findings [20] and recommended: ‘That the use of Luer connectors is restricted to devices intended to be connected: To the vascular system for delivery or sampling purposes or to assist in making some sort of measurement To a hypodermic syringe in order for the syringe or a connected device to achieve its intended purpose.’ European medical device manufacturers objected to the original report on the grounds that the proposals could generate new hazards and would be expensive to implement [21]. With these arguments still persisting, and accompanied by the claim that clinicians do not wish to change, and will not buy the new designs (personal communication – Association of British Healthcare Industries, ABHI), it still remains uncertain whether or not the recommendations will be accepted and whether therefore they will be progressed under the ISO (International Standards Organisation). With regard to UK initiatives, fatal wrong route errors between vascular and neuraxial administration [22] were critical to the formation of the NPSA (National Patient Safety Agency), and the government committed itself to eradication of this class of error by the end of 2001 [23]. Although the triggering event had numerous contributory procedural weaknesses, initial investigative reports promoted a technical solution in the form of non-Luer-compatible connectors for neuraxial interventions [24]. These proposals have been endorsed by key international figures in the field of patient safety [4] and individual members of the UK anaesthetic community [21, 25, 26], and have prompted published design solutions in Canada [27] and the UK [28]. The Department of Health supported investigation of design solutions and, on the basis of a Structured What-If Technique (SWIFT) for Hazard Identification undertaken by the NPSA, which favoured adoption of non-Luer connectors for all neuraxial interventions, commissioned an OJEC advert (Official Journal of the European Communities) for non-Luer prototypes in 2002. Under the supervision of the Department of Health Non-Luer Connector Steering Group (DHNLCSG), two prototypes have undergone non-clinical testing throughout 2006 and, following the completion of a second round of testing, one of the devices has been selected to progress to clinical trials in 2007. Parallel studies have been undertaken by the Psychology Department of the University of Leeds addressing not only user acceptability but issues such as new classes of hazard with novel equipment, professional support for changes in equipment, and logistical problems with supply and storage. Technical solutions in the form of unique connectors for neuraxial blockade are therefore on the national and European agenda, and it is noteworthy that the NPSA, in a recent alert notice on epidural drug administration error, anticipates and endorses these developments [29]. It is clearly unreasonable, however, to rely solely on unique connectors to eradicate or minimise wrong route error, and to ignore other appropriate measures during the inevitable delay whilst a definitive design gains professional acceptability, commercial productivity and regulatory endorsement. Such general measures have been defined in additional alert notice from the NPSA relating to oral/enteral therapies and on injectable medicines in general, [30, 31] as well as the epidural route, following an extensive consultation exercise in the light of incidents reported to the NRLS (National Reporting and Learning System) [29]. The implications of these recommendations for individual practising anaesthetists and clinical directors are highly significant, as there will be an obligation not only to implement changes in patient care, but also to embrace a raft of broader professional responsibilities including risk assessment of all areas of practice at least annually, the development of protocols for all aspects of practice, provision of training and accreditation for all relevant staff, and audit to confirm compliance with these procedures. Furthermore, the clinical obligations will extend beyond choice of disposables, labelling and colour coding, with significant inroads into autonomous prescribing. Readers will acknowledge the differing local anaesthetic concentrations and the vast range of additives for epidural infusion, noting the use of sufentanil in the relevant case report, and understand the arguments both for rationalisation to a limited number, and the purchase of these as distinctively marked standard solutions, thereby moving away from error-prone preparation at the point of administration. Such concepts also engage the professional bodies, as there are unassailable arguments for promoting a national consistent response, as demonstrated in the introduction of coloured syringe labels [30]. Individual units may not have the resources to competently interpret and implement these recommendations, provide education and audit the benefit from standardisation, but more particularly, the unavoidable mobility of the workforce creates a significant risk factor if the equipment and approaches are radically different between units. The challenge for the professional bodies will therefore be balancing the patient benefits of standardisation with a potential backlash at the loss of autonomous decision-making, threat to improvement through research or, in today's climate, the liability of individual practitioners who cannot or do not conform. It is important in this regard to separate out components in which there can be no argument against standardisation such as the design of labels either side of any connector, improved labelling of syringes and infusion bags, or for infusion bags of local anaesthetic to be easily distinguishable from crystalloid solutions. The involvement of design expertise from organisations completely outwith medicine such as the Royal College of Art is particularly welcome in such solutions [31]. One could also reasonably expect the professional bodies to endorse the principles set out in the alert notices promoting unique colouring of epidural infusion lines and dedicated volumetric devices for epidural infusions, rather than use of the syringe driver as described in the case report, with at least the increased potential for error and, indeed, infection with the inherently necessary frequent changes. The promotion of unique connectors for the three key routes of drug administration where error occurs, vascular, oral/enteral and epidural, is potentially the greatest professional challenge. Practitioners