O objetivo desta tese foi avaliar a influências da exposição a diversos fatores de risco e ocorrência de infecção hospitalar (IH) e examinar fatores de risco relacionados às pneumonias associadas ao uso de ventilador mecânico (PAV) em crianças críticas. Usamos os métodos de estudo prospectivo envolvendo uma coorte de 268 crianças menores de três anos, realizado em unidade de pacientes graves, de janeiro de 1997 e setembro de 2000. Aplicou-se técnica de regressão de Poisson para estimar razões de risco e estratégia de abordagem hierárquica para identificar fatores de risco associados à IH. Apenas para 179 crianças críticas que usaram ventilador mecânico, aplicou-se a regressão de Cox para estimar razões de risco e identificar fatores de risco associados à PAV. Os resultados apresentaram 74 infecções hospitalares diagnosticadas no total, com taxa de incidência de 48,1 IH por 1000 pacientes-dia. Foi determinante para ocorrência de infecção hospitalar, idade superior a dois anos (Razão de Risco) [RR]: 2,66; intervalo de confiança [IC]: 95%: 1,46-4,58), uso de sonda vesical (RR: 2,92; IC 95%: 1,47-5,80), uso de nutrição parenteral (RR: 1,90; IC 95%: 1,15-3,13), realização de broncoscopia (RR: 1,84; IC 95%: 1,03-3,27), tempo de exposição ao cateter vascular central (RR: 2,36; IC 95%: 1,18-4,71) e ao ventilador mecânico (RR: 1,72; IC 95%: 0,94-3,15). Observou-se PAV em 29 crianças (16,3%), com taxa de incidência de 29,3 casos por 1000 dias de ventilação mecânica (IC 95%: 20,34-42,11), dos quais 50% dos eventos ocorreram até o quinto dia de ventilação. A taxa de risco diária aumentou até um máximo de 2,3%, observada no 7 dia de ventilação, e reduziu a partir daí. Foram fatores de risco para PAV na análise multivariada hierarquizada, idade acima de 1 ano (RR: 3,49; IC 95%; 1,64-7,45), cirurgia do aparelho digestivo (RR: 5,05; IC 95%; 2,17-11,78) e nutrição parenteral (RR: 2,68; IC 95%: 1,24-5,8). /exposição a antibióticos antes da internação conferiu proteção (RR: 0,29; IC 95%: 0,13-0,66). Concluímos que os resultados encontrados neste estudo indicam que a influência do tempo de exposição é determinante para a ocorrência de infecções hospitalares e está associado aos processos de cuidados do paciente crítico. Políticas institucionais direcionadas ao controles e prevenção das IH devem fazer de estratégias fundamentais para a qualidade da assistência e segurança do paciente internado. Vigilância de Saúde Pública e componentes longitudinais de estudo de fatores de risco para infecções hospitalares e para pneumonias associadas ao ventilador podem ajudar avaliar prognósticos e planejar e testar medidas preventivas, concentrando-se esforços na primeira semana de ventilação. / The objective of this thesis is evaluate the influence of exposure to some risk factors and the occurrence of nosocomial infections (NI) and to examine risk factors for ventilator-associated pneumonia (VAP) in critically ill children. We use the prospective study was carried out on 268 children less than three years old, at a pediatric intensive care unit, from January 1997 to September 2000. Poisson regression and hierarchized multivariate analysis were performed to estimate incidence rate ratios and to identify risk factors associated to nosocomial infections. To 178 children only, who were ventilated, Cox regression was performed to estimate hazards ration and to identify risk factors for VAP. Our results were 74 nosocomial infections were diagnosed, and the overall incidence of NI cases was 48.1 per 1000 patient-days. The final model showed as predictors of nosocomial infections were age over two (incidence rate ratio [IRR]: 2.66, confidence interval [CI] 95%:1.46- 4.58), urinary tract catheter (IRR: 2.92; CI 95%: 1.47-5.8), parenteral nutrition use (IRR: 1.90; CI 95%: 1.15-3.13), bronchoscope use (IRR: 1.84; CI 95%: 1.03-3.27), central catheters exposure time (IRR: 1,72; CI 95%: 0.94-3.15), 29 childrem (16.3%) developed VAP and the overall incidence was 29.3 per 1000 ventilator-days (OC 95%: 29.34-42.11); 50% of all cases had happened until day 5 of ventilator. The daily hazard risk increased to 2.2% at day 7, and decreased after that. Risk factors of VAP in hierarchized multivariate analysis were age over one (hazards ratio [HR]: 3.49; IC 95%: 1.64-7.45), digestive surgery (HR: 5.05; CI 95%: 2.17-11.78), parenteral nutrition (HR: 2.68; IC 95% 1.24-5.80). Exposure to antibiotics conferred protection (HR: 0.29; IC 95%: 0.13-0.65). Our conclusion is: the results suggest that the influence of the exposure time is determinant of nosocomial infections occurrence, and it is healthcare associated in critically ill patients. Institutional politics might be aimed at prevention and control activities, as strategy to promote healthcare quality and patient safety. Public health surveillance and longitudinal studies of risk factors for ventilator-associated pneumonias can help to assess prognosis, and devise and test preventive strategies, and efforts might be concentrated on the first week of ventilation assistance.

