Pre‐operative gastric ultrasound – should we look inside Schrödinger's gut?

Abstract

Last year, Gagey et al. reported their prospective cohort study on the use of pre-operative gastric ultrasound before non-elective paediatric surgery 1. They concluded that the use of gastric ultrasound significantly increased the appropriateness of anaesthetic induction plans, and that it should be used more routinely in such circumstances. Their work has attracted discussion on statistical considerations such as the sensitivity/specificity and positive/negative predictive values of sequential tests 2. In the new 'Clinical Consequences' series of articles to be published in Anaesthesia, the implications of recently published research for clinically practicing anaesthetists will be discussed in depth. This first analysis will focus on the growing popularity of pre-operative gastric ultrasound and whether this should become part of routine clinical practice. The 4th National Audit Project of the Royal College of Anaesthetists and the Difficult Airway Society (NAP4) 3 highlighted pulmonary aspiration of gastric contents as the most common anaesthesia-related cause of death, with approximately half of all cases involving tracheal intubation 4. A common theme was poor pre-operative clinical judgement and inappropriate selection of anaesthetic technique. Pulmonary aspiration is a rare event, with NAP4 suggesting an incidence of around 1:350,000 cases 3, although a large number of anaesthetic, patient and surgical factors likely contribute. The incidence in children is greater, at around 9.3:10,000 cases 5, which is reasonable grounds for consideration of pre-operative gastric ultrasound. However, in addition to gastric content composition and volume, a range of other risk factors for pulmonary aspiration have been implicated 6, although those such as underdosing of anaesthesia, analgesia and/or neuromuscular blockade at the point of tracheal intubation are difficult to measure and study 7. The proposed treatment effect of pre-operative gastric ultrasound is the avoidance of pulmonary aspiration. However, there will never be a randomised control trial adequately powered to detect a reduction in this end-point. If the baseline risk of pulmonary aspiration in a non-obese, fasted patient is 1:10,000 (0.0001%) and assuming that gastric ultrasound could reduce this by 50% (which is very unlikely), then approximately 1.8 million patients would be required (assuming p < 0.01 and 90% power). There is also the potential for false negative examinations in non-fasted and/or emergent patient populations, where gastric ultrasound may wrongly predict a low risk of pulmonary aspiration and which results in the inappropriate use of a supraglottic airway device (SAD). In 2011, the European Society of Anaesthesiology issued their consensus guideline on peri-operative fasting 8. The message was broadly similar to previous guidance, that 2 h and 6 h should pass following the ingestion of clear fluids and solids, respectively. The evidence for this and other recommendations is limited by the rarity of pulmonary aspiration and the difficulties investigating the association between gastric volume, composition of ingested matter and the incidence of pulmonary aspiration. Given the problems conducting randomised controlled trials related to major airway complications 4, consensus is broadly determined by the degree of risk with which clinicians are willing to tolerate. New consensus goes further and states all children can take clear fluids up to 1 h before general anaesthesia due to the detrimental psychological, physiological and metabolic effects of prolonged fasting. In addition, it was felt the risk of pulmonary aspiration may have been overstated 9. Similar guidance should perhaps be considered in adults to prevent unnecessary fasting, as there is no reason why an older child is physiologically different to a fit and well adult. However, many clinicians still perceive the risk of pulmonary aspiration (and the potential for criticism and medicolegal claims) to be greater than any clinical benefits, especially given that adult fasting guidelines are long enshrined in routine practice. Proponents of gastric ultrasound suggest that its use would reassure clinicians and add support to novel fasting protocols. Establishing a link between gastric volume as measured with pre-operative gastric ultrasound and the risk of pulmonary aspiration may negate the requirement for traditional clinical assessment 10. The anaesthetic technique could be better planned for such patients, such as choosing a rapid sequence induction or avoiding the use of a SAD. This may nevertheless expose the patient to a new range of risks such as failure of tracheal intubation and its sequelae 4 and the requirement for neuromuscular blockade, which may increase the incidence of postoperative pulmonary complications 11 and accidental awareness during general anaesthesia 12. Alternatively, gastric ultrasound may reveal an empty stomach despite clinical assessment suggesting otherwise, which may support the avoidance of rapid sequence induction or tracheal intubation. Erroneous clinical assessments in the context of non-elective paediatric surgery may be due to variable gastric emptying, which is dependent on factors such as the underlying pathology, for example, the delayed emptying seen in trauma or sepsis. In adults, erroneous