Search

The design of rules (laws, norms, policies, procedures, protocols) and the implementation of a management that aimed at conforming the behaviors of operators with these rules has long been the dominant approach to improve safety in socio-technical systems. This approach has proven to be effective in enhancing safety by enabling organizations to cope with predictable risks and failures. However, in our modern and constantly evolving socio-technical systems, where the management of unforeseen situation has become the rule, this approach has shown its limitations: operators' initiatives, which sometimes deviate from rules, are equally important for maintaining safety. Therefore, while the rules remain a key feature of risk management, the challenge is not the search for total compliance with rules, but the permanent monitoring of their use to detect and distinguish gaps that constitute a drift towards the accident from gaps that highlight that the rule has become inappropriate and which the compliance with may prove dangerous for safety.

The design of rules (laws, norms, policies, procedures, protocols) and the implementation of a management that aimed at conforming the behaviors of operators with these rules has long been the dominant approach to improve safety in socio-technical systems. This approach has proven to be effective in enhancing safety by enabling organizations to cope with predictable risks and failures. However, in our modern and constantly evolving socio-technical systems, where the management of unforeseen situation has become the rule, this approach has shown its limitations: operators' initiatives, which sometimes deviate from rules, are equally important for maintaining safety. Therefore, while the rules remain a key feature of risk management, the challenge is not the search for total compliance with rules, but the permanent monitoring of their use to detect and distinguish gaps that constitute a drift towards the accident from gaps that highlight that the rule has become inappropriate and which the compliance with may prove dangerous for safety.

Resume Objectif Les erreurs medicamenteuses representent une cause importante d’evenements indesirables graves. En decembre 2006, l’Afssaps, en collaboration avec les professionnels de sante et les industriels, a diffuse des recommandations visant a harmoniser la presentation des etiquetages des solutions injectables dans le but de reduire les erreurs de preparation et d’injection. En 2007, une premiere « vague » d’harmonisation a concerne quatre substances actives et en 2008–2009, une seconde vague touchant 42 substances a ete lancee. Le present travail decrit l’appreciation des professionnels sur les modalites d’organisation nationale et locale de la seconde vague et leur appreciation globale de l’harmonisation des etiquetages. Methodes Une enquete utilisant un questionnaire electronique a ete distribuee aupres des professionnels medicaux et non medicaux d’anesthesie et de reanimation et des pharmaciens dans un echantillon representatif de 200 etablissements tires au sort. Resultats Le plan d’harmonisation est considere comme globalement satisfaisant pour une faible majorite de repondants (53 %). L’expression du dosage et la disposition preetablie des mentions sont accueillies positivement. L’application du code couleur est jugee plutot favorablement par les professionnels des services cliniques. Les professionnels de sante ont des doutes sur l’amelioration de la lisibilite des mentions. Les professionnels sont globalement critiques sur la mise en œuvre de la seconde vague au sein des etablissements et sur la communication qui l’a entouree. Une coexistence des anciens et des nouveaux etiquetages a ete observee dans 66 % des etablissements ayant repondu. Les repondants decrivent avoir eu connaissance d’erreurs medicamenteuses ou incidents survenus apres la diffusion de l’information et qui pourraient etre en rapport avec cette « transition » dans 28 % des cas. Pour plusieurs questions importantes, le point de vue des pharmaciens etait significativement different de celui des professionnels des services cliniques. Conclusion Cette enquete nationale sur l’appreciation des professionnels a mis en evidence un certain nombre de deficiences, notamment en matiere de communication et d’organisation au sein des etablissements de sante. Les resultats obtenus serviront a ameliorer les prochaines demarches de gestion des erreurs medicamenteuses.

