Cricoid pressure was used in 1774 by Monro to prevent gastric distension during inflation of the lungs in the recovery of persons 'drowned and seemingly dead' . The manoeuvre was re-introduced by Sellick to control regurgitation of stomach contents during induction of anaesthesia . Despite some controversy about the clinical usefulness of cricoid pressure [3,4], it is widely regarded as a mainstay of safe airway management when correctly applied [4,5].
Cricoid pressure can be applied incorrectly in many ways: by not identifying the cricoid cartilage, inappropriate timing, incorrect direction and the use of too little or too much force. The errors may render airway management difficult or impossible [6-9]. The sequelae include retching, aspiration, hypoxia, direct trauma to neck structures [5,10] and oesophageal rupture , which may prove fatal. Only trained staff should be allowed to perform this manoeuvre. Correct application of cricoid pressure involves accurate identification of the cricoid cartilage and applying the appropriate force in the correct direction. Current recommendations are to apply 10 N (1 kg) to the cricoid cartilage when the patient is awake and to increase the force to 30 N (3 kg) once the patient has lost consciousness .
In this study, we designed and tested a biofeedback cricoid pressure trainer. The device can be assembled with material available in most operating theatres and is easily calibrated. We used our cricoid pressure trainer to assess staff performance and to teach and simulate correctly applied cricoid force in real-time.
We took a roll of adhesive plaster, of 3 cm diameter and 7.5 cm length, to model the cricoid and larynx and the area of the imaginary cricoid cartilage was marked on the roll. The roll was mounted onto a 100 mL 0.9% saline bag completely vented of excess air to simulate the soft tissue of the neck. The bag was connected to a PX-260® pressure transducer (Edwards Lifescience LLC, Irvine CA, USA) with a vial access spike (Fig. 1). An infusion system connected to the side-port of the pressure transducer may be used. Pressures were displayed on a Datex A/S 3® monitor (Helsinki, Finland). A calibration curve was produced for the range of 0-40 N using electronic scales (Seca model 727®) and various custom-made weights (1 kg = 9.81 N) to establish the relationship between force and pressure for the system. Prior to every assessment session the system was calibrated with the two target forces of 10 N (1.020 kg) representing that recommended for 'awake' application, and 30 N (3.060 kg) that is recommended for 'asleep' application.
Initial assessment of cricoid pressure
Thirty-six anaesthesia nurses and operating department practitioners were recruited to the study and completed a questionnaire to assess prior training and experience (Table 1).
A standardized explanation was given to each participant. With the subject unable to see the monitor they were asked to apply appropriate 'awake' or 'asleep' cricoid pressure to the model with the hand normally used for this manoeuvre, and to maintain the force until instructed otherwise. The mean pre-training pressures were recorded once a stable reading over 15 s was achieved.
The participants were then allowed to see the screen. The targets for 'awake' and 'asleep' cricoid pressures were shown and real-time biofeedback training performed. The visible pressure curve enabled self-adjustment of the necessary force. Trainer guidance was gradually reduced according to training success. A person was regarded as trained when he was able to achieve the appropriate forces within 5 s and maintain it over 30 s to within 5% of the target pressure.
Assessment of training
A second assessment was carried out and the results recorded as in the pre-training assessment described above. Performance was classified as successful when the applied pressure was within 30% of recommended 'awake' (7-13 N), and 20% of recommended 'asleep' (24-36 N) cricoid pressure.
Statistical analysis was carried out with the F-test of variance, and McNemar's test for matched pairs . Data were processed using Microsoft Excel® 2000 (9.0.3821 SR-1).
Of the 36 anaesthesia assistants tested, 27 were female and nine were male. The majority (30/36, 83%) were anaesthesia nurses and the remaining six (17%) were operating department practitioners. The volunteers had a median of 9.2 yr (range 0-32 yr) experience in medical assistance and had been working as anaesthesia assistants for 3.9 yr (0-15 yr). All participants (36/36) had had practical in-theatre training and some also had a lecture (19/36, 53%). Only four (4/36, 11%) participants had been trained on a cricoid pressure model. Half of the group had been trained by a consultant anaesthetist (18/36), the other half (18/36) by a senior colleague. None had ever participated in refresher training for cricoid pressure. All subjects felt confident in applying cricoid pressure. The majority of the tested individuals (31/36, 86%) recognized the need for regular re-training. None of the participants were able to specify the recommended force in the awake patient and only two (6%) knew the correct force of 30 N in unconscious patients. Nine (25%) had observed complications of rapid sequence induction as a consequence of incorrectly applied cricoid pressure: too little force and regurgitation (4/36, 11%), too much force and difficulty in intubation (3/36, 8%), retching (1/36, 3%) and inaccurate application (1/36, 3%).