clinical assessments may result from the influence of comorbidities on gastric emptying such as labour, renal failure and diabetes mellitus. The question of whether or not gastric ultrasound is superior to clinical assessment alone with regard to predicting the risk of pulmonary aspiration is, therefore, valid and worthy of pursuit, not just for paediatric surgical emergencies, but for all patient populations. This is clearly a challenging situation, but if a gastric ultrasound had not been performed, the situation would have arguably been more straightforward. The increased popularity of point of care ultrasound, both clinically and as a research tool, presents other questions. First, how best to prove a treatment effect associated with pre-operative gastric ultrasound? As discussed above, we are unlikely to see evidence that gastric ultrasound prevents pulmonary aspiration, and surrogate outcomes are commonly selected. Gagey et al. chose the change in anaesthetic management following gastric ultrasound as their outcome measure 1, and this is reasonable, but we will never know if the correct course of action was taken. To assume gastric ultrasound leads to the correct course of action, every time, is to assume 100% sensitivity and specificity 2, which is highly unlikely. An interesting aspect of the methods used, however, was the combination of qualitative ultrasound assessment and gastric content quantification as measured with nasogastric tube aspirates 1. They determined the sensitivity and specificity to be 94% and 78%, respectively, even with inconclusive examinations included. Such methods have previously been employed by Bouvet et al. in the derivation of a model to determine gastric volume from antral cross-sectional area measurements in adults 14. Although it is also possible to quantitatively determine gastric volume through antral cross-sectional area measurement in children 13, 15, qualitative gastric content assessment may better lend itself to clinical practice due to its relative simplicity (Fig. 1). The clinical importance of the methods used and conclusions drawn by Gagey et al. are therefore extremely relevant to future research and clinical practice. Second, due to the limited availability of 'qualified' operators, Gagey et al. recruited participants only when operators were available 1. Although there are clearly defined criteria for accreditation in focussed cardiac ultrasound, and operators are usually described as having achieved this when used for research purposes 16, no such guidance exists for pre-operative gastric ultrasound. Given the precision of thresholds in gastric ultrasound (0.8 ml.kg−1 10 to 1.5 ml.kg−1 7), questions regarding inter- and intra-operator variability must be asked, more so when the only marker of experience is having conducted ≥ 50 gastric ultrasounds 1. Third, due to this limited pool of expertise and experience, governance processes around pre-operative gastric ultrasound must be considered. For example, it is unlikely that such images are stored, reported and peer reviewed, which would be necessary before implementation into routine clinical practice. A final question is of how the examination should be conducted. Pre-operative gastric ultrasound is an emerging science, but a recent algorithm for adults suggests qualitative examination to first determine if the stomach is empty and if not, whether the contents are solid or liquid 17. Solid contents are deemed 'high-risk' and fluid contents can be further evaluated with a volume assessment. Several volume assessment methods have been described with those reported by Bouvet et al. 14, Perlas et al. 18 and Spencer et al. (paediatric) 15 most commonly cited. There remain unanswered questions about: whether qualitative, quantitative or mixed methods should be used; which mathematical model should be selected; where to set the 'at-risk' gastric volume threshold and on the role of nasogastric tubes and gastric aspirates. A likely clinical consequence of the study by Gagey et al. 1 is an increase in the popularity and clinical use of pre-operative gastric ultrasound for non-elective paediatric surgical patients. This adds to a growing body of evidence suggesting gastric ultrasound may be superior to fasting consensus and clinical assessment for a range of scenarios and patient populations 17. If we think about gastric contents in a manner similar to Schrödinger's well-known thought experiment, the stomach is neither full nor empty before anaesthesia and surgery 19, and the temptation to look inside is hard to resist. Although gastric ultrasound seems to be an easy technique to learn and perform 20, there remain several unanswered questions relating to clinical practice standards, image acquisition and analysis, clinical decision making and clinical governance. Whether taking a look inside 'Schrödinger's gut' with an ultrasound probe is superior to clinical judgement requires, in our opinion, further study, more debate and careful thought on how best to integrate its use into routine clinical practice. MC is the Social Media Editor of Anaesthesia. MDW is an Editor of Anaesthesia and a council member of the Neuro-anaesthesia and Critical Care Society of Great Britain and Ireland. We thank Dr K. El-Boghdadly for the images provided that were obtained with patient consent.

The authors' abstract, as published at the source. Anaesthesia, 2018 · DOI ↗