P.-M. Mertes *, C. Karila , P. Demoly , Y. Auroy , C. Ponvert , M.-M. Lucas , J.-M. Malinovsky g a Service d’anesthesie-reanimation chirurgicale, hopital Central, CHU de Nancy, 29, avenue de Lattre-de-Tassigny, 54035 Nancy cedex, France b Service de pneumologie et allergologie pediatriques, hopital Necker–Enfants malades, AP–HP, 149, rue de Sevres, 75015 Paris, France c Exploration des allergies, maladies respiratoires et Inserm U657, hopital Arnaud-de-Villeneuve, CHU de Montpellier, 34295 Montpellier cedex 05, France d Service d’anesthesie reanimation, HIIA Val-de-Grâce, Paris, France e Service de pneumo-allergologie pediatrique, hopital Necker–Enfants-Malades, 149, rue de Sevres, 75015 Paris, France f Service d’anesthesie reanimation, hopital Pontchaillou, 35000 Rennes, France g Service d’anesthesie et reanimation, CHU de Reims, pole URAD, hopital Maison-Blanche, 45, rue Cognacq-Jay, 51092 Reims, France Unite Inserm U684, faculte de Medecine, 9, avenue de la Foret-de-Haye, 54505 Vandœuvre-les-Nancy, France

Safety in the field of transfusion medicine has greatly improved in France. The risk of viral transmission has decreased by a factor greater than 1500 within the last 20 years. In comparison, the risk related to ABO error has decreased only by half. The reporting of critical incidents, which occur at any step of the transfusion procedure is now mandatory in France and is subject to an in-depth analysis, using methods close to that used in aviation safety. The goal of these analyses is to better understand human factors in order to implement more adequate prevention measures.

Resume La pathologie iatrogene est un important probleme d’actualite. Les reanimations sont des services a risque de survenue d’evenements indesirables (EI) lies aux soins et d’erreurs medicales. L’incidence des EI en reanimation varie de 3 a 31 % selon les publications. Ces variations tiennent essentiellement a la methodologie du recueil des donnees. Celle-ci est primordiale. Les indicateurs doivent etre standardises (definitions consensuelles), faciles a colliger. La methode de recueil doit etre idealement prospective, non punitive, confidentielle, independante au sein d’une equipe compliante et realisee avec la participation de differents acteurs non seulement du service mais aussi exterieurs (biologistes, pharmaciens). Les facteurs de risque des EI en reanimation sont connus : âge et scores de gravite a l’admission eleves, avec prise en charge medicale et paramedicale plus importante. Les EI sont associes a une augmentation de la morbidite des patients en reanimation sans que l’on puisse affirmer formellement un rapport de causalite. Le surcout lie aux EI a ete chiffre a 3961 dollars aux Etats-Unis. La mortalite des patients presentant un EI est plus elevee mais aucune etude n’a demontre a ce jour que les EI constituaient un facteur de risque independant de mortalite en reanimation. Certains EI sont evitables (de 28 a 84 % selon les etudes). L’implementation de procedures de securisation (PS) est donc capitale. De nombreuses modalites souvent faciles a mettre en œuvre existent : procedures de soins, structurelles et manageriales. Le developpement d’une culture de securite dans les etablissements de soins est essentiel. C’est la premiere etape indispensable a une meilleure comprehension des professionnels de sante et du grand public.

Resume La pathologie iatrogene est un important probleme d’actualite. Les reanimations sont des services a risque de survenue d’evenements indesirables (EI) lies aux soins et d’erreurs medicales. L’incidence des EI en reanimation varie de 3 a 31 % selon les publications. Ces variations tiennent essentiellement a la methodologie du recueil des donnees. Celle-ci est primordiale. Les indicateurs doivent etre standardises (definitions consensuelles), faciles a colliger. La methode de recueil doit etre idealement prospective, non punitive, confidentielle, independante au sein d’une equipe compliante et realisee avec la participation de differents acteurs non seulement du service mais aussi exterieurs (biologistes, pharmaciens). Les facteurs de risque des EI en reanimation sont connus : âge et scores de gravite a l’admission eleves, avec prise en charge medicale et paramedicale plus importante. Les EI sont associes a une augmentation de la morbidite des patients en reanimation sans que l’on puisse affirmer formellement un rapport de causalite. Le surcout lie aux EI a ete chiffre a 3961 dollars aux Etats-Unis. La mortalite des patients presentant un EI est plus elevee, mais aucune etude n’a demontre a ce jour que les EI constituaient un facteur de risque independant de mortalite en reanimation. Certains EI sont evitables (de 28 a 84 % selon les etudes). L’implementation de procedures de securisation (PS) est donc capitale. De nombreuses modalites souvent faciles a mettre en œuvre existent : procedures de soins, structurelles et manageriales. Le developpement d’une culture de securite dans les etablissements de soins est essentiel. C’est la premiere etape indispensable a une meilleure comprehension des professionnels de sante et du grand public.

