Cricoid pressure: apply – but be ready to release

It is this last question that Andruszkiewicz et al. 6 have specifically addressed, and which will be the primary focus of this editorial. However, other subsidiary, but nonetheless important, questions include: is CP in obstetrics a special case?; do we perform CP optimally?; and does CP prevent stomach inflation during mask ventilation? Space does not permit us to consider these other aspects here, but we will discuss the primary questions we have posed. In 1946, Mendelson published his seminal paper on pulmonary aspiration in obstetric patients undergoing facemask anaesthesia for operative delivery, reporting that the incidence of aspiration was 1 in 667 (mortality 1 in 22,000) 7. More recent estimates in the general surgical population suggest an incidence of between 1 in 2000 and 1:3400, with mortality ranging between 1:45,000 and 1:70,000 8-10. The risk appears significantly greater in obstetric populations 11, and those undergoing emergency procedures (1 in 900) 8. ASA physical status also correlates with risk of aspiration, with ASA 3–4 patients at highest risk 12. Although some of these incidences seem low, if aspiration does occur in adults, morbidity appears to be high, with one study showing 47% of patients developing pneumonitis and 17% requiring artificial ventilation 9. In paediatrics, however, aspiration is uncommonly associated with pneumonitis 13. Data from the fourth National Audit Project (NAP4) revealed that aspiration was the commonest cause of mortality, with 42 cases in 2,872,600 general anaesthetics, 36 of which were aspiration of gastric contents. Mortality rates appeared to be greater than previously reported, with eight deaths from gastric aspiration 14. Morbidity was also significant in those who suffered aspiration, with 29 out of the 36 patients with gastric aspiration requiring prolonged admission to an intensive care unit (> 48 h). However, the majority of primary aspiration cases in NAP4 occurred during maintenance of anaesthesia or around the time of tracheal extubation (14/23), rather than at induction. The authors highlighted several cases where rapid sequence intubation with CP was indicated but not used, and where tracheal intubation was indicated but a supraglottic airway device (SAD) used. NAP4 did not report any case of harm arising from use of CP. In other studies, it is clear that aspiration can and does still occur despite the application of apparently correct CP 15-17. Survey data suggest that 11–14% of anaesthetists have witnessed aspiration during the application of CP at some point in their career 16. In a study of 297 patients undergoing emergency tracheal intubation, 12 patients had a new infiltrate on chest X-ray considered secondary to aspiration, nine of which had CP applied 18. In a review of nearly 5000 general anaesthetics for obstetric procedures in Malawi, 11 deaths were considered as a direct consequence of aspiration, nine of which had cricoid pressure performed 19. A further study from the obstetric population of 2114 parturients undergoing general anaesthesia without CP, and with careful bag mask ventilation, showed no episodes of regurgitation or aspiration, and no failed intubations 20. Thus, it is clear that CP should not be viewed as a panacea for all patients, but as part of an overall strategy to reduce the risk of aspiration. Since it is impossible to conduct randomised trials for so rare an event as aspiration, the question of whether CP is effective will remain an open one, amenable only to interrogation of large datasets (like NAP4) or expert opinion (like this editorial), which together might lead to consensus. In the absence of robust clinical data, it is necessary to consider other lines of evidence such as anatomical. Does the use of cricoid pressure actually occlude the oesophagus, as has been long supposed? The study by Andruszkiewicz et al. appears to be (perhaps remarkably) the first to use ultrasound to assess the efficacy of cricoid pressure 6. A number of other imaging techniques have previously been used anatomically to define the effect of CP. In his original study, Sellick assumed that the oesophagus sits directly posterior to the cricoid ring in the axial plane. However, a retrospective review of cervical computed tomography scans in healthy patients showed that in fact the oesophagus sits postero-lateral to the cricoid ring in 50% of subjects, predominantly leftwards 21, with lateral displacement increasing with the application of CP 22, findings confirmed by Andruszkiewicz et al. 6. Yet, all this may be irrelevant. According to Rice et al. 23, who analysed magnetic resonance imaging scans in awake volunteers, it is not the oesophagus which is compressed with CP, but the post-cricoid hypopharynx, one centimetre proximally. Their study suggested that even with oesophageal displacement, an average 35% reduction in the anteroposterior diameter of the post-cricoid hypopharynx occurred, with the authors suggesting this effectively represented complete occlusion of the lumen. They postulated that the cricoid ring and post-cricoid hypopharynx, connected by a complex network of muscular and ligamentous structures, act as a single anatomical unit. Thus, even if the cricoid ring is not fixed against a cervical vertebra, occlusion of the hypopharyngeal lumen still occurs against adjacent muscles. This is potentially relevant as an anatomical unit dependent on muscular structures is likely to be affected by the use of anaesthetic agents, although exactly how and to what extent is unclear 24. A study in 79 healthy anaesthetised and paralysed patients found that 30N of cricoid force made it impossible to insert a 4 mm nasogastric (NG) tube 25. They used videolaryngoscopy to confirm occlusion of the hypopharynx, providing a functional and anatomical demonstration of CP efficacy in anaesthetised patients. Although this study does not prove the effectiveness of CP against liquid matter, it does, at the very least, suggests CP results in a mechanical barrier to larger particles, with a partial barrier presumed to be superior than no barrier at all. Moreover with CP, there is a noted inability to pass several types of SAD whose tips enter the oesophagus, and this is also a marker of obliteration of the post-cricoid