[Perioperative conflicts between anaesthesiologists and surgeons: ethics and professionalism]. / Conflits en période périopératoire: un enjeu collectif, éthique et professionnel.

In the perioperative period, several potential conflicts between anaesthetists/intensive care specialists and surgeons may exist. They are detrimental to the quality of patient care and to the well-being of the teams. They are a source of medical errors and contribute to burn-out. Patients can become the victims of such conflicts, which deserve ethical reflection. Their resolution through analysis and shared solutions is necessary. This article seeks to analyse these conflicts, taking into account their specificities and constraints. In order to understand this context, it is important to consider the specificities of each group involved and the records of such situations. Several factors can prevent these conflicts, first and foremost the patients themselves and the quality of the care that is provided. Medical deontology aims mainly at preventing and resolving these conflicts. Generally speaking, the quality approach which is increasingly applied in health care institutions (involving declarations of adverse events, morbidity/mortality reviews, benchmarking, analysis and improvement of practices, etc.) also contributes to the prevention and resolution of disagreements. The teaching of communication techniques that begins with the initial training, the evaluation of team behaviours (through simulation training for example), the respect of others' constraints, particularly when it comes to learning, as well as transparency regarding conflicts of interests, are all additional elements of conflict prevention. Lastly, conflicts may at times be caused by deviant behaviours, which must be met with a clear and uncompromising collective and institutional approach. This article concludes by offering a standardised approach for conflict resolution.

Critical appraisal of organ procurement under Maastricht 3 condition.

The ethics committee of the French Society of Anesthesia and Intensive Care (Sfar) has been requested by the French Biomedical Agency to consider the issue of organ donation in patients after the decision to withdraw life-supportive therapies has been taken. This type of organ donation is performed in the USA, Canada, the United Kingdom, the Netherlands and Belgium. The three former countries have published recommendations formalizing procedures and operations. The French Society of Anesthesia and Intensive Care (Société française d'anesthésie et de reanimation [Sfar]) ethics committee has considered this issue and envisioned the different aspects of the whole process. Consequently, it sounded a note of caution regarding the applicability of this type of organ procurement in unselected patients following a decision to withdraw life-supportive therapies. According to French regulations concerning organ procurement in brain-dead patients, the committee stresses the need to restrict this specific way of procurement to severely brain-injured patients, once confirmatory investigations predicting a catastrophic prognosis have been performed. This suggests that the nature of the confirmatory investigation required should be formalized by the French Biomedical Agency on behalf of the French parliamentarians, which should help preserve population trust regarding organ procurement and provide a framework for medical decision. This text has been endorsed by the Sfar.

[Euthanasia, assisted suicide and palliative care: a review by the Ethics Committee of the French Society of Anaesthesia and Intensive Care]. / Fin de vie, euthanasie et suicide assisté : une mise au point de la Société française d'anesthésie et de réanimation (Sfar).

CONTEXT: Management of the end of life is a major social issue which was addressed in France by law, on April 22nd 2005. Nevertheless, a debate has emerged within French society about the legalization of euthanasia and/or assisted suicide (E/AS). This issue raises questions for doctors and most especially for anesthetists and intensive care physicians. OBJECTIVE: To highlight, dispassionately and without dogmatism, key points taken from the published literature and the experience of countries which have legislated for E/AS. RESULTS: The current French law addresses most of the end of life issues an intensive care physician might encounter. It is credited for imposing palliative care when therapies have become senseless and are withdrawn. However, this requirement for palliative care is generally applied too late in the course of a fatal illness. There is a great need for more education and stronger incentives for early action in this area. On the rare occasions when E/AS is requested, either by the patient or their loved-ones, it often results from a failure to consider that treatments have become senseless and conflict with patient's best interest. The implementation of E/AS cannot be reduced to a simple affirmation of the Principle of autonomy. Such procedures present genuine difficulties and the risk of drift. CONCLUSION: We deliver a message of prudence and caution. Should we address painful end of life and moral suffering issues, by suppressing the subject, i.e. ending the patient's life, when comprehensive palliative care has not first been fully granted to all patients in need of it ?

