The need for capnography standardisation

Patients continue to die from unrecognised oesophageal intubation despite decades of awareness, multiple national audits [1, 2], safety campaigns [3], educational initiatives [4] and clinical guidelines [5]. When Glenda Logsdail died during a routine appendicectomy in August 2020, the coroner's regulation 28 report identified confusion between capnography and other waveforms as a contributing factor [6]. Within the past 10 years, Joseph Parker, Sharon Grierson and Peter Saint also died from unrecognised oesophageal intubation, prompting other prevention of future deaths reports [7-9]. Most recently, an Australian inquest in January 2026 identified an unrecognised oesophageal intubation death where clinicians mistook a respiratory waveform for capnography [10]. A common feature in all of these deaths was clinicians failing to recognise oesophageal intubation in time to prevent catastrophic harm, despite a present and functional capnograph. A recent editorial describes the adoption of “sustained exhaled carbon dioxide” as the standard for interpreting capnography traces; an important advance in defining the only way to exclude oesophageal intubation [11]. This principle now forms a cornerstone of the 2025 Difficult Airway Society guidelines for management of unanticipated difficult tracheal intubation [5], underscoring the central role of capnography. However, even the most rigorous interpretive criteria cannot function if clinicians cannot identify reliably which waveform they should be observing. Our survey of capnography displays across UK hospitals revealed the scale of this problem [12]. We identified 36 distinct waveform configurations (Fig. 1), with individual hospitals displaying up to 13 different variants. In 32% of clinical areas, monitors displayed other waveforms with identical morphology to capnography, creating obvious potential for misidentification. Only 15% of 9052 capnographs examined conformed to the Safe Anaesthesia Liaison Group (SALG) standard [13]. Surveying approximately 30% of UK hospitals and identifying 36 distinct variants means the true national figure can only be higher. A solution exists through professional consensus, yet implementation has stalled. A survey of Difficult Airway Society members by Majumdar et al. examined strategies to reduce unrecognised oesophageal intubation and offers some insight into why progress has been limited [14]. Free-text comments mentioned display standardisation repeatedly as a necessary intervention, but the interpretation focused predominately on training initiatives and videolaryngoscopy adoption. Capnography display identification received only passing mention despite its direct relevance to the waveform interpretation failures. The hierarchy of controls in safety management places system design above administrative controls and education [15]. This framework is accepted widely in safety-critical industries [16], yet its application in anaesthesia has been inconsistent. Educational solutions are often implemented more readily than design changes. These require no negotiation with manufacturers, no procurement battles and can be developed within departments. Training programmes preserve existing workflows and avoid the implication that experienced clinicians need protection from poor design. When the respondents to the survey by Majumdar et al. suggested that unrecognised oesophageal intubation “could not happen to a competent anaesthetist” [14], they articulated a view that places confidence in individual vigilance over system robustness and exemplify an optimism bias that is an impediment to injury prevention. Glenda Logsdail died during a routine procedure in normal working hours [6]. It was not an emergency procedure or a difficult airway. Her death, and the death of Joseph Parker [9], suggest that relying primarily on clinician vigilance, however well-trained, may be insufficient when system design introduces unnecessary complexity. Consider the cognitive demands placed on clinicians who move between the operating theatre and emergency department. They encounter capnographs displayed in different colours, with different morphologies, in different screen positions. Sometimes these waveforms are visually identical to pressure or flow traces displayed simultaneously that also vary in screen position. During routine cases, this variation is manageable. During a crisis, when cognitive resources are fully deployed and seconds matter, the adaptation burden may contribute to delayed recognition. Critically, adopting “sustained exhaled carbon dioxide” as the interpretive standard [11] requires clinicians to identify which trace is the capnograph. When displays show multiple waveforms with identical morphology this preliminary step itself becomes prone to error. The coronial reports into the deaths of Glenda Logsdail [6] and Ms Guyula [10] identified confusion between capnography and other waveforms as contributing to delayed recognition. We cannot quantify exactly how often display confusion is the proximal cause of failed recognition, but coronial reports indicate it occurs. The suggestion for two-person verbal confirmation of capnography [14] is another example of administrative control. This also fails to note that both anaesthetists who were present during the attempts to intubate the trachea of Joseph Parker contemporaneously misinterpreted a capnography trace [9]. We seem to be repeating the pattern of identifying a systems problem, reaching for procedural solutions and leaving the underlying design unchallenged. The medication safety literature offers instructive parallels. Anaesthetists confront similar standardisation failures in drug packaging where markedly different medications with nearly identical presentations create latent conditions for error. Marshall and Chrimes describe how pharmaceutical manufacturers emphasise brand identity over distinctiveness between products, with positioning in storage drawers compounding misidentification risk when similar packaging is involved [17]. Technical solutions already exist, and Australian regulations now require distinctive packaging for neuromuscular blocking drugs, acknowledging that relying on clinician vigilance alone has proven insufficient. More fundamentally, Whitaker and Lomas found that prefilled syringes eliminated approximately 60% of human factors steps in medication preparation [18]. These interventions succeeded, not through educational campaigns, but through procurement specifications and regulatory requirements that made standardisation mandatory rather than aspirational. The capnography standardisation problem follows an identical pattern where there is a known hazard, an available solution and resistance rooted in