Contextualização: Na última década, houve um grande crescimento de ações de melhoria da qualidade assistencial e de promoção da segurança do paciente, com objetivo de diminuir a ocorrência de desfechos preveníveis, como a pneumonia associada à ventilação mecânica (PAVM). Embora existam várias diretrizes clínicas, ainda não há evidências suficientes para um consenso quanto aos cuidados não farmacológicos de prevenção (CNFP) a serem implantados e sua contribuição na redução da PAVM. Objetivos: Avaliar o impacto da implantação de CNFP na ocorrência da PAVM. Métodos: O estudo foi realizado com pacientes adultos, submetidos à ventilação mecânica (VM) por mais de 48 horas, internados em centro de tratamento intensivo clínico e cirúrgico, de 34 leitos. No primeiro artigo, é descrito um quase experimento com controles históricos, realizado no período de setembro de 2004 (pré-intervenção -20 meses) a dezembro 2007 (pós-intervenção – 20 meses). Constou do experimento a implantação do protocolo de prevenção da PAVM, com cuidados não farmacológicos. No segundo artigo, é descrito um estudo de coorte prospectiva de pacientes submetidos à VM no período de junho de 2006 a julho de 2007. Entre as onze medidas preventivas que fazem parte da rotina de cuidados, foram escolhidas seis para serem examinadas na rotina assistencial de pacientes submetidos à VM: manutenção da cabeceira elevada (HOB) de 30º a 45º; manutenção do circuito de VM (CVM) sem condensação e sujidade; manutenção de trocadores de umidade e calor (HME); monitorização da pressão do balonete (PB) do tubo endotraqueal entre 18 mmHg a 25 mmHg; realização da higiene oral (HO) e realização fisioterapia respiratória (FR). Estabeleceu-se um ponto de corte de 80% para adesão aos cuidados de prevenção. Considerou-se o CNFP como adequado quando, durante a internação, o paciente foi observado em frequência, igual ou maior de 80%. Resultados: No primeiro estudo, foram diagnosticados 367 episódios de PAVM durante os 40 meses, em um total de 20285 dias de ventilação mecânica. A taxa de PAVM baixou em 28,7%, ou seja, de 20,6 casos por 1.000 dias de ventilação no período pré-intervenção para 16 casos por 1.000 dias de ventilação nos 20 meses seguintes à intervenção (P<0,001). Na análise de regressão segmentada verificou-se uma redução abruta de nível significativa do desfecho (t = -2,506; p= 0,017), mas não foi observado uma redução de tendência significativa do desfecho (t= -1,670; p= 0,104). No segundo estudo, a adesão aos seis CNFP foi aferida em 5.781 observações, em 541 internações de pacientes submetidos à ventilação mecânica, com 111 casos de PAVM. Nas internações em que os pacientes receberam cuidados adequados de prevenção, houve redução na ocorrência de PAVM de 61% para a FR (OR=0,39; IC95%= 0,18 a 0,84); 43% para a HOB (OR=0,57; IC95%= 0,31 a 0,99). Houve interação entre HO e PB, com uma redução da razão de chances de 56% para HO (OR=0,44; IC95%= 0,24 a 0,82) e 58% para PB (OR=0,42; IC95%= 0,21 a 0,85). Na ausência de um desses dois cuidados, o outro presente passa a ser fator de risco. Estimouse a fração de risco atribuível para não realização de FR, HO, monitorização PB e manutenção da HOB em frequência adequada, implicando a ocorrência de, respectivamente, 49% (IC95%= 13% a 65%), 29% (IC95%= 9% a 39%), 32% (IC95%= 8% a 44%) e 9% (IC95%= 0% a 13%) das PAVM. Conclusão: Os cuidados de prevenção da PAVM implantados mostraram-se medidas tecnologicamente simples, exequíveis em qualquer realidade e de baixo custo, possibilitando redução no desenvolvimento de PAVM e aumento na segurança do paciente. / Context: In the past decade there has been a large increase in the number of measures aimed at improving service quality and promoting patient safety. Most of these measures have been aimed at reducing the incidence of preventable iatrogenic morbidities such as ventilator-associated pneumonia (VAP). While related clinical guidelines do exist, there is still little evidence to indicate which of the many non-pharmacological preventive care (NPPC) measures provide the best protection against VAP. Objectives: To evaluate the impact of NPPC interventions on the incidence of VAP. Methods: The population represented in this study are adult, intensive care patients, admitted without pneumonia, which required mechanical ventilation (MV) for at least 48 hours. The 34 bed intensive care center clinic and surgical is part of a public, university hospital in Porto Alegre, Brazil. The first article describes a quasi-experimental study done during the 40 months between September, 2004 and December, 2007. Halfway through this period a non-pharmacological, VAP prevention protocol was introduced to the routine care intubated patients. As such the control group are those patient treated prior to introduction of the protocol and the experimental group, those patients treated after its introduction. The second article describes a prospective study of cohort of patients requiring mechanical ventilation between June, 2006 and July 2007. Of the 11 NPPC interventions that make up the protocol, 6 were selected for this study: maintaining the head-of-bed elevation between 30º and 45º (HOB), maintaining the mechanical ventilation circuit (MVC) clean and dry, maintaining the heat moisture exchange (HME) equipment; maintaining endotracheal tube cuff pressure (CP) between 18 and 25 mmHg, realizing oral hygiene (OH) and doing respiratory physiotherapy (RP). These interventions were audited, for each patient during their entire length of stay, to measure compliance with the protocol. For each non-pharmacological preventive intervention, any patient receiving ≥8O% of the recommended frequency of care was considered to have received that intervention. Results: During a total of 20285 days of MV, 367 cases of VAP were diagnosed. The incidence of VAP fell 28.7% (p<0.001), between the 20 month pre-intervention period (20.6 cases/1.000 MV days) and the 20 month post-intervention periods (16 cases/1.000 MV days). In the analysis of segmented regression there was a significant abrupt