Anaesthesiologists are often compared with airline pilots; both operate technologically advanced equipment and use monitoring instruments to ensure that the equipment is operating properly. The introduction of a check procedure for the anaesthetic machine by the anaesthesiologists' associations is an attempt to improve the performance of anaesthesiologists in this practice [1-10]. Nevertheless, and in comparison with aviation, the effectiveness and application of such recommendations are different in both professions. In aviation, checking the equipment is strongly reinforced and each pilot is periodically trained; in contrast, in anaesthesia, it is not so regularly emphazised during postgraduate training and practice varies between institutions . Moreover, patients continue to be harmed because anaesthetic equipment has not been checked properly before use [11-16].
Thus, in order to develop a quality assurance programme for the safe function of the anaesthetic equipment in our institution, a study was designed to evaluate anonymously with a questionnaire how frequently the anaesthetists in a University Hospital assess their daily checking of the anaesthetic machine according to a check list used in the institution, and to compare the responses by nurses, residents in training and staff members. We hypothesized that the nurses would assess each component of the anaesthetic machine better than physicians.
At the University Hospital of Geneva, which is a teaching and referral hospital, working as the sole public institution in an area of about 400 000 inhabitants, more than 18 000 anaesthetic procedures are performed yearly in over 50 work places, representing 10 main operating room facilities localized in different buildings. Because no standard, for anaesthetic machine checking procedures, has been adopted in Switzerland, we defined a check list based on the existing checking procedures used in other countries and on the recommendation of the manufacturers. This check list contained 16 items with their different elements corresponding to the main components of the anaesthetic machine (cf. appendix). A regular check procedure lasts no more than 12-15 min [2-4]. This check list is an approved guideline included in the curriculum for post-graduate training on anaesthetic equipment and safety, organized yearly for all anaesthetists (physicians and nurses). Each anaesthetist and nurse, responsible for the equipment is required to check the equipment used.
Having received the ethics committee's approval, a questionnaire was sent to the 63 anaesthetic nurses, the 34 residents in training and the 33 staff physician anaesthesiologists in our institution. Sixty-three questions, grouped under 16 headings, concerned the main components of the anaesthetic machine, and asked the participant how frequently each component of the anaesthetic machine was checked. The scoring of the answers is reported in Table 1. Data were collected anonymously.
In order to test the participants critical sense, seven nonsense questions were inserted in the questionnaire, such as: 'checking the pressure inside the spirometer?'.
The answers to the questionnaire were analysed as follows:
The total score for the questionnaire, defined as the sum of the scores of each question: maximum 189 (3 × 63) for relevant questions and 21 (3 × 7) for nonsense questions.
The individual score for each item, defined as the sum of the score of the relevant questions for each of the 16 items.
The incidence of score 0 or 1 for each item checked.
The scores of the answers are expressed as median and ranges. To compare the results collected from the three groups of participants, a χ2 test was used for parametric variables and a Kruskal-Wallis non-parametric ANOVA test was associated with a Dunn's Multiple Comparison Test for non-parametric variables. P<0.05 was considered significant.
Table 2 summarizes the response rate to the questionnaire and the median for the total score following both relevant and nonsense questions. The overall answer rate was 78% (101/130), but significantly fewer nurses compared with medical staff answered the survey (P<0.05). The nurses had a significantly higher total score for the relevant questions, when compared with the residents (P<0.01) and with the senior staff (P<0.001), whereas the nonsense questions were scored comparably by all three groups of participants.
The individual score related to the 16 items defined were comparable for the three groups of participants, except for the six items presented in Table 3, which were different (P<0.05). The nurses had a significantly higher score for these six items when compared with the senior staff (circle, gas-scavenger, spirometer, ventilator, water-set and Mapleson-C system and manual resuscitator), and for four out of six items when compared with the residents (spirometer, ventilator, Mapleson-C system and manual resuscitator).
Table 4 ranks individually the 16 items of the anaesthetic machine according to the percentage of scores 0 and 1 obtained among all participants. A high percentage of scores 0 and 1 were particularly noted for the gas-cylinder, oxygen-pressure failure system and manual resuscitator, whereas for circle, gas-scavenger, oxygen-bypass and suction, the percentage was low. When the results of the three groups of participants were compared, the incidence of scores 0 and 1 were significantly lower for the nurses compared with physicians, when gas-cylinder, ventilator, spirometer Mapleson-C system and manual resuscitator were considered.