Among the adverse events in the blood transfusion process, transfusion to a "wrong" patient is potentially dangerous, as it can lead to an adverse reaction at least in case of ABO incompatible red cell concentrate.The "Root Cause Analysis" working party of the National Hemovigilance Commission developed a tool to collect this type of adverse event, and tested it on a sample of 43 cases involving red cell concentrates notified between March, 2009 and February, 2010.One hundred and nine failures of a step in the transfusion process were observed, i.e. 2.5 failures per adverse event. Failures may occur early in the process. However, they are mainly found at the time of issuing of the blood component, and further, in the clinical ward. How the failure is eventually detected is not always described when the blood component has been fully transfused, in contrast with the cases where actual transfusion to the wrong patient has been prevented. Knowing the way of failure detection enables an objective approach of the efficacy of the numerous existing safety measures. In this sample, bedside controls (documents check as well as the use of anti-A and anti-B reagents with patient's blood and red cell concentrates) detected the failure in three cases out of 34, which were not detected before, showing an efficacy similar to the administrative control done at reception in the clinical ward.The document, set up to analyse step by step these cases of patient errors, will be used in the future to analyse all similar cases, not only with red cell concentrates, but also with platelet concentrates and fresh frozen plasma, ultimately in order to improve their prevention.

Hyperglycemia is significantly associated with increased mortality in critically ill patients and then, strict control of blood glucose (BG) concentration is important. Lowering of BG levels with intensive insulin therapy (IIT) was recommended in order to improve patient outcomes. But recently, some recent prospective trials failed to confirm the initial data, showing conflicting results (significantly increased mortality with IIT, more hypoglycemic episodes). So there is no consensus about efficiency and safety of IIT. Significant associations between glucose variability and mortality have been confirmed by several recent studies. A difference in variability of BG control could explain why the effect of IIT varied from beneficial to harmful. Managing and decreasing this BG variability could be an important goal of BG control in critically ill patients. Clinicians have to consider definitions, physiopathology and impacts of glucose variability, in order to improve patient outcomes.

Medication errors are a significant cause of severe healthcare-associated complications. In December 2006, the French Health Products Agency (Afssaps) has issued a protocol to harmonise labeling of injectable drugs vials. In 2007, a first change was launched for four drugs and was followed in 2008-2009 by a second wave concerning 42 active drugs.The present study describes how healthcare professionals have perceived this change and their overall appreciation of the drug harmonisation programme. A survey using an electronic questionnaire was distributed to medical and non-medical professionals in anaesthesia and intensive care and pharmacists in a representative sample of 200 French hospitals.The harmonisation procedure was felt as being overall satisfactory by 53% of professionals who had responded but it was recognised that the new procedure is associated with improved readability and understanding of drug dosage. The use of colour coding was also well accepted by the personnel of clinical units. Respondents expressed significant criticisms regarding both the communication plan and the way the plan was implemented locally in hospitals. Old and new labeling coexisted in 66% of responding hospitals and many respondents described being aware of errors or near-misses that were considered related to the transition. For many important topics, pharmacists had views that were significantly different from clinicians.This national survey describing the perception of healthcare professionals regarding the new harmonisation procedure for injectable drugs highlighted some progress but also a number of deficiencies, notably regarding communication and implementation of the change in clinical units. This survey will be used by the French Health Products Agency to improve future steps of the long-lasting campaign against medication errors.

