No description available.

Answers to the most complex biological questions are rarely determined solely from the experimental evidence. It requires subsequent analysis of many data sources that are often heterogeneous. Most biological data repositories focus on providing only one particular type of data, such as sequences, molecular interactions, protein structure, or gene expression. In many cases, it is required for researchers to visit several different databases to answer one scientific question. It is essential to develop strategies to integrate disparate biological data sources that are efficient and seamless to facilitate the discovery of novel associations and validate existing hypotheses. This thesis presents the design and development of different integration strategies of biological and clinical systems. The BioSPIDA system is a data warehousing solution that integrates many NCBI databases and other biological sources on protein sequences, protein domains, and biological pathways. It utilizes a universal parser facilitating integration without developing separate source code for each data site. This enables users to execute fine-grained queries that can filter genes by their protein interactions, gene expressions, functional annotation, and protein domain representation. Relational databases can powerfully return and generate quickly filtered results to research questions, but they are not the most suitable solution in all cases. Clinical patients and genes are typically annotated by concepts in hierarchical ontologies and performance of relational databases are weakened considerably when traversing and representing graph structures. This thesis illustrates when relational databases are most suitable as well as comparing the performance benchmarks of semantic web technologies and graph databases when comparing ontological concepts. Several approaches of analyzing integrated data will be discussed to demonstrate the advantages over dependencies on remote data centers. Intensive Care Patients are prioritized by their length of stay and their severity class is estimated by their diagnosis to help minimize wait time and preferentially treat patients by their condition. In a separate study, semantic clustering of patients is conducted by integrating a clinical database and a medical ontology to help identify multi-morbidity patterns. In the biological area, gene pathways, protein interaction networks, and functional annotation are integrated to help predict and prioritize candidate disease genes. This thesis will present the results that were able to be generated from each project through utilizing a local repository of genes, functional annotations, protein interactions, clinical patients, and medical ontologies.

Bakgrund: Organdonation och transplantation har gjort det möjligt att både rädda liv och även förbättra livskvaliteten för ett stort antal personer. Problematiken som idag råder är att det finns ett stort glapp mellan behovet av organ och tillgången till organgivare. Intensivvårdssjuksköterskan har ett ansvar att se till att potentiella donatorer fångas upp och uppmärksammas på intensiven. Att vårda en potentiell donator samt bemöta anhöriga är en utmaning och kräver både kunskap och erfarenhet för intensivvårdssjuksköterskan. Syfte: Att belysa hur intensivvårdssjuksköterskan upplever sin kompetens gällande vårdandet av en donator och bemötandet av anhöriga under donationsprocessen på en intensivvårdsavdelning. Metod: Studien genomfördes med en kvalitativ metod genom semistrukturerade intervjuer med tio intensivvårdssjuksköterskor. Data analyserades med hjälp av en kvalitativ latent innehållsanalys. Resultat: Det ansågs viktigt att intensivvårdssjuksköterskan bemötte både donatorn och anhöriga på ett professionellt sätt, dels bemöta donatorn med värdighet och respekt samt att vara lyhörd och visa empati gentemot anhöriga. Intensivvårdssjuksköterskorna upplevde svårigheter när donatorns önskan inte var känd, att vården var tidskrävande, få anhöriga att förstå att deras närmaste avlidit, samt att några intensivvårdssjuksköterskor tog upp punkter som kunde förbättras för att öka antalet donationer. Samtliga intensivvårdssjuksköterskor upplevde att de hade tillräckligt med kunskap för att vårda en donator men att uppdatering och utbildning behövs alltid. Debriefing eller samtal var något annat som togs upp vilket de flesta upplevde som viktigt för att inte ta med arbetet hem. Slutsats: I studien identifierades brister inom donationsvården och att mer utbildning och riktlinjer kan medföra att dessa skulle kunna minskas. Studien har också bidragit till ökad förståelse och kunskap kring donationsprocessen inom intensivvården. Detta är något som författarna tros kunna ha nytta av i sitt framtida yrke som intensivvårdssjuksköterskor, men också att resultatet i studien kan vara till nytta för andra intensivvårdsavdelningar i Sverige. / Background: Organ donation and transplantation has made it possible to both save lives and also improve the quality of life for a large number of people. In recent years there has been a large gap between the need for organs and the availability of organ donors. Intensive care nurses have a responsibility to ensure that potential donors are identified and highlighted in the intensive care unit. Caring for a potential donor and meet relatives is a challenge and requires both knowledge and experience of intensive care nurses. Aim: To show how intensive care nurses perceive their expertise regarding the care of a donor and the hospitality of relatives during the donation process in the intensive care unit. Method: The study was conducted with a qualitative approach through semi-structured interviews with ten intensive care nurses. Data were analyzed using a qualitative latent content analysis. Results: It was important that the intensive care nurses responded to both the donor and family members in a professional manner, but also responding to donors with dignity and respect and to be sensitive and show empathy towards their families. They experienced difficulties when the donor's wishes are not known, the treatment was time-consuming, getting families to understand their loved one has died. Some intensive care nurses brought up a point of view which could increase the numbers of donations. All intensive care nurses felt that they had enough knowledge to take care of a donor but being up to date and training is always needed. Debriefing or conversation was also something else that was raised which most felt was important not to bring work home. Conclusion: The study identified deficiencies in the donation care but with more training and guidelines this could be improved. The study has also contributed to greater understanding and knowledge surrounding the donation process in intensive care unit. This is which the authors believed to be useful in their future profession as intensive care nurses, but also the results of the study can be useful to other intensive care unit in Sweden.