may argue that the low frequency of error in the theatre environment does not justify the hazard of trialling new equipment, that a connector alone without uniquely compatible preparations of the active drug does not eradicate error, or may highlight the difficulties in duplicating stocks of unique syringes etc. for each of the three routes. However, fatal wrong-route errors have occurred at the hands of anaesthetists in theatre [8, 9] and predictably will continue to occur. More importantly, anaesthetists are inserting these systems for longer-term use in an environment over which they have no immediate control and within which many other health-care professionals will be involved in delivering therapies, as the case reports illustrate, and where fatalities continue to occur [32]. This constitutes the most potent argument, namely, the greater good, for anaesthetists accommodating new equipment and subsequently ensuring that all associated devices such as ampoules and infusion bags are similarly uniquely compatible. The precise choice of unique connector is a complex amalgam of design, engineering, manufacture, usability, safety trials, CE marking, marketing and professional and regulatory endorsement. One solution for spinal needles has been described in the Canadian literature [24] and the reverse Luer trialled within the UK for epidural systems [25], but it should be noted that the reverse Luer is a current favoured solution for oral/enteral connectors and has already been adopted for blood pressure cuffs in view of previous incidents related to air insufflation via the wrong route [12]. Furthermore, given the vulnerability of the system to access by a conventional male Luer syringe by the simple addition of a three-way tap, it appears prudent to promote a unique solution if feasible, in line with the principle of the pin-index system for gas cylinders or pipeline connectors. It is inevitable, however, that pursuit of this goal will require modification of the product after clinical trials, as other innovations in anaesthesia such as the laryngeal mask have demonstrated. Practitioners should therefore assist this process of evaluation and development through engagement and constructive criticism rather than simply listing potential hazards as barriers to change. It would appear perverse for significant investment in other technical solutions to reduce medication errors such as ‘smart’ infusion devices [33, 34], without support for an accompanying mechanism to prevent this then being applied by the wrong route. Ultimately, practitioners need to know where authority for change lies and where liability will rest in the event of future wrong-route errors. The difficulties for the NPSA in co-ordinating all aspects of a national safety agenda were identified by the National Audit Office [35] and triggered a report by the CMO (Chief Medical Officer), Safety First, in December 2006 [36]. A new strategic body is to be established, the National Patient Safety Forum (NPSF), jointly chaired by the Chief Executive of the NHS and the CMO, to co-ordinate the other organisations with responsibility for patient safety, such as NICE (National Institute for Health and Clinical Excellence), the MHRA (Medicines and Healthcare products Regulatory Authority), the Healthcare Commission and NHS Litigation Authority. The NPSA is to be ‘refocused on its core objective of collecting and analysing patient safety data to inform rapid patient safety learning, priority setting and coordinated activity across the NHS’. Responsibility for technical solutions will be contracted to NICE and the NHS Institute for Innovation and Improvement will be liaising with the medical Royal Colleges and other educational providers ‘to ensure that advances are made in education and training to support patient safety’. It can be seen that the professional bodies will have greater responsibility in this field, which will inevitably include taking a position on technical solutions such as the non-Luer connector. In the meantime, practitioners need to consider the environment in which individuals with epidural catheters or invasive monitoring, and particularly both, will be cared for postoperatively and whether procedures are in place to minimise both the frequency of error and the sequelae. Clear labelling of lines, using standards that have already been defined [37], appears mandatory. Rationalising epidural infusions within an institution and using those which are commercially available in distinctive bags as a licensed product, rather than accommodating individual idiosyncratic preparations, appears inescapable. With regards to equipment, it should be noted that the alert notice falls short of mandating dedicated volumetric devices for epidurals or colour-coded catheters, or adoption of the NPSA recommendations for oral/enteral tubes and medications as, not unreasonably, the Department of Health is mindful of cost implications, potential dependence on one manufacturer and the future hazards of limited competition in this market. Institutions which do not conform will clearly be syndicating risk and this should be made explicit for the protection of individual practitioners. However, with the Corporate Manslaughter and Corporate Homicide Bill having passed through the Commons and now going through the House of Lords [38], it is predictable that Trusts will be mindful of their vulnerabilities. With this information on fatal error in the public domain, practitioners also have an obligation to include these considerations when evaluating risk/benefit balance of epidural blockade and to inform patients under the principle of informed consent. These may appear as three isolated events, therefore, but they carry significant implications for all practising anaesthetists, employing institutions and professional bodies if we are to meet the expectations of the CMO and the public. It is vital that there is clarity as to where responsibility lies and that the efforts of all parties are co-ordinated. Integral to the safety agenda is the refinement of data collection and critical incident reporting to ensure that there is adequate information on the relative contribution of human factors and system failures to drive the most appropriate solutions. At a time when the concept of professionalism and self-regulation is in the spotlight, the specialty will be judged on how it responds to these recurrent events and associated challenges.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.