Although this has been challenged by some authors,’ it is very likely that the presence of anesthetists in obstetric suites has substantially contributed to the improved quality of care and reduced maternal mortality and morbidity. Fasting guidelines and use of i.v. fluids during labour together with maternal and fetal monitoring represent cornerstones of improved quality of care. However, since aspirationrelated maternal mortality has significantly decreased in recent years2 there has been a trend initiated by obstetricians3 to relax the restriction of oral intake during labour while iatrogenic complications related to glucose infusion during labour have been recognised. This paper is aimed at encouraging strict safety guidelines for the correct use of iv fluids during labour.

Since March 2003, military operations in Iraq "Operation Iraqi Freedom" (OIF) and in Afghanistan "Operation Enduring Freedom" (OEF), have made many wounded and killed in action (KIA). This article proposes to highlight the specific epidemiology of combat casualties, met in these both non-conventional and asymmetric conflicts. Personal protective equipments, Kevlar helmet and body armor, proved their efficiency in changing features of war injuries. Health Force Services organized trauma care system in different levels, with three main objectives: immediate basic medical care in battalion aid station, forward surgery and early aeromedical evacuation. The Joint Theater Trauma Registry (JTTR), a war injury registry, provides medical data, analyzed from the combat theater to the military hospital in United States. This analysis concluded that during modern conflicts, most injuries are caused by explosive devices; injuries are more severe and interestingly more specifically the head region and extremities than the trunk. Hemorrhage is the first cause of death, leading to the concept of avoidable death. Specific databases focused on mechanisms and severity of injuries, diagnostic and treatment difficulties, outcomes can guide research programs to improve war injuries prevention and treatment.

Safety in the field of transfusion medicine has greatly improved in France. The risk of viral transmission has decreased by a factor greater than 1500 within the last 20 years. In comparison, the risk related to ABO error has decreased only by half. The reporting of critical incidents, which occur at any step of the transfusion procedure is now mandatory in France and is subject to an in-depth analysis, using methods close to that used in aviation safety. The goal of these analyses is to better understand human factors in order to implement more adequate prevention measures.

The mistake-proofing concept often refers to physical devices that prevent actors from making a wrong action. In anaesthesiology, one immediately thinks to specific design of outlets for medical gases. More generally, the principle of mistake-proofing is to avoid an error, by placing knowledge in the world rather than knowledge in the head. As it often happens in risk management, healthcare has received information transfers from the industry. Computer is changing the concept of mistake-proofing, initially based on physical design, such as aerospace and automotive industry. The mistake-proofing concept may be applied to prevention, detection, and mitigation of errors. The forcing functions are a specific part of mistake-proofing: they prevent a wrong action or they force a virtuous one. Grout proposes a little shortcut to identify mistake-proofing devices: "If it is not possible to picture it in action, it is probably not a mistake-proofing device".

Death certificates from the French national mortality database for the calendar year 1999 were reviewed to analyse cases in which airway complications had contributed to peri-operative death. Respiratory deaths (and comas) found in a previous national 1978-82 French survey (1:7960; 95% CI 1:12,700 to 1:5400) were compared with the death rate found in the present one: 1:48,200 (95% CI 1:140,000 to 1:27,500). In 1999, deaths associated with failure of the breathing circuit and equipment were no longer encountered and no death was found to be related to undetected hypoxia in the recovery unit. Deaths related to difficult intubation also occurred at a lower rate than in the previous report (1:46,000; 95% CI 1:386,000 to 1:13,000) in 1978-82 vs 1:176,000 (95% CI 1:714,000 to 1:46,000) in 1999, a fourfold reduction. In most cases, there were both inadequate practice and systems failure (inappropriate communication between staff, inadequate supervision, poor organisation). This large French survey shows that deaths associated with respiratory complications during anaesthesia have been strikingly reduced during this 15-year period.