[Critical appraisal of organ procurement under Maastricht 3 condition]. / Analyse critique du prélèvement en condition M3 de Maastricht.

The ethics committee of the French Society of Anesthesia and Intensive Care (Sfar) has been requested by the French Biomedical Agency to consider the issue of organ donation in patients after a decision of withdrawing life supporting therapies has been taken. This category of organ donation is performed in the USA, Canada, United Kingdom, the Netherlands and Belgium. The three former countries have published recommendations, which formalize procedures and operations. The Sfar ethics committee has considered this issue and envisioned the different aspects of the whole process. Consequently, it sounds a note of caution regarding the applicability of this type of organ procurement in unselected patient following a decision to withdraw life supporting therapies. According to the French regulation concerning organ procurement in brain dead patients, the committee stresses the need to restrict this specific way of procurement to severely brain injured patients, once confirmatory investigations predicting a catastrophic prognosis have been performed. It suggests that the nature of the confirmatory investigation required should be formalized by the French Biomedical Agency on behalf of the French parliamentarians. This should help preserving population trust regarding organ procurement and provide a framework to medical decision. This text has been endorsed by the Sfar.

Kenneth J. Rothman and multicausality in epidemiology.

BACKGROUND: The sufficient-component cause model is one of several conceptual models for causation that appeared in the 1970s in response to the problem of multicausality in chronic diseases. METHODS: The aim of this article is to present the Rothman model as he introduced it in his seminal article "Causes," written in 1976. RESULTS: We show that: the notion of sufficiency and necessity, as opposed to the notion of probability, succeeded in forming a valid concept of cause; that this theoretical model recently introduced in other models of causality in epidemiology did not always succeed in solving the several practical problems related to multicausality, which Rothman wanted to solve by defending a working definition of causality. CONCLUSION: Despite its weaknesses, the Rothman model has contributed significantly to the understanding of what a cause is in epidemiology, making it possible to address this question from a point of view unfamiliar to the "risk factor" approach to diseases.

[Evidence-based medicine: methods and critics]. / Evidence-based medicine: méthodes et critiques.

The evaluation of professional practices is now obligatory. Evidence-Based Medicine, defined as the use of current best evidence in making well-informed decisions, could play a major role for that purpose, as it allows an objective evaluation of the clinical practices by the use of there commendations it gives rise to. The aim of that study was to describe the EBM decision procedure with examples in anaesthesia, and to present its limits, particularly due to the problems of hierarchies of evidence.

Role of hospital stay and antibiotic use on Pseudomonas aeruginosa gastrointestinal colonization in hospitalized patients.

The prospective cohort study presented here assessed the risk factors associated with Pseudomonas aeruginosa gastrointestinal colonization (PAGIC) in 933 patients hospitalized in five different wards in a French university hospital. A total of 195 patients were colonized. By logistic regression, hospitalization in an intensive care unit and length of hospital stay were independent risk factors. A significant association was observed between fluoroquinolone use and PAGIC caused by an ofloxacin-resistant strain (p < 0.0001), imipenem use and PAGIC caused by an imipenem-resistant strain (p < 0.0002) and ceftazidime use and PAGIC caused by a ceftazidime-resistant strain (p < 0.02). The ecological impact of antibiotic use is of great clinical relevance and clinicians should consider antimicrobial resistance in order to limit the development and dissemination of resistant microorganisms.

Risk-factors for gastrointestinal colonisation with resistant Enterobacteriaceae among hospitalised patients: a prospective study.

