The effective introduction of Lifebox pulse oximetry to Malawi

Pulse oximetry was introduced into anaesthesia practice in the early 1980s, with the first practical commercial units becoming widely available from 1982. Before pulse oximetry, assessment of oxygenation in routine cases was usually based on visual inspection of the skin and mucosa for cyanosis, which was notoriously insensitive and, in many circumstances, not possible. This was especially problematic in poor lighting conditions or with darker skinned individuals. For complex cases, intermittent blood gas analysis from arterial blood samples was used, but blood gas machine access was costly, often restricted, and intervals between samples were often long. Pulse oximetry heralded the possibility of prompt identification of a whole range of problems including failure of adequate oxygen delivery, ventilatory insufficiency, and shunting. Detection of failure to oxygenate, such as with oesophageal intubation, whilst a late sign, could be corrected before hypoxic injury occurred. Likewise, the importance of pre-oxygenation and minimal delay in intubation in obstetrics, the identification of endobronchial intubation, and the management of one-lung anaesthesia in thoracic surgery, all became optimised as part of routine clinical practice with pulse oximetry. Pulse oximetry has therefore been a ‘mandatory’ monitoring requirement, where available, world-wide for over 30 years 1. Qualitative review of anaesthesia incidents has suggested that the pulse oximeter is a life-saving device, with Runciman concluding that oximetry could have prevented up to one-third of the anaesthesia-related deaths that were reported to the landmark Australian Incident Monitoring Study 2, 3. It is surprising then, that in practice, many studies that have been done do not demonstrate a benefit in clinical outcomes with pulse oximetry 4 – despite clear evidence that it reliably detects arterial hypoxia. Is this a failure of evidence-based medicine to prove what we ‘know’ or is it a result of complex confounders, ethical barriers, and unavoidable bias? This failure to confirm qualitative evidence with quantitative data most likely reflects a persistent detection bias (an ethical need to intervene if a known potentially harmful event occurs), and also the low frequency of important, reversible, but potentially catastrophic events (e.g. undetected oesophageal intubation, ventilation failure, or oxygen supply disconnection). We know that hypoxia can be bad if prolonged and profound, but the impact of shorter undetected periods may not be clinically relevant in the majority of patients. However, most studies simply examined the frequency of desaturation and, not unreasonably, assumed that a sub-set of these would be more severe, and have negative clinical consequences. In most high-income countries, we practice with the benefit of many resources to quickly address the causes of desaturation, and thus again avoid most preventable serious outcomes. Pederson et al. reflected this when they observed that it is possible that pulse oximetry may have a greater impact on outcomes in other geographical areas with less comprehensive provision of healthcare 4. Because pulse oximetry is accepted, prima facie, as an essential tool for safety of anaesthesia, there is no ethical possibility of including patients in a study where they would be at risk of hypoxia but might be randomly allocated to not be monitored with pulse oximetry 5. Unfortunately, despite the universal acceptance of the importance of pulse oximetry in anaesthesia, there are still many countries where it has not been widely introduced. The paper by Albert et al. in this issue of Anaesthesia 6 provides us with a unique insight into the real benefits that were achieved following the introduction of Lifebox pulse oximeters combined with appropriate training in Malawi, as measured by a reduction in hypoxic episodes (SpO2 < 90%). Before the training course around which this study is based, only one-third of hospitals had pulse oximeters available in every operating room. Desaturation episodes decreased from 17.2% to 6.5% over the first hundred cases following the intervention, which is a significant and clinically relevant improvement. This reduction in desaturation frequency suggests earlier detection and response – a hallmark of vigilance and good clinical practice. The authors note that with the limited numbers studied and the crude outcomes measured, there can still be no conclusion regarding morbidity or mortality. Despite this, in the small sub-set of case examples provided in the clinical environment of anaesthesia practice in Malawi, the monitoring of pulse oximetry allowed the detection and rectification of a machine disconnect, a tube obstruction, and also procedurally-related desaturations – all of which could plausibly have led to severe adverse outcomes. This study also provides insight into the real-world practical impact of the Lifebox programme. After eight months, 82% of oximeters were locatable and 97% of these were in routine use. This is a remarkable uptake and retention of a clinical programme, and attests to the ‘face validity’ of the intervention. Malawi is a sub-Saharan country with significant geographic and economic challenges. One of the catalysts for the Lifebox programme was the inclusion of pulse oximetry in the World Health Organization Surgical Safety Checklist 7, which recommended the use of a pulse oximeter for all procedures involving anaesthesia. At the time of the introduction of the checklist it was estimated that there were more than 77,000 operating theatres without a pulse oximeter 8, and it has previously been noted that this number did not include recovery areas, high dependency units and intensive care units 9 – all areas which should benefit from oximetry. The Lancet Commission on Global Surgery 10 has called for growth in the provision of safe surgery and anaesthesia in low- and middle-income countries and this will mean more operating theatres and more oximeters. Programmes such as Lifebox are an important part of this upscaling of safe surgical activity. The retention of the equipment and knowledge in Malawi, despite many challenges and recent major floods 6, reflects the relevance and robustness of the Lifebox program. Implementation of the checklist in sub-saharan Africa has been found to be a challenge, as described by Lilaonitkul et al. 11, although the high acuity of patients presenting for surgery in this part of the world 12 means that the opportunity for significant potential benefit in decreased morbidity and mortality by checklist and oximetry use is evident. In the article by Albert et al. 6, a ‘preventable anaesthesia mortality’ in Malawi of approximately 1 in 500 is described, which is orders of magnitude higher than those published for high income countries, which are typically 1 in 50,000 or lower (the ‘preventability’ at this extreme is of course an ongoing discussion) 13. This highlights the potential benefits of what is taken for granted and difficult to ‘test’ in better resourced environments. Demonstrating effectiveness of an education program in resource-poor settings is challenging. There are many short-course programmes that have been designed for low income countries. A number of these courses have demonstrated an improvement in participant's knowledge at the end of the course 14. Few have demonstrated sustained improvement in knowledge, let alone patient care beyond the short term 15. This study has demonstrated the effectiveness of a standardised training programme, in the form of a short course, and post-training logbook use. The training was effective in improving knowledge from baseline, even though the participants were already trained and practicing as Anaesthesia Clinical Officers, and at eight months there was retention of knowledge and evidence of continuing practice improvement, as measured by reduction in oxygen desaturation events. Why did this intervention work? The logbook may serve as a reminder of the principles of oximetry and management of hypoxia, as taught during the training program, or as a tool for reflective practice. This study is evidence that those planning an educational project in low-resource areas should plan beyond a short course and ensure that knowledge and clinical decision making taught during their programme is reinforced in the clinical environment in the months following. This study also highlights the importance of training for equipment donations. Lack of user training has been cited by the World Health Organization as a major reason why medical equipment donations often do not become operational, let alone effective, in strengthening health systems 16. Probably we will never know what the true cost-benefit is for pulse oximetry in anaesthesia beyond qualitative data, or from uncontrolled pre- and post-intervention studies looking at high-level outcomes such as crude mortality. We should recognise this, and acknowledge that it is a powerful and now relatively inexpensive tool that can be effectively introduced worldwide to provide an opportunity for early clinical intervention and management of hypoxic episodes, and also most likely heightens vigilance and detection of a larger range of related ventilatory and circulatory disturbances. RM has contributed to the development of Lifebox education materials and has worked as a volunteer on Lifebox projects. No external funding or other competing interests declared.