Iatrogenic pathology is currently a serious problem. Intensive care units (ICU) are wards with a high risk of occurrence of adverse events (AE) related to the care and medical errors. The incidence of AE in ICU varies from 3 to 31% according to the publications. These variations are mainly due to the methodology of data collection. The latter is essential. The indicators must be standardized (consensual definitions), and easily collected. The method of collection must be ideally prospective, nonpunitive, confidential, independent within a compliant team, and realized with the participation of various actors not only of the unit but also external (biologists, pharmacists). The risk factors of AE in ICU are known: old age and high severity scores at admission, with medical and nurse workload more important. AE are associated with an increased patients' morbidity in ICU with no evident causality. The over cost related to AE in ICU was quantified to 3961 dollars in the United States. The mortality of patients with an AE is higher but no study showed to date that AE constituted an independent risk factor of mortality in ICU. Some AE are preventable (from 28 to 84% according to studies). Therefore, the implementation of procedures of security (PS) is capital. Many methods often easy to implement exist such as in care, structural and managerial procedures. The development of a safety culture in hospitals and other delivery care settings is essential. It is the first essential step in a better comprehension of the health care professionals and the public opinion.

Aspiration of gastric contents is a major complication in relation with the practice of anaesthesia. The present article is aimed at describing detailed data related to aspiration which were obtained during the French national survey on anaesthesia-related mortality conducted by both Sfar and CépiDC-Inserm. Information regarding methods of the survey and the main results has been previously published. In brief, the first part of the survey described the number and characteristics of anaesthetic procedures performed in 1996 (denominator). The second survey analysed deaths related to anaesthesia which were identified from death certificates of the calendar year 1999. Because of the numerical importance of aspiration among the causes of deaths, a secondary analysis was undertaken to assess into details factors leading to the occurrence of this complication. Eighty-three cases of death were found related to aspiration, i.e. one-fifth of deaths related completely or partially to anaesthesia, implying a death rate of one for 221,368 general anaesthetic procedures or 4.5 x 10(-6) (95% IC: 0.8 x 10(-6)-14 x 10(-6)). Patients involved were all in a severe clinical condition (ASAor =3: 92%), very old and often scheduled for urgent abdominal surgery. Two cases of death occurred during colonoscopy but none in obstetric patients. Aspiration almost always occurred during induction of anaesthesia. Analysis of practice patterns disclosed significant deviations from recommendations. French anaesthetists should voluntarily move through a personal and active process toward an improvement of their practice to reduce the incidence of aspiration.

The assessment of postoperative care relies mainly on spontaneous reporting of major adverse events (MAE). We propose an alternative approach that we have tested on the measurement of the risk of postoperative anemia which is frequentely associated with anaesthethic related deaths.The procedure consisted in extracting the electronic sheet generated by the biological analyser and merging it with the anesthesia data base with a filter based on patient's location in the hospital. The orthopedic surgery ward was chosen because of the frequency of full blood count (FBC) associated with the prescription of low weight molecular heparin in this setting. All FBC of the year 2005 were analysed. A risk stratification was achieved according to the conbination of age and depth of anemia. Medical charts of patients exposed to the greatest risk were reviewed.FBC of 691 orthopaedic surgery patients were analysed. Haemoglobin levels (Hb) less than 8g/dl were observed in 41 (5.9%) patients. Two consecutive Hb less than 8g/dl were founded in 18 patients (2.6%), eight (1.2%) charts of patients exposed to the highest risk were reviewed and six cases of no-transfusion or delayed transfusion were identified.A global approach to the risk of postoperative anemia but also of MAE related to postoperative haemorrage can be obtained by monitoring electronic databases.

Assessing the quality of anaesthesia records according to the criteria of the reference frame of professional practices evaluation proposed by the French Anaesthetists College (Cfar) in 2005.Retrospective, multicentric study.Fifty anaesthesia records were randomly selected from each of the 64 health care settings in Aquitaine with an inpatient surgery activity. Survey sampling was based on hospital stays longer than 24h occurring during the second half of 2005 and including a surgical procedure under anaesthesia, excepting local anaesthesia or regional analgesia for labour and delivery. Information regarding the 10 evaluation criteria defined by the Cfar and the French National Authority for Health were gathered. For each setting, the mean completion of the 50 records was rated by a global indicator defined by the number of criteria in conformity divided by the number of relevant criteria.Three thousand one hundred and ninety-three anaesthetic files were screened. The completion of criteria varied between 1 and 92% with a wide dispersion for some criteria according to settings. Completions of airway assessment, preanaesthetic visit and perianaesthetic incidents/accidents were lower than 50%. The median conformity rate of the global indicator was 57%.This study brings information concerning the clinical relevance of anaesthesia records and preoperative evaluation practices.