reduction in the level of outcome (t = - 2.506; p=0.017), but reduction of tendency of the outcome was not significant (t= - 1.670; p= 0.104). In the second study compliance with the 6 non-pharmacological preventive interventions was measure by means of 5.781 observations in 541 patients of whom 111 developed VAP. Patients that received ≥8O% of the recommended frequency of a specific intervention exhibited a reduction in the risk of developing VAP: 61% for RP (OR= 0.39; CI95%= 0.18 to 0.84); 43% for HOB (OR= 0.57; CI95%= 0.31 to 0.99); 56% for OH (OR= 0.44; CI95%= 0.24 to 0.82) and 58% for CP (OR= 0.42; CI95%= 0.21 to 0.85). OH and CP maintenance are interdependent. In the absence of the other intervention each goes from being protective to being a risk factor. The VAP risk attributable to non-compliance in each of RP, OH, CP and HOB is estimated to be: 49% (CI95%= 13% to 65%), 29% (CI95%= 9% to 39%), 32% (CI95%= 8% to 44%) and 9% (CI95%= 0% to 13%) respectively. Conclusion: NPPC interventions are technologically simple, low cost that, in combination, provide a patient security and substantial level of protection against VAP.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-08-22T18:18:37Z No. of bitstreams: 1 fredericocastrocostapovoa.pdf: 1118829 bytes, checksum: 503d88e552bed3067b98ec8516596272 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-08-28T13:16:01Z (GMT) No. of bitstreams: 1 fredericocastrocostapovoa.pdf: 1118829 bytes, checksum: 503d88e552bed3067b98ec8516596272 (MD5) / Made available in DSpace on 2018-08-28T13:16:01Z (GMT). No. of bitstreams: 1 fredericocastrocostapovoa.pdf: 1118829 bytes, checksum: 503d88e552bed3067b98ec8516596272 (MD5) Previous issue date: 2018-06-29 / Introdução: A pneumonia associada à ventilação mecânica (PAV) correlaciona-se com um aumento da mortalidade e dos custos na unidade de terapia intensiva (UTI). Muitas estratégias preventivas foram estudadas e, entre as mesmas, encontra-se a administração de antibióticos pelo trato respiratório. Objetivo: Avaliar o efeito de antibióticos administrados pelo trato respiratório na prevenção da PAV em pacientes em ventilação mecânica (VM). Métodos: Foram pesquisados estudos que avaliaram o impacto da administração profilática de antibióticos através do trato respiratório na ocorrência de PAV, em bases eletrônicas de dados (PubMed, Embase, Web of Science, Cochrane e Scielo), e em referências de artigos relevantes sobre o tema. Não houve restrição em relação a data de publicação e selecionamos artigos publicados em inglês, espanhol ou português. Três autores avaliaram independentemente os títulos, resumos e os textos completos e extraíram os dados utilizando formulários padronizados. Foi utilizado o modelo de efeitos randômicos para o agrupamento dos estudos individuais. O desfecho primário foi a ocorrência de PAV e o secundários, a mortalidade hospitalar e a ocorrência de PAV por patógenos resistentes. Resultados: Foram incluídos seis estudos envolvendo 1158 pacientes (632 receberam antibióticos profiláticos). Nesta metanálise foi demonstrado que antibióticos administrados de forma profilática, pelo trato respiratório, reduziram a incidência de PAV quando comparado ao placebo ou a nenhum tratamento (OR 0,53; 95% IC 0,34–0,84). Esse efeito foi identificado quando os antibióticos eram administrados por meio de nebulização (OR 0,46; 95% IC 0,22–0,97), mas não quando eles eram administrados por meio de instilação intra-traqueal (OR 0,57; 95% IC 0,28–1,15). Não houve diferença significativa, entre os grupos, na mortalidade na unidade de terapia intensiva (UTI) (OR 0,89; 95% IC 0,64–1,25). A profilaxia com uso de antibiótico inalatório não alterou a incidência de PAV por bactérias resistentes (OR 0,67; 95% IC 0,17–2,62). Conclusão: Antibióticos administrados de maneira profilática pelo trato respiratório através de nebulização reduziram a incidência de PAV sem um efeito significativo na mortalidade em UTI ou na incidência de patógenos resistentes. / Introduction: Ventilator-associated pneumonia (VAP) is correlated with increased mortality and costs in the intensive care unit (ICU). Several preventive strategies have been studied, and among them is the antibiotic administration via the respiratory tract. Objective: We evaluated the effect of antibiotics administered via the respiratory tract to prevent the VAP in mechanically ventilated (MV) patients. Methods: We searched PubMed, Scopus, Embase, Web of Science, Cochrane, Scielo and references of relevant articles for trials that evaluated the impact of prophylactic antibiotics administered through the respiratory tract on the occurrence of VAP. No restrictions based on the date of publication were imposed, and we selected articles written in English, Spanish or Portuguese. Three authors independently screened titles, abstracts and full texts, and extracted data using standardized forms. The random effects model was used to pool individual study results. The primary end-point was the ocurrence of VAP, and the secondary end-points were hospital mortality and occurrence of VAP due to multidrug resistant (MDR) pathogens. Results:We included six comparative trials involving 1158 patients (632 received prophylactic antibiotic). Our metaanalysis revealed that prophylactic antibiotics administered through the respiratory tract reduced the occurrence of VAP when compared to placebo or no treatment (OR 0.53; 95% CI 0.34–0.84). This effect was seen when the antibiotics were given by nebulization (OR 0.46; 95% CI 0.22–0.97), but not when they were administered by intratracheal instillation (OR 0.57; 95% CI 0.28–1.15). We did not find a significant difference between the compared groups in the ICU mortality (OR 0.89; 95% CI 0.64–1.25). Antibiotic prophylaxis did not impact occurrence of VAP due to MDR pathogens (OR 0.67; 95% CI 0.17–2.62). Conclusions: Prophylactic antibiotics administered through the respiratory tract by nebulization reduce the occurrence of VAP, without a significant effect on either the ICU mortality or occurrence of VAP due to MDR pathogens.