space; the mechanism and implications are similar to failure to pass an NG tube with CP 26-28. Interestingly, there appears a lesser effect on the i-gel, whose tip is more rounded and which does not extend into the top of the oesophagus; this perhaps confirming the anatomical argument 29. Intriguingly, Andruszkiewicz et al. 6 found that a novel pressure site (paralaryngeal) achieved the desired oesophageal compression, which CP itself did not. Taking these findings together therefore, it is possible that (a) appropriate cricoid pressure compresses the post-cricoid hypopharynx in an efficacious way; (b) inappropriate cricoid pressure (eg, perhaps something that is fact paralaryngeal) could also achieve effective oesophageal compression. At a more fundamental level, these results raise the questions: "what is proper cricoid pressure, and how is its effect best assessed?" We are not in a position to answer these yet. It seems Sellick (and many subsequent authors) did not entertain the possibility that the manoeuvre might be harmful. However, for any intervention whose efficacy is debatable, it is important to consider the overall risk-benefit balance. There is little doubt that CP interacts with all aspects of airway management, even when properly applied in the relevant studies 5. It may increase inspiratory pressures, reduce tidal volumes, and result in partial or even complete airway occlusion 30-32. On laryngoscopy, laryngeal structures can appear distorted during CP 33, with failed intubations almost eight times as frequent 14. Following failed intubation, keeping CP on can increase the risk of difficult or impossible facemask ventilation 34. All these are likely exacerbated by improper application. Fibreoptic studies have shown vocal cord pinching in over 50% of patients at 30N 30, with complete cord closure in 50% of subjects at 40N, a finding replicated in other adult and paediatric populations 35, 36. However, as Cook highlights in recent correspondence 2, many of these problems can be negated with the use of concurrent videolaryngoscopy. In a meta-analysis, CP reduced the chance of successful SAD insertion from 94% to 67% 37. This is expected given that the tip of a classic laryngeal mask airway should sit in the lumen of the post-cricoid hypopharynx. Personal experience and case reports testify to difficult or impossible tracheal intubation with CP applied, becoming easy on its removal 38, 39. Indeed, authoritative guidelines show complete consensus: if CP is impairing airway management or oxygenation, it must be removed 40, 41. There is some evidence to suggest that CP may actually increase the risk of nausea and vomiting. At present, the current guidelines recommend 'gentle' cricoid force pressure (10N) with the patient awake, increasing to 30N following loss of consciousness 40. However, studies have shown that applied forces of just 20N may cause relaxation of the lower oesophageal sphincter 42, potentially triggering vomiting in a patient deemed high risk of aspiration. This has been confirmed in anaesthetised patients, with one study showing a 44% reduction in barrier pressure when CP is applied at a force of 30N 43. If active vomiting occurs with CP applied, and the patient is either fully or partially awake, CP should be removed immediately, due to the risk of oesophageal rupture 44, 45. Thus, the very thing we have set out to prevent may in part be triggered by our actions. In the light of the uncertainties, it is perhaps surprising that the courts appear to favour the use of CP. Athanassoglou and Pandit 4 cite the judgement in M vs. E Hertfordshire HA 1991, where an anaesthetist failed to apply CP and the patient suffered aspiration pneumonitis. The judge ruled that: "We cannot assert that CP is not effective until trials have been performed (i.e. we must assume its efficacy), especially as it is an integral part of anaesthetic technique…that has been associated with a reduced maternal death rate from aspiration since the 1960's." United States lawsuits appear more common after aspiration than in other settings, and are settled for higher amounts if CP is not applied 46. The legal onus seems to place the burden of proof on those who wish to abandon the use of CP. Yet, courts are advised by medical experts, who in turn need to be persuasive, and it is apposite for us to ask: if CP were to be introduced now as a novel technique, what evidence would be needed for its universal adoption? We suggest three things: (a) a plausible functional anatomical mechanism; (b) evidence from large datasets that its use is associated with fewer cases of regurgitation/aspiration; and (c) lack of harm from its use. As discussed above, each of these is open to question and in particular, Andruszkiewicz et al. 6 have provided fresh insight into the first of these. The topic is ripe for further research, and the paralaryngeal pressure effects discovered by Andruszkiewicz et al. 4 are worthy of specific study, but the current uncertainties do not stop us from reaching a pragmatic balance in our practice. There is indeed some evidence that cricoid pressure does occlude to good effect the relevant structures in the neck that plausibly prevent regurgitation. Therefore, there is little reason not to apply it 'by default' in the context of a rapid sequence induction. Little or no direct damage will result; but any expectation that it is a panacea, or that proper application alone avoids adverse effects, would be misplaced. However, if there is any hint of difficulty with airway management (and that extends to active vomiting in the partially anaesthetised), then the first action must be to release cricoid pressure immediately. This advice is already within published guidelines 41 and has recently entered correspondence pages 2, and we think it worth reiterating. This balanced approach, we think, will enable the anaesthetist to show good intent as desired by some peers and by the courts, whilst not endangering the patient should that intent be misplaced or counterproductive. JP is an editor of Anaesthesia, and this editorial was therefore subject to additional external review. AP is President and JP Scientific Officer, Difficult Airway Society: the views expressed are their own and not those of the Society. No other conflicts of interest declared.