This study assessed the incidence of gastrointestinal colonisation by resistant Enterobacteriaceae among hospitalised patients, and identified risk-factors for ceftazidime and ofloxacin resistance. A prospective cohort study was performed in five wards in a French teaching hospital during a 2-year period. Patients hospitalised for > 48 h were enrolled between 17 April 2000 and 30 April 2002. A rectal swab was taken at admission, then once-weekly and/or on the day of discharge. In total, 933 patients were investigated and 585 amoxycillin-resistant isolates were obtained. Resistance rates for ceftazidime and ofloxacin were 9.4% and 4.8%, respectively. Multivariate analysis indicated that previous hospitalisation (p < 0.004) and exposure to amoxycillin-clavulanate (p < 0.003) and ceftriaxone (p < 0.002) were associated significantly with ceftazidime resistance. Hospitalisation in the urology ward (p < 0.02) and previous exposure to fluoroquinolones (p < 0.01) were the two independent risk-factors associated with ofloxacin resistance. The results of the study confirmed that antibiotic use selected resistant Enterobacteriaceae from the gut flora. Resistance was observed mostly in patients with previous antibiotic exposure and previous hospitalisation in wards with a high antibiotic selection pressure.

[Body mass index and albuminemia in patients under mechanical circulatory assistance before cardiac transplantation]. / Indice de masse corporelle et albuminémie péri-implantatoires de patients sous assistance circulatoire mécanique (ACM) en attente d'une transplantation cardiaque.

OBJECTIVE: To evaluate the prognostic influence of peri-implantation nutritional status of patients under mechanical circulatory assist (MCA) prior to cardiac transplantation (CT). STUDY DESIGN: Retrospective analysis of patients with cardiogenic shock included from June 1997 to December 2002. PATIENTS AND METHODS: Evaluation at MCA's implantation, at day (D) 30 and at CT or patient's death (D) of body mass index (BMI=body weight (kg)/size (m(2)), albuminemia (Alb g/l), expressed as median values (med) and range (min-max). Odds ratio (OR) and CI 95%) were calculated. A multivariate analysis was performed to determine variables related to D or CT success. RESULTS: Thirty-four patients (30 men), median age 40.5 years (10-63), were included. MCA types were cardiac pumps (N=3); pneumatic (N=18) or electric (N=5) ventricular assist devices and artificial heart (N=8). Global mortality was 56% (19 over 34 patients) and at implantation in the global population (N=34) BMI was 19.4 (9.3-28.1) and Alb 24.6 g/l (15-37.5). At MCA's implantation and D30 respectively, 38 and 42% of the patients had a severe hypoAlb (Alb<30 g/l) and a BMI<19 attesting of a seriously deteriorated nutritional status. No significant statistical difference was observed on median BMI of transplanted patients T (N=15) and expired patients D (N=19). Alb was significantly different (p<10(-4)) between T and D patients: median Alb: 30 g/l (20-37.5) in T patients, 20 g/l (15-31) in D patients. HypoAlb<21 g/l was an independent prognostic factor of death (p=0.004; OR: 0.541; IC95% : 0.36-0.82) and Alb>33 g/l an independent prognostic factor of CT success (p=0.003; OR:1.38; IC95% : 1.12-1.71). CONCLUSION: These results seem to demonstrate that at MCA implantation, a seriously deteriorated albuminemia level (<30 g/l) negatively impacts patients overall survival after CT.

[Intensive care unit hospitalization after colorectal surgery]. / Recours à l'hospitalisation en unité de soins intensifs après chirurgie colorectale.