Figure 2 shows the calibration curve between force in N and pressure in mmHg for the system used in this study. The relationship is described by the linear equation: force (N) = 0.428 × pressure (mmHg) + 0.05 (r2 = 1.000). All pressures were converted into force prior to analysis.
Figures 3 and 4 show the results of the group assessment before and after training for 'awake' and 'asleep' cricoid pressure.
For the target force of 10 N in this test series the mean (SD; range) force achieved was 12.6 N (6.6; 4.3-35.6) before training and 11.4 N (1.6; 7.8-16.7) after training. The variance of forces produced was reduced from 43.8 N2 before training to 2.7 N2 after training (P < 0.0001). Before training, 20 subjects (20/36, 56%) failed to be within the target range of 10 N ± 30%. After training only three (3/36, 8%) failed to achieve this objective (P < 0.0005).
For the target force of 30 N the mean (SD; range) force achieved was 27.7 N (13.1; 12.0-75.8) before training and 30.9 N (2.2; 24.9-35.6) after training. The variance was reduced from 171.3 N2 before training to 4.7 N2 after training (P < 0.0001). Before training, 26 (26/36, 72%) failed to apply pressure within 30 N ± 20%. After training none of the participants failed the assessment (P < 0.0005).
A Review article and an Editorial underlined the need for regular staff training in the correct application of cricoid pressure [4,5]. There are complicated and expensive devices for teaching the correct force for cricoid pressure [13-15]. On the other hand, a simple syringe or scale can be used for effective training [16,17]. Recommended training intervals range from weekly to every 3-6 months [16,18]. Meek and colleagues assessed cricoid pressure performance of 135 anaesthesia assistants attending the annual conference of the British Association of Operating Assistants in May 1997 and demonstrated that incorrectly applied cricoid pressure was common . This is consistent with the results of the 36 participants tested in our study. Before training, 56% of our participants failed to meet the accepted criteria for 'awake' cricoid pressure, and, more importantly, 72% applied incorrect 'asleep' cricoid pressure. Similar or even worse results have been described [15,17,19].
We used biofeedback training to improve our staff's performance. The participants were initially assessed, then trained to our defined criteria and consequently re-assessed according to an accepted standard . For training purposes we set strict pressure limits, but used the limits recommended in the literature for the follow-up assessment. While unblinded, each candidate was requested to apply the force to within 5% of the target pressure using visual feedback. We considered that the ability to achieve this in 5 s and maintain it for 30 s would simulate the demands of a normal rapid sequence induction. Following a short training session, only 8% of the tested individuals failed the 'awake' cricoid pressure assessment. All participants passed post-training assessment for 'asleep' cricoid pressure. For both 'awake' and 'asleep' cricoid pressure, we did not find the group means to differ much from the target. However, it is individual performance that matters and the group variance was significantly reduced after training (Figs. 3 and 4).
We favour the cricoid pressure trainer used in this study, as it is quick, easy and inexpensive to assemble with material present in most operating theatres. This would allow such set-ups to be implemented at different sites throughout a hospital at reasonably low cost. The device allows the individual to directly follow training success on a monitor and enables self-control and biofeedback adjustment of the target force without distraction. This concern was expressed by staff taking part in a recent study .
Our cricoid pressure trainer allows simulation and effective training of the correct perpendicular force recommended to occlude the oesophagus. It does not attempt to teach the correct identification of the cricoid cartilage nor correct direction of force. These skills can be best taught on volunteers or patients in the operating room.
Since force applied to the model cannot be converted directly into pressure in the system we calibrated our system using various weights. There was a close linear relationship between measured pressures and applied weight (Fig. 2). Each system has a characteristic slope. The regression line can change if the same system is used for prolonged periods or at different temperatures. We did not find this to be a problem within the same assessment session, but the system should be calibrated with the target forces before each assessment session. As the pressure-force relationship in our model was linear, calibration could be done quickly by applying only the 10 N = 1.020 kg and 30 N = 3.060 kg weights to the model.