En infectiologie, comme en cancérologie, la médecine personnalisée se développe. Ainsi, les traitements antibiotiques probabilistes cèdent leur place à des traitements adaptés aux pathologies et aux patients. En plus des risques d’échecs liés à une thérapie non adaptée, le traitement probabiliste d’une infection est associé à l’augmentation des résistances acquises chez les bactéries. Cependant, cette orientation nécessite de disposer de tests compagnons, c’est-à-dire des tests diagnostiques sensibles et spécifiques pouvant précocement identifier les bactéries et les marqueurs de résistance à partir des liquides biologiques. A côté des méthodes moléculaires largement développées mais ayant des limites de multiplexage, les techniques protéomiques ont récemment été intégrées dans le diagnostic en infectiologie. Ce travail de thèse a consisté à développer des méthodes de spéctrométrie de masse et à les appliquer à la détection de pathogènes et de marqueurs de résistance. Ce travail s’est focalisé sur trois applications : 1) l’identification, la caractérisation et la quantification précoce de micro-organismes dans des échantillons primaires (aspirats endotrachéaux (AET)) de patients atteints de pneumopathies acquises sous ventilation mécanique (PAVM), 2) la détection d’éléments génétiques de la résistance aux antibiotiques : les intégrons, 3) la détection de phénotypes de résistance aux antibiotiques chez Staphylococcus aureus. / Personalized medicine for infectious diseases or cancer becomes more and more important in modern therapy. Furthermore, probabilistic treatment has been associated with the development of resistant bacteria causing infectious diseases. As a result, probabilistic treatments are replaced by adapted treatment for pathologies and patients. However, this new approach needs available companion diagnosis tests that are sensitive but also specific tests able to provide rapid pathogen and resistance markers identification in biological fluids. Beside molecular methods, widely developed but with multiplex limits, proteomic technics have recently joined the infectious diagnosis. This work consisted in developing mass spectrometry technics for bacteria and resistance marker identifications. This work focused on 3 applications: 1) identification and quantitation of microorganisms in crude samples (endotracheal aspirates (ETA)) from patient suffering of ventilator associated pneumonia (VAP), 2) detection of genetic elements involved in antibiotic resistance : the integrons, 3) detection of antibiotic resistance phenotypes in S. aureus.

Abstract Professional practice in critical care settings is characterized by the application of relevant theories, research and evidence-based guidelines to clinical practice. However, critical care nurses’ knowledge and skills in adhering to evidence-based protocols and guidelines for avoiding ventilator-associated pneumonia are inadequate. The aim of the study was to evaluate critical care nurses’ knowledge and skills in adhering to best-practice endotracheal suctioning recommendations and ventilator bundles, to develop and validate instruments to evaluate the care of mechanically ventilated patients, and to evaluate the effectiveness of continuing education on critical care nurses’ knowledge and skills, with a focus on ventilator bundles. In the first study, a descriptive and cross-sectional correlation study was conducted to evaluate critical care nurses’ (n&#160;=&#160;40) endotracheal suctioning practices in relation to current best-practice recommendations. In the second study, a descriptive design with a literature review was conducted to assess the effectiveness of educational programmes in preventing ventilator-associated pneumonia. In the third study, an instruments validation study was conducted to develop and test the psychometric properties of the Ventilator Bundle Questionnaire (VBQ) and Observation Schedule (VBOS). In the fourth study, the effectiveness of human patient simulation education was evaluated among thirty (n&#160;=&#160;30) critical care nurses who were randomly allocated to intervention and control groups (n&#160;=&#160;15 each). Critical care nurses’ knowledge and skills in adhering to best-practice endotracheal suctioning recommendations and ventilator bundles continues to be inadequate. However, educational programmes were linked to significant improvements in learning and clinical outcomes. The VBQ and VBOS were developed and shown to have acceptable psychometric properties (CVI 0.99–1.0, ICC 0.93–1.0). After human patient simulation education, the mean skill scores of the intervention group increased significantly (pt*g&#160;=&#160;0.02). Educational programmes may have a significant impact on clinical outcomes and thus, patients’ safety and quality of care, through improvements in nurses’ knowledge and skills in adhering to evidence-based guidelines in critical care settings. The VBQ and VBOS can provide an objective method measuring whether evidence-based guidelines are being used in clinical practice. In addition, there was a significant transfer of learned skills to clinical practice following human patient simulation education. / Tiivistelmä Teho-osastoilla ammatillinen erityisosaaminen edellyttää tutkitun tiedon, teorioiden sekä näyttöön perustuvien hoitosuositusten soveltamista kliiniseen käytäntöön. Kuitenkin tehohoitajien tiedot ja taidot noudattaa näyttöön perustuvia hoitokäytäntöjä ja suosituksia hengityslaitehoitoon liittyvän keuhkokuumeen ehkäisyksi ovat olleet puutteellisia. Tutkimuksen tarkoituksena oli arvioida tehohoitajien tietoa ja taitoa noudattaa hyväksi havaittuja hengitysteiden imukäytäntöjä sekä hengityslaitehoitoon liittyviä hoitosarjakäytäntöjä, kehittää ja validoida mittareita hengityslaitehoitoa saavien potilaiden hoidon laadun arvioimiseksi sekä arvioida täydennyskoulutuksen vaikuttavuutta tehohoitajien tietoihin ja taitoihin noudattaa hengityslaitehoitoon liittyviä hoitosarjakäytäntöjä. Ensimmäisessä osatyössä arvioitiin kuvailevan ja korrelatiivisen tutkimusasetelman avulla tehohoitajien (n = 40) alahengitysteiden imukäytäntöjä suhteessa hyväksi havaittuihin toimintakäytäntöihin. Toisessa osatyössä arvioitiin kuvailevan kirjallisuuskatsauksen