Replacing Luer connectors: still work in progress

At the time of writing, all NHS trusts and independent healthcare institutions in England and Wales will be required to take action to use spinal needles with non-Luer connectors by 1 April 2011 [1]. This important target follows repeated calls to design a solution to make it physically impossible to connect medications intended for intravenous administration to spinal needles [2–5]. However, < 6 months from the implementation date, new devices are only just beginning to appear on the market and there is only one clinical study of a connector prototype in the peerreviewed literature [6]. In this issue of Anaesthesia, Cook et al. describe the results of bench tests undertaken by experienced clinicians that demonstrate that two new connectors required further modification (n.b. both have been modified since the study) [7]. Why has it been so hard to find a simple engineering solution to a problem that was first described more than 40 years ago [8]? German medical instrument maker Hermann Wülfing Luer designed the Luer connector in 1896 and suggested that the simple 6% taper for glass bottle stoppers be used to provide a secure connector for intravenous devices. A threaded collar was described by Fairleigh Dickenson in 1925, and by the 1960s, the Luer-Slip and Luer-Lok connectors were the most common in clinical use [9]. Unfortunately, the success of the Luer connectors in healthcare has also been a major failure as they risk misconnection between a multiplicity of different devices and provide no protection against wrong-route drug errors [10]. The first death due to accidental intrathecal vincristine was in 1968 and since then, there have been at least 58 deaths worldwide, probably with many more unreported [11]. Enteral feeds, epidural local anaesthetic infusions, capnography tubing, noninvasive blood pressure devices and oxygen tubing have all been connected to intravenous devices, resulting in numerous fatalities. In the words of a recent Sentinal Event Alert from the Joint Commission, ‘if it can be done it will be done’ [12]. In the UK, the most recent case of intrathecal vincristine was in 2001 and resulted in the death of Wayne Jowett, a teenager who was in remission from leukaemia and was to receive his last treatment [2, 3]. This tragic accident occurred the year after the publication of the Department of Health’s report ‘An Organisation With a Memory’, which highlighted the importance of learning from rare, but serious, adverse events in healthcare such as vincristine medication errors [13]. The external accident inquiry into the death of Wayne Jowett carefully documented the latent conditions and active failures that enabled the intravenous chemotherapy to be transferred from the pharmacy to the room where the intrathecal chemotherapy was to be given. The universal Luer connector enabled the vincristine syringe to be attached to the spinal needle, thus removing the final safeguard for the patient [2]. Awareness of human factors and system errors will never provide fail-safe protection and a physical solution to prevent accidental misconnections would seem essential. Since then, national guidance on the administration of chemotherapy has been updated, and the UK has followed other countries in stipulating that intravenous minibags should be used for all vinca alkaloids, although in the UK, vinca alkaloids are still administered by syringe for children under 10 years as there are concerns about extravasation injury [14]. Intrathecal injection from a minibag would be exceptionally unlikely due to the formulation and volume contained. However, a non-interchangeable system to ensure that drugs intended for the intravenous route cannot be given intrathecally still remains appropriate. In anaesthetic practice, wrong-route errors are much more common when epidural drugs are given intravenously or vice versa. A recent tragedy in Australia, in which chlorhexidine cleaning solution was drawn into an epidural loss-of-resistance syringe reminds us that a new connector cannot prevent the wrong drug from being given [15]. There were 20 wrong-route epidural injections reported to the National Patient Safety Agency (NPSA) between 1 January 2005 and 31 May 2006, including six incidents in which epidural medications were given via the intravenous route [16]. In recent years, there have been four deaths reported in the UK due to intravenous bupivacaine, one occurring after the NPSA issued a Patient Safety Alert concerning wrongroute epidural infusions [17, 18]. Again, an engineering solution would seem essential, and this is the subject of ‘Part B’ of the NPSA Safety Alert to replace all epidural, spinal and regional anaesthesia connectors with non-Luer connectors by April 2013 [18]. The best example of an engineering solution to solve risk in anaesthesia is the hypoxic guard developed to prevent the administration of hypoxic gas mixtures to patients. This practical solution has undoubtedly saved many lives. There have been efforts to encourage development of new non-Luer connectors for spinal needles, and a number of prototypes have been put forward for testing. The Department of Health proposed that European manufacturers should design a replacement for the Luer connector in 2002 [4] and commissioned research through the Department of Health Patient Safety Research programme [19]. This included a hazard analysis, evaluation of prototypes in simulators, and pilot evaluations of new devices in clinical practice. In 2004, the NPSA conducted a risk assessment that concluded that all spinal Anaesthesia, 2010, 65, pages 1059–1068