Violations are deliberate deviations from standard procedure. The usual reaction is to attempt to eliminate them and reprimand those concerned. However, the situation is not that simple. Firstly, violations paradoxically may be markers of high levels of safety because they need constraints and defences to exist. They may even become more frequent than errors in ultrasafe systems. Secondly, violations have both positive and negative aspects. On the one hand they occur frequently, increase system performance and individual satisfaction, are mostly limited to practices with limited safety consequences, and therefore are often tolerated or even encouraged by the hierarchy. On the other hand, extreme violations can lead to real danger or actual harm. This paper proposes a three phase model derived from Rasmussen's theory of migration to boundaries to explain the mechanism by which the deviance occurs, stabilizes, regresses, or progresses to harm. The model suggests that violations are unavoidable because system dynamics and deviances are markers of adaptation to this dynamicity. Violations cannot be eliminated but they can be managed. Solutions are specific to each step of the model, with a mix of relaxing constraints, increasing peer control (staff), and constraining dangerous individuals.

More and more often, the anaesthesiologist may have to perform lumbar epidural anaesthesia in a patient with a central lumbar tattoo, and this can occur in an urgent obstetric setting. Before managing two uneventful cases of epidural analgesia for labour, we have performed a literature review and noted that no serious complication has been reported. Nonetheless, a needle passed through a tattoo can entrap pigmented tissue fragments (cores) into the epidural or subarachnoid space. This could theoretically induce risk of late neurological complications, related to an inflammatory or granulomatous response to the pigmented cores introduced in these spaces. To avoid this theoretical risk, the anesthesiologist should try to avoid puncturing through the tattoo, either by selecting a different vertebral interspace, or by using a paramedian approach or by finding a pigment free skin spot within the area of the tattoo. When these options cannot be implemented, a superficial skin incision prior to needle insertion should prevent from coring tattoo pigment when entering the skin. Whatever the final choice, the technique to be implemented should be determined as early as the antenatal visit, after informed consent.

To determine on a national level the factors associated with the use of laparoscopy for digestive surgery.Nation wide study using a large representative sample (3 days of anaesthesia in France).Univariate followed by multivariate analyses of data gathered in 1996 during the survey led by the French Society of Anaesthesia and Intensive care ("SFAR") including 2847 surgical procedures for cholecystectomy, appendicectomy or inguinal herniorraphy.Independent factors associated with the use of laparoscopy were: for cholecystectomy: age (less frequent whenor =71 years: adjusted Odds ratio [AOR] 0.4), sex (more frequent in female: AOR 1.7), ASA physical status (less frequent whenor =3: AOR 0.5), private hospital (AOR 2.0), procedure scheduled at least the night before (AOR 2.1), and use of closed circuit general anaesthesia (AOR 1.6); for appendectomy: age15 years (AOR 1.9-2.2), female (AOR 2.1), private hospital (AOR 2.7), scheduled procedure (AOR 2.1), prolonged procedure (AOR 8.4), endotracheal intubation (AOR 16.7), and closed circuit (AOR 2.7); for inguinal herniorraphy: ASA physical status (less frequent whenor =3: AOR 0.4), private hospital (AOR 3.4), prolonged procedure (AOR 5.6), and endotracheal intubation (AOR 21.6). Association with a closed circuit was confirmed for general anaesthesia using a volatile agent (AOR 1.5). Overall, ambulatory surgery was rarely performed and used only for open procedures. Regional anaesthesia was used only for inguinal open herniorraphy.These data obtained from a large national survey confirmed the higher frequency of laparoscopy in middle aged patients, female (except for inguinal herniorraphy), without important comorbidity, in private hospitals. Laparoscopy was associated with prolonged procedures and with a change in the anaesthetic technique for appendicectomy and inguinal herniorraphy: tracheal intubation was almost constantly used. Whatever the procedure, closed circuit anaesthesia was more frequently used when surgery was performed under laparoscopy, reflecting newer equipment of the hospital, private or public.