avulla koulutusinterventioiden vaikuttavuutta hengityslaitehoitoon liittyvän keuhkokuumeen ehkäisyssä. Kolmannessa osatyössä kehitettiin ja testattiin hengityslaitehoitoon liittyvä hoitosarjakysely (VBQ) sekä havainnointimittari (VBOS). Neljännessä osatyössä arvioitiin simulaatiokoulutuksen vaikuttavuutta satunnaistetussa koeasetelmassa interventio- (n&#160;=&#160;15) ja kontrolliryhmän (n&#160;=&#160;15) välillä. Tehohoitajien tiedot ja taidot noudattaa hyväksi havaittuja hengitysteiden imukäytäntöjä sekä hengityslaitehoitoon liittyviä hoitosarjakäytäntöjä olivat edelleen puutteellisia. Kuitenkin koulutusinterventioiden vaikuttavuus kliinisiin hoitotuloksiin sekä oppimistuloksiin oli merkittävä. VBQ- ja VBOS-mittareiden psykometriset ominaisuudet osoittautuivat hyväksyttäviksi (CVI 0,99–1,0, ICC 0,93–1,0).Simulaatiokoulutuksen jälkeen interventioryhmän taidot noudattaa hoitosuosituksia lisääntyivät merkittävästi (pt*g&#160;=&#160;0,02). Koulutusinterventioiden kliininen vaikuttavuus potilasturvallisuuden ja hoidon laadun kehittämisessä voi olla merkittävää, kun hoitajien tietoa ja taitoa noudattaa näyttöön perustuvia hoitosuosituksia lisätään kliinisessä tehohoitotyössä. VBQ- ja VBOS-mittarit voivat tarjota objektiivisen tavan arvioida tutkitun tiedon siirtymistä kliiniseen käytäntöön. Simulaatiokoulutuksen jälkeen opittujen taitojen siirtovaikutus kliiniseen käytäntöön oli merkittävä.

Ventilator-associated pneumonia (VAP) is the most common healthcare-associated infection in the intensive care unit, resulting in a high burden of illness, mortality and increased cost. The literature around the cost-effectiveness of probiotics in prevention of health-care associated infections has not been previously well-described, and a definitive health economic evaluation alongside a well-designed randomized control trial assessing probiotic prophylaxis has not been previously performed. This thesis consists of 3 separate manuscripts (with 2 published in peer-reviewed journals and 1 pending). The theme of this thesis was to: (1) describe the literature about the cost-effectiveness of probiotics in hospitalized patients in preventing healthcare-associated infections; (2) design a protocol for an economic evaluation alongside a randomized control trial (RCT) examining probiotic prophylaxis of VAP; and then (3) perform and analyze the health economic evaluation presented in the protocol. The first component of this thesis is a systematic review of probiotic prophylaxis of healthcare-associated infections in hospitalized patients. We performed an extensive search including multiple databases which found 7 studies. Probiotics demonstrated favourable cost-effectiveness in 6 of 7 (86%) economic evaluations, with 3 studies being manufacturer-supported, all suggesting cost-effectiveness. Certainty of cost-effectiveness evidence was very low due to risk of bias, imprecision and inconsistency using the GRADE approach. Hence further RCTs with economic evaluations were stated as a solution. The second component of this thesis is a study protocol for an economic evaluation alongside the Probiotics to Prevent Severe Pneumonia and Endotracheal Colonization Trial (PROSPECT), which assessed the efficacy of probiotic prophylaxis in the prevention of healthcare-associated infections (specifically VAP). The third component of this thesis is the cost-effectiveness analysis performed utilizing the individual patient data from PROSPECT to produce the economic evaluation (E-PROSPECT). As of the date of thesis submission, PROSPECT is still pending publication, and hence E-PROSPECT is also pending analysis and publication. However, I have prepared a draft manuscript (along with figures and tables) that will be produced at the conclusion of E-PROSPECT for thesis committee review. / Thesis / Master of Health Sciences (MSc) / Ventilator-associated pneumonia (VAP) is the most common healthcare-associated infection in the intensive care unit, resulting in a high burden of illness, mortality and increased cost. The literature around the cost-effectiveness of probiotics in prevention of health-care associated infections has not been previously well-described, and a definitive health economic evaluation alongside a well-designed randomized control trial assessing probiotic prophylaxis has not been previously performed. This thesis consists of 3 separate manuscripts (with 2 published in peer-reviewed journals and 1 pending). The theme of this thesis was to: (1) describe the literature about the cost-effectiveness of probiotics in hospitalized patients in preventing healthcare-associated infections; (2) design a protocol for an economic evaluation alongside a randomized control trial (RCT) examining probiotic prophylaxis of VAP; and then (3) perform and analyze the health economic evaluation presented in the protocol. The first component of this thesis is a systematic review of probiotic prophylaxis of healthcare-associated infections in hospitalized patients. We performed an extensive search including multiple databases which found 7 studies. Probiotics demonstrated favourable cost-effectiveness in 6 of 7 (86%) economic evaluations, with 3 studies being manufacturer-supported, all suggesting cost-effectiveness. Certainty of cost-effectiveness evidence was very low due to risk of bias, imprecision and inconsistency using the GRADE approach. Hence further RCTs with economic evaluations were stated as a solution. The second component of this thesis is a study protocol for an economic evaluation alongside the Probiotics to Prevent Severe Pneumonia and Endotracheal Colonization Trial (PROSPECT), which assessed the efficacy of probiotic prophylaxis in the prevention of healthcare-associated infections (specifically VAP). The third component of this thesis is the cost-effectiveness analysis performed utilizing the individual patient data from PROSPECT to produce the economic evaluation (E-PROSPECT). As of the date of thesis submission, PROSPECT is still pending publication, and hence E-PROSPECT is also pending analysis and publication. However, I have prepared a draft manuscript (along with figures and tables) that will be produced at the conclusion of E-PROSPECT for thesis committee review.

