Intended for healthcare professionals


Observational study of defibrillation in theatre

BMJ 1999; 318 doi: (Published 23 January 1999) Cite this as: BMJ 1999;318:232
  1. Jonathan M Fielden, specialist registrar anaesthetist (jonathan{at},
  2. Neil S Bradbury, resuscitation training officer
  1. Royal United Hospital NHS Trust, Bath BA1 3NG
  1. Correspondence to: Dr J M Fielden, Royal Berkshire Hospital, Reading RG1 5AN
  • Accepted 11 June 1998

Prompt, safe defibrillation is the treatment most likely to improve survival after ventricular fibrillation.1 Anaesthetists and surgeons need adequate skills to treat cardiac arrest.2 This observational study, set in the operating department of an acute hospital, tested whether surgeons and anaesthetists could manage ventricular fibrillation in accordance with advanced life support protocols.3

Subjects, methods, and results

Over two separate days 23surgeons and 25anaesthetists were asked, without warning or apparent prior knowledge, to manage simulated ventricular fibrillation (Laerdal skill master, Laerdal Heartstim 2000,and Laerdal monitor interface, Laerdal Medical, Orpington, Kent). Candidates were randomised (by tossing a coin) to either the S&W defibrillator (Simonson and Weald DMS 930,Vickers Medical, Sidcup, Kent) or the Lifepack 9(Physio Control Corporation, Redmond, WA, USA). Nineteen consultants, four staff grades, and 25trainees (12senior house officers and 13registrars, senior registrars, and specialist registrars) were studied from initial assessment to the third defibrillation. Results were analysed with Mann-Whitney U test and Fisher's exact tests (Analyse-it for Microsoft Excel).

Of all the candidates, 83% (40/48) failed to defibrillate according to advanced life support protocols (table). The Lifepack 9was easier to turn on (median (range) 61(11-113) seconds v 82(14-196) seconds for the S&W; P=0.03; n=44), and the first shock was delivered more rapidly (72(16-123) seconds v 102(40-201) seconds; P=0.006; n=44).This was not significant, however, after three shocks (129(33-218) seconds v 152.5(85-278) seconds; P=0.15; n=43). Forty candidates failed to deliver the first shock within 60seconds, 24by 90seconds (range 11-201; n=44). Four candidates failed to turn on the defibrillator, five candidates failed to deliver three shocks, and only four candidates delivered three shocks within 90seconds; seven took over 180seconds (range 33-278). Median (range) times to confirm arrest and call the arrest team were 10(0-87) seconds and 10.5(0-120) seconds, respectively.

Reasons for failure of staff to defibrillate in simulated setting in operating theatres

View this table:

All candidates were content with the method of testing and were happy to be tested in this manner in future.

Anaesthetists fared better than surgeons, although because of the small sample this was not significant (7/25 v 1/23; P=0.06). There was no difference between trainees and consultants (3/25 v 3/19; P=0.71).


Defibrillation skills are poor across a cross section of grades of anaesthetists and surgeons, the main reasons being lack of safety procedures and lack of knowledge. Operating theatres are often inaccessible to non-theatre staff, thus reliance on early arrival of the hospital resuscitation team may adversely affect outcome. This study confirms that the key advanced life support skill of defibrillation is still inadequate across a range of clinical experience, despite previous reports. 4 5

It is of some concern that 69% of attempts failed because of inadequate safety, replicating the findings of Bell et al.5 If used injudiciously, charged defibrillator paddles are dangerous to patients and staff. The 62% failure from lack of knowledge reflects inadequate training and skill retention. The absence of an initial pulse check to confirm arrest by 46% of candidates is worrying as interference from electrocardiographs in theatres may mimic ventricular fibrillation.

Optimum effect from defibrillation occurs within 90seconds of onset of ventricular fibrillation1; only half of the candidates achieved this. Defibrillators are used infrequently and thus need to be “self explanatory.” The covers and position of the buttons on the S&W defibrillator resulted in a significant delay. Unnecessary breaks between shocks for administration of drugs and cardiac massage caused further delay. Training and the use of (semi) automatic defibrillators might improve this.

All doctors in theatre who might operate alone should be competent in advanced life support. Such training is time consuming, and resources are not available to retest with sufficient frequency. The 100% positive response suggests there should be further study of the efficacy of random testing and use of “mock arrests” on maintenance of skills in advanced life support techniques.


This research was carried out before the publication of the 1997guidelines from the Advanced Life Support Working Party of the European Resuscitation Council.

Contributors: JF and NB jointly tested all subjects; JF analysed the data and wrote the paper and is the guarantor of the study.

Funding: None.

Conflict of interest: None.


View Abstract