No description available.

No description available.

No description available.

Introduction: Haemorrhage is the leading cause of death from battlefield trauma and the second leading cause of death after civilian trauma. Blood loss causes 80% of all early trauma deaths. Control of non-compressible, truncal haemorrhage prior to surgical correction may save many lives especially in the case of military casualties where evacuation times to surgical care can be prolonged. A number of approaches have been advocated to limit blood loss. Several anecdotal reports suggest that recombinant activated factor VII (rFVlla) may arrest uncontrolled bleeding after trauma. However, the majority of prospective randomised controlled trials show little benefit in survival. The aim of this study was to determine whether rFVila could increase survival time and reduce the volume of blood loss in a model of noncompressible arterial haemorrhage over a prolonged prehospital phase, relevant to battlefield evacuation. A secondary aim was to determine the effects of hypotensive vs normotensive resuscitation on the efficacy of rFVlla. Methods: A prospective randomised controlled trial was conducted on 27 terminally anaesthetised pigs. The animals were randomly allocated to one of four treatment groups. All animals were subjected to a controlled haemorrhage of 40% of their estimated blood volume. They were then given either rFVlla (180 Jig/kg) or placebo (saline 0.3mUkg) intravenously and a 4-5 mm aortotomy created in the infra-renal aorta using a pre-implanted wire. Resuscitation was commenced with 0.9% saline to one of two target systolic arterial blood pressures: 110 mmHg (normotensive) or 80 mmHg (hypotensive). Results: Recombinant FVlIa was associated with a significantly prolonged survival time in animals managed hypotensively (214 [79-349] vs 35 [19-52] minutes mean [95% confidence interval], rFVlla vs placebo, p =0.03, Peto log rank test). There was no significant difference in survival time between those given rFVlla and placebo in groups managed normotensively (128 [6-250] vs 40 [15-66] minutes mean respectively, P =0.27). Both rFVlla and hypotensive management were associated with reduced volumes of uncontrolled haemorrhage. Post mortem evaluation revealed no evidence of inappropriate intravascular thrombi or microthrombi associated with the use of rFVlla. Conclusions: Recombinant FVlIa, combined with hypotensive resuscitation, can increase survival time and reduce blood loss in a model of uncontrolled arterial haemorrhage. The increase in survival time is clinically relevant for military evacuation of battlefield casualties to surgical care.

I have addressed in this thesis several major areas impacting upon the practice of regional anaesthesia in clinical practice. These include patient-centred issues (consent, anxiety and satisfaction) and technical issues affecting the delivery of service (needling and the use of ultrasound-guidance). While disparate, these issues combine to affect the efficient, safe and humane delivery of an important part of modern healthcare. These areas have been chosen as those most amenable to study and those with the greatest potential for real-world impact. Their unifying theme is the improvement in the delivery of regional anaesthesia.

Annually in Britain, iatrogenic harm results in patient deaths, increased morbidity, and millions of pounds spent on additional healthcare. Errors in the administration of drugs have been identified as a leading cause of patient harm in major international reports,1 2 and the literature also suggests that most practicing anaesthetists have experienced at least one drug error.34 Methods of conventional drug administration in anaesthesia are idiosyncratic, relatively error prone, and make little use of technology to support manual checking. While there is support for the use of double-checking during anaesthesia practice, the availability of a second person during every drug administration, and issues around hierarchy and recognised automaticity in checking 5 can potentially be the limitations. Currently there has been little work carried out in the UK in relation to the use of double checking protocols and there remains a need for a robust check that can be implemented within the National Health Service (NHS). The first study explored the feasibility of introducing a double check methodology, either second-person confirmation or electronic confirmation into clinical practice within the NHS. This was the first study of this nature within the NHS and explored the attitudes, barriers and benefits of each method. The second study was designed to explore the beliefs and attitudes of anaesthetists and Operating Department Practitioners (ODPs) on introducing technology which is designed to reduce drug error. This study also explored in greater depth the culture issues raised in the first study and the impact of introducing the electronic confirmation on the anaesthetist’s workload. The findings suggested that while many participants acknowledged that the process of second person double checking was an important factor to minimise the opportunity of any unsafe medication administration, the process of second person confirmation could be prone to human manipulation and could alter the behaviour and practice of the anaesthetist, resulting in a reluctance to adopt it. The electronic confirmation method was found to be more feasible. It did not rely on the presence of a second person at the time of drug administration, and did not impact on the anaesthetist’s workload. This thesis has shown that technology was more readily accepted and seen as more feasible to use by anaesthetists within their clinical practice. However, these studies have also shown that the culture and beliefs of the organisation and individuals, in particular of ‘blame and shame’, has such a strong influence that it continues to prevent a true safety culture developing into an open culture of reporting incidents, recognising that drug errors remain a problem, and that corrective measures are required to prevent them.