OBJECTIVE: Intensive care units (ICU) support critically ill patients during the perioperative period. Few studies exist focusing on ICU hospitalisation after colorectal surgery. The objective of the study was to 1) detect predictive factors of mortality and length of stay in ICU after colorectal procedures, and 2) compare the autonomy status of the patients before and 30 days after their ICU stay. PATIENTS AND METHODS: This study followed a prospective non randomized cohort in our colorectal surgery unit. During a period of one year (January 1st to December 31th, 2000) 351 colorectal procedures were performed and 54 patients were admitted to ICU after surgery. For each patient, 37 parameters were collected on a standardized register. Predictive factors of mortality (30 days after the procedure) and ICU stay (up to 3 days) were studied by univariate and multivariate statistical analysis. Self autonomy before surgery and 30 days after was also investigated. RESULTS: "Multiple-intervention" was the only independent factor influencing mortality. Both "low autonomy status before surgery" and "pulmonary comorbidity" increased the length of stay. Regarding the 48 survivors, 45 (94%) recovered the same autonomy index as in the preoperative period 30 days after the procedure. CONCLUSION: This study highlights the poor predictive factors influencing mortality during or after ICU stay following colorectal surgery, and emphasizes two preoperative parameters increasing the length of stay up to 3 days. This should guide the informations given to the patients families. Finally, this study confirms the good quality of self-sufficiency after ICU stay even for a long time (over 3 days).

Isolated mitral valve replacement with St. Jude medical prosthesis: long-term results: a follow-up of 19 years.

BACKGROUND: In this retrospective study, approximately 440 patients received mitral valve replacements with the St Jude Medical prosthesis. The last patient was operated on 10 years before the beginning of the follow-up. The extended follow-up was 19 years. METHODS AND RESULTS: Four hundred forty patients (sex ratio, 1.32 [men to women]; age, 60+/-11.4 years; age range, 7 to 75 years) were operated on from 1979 to 1987. All patients underwent isolated mitral valve replacement. Tricuspid plasty was the only associated procedure. The follow-up at 19 years was 98% complete. The overall actuarial survival rate was 63+/-3.3% at 19 years, and the actuarial survival rate (only valve related) was 83+/-2.7%. The operative mortality rate (0 to 30 days) was 4.09%. We found that 89.4% of the patients alive at 19 years were in NYHA class I/II. Multivariate analysis showed that age and sex were significantly correlated with valve-related mortality and that age, sex, NYHA class, and atrial fibrillation were significantly correlated with overall mortality. The linearized rates (percent patient-years) of thromboembolism, thrombosis, and hemorrhage were 0.69, 0.2, and 1, respectively. At 19 years, freedom from endocarditis and reoperation was 98.6+/-1% and 90+/-3%, respectively. CONCLUSIONS: In this study, the very-long-term results confirm the excellent durability of the St Jude Medical prosthesis in the mitral position and show the difficulty of adjusting the anticoagulation protocol, even after long-term treatment.

[Rational choice, prediction, and medical decision. Contribution of severity scores]. / Choix rationnel, prédiction et décision médicale: apport des scores de gravité.

The aim of this study was to determine what type of representation the medical doctor adopted concerning the uncertainty about the future in critically ill patients in the context of preoperative evaluation and intensive care medicine and to explore through the representation of the patient health status the different possibilities of choice he was able to make. The role played by the severity classification systems in the process of medical decision-making under probabilistic uncertainty was assessed according to the theories of rational behaviour. In this context, a medical rationality needed to be discovered, going beyond the instrumental status of the objective and/or subjective constructions of rational choice theories and reaching a dimension where means and expected ends could be included.

Postoperative analgesia by epidural methylprednisolone after posterolateral thoracotomy.