The present study is the first describing self-reported assessment and comparing the results collected from senior medical staff, residents and anaesthetic nurses. The results of this survey supported only partially our hypothesis. Indeed, nurses obtained a significantly higher score in self-reporting the checking of their anaesthetic machine, when compared with physicians; nevertheless, items in relation to gas supply were badly scored by all participants.
The problems of anaesthetic equipment failure are widely documented and the conclusions of many studies emphasize the importance of checking the anaesthetic machine before use [6-7,10-18, 19, 21]. Indeed, observations were made that failure to perform an adequate check of the anaesthetic machine was the commonest associated factor in critical incidents , ranging between 14 and 30% of incidents [12-14, 16].
The methodology used in this survey possibly overestimates the actual checking level of the anaesthetic machine in our institution, as it examined self-reported assessment anonymously, without any pressure from the institution. The data may have been much less optimistic if the study had been designed to evaluate theoretical knowledge with multiple choice questions or to control the actual performance by an observer or a video camera, which raises methodological bias [9,17,21].
The overall response rate to the survey was 78%, which confirms the interest in the subject by the participants. The higher total score noted in the nurse group (median = 144), compared with the residents (median = 126.5) and the staff (median = 129.5), suggests that the nurses are more aware than physicians of the importance of checking the anaesthetic machine. This could be a consequence of the nurses' greater technological interest, when compared with physicians, which may be related to the fact that daily preparation of the anaesthetic equipment is an important part of their task. Moreover, nurses may adhere to guidelines and policies rather better than physicians. Physicians are encouraged to think freely and to have a greater autonomy, thus are less compliant when faced with published recommendations. This has been demonstrated in recent publications in which physicians have shown resistance to clinical guidelines  and to voluntary reporting of defined intra-operative incidents .
Although the nonsense questions were obviously irrelevant, they were scored surprisingly highly by all participants (median: 2) when compared with the relevant questions (median: 3). It is surprising that no difference in the nonsense questions score was observed for the three groups, a lower score for nonsense questions by those scoring the highest to relevant questions could theoretically have been expected. This suggests that neither group demonstrated a high critical sense, or questioned the validity of an anonymous questionnaire used to investigate self-reported assessment.
When the three groups of participants individual scores for the 16 items were compared, nurses reported more frequently than physicians; checking the circle, the gas-scavenger, the spirometer, the ventilator, the Mapleson-C system and the manual-resuscitator. Assessment which represents more than 50% of the time spent in the checking procedure. Independently of the lower adherence of physicians to guidelines, their major criticism of a long and time-consuming checkout procedure could explain the differences . Nevertheless, the checkout procedure used in this institution lasts no longer than 12-15 min, which seems comparable with other checkout procedures [2,25]. The suggestion that checkout procedures be shortened, and that the checking of certain items should be omitted, could only be carried out at the expense of the patient's safety .
The ranking of the 16 items tested according to the percentage of scores 0 and 1 obtained for all participants should give an indication of deficient checking. Although nurses scored generally higher than physicians, items in relation to gas supply or rescue ventilatory devices, such as gas cylinders, O2-pressure failure systems and manual-resuscitators, are too rarely or not checked at all by 50% of the participants. This should be a warning to an-aesthesiologists because these items have been reported frequently in previous studies to be implicated in critical incidents related to anaesthetic equipment failure [11,20,27].
In conclusion, nurses generally have a higher score for checking when compared with physicians. This raises the question of a difference in adherence to guidelines by nurses compared with physicians. An attempt is required to reinforce physicians' compliance with guidelines. Nevertheless, because nurses had a poor response rate to the survey, the conclusions must be considered cautiously. Some components of the anaesthetic machine related to gas supply and rescue ventilation devices were too rarely checked, increasing the risk of critical incidents. A reinforced and targeted theoretical and practical training should rapidly improve the inadequate checking. Finally, the methodology used in this survey raises the question of the best way to evaluate the actual checkout procedure by each anaesthetist (nurse and physician). Though a multiple choice questionnaire, observer, or video evaluation are recognized methods, self-reported assessment is an alternative: the advantages are the anonymity, the feasibility, the possibility to reuse the same questionnaire after training re-inforcement, and the low expense; it could thus be an effective tool in a quality assurance programme.
We thank M. Bernstein, MD, from the Epidemiology Unity of the Department of Medicine, University Hospital, Geneva, for her help in the elaboration of the questionnaire and her advice on the feasibility of our survey.