A pneumonia associada à ventilação mecânica (PAVM) é uma infecção freqüente nas Unidades de Terapia Intensiva (UTI), acarretando aumento no período de hospitalização, nos índices de morbimortalidade e com repercussão significativa nos custos. A implementação de medidas específicas para a prevenção da PAVM é baseada em diretrizes para a prática clínica, elaboradas por órgãos governamentais e associações de especialistas. Nesse sentido, é importante destacar a necessidade de atualização permanente dos profissionais da saúde. Frente ao exposto, objetivouse avaliar e descrever as evidências científicas disponíveis sobre as práticas de prevenção da PAVM, em pacientes adultos, hospitalizados em UTI. A prática baseada em evidências representou o referencial teórico-metodológico. E, para a obtenção das evidências de Níveis I e II, publicadas posteriormente à diretriz do CDC, realizou-se a revisão integrativa da literatura nas bases de dados MEDLINE, LILACS, CINAHL e Biblioteca Cochrane. Totalizou-se 23 publicações, agrupadas nas categorias temáticas: 5 (22%) higienização bucal, 7 (30%) aspiração de secreções, 5 (22%) umidificação das vias aéreas, 3 (13%) posicionamento do paciente e 3 (13%) diretrizes para a prática clínica. O uso da clorexidina na higienização bucal de pacientes sob ventilação mecânica diminuiu a colonização da orofaringe, o que pode reduzir a incidência de PAVM. Em adição, a aspiração da secreção subglótica e a terapia cinética mostraram-se medidas eficazes na prevenção da PAVM. Por outro lado, o uso do sistema fechado para a aspiração endotraqueal, a umidificação das vias aéreas com o dispositivo HME (heat and moisture exchanger), o controle da pressão do balonete do tubo endotraqueal, bem como, o posicionamento semirecumbente do paciente não apresentaram impacto na prevenção da PAVM e configuram como questões controversas. Assim, outras pesquisas são necessárias, especialmente, para elucidar questionamentos e implementar novas tecnologias acerca das medidas de prevenção da PAVM, o que sem dúvida repercutirá na qualidade da assistência de pacientes submetidos à ventilação mecânica. / Ventilator-associated pneumonia (VAP) is a common infection in the Intensive Care Unit (ICU), which leads to a longer period of hospitalization, higher rates of morbidmortality and a significant repercussion on the costs. The implementation of specific measures to prevent VAP is based on clinical practice guidelines elaborated by governmental organizations and expert committees. Thus, it is important to stand out the permanent actualization of health care professionals. Therefore, this study aimed to evaluate and describe the available scientifical evidences on VAP prevention practices in adult patients hospitalized in the ICU. The evidence based practice represented the theoretical-methodological reference. And, to obtain the evidences Levels I and II, published after the CDC guideline, an integrative review of the literature of MEDLINE, LILACS, CINAHL and Cochrane Library databases was realized. A total of 23 publishing grouped in categories: 5 (22%) oral hygiene, 7 (30%) aspiration of secretions, 5 (22%) airways moisturizing, 3 (13%) patient positioning and 3 (13%) clinical practice guidelines. The use of chlorhexidine in the oral hygiene of the mechanical ventilated patients decreased the oropharyngeal colonization, which may decrease VAP incidence. Besides, the subglottic secretions drainage and the kinetic therapy proved to be efficient on VAP prevention. However, the use of the closed system to the endotracheal aspiration, moisturizing the airways with HME (heat and moisture exchanger), the control of the pressure of the endotracheal tube cuff as well as the semirecumbent positioning of the patient, did not present any impact on the VAP prevention and are controversial matters. Thus, further researches are required mainly to clarify some questions and implement new technologies on measures to prevent VAP, which will certainly reflect on the quality of the assistance given to patients on mechanical ventilation.