concerns about cost, familiarity and implementation complexity rather than technical feasibility. The anaesthesia community has implemented design-level interventions in the past when harm became undeniable. Wrong-gas delivery caused patient deaths until the pin index safety system made incorrect cylinder-to-yoke fitting mechanically impossible. Drug labelling errors led to the establishing of syringe label standards (ISO 26825). Widespread adoption required procurement mandates, not merely professional endorsement. These interventions are not perfect analogies to display standardisation, which operates at a different level of the safety hierarchy. They do, however, show that moving beyond educational solutions requires regulatory and procurement leverage. Each intervention took years to implement and occurred only after multiple deaths. Multiple deaths have been associated with capnography traces; we must act now before there are more. The technical capacity for standardisation exists. Many monitors display ECG traces in green and arterial pressure in red through market-driven standardisation that occurred without regulatory mandate. Clinicians specified these configurations in procurement, manufacturers responded and consistency emerged organically. This has not happened for capnography despite manufacturers having identical technical capability. For a safety-critical monitor where misidentification may contribute to deaths, relying solely on market forces appears insufficient. Our data suggest that over 44% of monitors would require manufacturer software modifications to achieve the standard [12]. While this represents real development costs, manufacturers already provide ongoing software updates and support throughout equipment lifecycles. Regulatory requirements or procurement specifications mandating SALG-conformant displays would compel manufacturers to prioritise these modifications, transforming standardisation from optional to inevitable. The balance between development costs and potential safety benefits favours modification, though we lack precise data on the magnitude of benefit to be gained. Implementation will require co-ordination across multiple stakeholders. International Organization for Standardization development typically takes 18–36 months through staged consensus processes involving national standards bodies, manufacturers, regulators and clinicians [19]. Once published, international standards remain voluntary unless they are mandated through procurement or referenced in national regulatory frameworks. Procurement offers the most direct lever to instigate change. However, even with immediate action, non-conforming devices may remain in service for many years due to equipment replacement cycles. If standardisation began today, widespread implementation across the NHS might not occur until 2030 or beyond. Standardisation will not eliminate unrecognised oesophageal intubation. Even with perfectly uniform displays, clinicians will fail to recognise the problem in some cases through inattention, misinterpretation or ambiguous clinical pictures. In safety-critical industries, standardisation is intended to mitigate predictable human limitations rather than compensate for individual error. Drug errors continue despite standardised label colours. Wrong-gas incidents occur despite pin-index systems [20]. This cannot be used as an argument against standardisation. The relevant outcome is, therefore, not absolute prevention, but reduction in the time to recognition. In addition, we should consider potential unintended consequences. Could excessive confidence in standardised displays reduce vigilance in other aspects of airway management? Might homogeneous displays create new vulnerabilities if clinicians over-rely on expected appearance? These concerns do deserve consideration but they do not justify maintaining the current level of variation. Standardisation removes one source of cognitive load and reduces misidentification risk when morphologically identical waveforms are displayed simultaneously. If standardisation prevents delayed recognition in a meaningful proportion of cases, it represents a justified and proportionate intervention, even if unrecognised oesophageal intubation is not eliminated entirely. The Royal College of Anaesthetists could consider incorporating compliance into the Guidelines for Provision of Anaesthetic Services [21]. More immediately, equipment leads at individual institutions should audit current capnography displays and take ownership of standardisation within their departments. Clinical governance and safety leads have a vital role in driving on-the-ground improvements, not only for unrecognised oesophageal intubation but also for other potential adverse events such as wrong site blocks. This approach is achievable rapidly and complements longer-term regulatory solutions. The SALG must continue to engage productively with the British Standards Institution to propose capnography display standardisation and inform the pending update of ISO 80601-2-55 (respiratory gas monitoring). Early manufacturer engagement would help establish realistic implementation timelines and identify technical constraints. Procurement teams across the NHS could begin specifying SALG-conformant displays in new equipment tenders, creating market pressure that signals regulatory change is likely. Individual departments should audit their current equipment and implement user-modifiable changes where feasible rather than waiting for mandates. Resistance is likely. Clinicians may prefer displays they find familiar, particularly if local incident rates appear low. Departments may argue that local variation has not caused identifiable problems. Manufacturers face genuine development costs and regulatory approval requirements. These are all legitimate concerns requiring practical solutions, not dismissal. Transition periods may create temporary increases in variation as some equipment is modified before others. The 1% of equipment with hardware limitations preventing conformity will eventually require replacement [12]. These challenges are real, but manageable, and have been navigated by our profession in other standardisations. They should inform implementation strategy, not prevent action. The technical capability for implementation is available. What appears lacking is the collective will to prioritise system redesign over continued reliance on training and procedural solutions [14]. The deaths have occurred. The question now is whether we will act on what we already know. We would like to acknowledge Dr Jennifer Proc for her continued support and the CaVa collaborators across the UK. ML is a Trainee Fellow of Anaesthesia. No external funding or other competing interests declared.