In order to evaluate the benefits arising from regular recording of intraoperative adverse events, we extracted from our database all episodes of intraoperative hypotension and studied the risk factors of this event.Retrospective analysis of a large database from two university teaching hospitals evaluating the incidence and the risk factors of intraoperative hypotension by logistic regression.A data collection chart describing the patient's characteristics, the anaesthetic technique and selected intraoperative incidents was filled for each anaesthetised patient in the operating room and then recorded in the computer database. Data collected in 2001 in patients undergoing general anaesthesia for orthopaedic and general surgery were reviewed and univariate and multivariate analysis were performed using Statview 5.0 and Stata 7.0.Among 11 820 patients who underwent anaesthesia, 2691 were selected. The incidence of intraoperative arterial hypotension was 16.8%. The associated factors were duration of surgery, age and ASA status of the patients. The use of etomidate for induction was not associated with a decreased risk of intraoperative hypotension.Systematic recording of intraoperative events in a database has been suggested as useful by many experts for quality-assurance and safety analysis purposes. Analysis of a frequent anaesthesia-related (i.e. hypotension) event did not disclose any relevant factor that might lead to improvement. Running such databases is time-consuming and may be expensive. This leads us to question the efficiency of such databases.

To determine over a whole country what are the factors associated with an intraoperative homologous blood transfusion and with the use of autologous techniques (preoperative autologous blood donation: PABD; acute normovolemic hemodilution: ANVH; intraoperative red cell salvage: IRCS).National enquiry using a large representative sample (3 days of anaesthesia in France).Univariate followed by multivariate analyses of data gathered in 1996 during the survey leaded by the French society of anaesthesia and intensive care (Sfar) and corresponding to 884 scheduled hip and knee prosthesis surgical procedures.Factors associated with a decreased use of PABD programme were: 1--old age and high ASA physical status; 2--procedures of short duration. By contrast, an increased use of PABD was associated with anaesthetics in which a closed circuit had been used. Except for a significant association with increasing age and with absence of PABD used, no additional factor was found to be linked with ANVH. No factor among those studied was found related to the use of IRCS. Homologous blood transfusion was more frequently used in ASAor = 3 patients, in long duration surgeries while its use was decreased in patients with PABD (odds ratio--for reduction by PABD: 4.4 [95% confidence interval: 2.2-8.8]). Homologous blood transfusion was not related to the use of ANVH or IRCS.These data obtained from a large national survey confirm previously published studies and meta-analyses and are in agreement with current recommendations. An unexpected relation between PABD and closed circuit anaesthesia has been found.

Anaesthesiology is a recent medical speciality, established in 1965, that is formally termed in French Anesthésie-réanimation (Anaesthesiology-intensive care). Since its creation, the discipline has had a range of activity including anaesthesia, intensive care, prehospitalisation emergencies, acute and chronic pain and diverse follow-up. In France 8 million anaesthesias are performed per year, by 8,500 physicians specialised in anaesthesiology, assisted by 7,500 anaesthesiology nurses. It is estimated that each year 13.5% of the French population undergoes anaesthesia, with variations related to sex and age. Over 15 years, the number of anaesthesias performed increased 2.2 fold, due to the increase in certain types of surgery (especially orthopaedics) but also obstetrics (35 times more peridural anaesthesias) and the explosion of GI endoscopy (50 times increase). This increased activity involves patients at the extreme ranges of life, young women and fragile persons. Local and regional anaesthesias represent 21% of those performed, and outpatient anaesthesia is 12 times that of 15 years ago.