The aim of this study was to evaluate the potential analgesic effect of epidural methylprednisolone (MP) after posterolateral thoracotomy (PLT). Adult male patients undergoing PLT for lung surgery were included in a prospective, randomized, double blind study. Peroperative analgesia (bupivacaine plus sufentanil) was given by a thoracic epidural catheter associated with general anaesthesia. After surgery, patients received either MP 1 mg kg(-1) followed by a continuous epidural infusion of MP 1.5 mg kg(-1) during 48 h (MP group) or 0.9% saline as a bolus injection and continuous epidural infusion (P group). Additional morphine analgesia was administered by i.v. patient-controlled analgesia. Pain was assessed at rest and with mobilization every 4 h after operation during 48 h with a visual analogue scale (VAS). The primary end-point was the total morphine requirements during the 48 first postoperative hour. Twenty-four patients were allocated to MP (n=12) and P (n=12) groups. Characteristics of the two groups were similar. There were no differences between groups for morphine requirements (median and interquartile range) during the 48 h: 59 mg (40-78) in MP group vs 65 mg (59-93) in P group. There were no differences between groups for morphine requirements every 4 h during the 48 h and VAS for pain at rest and evoked pain. No side effects were reported. It was concluded in this small study that these results did not support the use of epidural steroids for postoperative analgesia after PLT.

Hepatic resections for non-colorectal metastases: forty resections in 35 patients.

BACKGROUND/AIMS: Hepatic resection, though now an accepted practice for colorectal primary tumors, is poorly documented for non-colorectal metastases. However, the few series reported suggest that this approach may lead to a significant increase in survival. METHODOLOGY: Study of 40 cases of resection in 35 patients with non-colorectal hepatic metastasis to define the role of hepatic resection between 1986 and 1997. RESULTS: Resection was performed for 5 metastases of ovarian and fallopian tube carcinoma, 8 gastrointestinal tract adenocarcinomas, 8 endocrine tumors, 8 sarcomas and 6 miscellaneous metastases, involving 17 lobectomies, 3 trisegmentectomies, 5 lateral segmentectomies and 15 non-anatomical local resections. Survival at 1, 2 and 5 years was 54 +/- 8, 42 +/- 8 and 27 +/- 8%, respectively. Hepatic metastases of gastrointestinal tract adenocarcinomas were found to have the poorest prognosis (median time: 13 months), and genital tract adenocarcinomas the best (27 months). CONCLUSIONS: Some carefully selected patients may benefit from liver resection for non-colorectal metastases.

Influence of respiratory disease on perioperative cardiac risk in patients undergoing elective surgery for abdominal aortic aneurysm.

We retrospectively reviewed perioperative cardiac complications in a series of 214 patients who underwent surgical treatment for infrarenal aortic aneurysm between 1992 and 1996. There were 192 men and 22 women, with a mean age of 68.3 years. Cardiac risk factors included angina in 28% of patients and previous myocardial infarction in 25%. Resting electrocardiography was normal in 80 patients (37.5%). Depending on clinical findings, thallium-201 scintigraphy was undertaken in 76 patients (35.5%) and led to elective coronary arteriography in 22 patients (10%). Results of coronary arteriography revealed lesions in 14 patients. Aortic reconstruction was performed by the transperitoneal route in all patients. Procedures consisted of aortoaortic bypass (63%), aortobiiliac bypass (27.5%), or aortobifemoral bypass (9.5%). Nine patients (4.2%) died within the first 30 postoperative days. The cause of death was myocardial infarction (MI) in two patients (1%), colonic necrosis in two (1%), acute pancreatitis in one (0.5%), acute renal insufficiency in three (1.4%), and multiple organ failure in one patient (0.5%). Nonfatal cardiac complications were observed in 15 patients (7%). Statistical analysis of risk factors revealed two predictors of perioperative cardiac complications, i.e., history of chronic bronchitis and reoperation. On review of the literature, we cannot propose a routine preoperative work-up. Prospective multicentric studies are needed to determine the predictive value of current preoperative screening methods.

[Hepatic resections for non-colorectal, non-neuroendocrine metastases. Results of 32 hepatectomies in 27 patients]. / Résections hépatiques pour métastases non colorectales, non neuroendocrines. Résultats de 32 hépatectomies chez 27 patients.