Made available in DSpace on 2016-01-26T12:51:44Z (GMT). No. of bitstreams: 1 mariacristinadeavilameinberg_dissert.pdf: 2412515 bytes, checksum: a1233bb3e2f1e020476d33588e32153f (MD5) Previous issue date: 2013-06-07 / Introduction: Nosocomial pneumonias determine a significant increase length of Stay, in hospital costs and mortality. Oral hygiene with chlorhexidine has been considered a tool in the prevention of nosocomial pneumonia. However, current data suggest that such benefits are more significant in cardiac surgeries patients. Objective: To evaluate the effects of oral chlorhexidine hygiene with tooth brushing on the rate of ventilator-associated pneumonia in a mixed population of critically ill patients under prolonged mechanical ventilation. Methods: Prospective, randomized, and placebo-controlled pilot study. Patients who were receiving mechanical ventilation, had been admitted in the intensive Care Unit in the previous 24 hours, and were anticipated to require mechanical ventilation for more than 72 hours were included in the study. The patients were randomly divided in to one of the following groups: chlorhexidine hygiene with toothbrushing or a placebo group (gel with the same color and consistency and tooth brushing). Results: The planned interim analysis was conducted using 52 patients, and the study was terminated prematurely. In total, 28 patients were included in the chlorhexidine/tooth brushing group, and 24 patients were included in the placebo/tooth brushing group. Ventilator-associated pneumonia occurred in 45.8% of the placebo group and in 64.3% of the chlorhexidine hygiene with toothbrushing group (RR=1.4; 95% CI=0.83-2.34; p=0.29). Conclusion: The use of gel with chlorhexidine 2% and toothbrushing for oral hygiene did not have effect on the rate of VAP in this heterogeneous population of critically ill patients under prolonged mechanical ventilation. / Introdução: As pneumonias nosocomiais determinam significativo aumento em tempo de internação, custos hospitalares e mortalidade. A higiene bucal com clorexidina é considerada de grande importância na prevenção de pneumonia nosocomial. Contudo os dados atuais mostram que tais benefícios são mais significativos em pacientes submetidos à cirurgia cardíaca. Objetivo: Avaliar os efeitos da higiene bucal com clorexidina 2% e escovação mecânica sobre a taxa de pneumonia associada a ventilador (PAV) em uma população mista de pacientes sob ventilação mecânica prolongada. Método: Estudo piloto prospectivo, aleatório e placebo-controlado. Foram incluídos pacientes sob ventilação mecânica, com menos de 24 horas de internação e cuja perspectiva de duração da ventilação mecânica era a de um período > 72 horas. Os pacientes foram randomizados para o grupo clorexidina (gel com clorexidina a 2%) e escovação mecânica ou grupo placebo (gel da mesma cor e consistência e escovação mecânica) na higiene bucal. Resultados: A análise interina planejada foi realizada quando 52 pacientes foram incluídos e o estudo foi interrompido precocemente. Um total de 28 pacientes foi incluído no grupo clorexidina e 24 no grupo placebo. As taxas de PAV foram de 45,8% no grupo placebo/escovação mecânica e de 64,3% no grupo clorexidina/escovação mecânica (RR=1,4; IC95%= 0,83-2,34; p=0,29). Conclusão: Os resultados deste estudo não evidenciaram beneficio do uso de clorexidina a 2% e escovação mecânica na higiene bucal nas taxas de PAV nesta população heterogênea de pacientes críticos sob ventilação mecânica prolongada.

Made available in DSpace on 2016-01-26T12:51:45Z (GMT). No. of bitstreams: 1 marcelospegiorinmoreno_tese.pdf: 1082572 bytes, checksum: d3881298debccf50a2a1f76433f765a4 (MD5) Previous issue date: 2011-02-24 / Resumo:Introdução: Pneumonias adquiridas no hospital (PAH) e pneumonias associadas à ventilação mecânica (PAVM) são complicações frequentes e importantes causas de aumento da morbidade, mortalidade e dos custos. Objetivo: O objetivo do estudo foi avaliar o valor prognóstico da dosagem seriada da PCR nos pacientes com pneumonia nosocomial (PN). Casuística e Método: Estudo prospectivo e observacional, conduzido em uma UTI de 24 leitos em um hospital terciário. Pacientes com PN, incluindo pacientes não ventilados e pacientes com pneumonia associada à ventilação mecânica foram incluídos. Dosagens diárias da PCR e cálculo da Taxa da PCR (TPCR) eram realizadas do dia da prescrição do antibiótico (D0) até o décimo dia. Pacientes foram classificados de acordo com a TPCR em 2 grupos: boa resposta (TPCR < 0,67 no D10) e má resposta (não-resposta ou resposta bifásica - TPCR 0,67 no D10). Resultados: Cento e cinquenta e cinco pacientes com PN foram avaliados e 64 foram incluídos. O grupo má resposta (n= 34) teve uma taxa de mortalidade geral de 53% em comparação a 20% no grupo boa resposta (n= 30) (RR = 2,65; IC 95%, 1,21-5,79, p = 0,01). Diferenças significantes entre os xii dois grupos foram encontradas a partir do Dia 4 (p= 0,01). A adequacidade da antibioticoterapia foi muito menor no grupo má resposta em comparação ao grupo boa resposta, 14,3% vs 66,7% (p = 0,008), respectivamente. Conclusão: Dosagens seriadas de PCR e a análise da cinética da TPCR são úteis no seguimento de pacientes com pneumonia nosocomial.

