• André Lienhart, Yves Auroy, Françoise Péquignot, Dan Benhamou, Josiane Warszawski, Martine Bovet, and Eric Jougla.
• Service d'anesthésie-réanimation, CHU Saint-Antoine, 75012 Paris.
• B Acad Nat Med Paris. 2004 Jan 1;188(8):1429-37; discussion 1437-41.

AbstractA National Confidential Inquiry was conducted among death certificators and anaesthetists. A sample of 3700 death certificates from the year 1999 were randomised, after selection of words relating to anaesthesia, surgery, obstetrics, endoscopy, procedural complications, and violent death, with different ratios according to the words and the age; 500 additional certificates relating to deaths in hospital were evaluated to verify the exhaustive nature of the mention of procedures in the certificates. The certificator was sent a simplified form each time the role of the procedure in death could not be excluded (response rate 97%). The anaesthetist was offered a peer review whenever the role of the anaesthetic procedure could not be ruled out (uptake rate 97%). An expert committee analysed the (anonymized) files to determinate the mechanism of the accident and its relationship to anaesthesia. The mortality rates were estimated from the 1996 "Anaesthesia in France" survey. The annual rates of deaths that were totally or partially related to anaesthesia were respectively 7 (CI95%: 2-12) and 47 (31-63) per million. These mortality rates increased with comorbidity, from 4 per million in patients of ASA physical status class 1 to 554 per million in class 4. Similarly, these rates increased with age, from 7 per million in patients less than 45 years old, to 32 in older patients. Most accidents were of ventilatory (38%: airway management: 6%, aspiration pneumonitis: 9%), cardiac (31%: ischaemia: 25%, including anaemia-related), and vascular origin (30%: hemorrhage: 12%, vasodilation by spinal anaesthesia: 6%, anaphylaxis: 3%). The main surgical procedures involved were orthopaedic (50%: hip fracture, haemorrhagic surgery) and digestive (24%: occlusion, peritonitis). INSERM had previously collected data on complications associated with anaesthesia between 1978 and 1982: the annual rates of deaths that were totally or partially related to anaesthesia were respectively 76 and 263 per million. Compared to these previous data, the anaesthesia-related mortality rate fell ten-fold over the last two decades, while the number of anaesthetic procedures at least doubled. In addition, the number of procedures involving old people and patients with poor physical status was multiplied by four. It seems logical to attribute these results to safety and practice guidelines published after the previous inquiry. Progress remains to be made: the present rate of 1/145000 will serve as a basis for systematic analysis of accidents.

16 keywords...

hide…