STUDY AIM: Liver resections for metastases are commonly performed in colorectal primary tumors and poorly documented in non colorectal tumors. The aim of this study was to report a series of 32 liver resections in 27 patients for different types of non colorectal, non neuroendocrine liver metastases. PATIENTS AND METHOD: From 1986 to 1997, 27 patients (20 women and 7 men, mean age: 56.8 years) were operated on in the same center for liver metastases. Initial cancer was female genital tract (ovarian and fallopian tube) adenocarcinomas (n = 5), gastrointestinal tract adenocarcinomas (n = 8), sarcomas (n = 8), and miscellaneous cancers (n = 6). Liver resections included atypical resections (n = 9), right hepatectomies (n = 11), extended right hepatectomies (n = 2), left hepatectomies (n = 4) and resections of 2 or 3 segments (n = 6). RESULTS: There was no perioperative death. Postoperative morbidity included 8 complications in seven patients, requiring reintervention in three patients. Follow-up was complete for all patients. Survival rate at one, two and five years was 59, 44 and 29% respectively. The longest median survival time was observed in genital tract adenocarcinomas (27 months), whereas the other types of malignancies had a 13- to 17-month mean survival rate. CONCLUSION: These results are almost similar to those observed in liver resections for colorectal metastases. Some carefully selected patients may benefit from liver resection for non colorectal, non neuro-endocrine metastases.

Intraoperative autotransfusion for repair of unruptured aneurysms of the infrarenal abdominal aorta. A multicenter study of 203 patients. Association Universitaire de Recherche en Chirurgie (AURC).

BACKGROUND: The goals of this study were to evaluate the costs and savings of intra- and postoperative blood transfusions as well as the potential biological modifications associated with the use of intraoperative blood salvage. METHODS: Intraoperative autotransfusion (IOAT) with wash-out was prospectively studied during the repair of unruptured aneurysms of infrarenal abdominal aorta in 203 patients operated on in 13 institutions. RESULTS: The mean quantity of blood retrieved was 688+/-468 mL The mean quantity of blood derivatives and intraoperative solutes used for repletion was 4,261 ml, ranging from 1,723 ml between days 0 to D2 to 562 ml from D3 to D8. Ninety-eight patients did not receive any blood derivatives at all. Thirty-five patients received plasma to correct coagulation factors. The quantity of autotransfused globular concentrate was less than 500 ml in 89 patients. CONCLUSIONS: IOAT precluded the need for transfusion of homologous globular concentrates, particularly in those patients who had bled most. On average, more than two globular concentrates were recuperated. Use of IOAT led to financial savings. Perioperative bleeding is not the only factor that intervenes in the decision to transfuse globular concentrates. Postoperative dilution is the most important factor as attested by the amount of protides and the hematocrit. Coagulation factors are modified but remain compatible with normal hematosis in 83% of patients undergoing operation.

Early and late outcome after elective cardiac surgery in patients with cirrhosis.

BACKGROUND: A prospective study was performed to evaluate the early and late outcome after elective cardiac surgery in patients with cirrhosis. METHODS: All patients who underwent elective cardiac surgery between 1995 and 1997, and were suspected of having a history of cirrhosis, were followed in the intensive care unit (ICU), during hospitalization and after hospital discharge. All patients received high doses of aprotinin during surgery. RESULTS: Ten patients of Child-Pugh class A and 2 patients of Child-Pugh class B were studied. All patients had signs of portal hypertension, and 11 of 12 patients had thrombocytopenia. In the first 24 h after operation, the median chest tube output was 810 mL (range 350 to 1,500 mL). Median ICU and hospital stays were 3 and 15 days, respectively (range 2 to 10 and 7 to 36 days, respectively). Seven patients experienced postoperative morbidity and 7 patients had significant complications after their hospital discharge. One death occurred in the ICU. Two deaths occurred after hospital discharge and were related to further hepatic damage. CONCLUSIONS: These results suggest that, in patients with mild or moderate cirrhosis, the incidence of significant complications was high after elective cardiac surgery, increasing the length of stay in ICU and overall hospitalization time and compromising the health status, even well after the operation.