Nonobstant d’immenses progrès accomplis depuis des décennies dans la prise en charge des patients soumis à la ventilation mécanique, les pneumonies acquises sous ventilation mécanique continuent de compliquer le séjour en réanimation de près de 28% des patients recevant une assistance respiratoire invasive prolongée. Parmi les malades des unités de soins intensifs, le risque de développer une pneumonie est de 3 à 10 fois supérieur chez les intubés sous ventilation. Elle reste cependant bien souvent le seul moyen de venir en aide aux patients souffrant de graves détresses respiratoires. Il a maintenant été clairement démontré que la ventilation mécanique, en particulier lorsqu’elle est mise en place selon des stratégies dites agressives, active les cellules pulmonaires conduisant alors à une réponse pro-inflammatoire même en l’absence de pathogène. Ce phénomène est connu sous le terme de biotrauma, et serait responsable en partie des lésions induites sur le poumon par la ventilation mécanique. En quelques sortes, la ventilation mécanique prépare les cellules épithéliales pulmonaires à répondre massivement à une seconde agression pro-inflammatoire par la libération de grandes quantités de cytokines (comme l’IL-8 notamment), accentuant alors les lésions du tissu pulmonaire essentiellement par le recrutement de polynucléaires neutrophiles attirés par la sécrétion massive d’IL-8. L’immunité innée joue donc un rôle très important dans le développement du VILI. L’implication des Toll Like Récepteurs a été suggérée par plusieurs études expérimentales. Par ailleurs, la ventilation en décubitus ventral a été décrite pour avoir des effets bénéfiques sur les patients ventilés souffrant de graves lésions pulmonaires particulièrement chez ceux souffrant du syndrome de détresse respiratoire aiguë. Notre équipe s’est particulièrement intéressée au TLR2, qui reconnait les bactéries à Gram-positif, car elle a montré dans des études précédentes in vitro que l’étirement cyclique de cellules pulmonaires humaines augmentait principalement l’expression de TLR2 ainsi que la réactivité de cellules pulmonaires à des composants de la paroi de bactéries à Gram positif. Ces données ont par la suite été confirmées dans un modèle in vivo de lapins ventilés dont la réponse immune innée était stimulée par du Pam3CSK4.Dans un premier projet, nous avons évalué l’impact d’une ventilation mécanique en décubitus ventral chez des lapins avec pneumonie unilatérale à Enterobacter aerogenes soumis à la ventilation mécanique. Nos résultats montrent que le décubitus ventral peut être protecteur si l’hôte est soumis à la ventilation mécanique dans le contexte d’une pneumonie bactérienne unilatérale.Pour vérifier la pertinence de nos hypothèses sur le TLR2 dans notre modèle animal de pneumonie acquise sous ventilation mécanique, nous avons mené des expériences avec des bactéries vivantes reconnues par le TLR2 (une souche de Staphylococcus aureus résistante à la methicilline SARM). Notre étude met en évidence qu’une ventilation mécanique modérément agressive impacte sur la clairance bactérienne pulmonaire en la diminuant, aggrave les lésions sur le tissu pulmonaire et favorise une réponse inflammatoire systémique. La surexpression du TLR2 tant au niveau pulmonaire que systémique pourrait expliquer ces résultats.Le troisième projet s’est attaché à évaluer l’impact d’une thérapie aux statines dans le contexte d’une pneumonie acquise sous ventilation mécanique à SARM, conjointement traitée par le linezolide, dans notre modèle animal de lapins ventilés. Nos résultats suggèrent qu’une pré-­‐exposition aux statines pourrait avoir un effet anti-inflammatoire au niveau pulmonaire et systémique dans ce modèle, qui pourrait passer par une régulation négative de l’expression de TLR2, contre-balançant les effets de l’étirement cyclique. / Despite major advances since decades in the management of ventilated patients, ventilator-associated pneumonia (VAP) continues to complicate the course of approximately 28% of the patients receiving mechanical ventilation (MV). Among patients hospitalized in intensive care units, the risk of pneumonia is 3- to 10- fold increased in MV patients. However, MV is often the only way to care for critically ill patients with respiratory failure. It has now been clearly demonstrated that MV, in particular adverse ventilatory strategies could activate lung cells, thus leading to a proinflammatory response, even in the absence of pathogen. This is the biotrauma paradigm, which accounts, at least in part, for the ventilator induced lung injury (VILI). In one way, MV primes airway cells to respond massively to a second proinflammatory insult, through the subsequent release of large amounts of cytokines (as interleukin (IL)‐ 8), thus leading to additional lung injury, particularly through the recruitment of neutrophils attracted by the massive release of IL-8. Accordingly, innate immunity plays an important role in the developement of VILI. The involvement of Toll-like receptors has been suggested by several experimental studies. Ventilation in the prone position (PP) has been described to have beneficial effects on patients under MV, especially in those with lobar involvement. Our team focused particularly on the TLR2, which interacts with Gram-positive bacteria, and we have previously demonstrated in vitro that cyclic stretch of human pulmonary cells resulted in TLR2 overexpression and enhanced TLR2 reactivity to Gram-positive cell wall components. We confirmed these datas in an in vivo model of ventilated rabbits which immune response had been stimulated with Pam3CSK4. In a first project, we assessed the impact of the PP on unilateral pneumonia to Enterobacter aerogenes in rabbits subjected to MV. Our results shows that the prone position could be protective if the host is subjected to MV and unilateral bacterial pneumonia. To ensure the relevance of our hypothesis on TLR2 in our animal model of VAP, we conducted experiments using live bacteria specifically recognized by TLR2 (Methicilin resist. aureus). We demonstrate that mild-­‐stretch MV impaired lung bacterial clearance, hastened tissue injury and promoted a systemic inflammatory response. Both pulmonary and peripheral blood TLR2 overexpression could account for such an impact. The third project assessed the impact of a statins therapy in the context of MRSA VAP, treated with linezolid, in our model of ventilated rabbits. Our results suggest that statin exposure prior to pneumonia provides an anti-­‐inflammatory effect within the lung and the systemic compartment of rabbits with MRSA VAP. Although LNZ enhances pulmonary bacterial clearance, dampening the host systemic inflammatory response with statin could impede defense against MRSA in this compartment. It could be subsequent to enhanced antibacterial defences and improvements in lung mechanics, thereby blunting overwhelming inflammation. In the last project, in collaboration with the University of Geneva, we assessed whether mitochondrial alarmins are released during VILI and can generate lung inflammation. Our results confirmed the hypothesis made and showed indeed that alarmins are released during during cyclic stretch of human epithelial cells, as well as in BAL fluids from rabbits ventilated with an injurious ventilatory regimen. These alarmins stimulate lung cells to produce bioactive IL-­‐1, and are likely to represent the proximal endogenous mediators of VILI and ARDS, released by injured pulmonary cells.