Background Routine induction of anaesthesia and maintenance haemodialysis are two examples of clinical procedures that exert a direct effect on the cardiovascular system. The exact incidence of haemodynamic instability during such procedures is not well described, as it would have required invasive intra-arterial monitoring, which is not justified for routine use. As part of two service evaluations, i.e. routine induction of anaesthesia and maintenance haemodialysis, we utilised a noninvasive continuous beat-by-beat haemodynamic monitor, which works using a finger cuff (Finometer), to assess the incidence of haemodynamic instability encountered during these procedures in comparison to the conventionally used intermittent noninvasive blood pressure (NIBP) measurement protocols. Methods Using the Finometer, we recorded haemodynamic variables during induction of anaesthesia in 100 patients undergoing elective surgery, and during maintenance haemodialysis in 25 patients with established renal failure. Firstly, we assessed the feasibility of using the Finometer during induction of anaesthesia and haemodialysis by evaluating its success rate in providing measurements of haemodynamic variables, and by assessing its accuracy in comparison to the readings obtained by the conventional NIBP devices during our service evaluations. Secondly, we assessed the incidence of haemodynamic instability during both procedures as detected by the Finometer in comparison to the existing conventional intermittent NIBP measurement protocols. Results and discussion The Finometer was successful in providing adequate haemodynamic monitoring in 96% and 86% of the attempts to use it in our service evaluations during induction of anaesthesia and haemodialysis respectively. The Finometer showed comparable accuracy in terms of BP monitoring to the conventional NIBP monitors during induction of anaesthesia and haemodialysis. A high incidence of significant hypotension as well as significant hypertension was shown during both, routine induction of anaesthesia and maintenance haemodialysis, which were underestimated or even missed by the conventionally used intermittent NIBP monitoring protocols. During induction of anaesthesia, 19% of the patients sustained an episode of hypotension defined as SBP less than 80 mmHg for more than 1 min, and 53% showed a transient increase of the SBP of more than 20% from baseline values. During Haemodialysis, 28% of the patients sustained an episode of hypotension defined as SBP less than 90 mmHg for more than 10 min, and 16% sustained an episode of severe hypertension defined as SBP more than 180 mmHg for more than 10 min. Conclusion Haemodynamic instability is commonly encountered during routine induction of anaesthesia and maintenance haemodialysis. Continuous noninvasive finger arterial haemodynamic monitoring is more reliable than the conventionally used protocols of intermittent NIBP monitoring in detecting such haemodynamic instability, thus providing higher levels of patient safety. Extra and early information about haemodynamic variables, as provided by the Finometer, may provide a better insight on the exact cause of haemodynamic instability, which may aid the physicians in prompt and targeted management.

- / -

L'hyperglycémie est très souvent présente chez le patient de réanimation, liée à une insulinorésistance et une production accrue de glucose par le foie. La relation entre hyperglycémie en réanimation et pronostic défavorable était déjà établie lorsque Van den Berghe et al montra une réduction de mortalité associée à une insulinothérapie intensive visant à normaliser la glycémie. Du fait de l'impossibilité à reproduire cet effet bénéfique et de la mise en évidence d'une surmortalité associée à un contrôle glycémique strict par rapport à un contrôle visant une glycémie < 10mmol/L (étude NICE-SUGAR), le premier objectif de cette thèse fut de rechercher à l'aide d'une étude multicentrique (étude CGAO-REA) une association entre utilisation d'un logiciel facilitant le contrôle glycémique strict et réduction de mortalité. Malgré le résultat négatif de l'étude CGAO-REA, l'intérêt d'un système informatisé demeure en raison de la complexité du contrôle glycémique. Sont ainsi discutés les sujets suivants appliqués à la réanimation : aléas du contrôle glycémique, méthodes de mesure de la glycémie, contrôleurs glycémiques, insulinorésistance, variabilité glycémique, impact du statut diabétique et systèmes informatisés. Le deuxième objectif de cette thèse est de détailler les nouveaux objectifs du contrôle glycémique en réanimation et d'en présenter l'évolution et les perspectives. En s'inspirant des diabétologues engagés dans le développement d'un pancréas artificiel et en généralisant les évaluations in silico, des équipes pluridisciplinaires pourraient construire une station de contrôle informatisée afin d'obtenir un contrôle glycémique optimisé et individualisé en réanimation. / Stress-induced hyperglycemia is common in critically ill patients due to insulin resistance and increased hepatic output of glucose. The relationship between stress hyperglycemia and poor outcome for patients hospitalized in the intensive care unit (ICU) was already established when Van den Berghe et al demonstrated that tight glucose control (TGC) by intensive insulin therapy was associated with decreased mortality and rate of complications in surgical ICU patients. Because further randomized controlled studies have failed to replicate any mortality benefit and have even shown an increased mortality (the NICE -SUGAR study), the primary objective of this thesis was to test the hypothesis that TGC with a computerized decision-support system (CDSS) reduces the mortality in adult critically ill patients as compared to conventional glucose control targeting BG levels to <10 mmol/L (the CGAO-REA study). We discussed several aspects of glucose control in the ICU (integration in the clinical workflow, confounding issues in BG measurements, insulin resistance, glucose variability, impact of diabetic status) and reviewed existing computerized tools. Despite the negative result of the CGAO-REA study, the interest for CDSSs remains because meanwhile new therapeutic goals for glucose control in the ICU have emerged. The second objective of this thesis is to present ongoing developments. Based on research programs targeting an artificial pancreas for diabetic patients, in silico trials, multidisciplinary approaches integrating human factors, a computerized glucose control station could be developed to allow clinicians to achieve an optimized, individualized glucose control in the ICU.

Bibliography :leaves 350-384. / xv, 384 leaves : ill. ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Introduces possible neural mechanisms of action of general anaesthesia. / Thesis (Ph.D.)--University of Adelaide, Dept. of Physiology, 1994