Previous studies used arbitrary force limits to judge the success of the applied cricoid force. In an assessment of 20 qualified anaesthesia assistants, Cook and colleagues defined adequate cricoid pressure 'awake' as 5-20 N and 'anaesthetized' as 20-45 N . A force greater than 20 N can cause retching in conscious volunteers  with potentially catastrophic sequelae. There is very little published evidence for what might be a safe lower limit of cricoid pressure in awake patients; some would argue that none is required. In our study we considered the force range of 7-13 N (i.e. 10 N ± 30%) to be appropriate for awake patients. More studies have been carried out to evaluate a safe range for 'asleep' cricoid pressure [7-9]. Twenty newtons of cricoid pressure appear to be enough to prevent regurgitation into the pharynx and 30 N seems to be more than enough .
Airway deformation starts with as little as 20 N, and increasing cricoid pressure can lead to a clinically significant increase in airway obstruction and potentially difficult airway management [7,8]. In our study, we employed the amount of cricoid pressure recommended for unconscious patients . We assumed that the range of 24-36 N, i.e. 30 N ± 20% would be a clinically effective and safe cricoid pressure.
In summary, we described a cricoid pressure trainer assembled with material available in most operating theatres. Once calibrated for the target pressures it can be used to assess and train staff in the application of the correct cricoid force. Our cricoid pressure trainer allows individuals to learn the correct force via self-directed, visual biofeedback control.
We would like to thank all participants at The Royal Alexandra Hospital, Paisley, and Gartnavel General Hospital, Glasgow, UK, for taking part in the cricoid pressure assessment. We are also obliged to Dr. Ian Kestin for his contributions in the planning phase of this study.
1. Salem MR, Sellick BA, Elam JO. The historical background of cricoid pressure in anaesthesia and resuscitation. Anesth Analg
2. Sellick BA. Cricoid pressure to control regurgitation of stomach contents during induction of anaesthesia. Lancet
3. Benhamou D. Cricoid pressure is unnecessary in obstetric general anaesthesia. Int J Obstet Anesth
4. Brimacombe JR, Berry AM. Cricoid pressure. Can J Anesth
5. Vanner RG, Asai T. Safe use of cricoid pressure. Anaesthesia
6. Allman KG. The effect of cricoid pressure application on airway patency. J Clin Anesth
7. Hartsilver EL, Vanner RG. Airway obstruction with cricoid pressure. Anaesthesia
8. Palmer JHM, Ball DR. The effect of cricoid pressure on the cricoid cartilage and vocal cords: an endoscopic study in anaesthetised patients. Anaesthesia
9. Hocking G, Roberts FL, Thew ME. Airway obstruction with cricoid pressure and lateral tilt. Anaesthesia
10. Heath KJ, Palmer M, Fletcher SJ. Fracture of the cricoid cartilage after Sellick's manoeuvre. Br J Anaesth
11. Ralph SJ, Wareham CA. Rupture of the oesophagus during cricoid pressure. Anaesthesia
12. Riffenburgh RH. Statistics in Medicine.
San Diego, USA: Academic Press, 1999.
13. Lawes EG, Duncan PW, Bland B, Gemmel L, Downing JW. The cricoid yoke - a device for providing consistent and reproducible cricoid pressure. Br J Anaesth
14. Ashurst N, Rout CC, Rocke DA, Gouws E. Use of a mechanical simulator for training in applying cricoid pressure. Br J Anaesth
15. Owen H, Follows V, Reynolds KJ, Burgess G, Plummer J. Learning to apply effective cricoid pressure using a part task trainer. Anaesthesia
16. Flucker CJ, Hart E, Weisz M, Griffiths R, Ruth M. The 50-millilitre syringe as an inexpensive training aid in the application of cricoid pressure. Eur J Anaesthesiol
17. Clayton TJ, Vanner RG. A novel method of measuring cricoid force. Anaesthesia
18. Herman NL, Carter B, Van Decar TK. Cricoid pressure: teaching the recommended level. Anesth Analg
19. Meek T, Gittins N, Duggan JE. Cricoid pressure: knowledge and performance amongst anaesthetic assistants. Anaesthesia
20. Cook TM, Godfrey I, Rockett M, Vanner RG. Cricoid pressure: which hand? Anaesthesia
21. Vanner RG. Tolerance of cricoid pressure by conscious volunteers. Int J Obst Anesth