1 Crosby WM. Checking the anaesthetic machine, drugs and monitoring devices. Anaesth Intens Care
2 Barthram C, McClymont W. The use of a checklist for anaesthetic machine. Anaesthesia
3 Morrison L. FDA anesthesia apparatus checkout recommandations, 1993. American Society of Anesthesiologists, Newsletter
4 SFAR. Recommandations concernat l'appareil d'anesthésie et sa vérification avant utilisation. Socièté Française d'Anesthésie et de Réanimation
5 Lunn JN. The law and the practice of anaesthesia. Anaesthesia
6 Checklist for Anaesthetis Machines: A recommanded procedure based on the use of an oxygen analyser. Issued 1990. London: Association of Anaesthetists of Great Britain and lreland.
7 FDA. Checkout procedure. Anesthesia Apparatur Checkout Recommandations. August 1986. Anesthesia Patient Safety Foundation Newsletter
8 Eagle C. Anaesthesia and education. Can J Anaesth
9 Olympio MA, Goldstein MM, Mathes DD. Instructional review improves performance of anesthesia apparatus checkout procedure. Anesth Analg
10 Adams AP. Checking anaesthetic machines - checklists or visual aids? Editorial. Anaesthesia
11 Cooper JB, Newbower RS, Kitz RJ. An analysis of major errors and equipment failures in anesthesia management: considerations for prevention and detection. Anesthesiology
12 Craig J and Wilson ME. A survey of anaesthesia misadventures. Anaesthesia
13 Chopra V, Bovill JG, Spierdijk J, Koorneef F. Reported significant observations during anaesthesia: a prospective analysis over 18-month period. Br J Anaesth
14 Short TG, O'Regan A, Law J, Oh Te. Critical incident reported in anaesthetic department quality assurance programme. Anaesthesia
15 Lagasse RL, Steinberg AS, Katz RI, Saubermann AJ. Defining quality of perioperative care by statistical process control of adverse outcomes. Anesthesiology
16 Clayton DG, Barker L, Runciman WB. Evaluation of safety procedures in anaesthesia and intensive care. Anaesth Intens Care
17 Major AH and Eaton JM. Anaesthetic machine checking pratices. Anaesthesia
18 Charlton JE. Checklists and patient safety. Editorial. Anaesthesia
19 Berge JA, Gramstad L, Jensen O. A training simulator for detecting equipment failure in the anaesthetic machine. Eur J Anaesthesiology;
20 Buffington CW, Ramanathan S, Turndorf H. Detection of anesthesia machine faults. Anesth Analg
21 Gaba DM. Improving anesthesiologists' performance by stimulating reality. Editorial. Anesthesiology
22 Ellrodt AG, Conner L, Riedlinger M, Weingarten S. Measuring and improving physicians compliance with clinical practice guidelines. A controlled interventional trial. Ann Intern Med
23 Sanborn KV, Castro J, Kuroda M, Thys DM. Detection on intraoperative incidents by electronic scanning of computerized anesthesia records. Anesthesiology
24 Higham H, Back GN. Checking anaesthetic machine. Anaesthesia
25 March MG, Crowley JJ. An evaluation of anesthesiologists' present checkout methods and the validity of the FDA checklist. Anesthesiology
26 Berge JA, Gramstad I, Grimnes S. An evaluation of a time-saving anaesthetic machine checkout procedure. Eur J Anaesthesiol
27 Hatton F, Tiret L, Manjol L. Enquête épidémiologique sur les anesthésies. Ann Fr Anesth Réanim
The 16 items of the check list of the anaesthetic machine (and the number of questions for each item)
- Gas cylinder (9): identfication, colour of cylinder, fastening to the anaesthetic machine, gas content, gas flow control, leak control
- Pipe line (7): connnection to the pipe line and the anaesthetic machine, leak control
- O2-pressure failure system (2): control with gas cylinder and with pipe-line supply
- ORC (1): proportioning system for oxygen/nitrous oxide delivery
- Flow meter (6): control of the float and flow faucet
- O2-bypass (1): on/off
- Vaporizer (5): fitting to the anaesthetic machine, dosing screw, filled or empty, leaks, absence of gass in the circuit when closed
- Circle (6): adequate assembly; inspiratory, expiratory and pressure relief valve (or APL valve)
- Gas scavenger (1): on/off
- O2-analyser (3): calibration at 21%, 100%, alarm
- Spirometer (3): on the expiratory part of the circle, function on/off
- Manometer (5): low-pressure alarm, high pressure alarm
- Ventilator (7): adequate assembly, driving-gas supply (air), tidal volume, respiratory rate and PEEP adjustment, pressure relief valve
- Suction (2): on/off, intensity adjustment
- Mapelson C system (3): separate O2-supply
- Manual-resuscitator (2): Ruben-valve control