Bakgrund: Vården på en intensivvårdsavdelning är tänkt att vara kurativ, det vill säga botande. Trots det, så inträffar ca 20 procent av alla dödsfall på intensivvårdsavdelningar i Sverige och omställningen till palliativ vård sker ofta snabbt. Döden inträffar ofta kort därpå, och denna transition kan upplevas på olika sätt beroende på hur personalen informerar och kommunicerar med de närstående om vad som sker. Syfte: Syftet med litteraturöversikten är att undersöka hur närstående upplever den palliativa vården på intensivvårdsavdelningar. Metod: En litteraturöversikt gjordes och tio vetenskapliga artiklar valdes ut, varav sex var kvantitativa och fyra kvalitativa. Resultat: Fem huvudteman med underteman identifierades. Information om diagnos och prognos var viktigt för de närstående och var avgörande för hur passförberedda de var inför den sjukes död. Kommunikation, det vill säga hur informationen förmedlades, var också en viktig parameter för upplevelsen. Närstående eftersökte också att få stöd från personalen i form av empati eller existentiellt stöd. Med interaktion menas såväl den praktiska delaktigheten i omvårdnaden av den sjuke, som den fysiska och existentiella närvaron som visade sig vara mycket betydelsefull. Slutligen behandlas hur den fysiska miljön påverkar de närståendes upplevelse, med mycket teknisk utrustning och lite utrymme för avskildhet och lugn. Diskussion: Resultaten diskuterades utifrån Andersheds teoretiska ramverk och hennes huvudbegrepp “att veta”, “att vara” och “att göra”, samt utifrån hennes tankar kring det humanistiska förhållningssättet hos personalen och vad “extraresurser” kan innebära. / Background: The care in the intensive care units intends to save lives. In spite of that aim, 20 percent of the deaths in Sweden occur at the intensive care units and the transition between curative and palliative care can be very quick. Death often occurs shortly thereafter, and this can be experienced in different ways depending on how the staff informs and communicates with the next-of-kin. Aim: The purpose of the literature review is to examine how the next-of-kin experience palliative care in intensive care units. Method: A literature review was made and ten scientific articles were selected. Six were quantitative and four qualitative. Results: Five main themes with subthemes were identified. Information on diagnosis and prognosis was crucial. Communication was also important. The next of kin appreciated emotional and existential support. Interaction, as in practical involvement and as in the physical and existential presence, turned out to be very important. Finally, results show that the environment with a lot of technical equipment and little space for privacy and tranquility affects the experience for next-of-kin. Discussion: The results are discussed in relation to Andershed’s theoretical framework “Involvement in the light - Involvement in the dark.” and her main concepts“ to know”, “to be” and “to do”. In addition, Andershed’s concepts of humanistic approach and “extra resources” are being discussed.

Rationnelle. La ventilation mécanique invasive (VI) s’accompagne lorsqu’elle se prolonge, d’une augmentation de la morbimortalité. Jusqu’à 64% des enfants hospitalisés aux soins intensifs sont ventilés et peu de données épidémiologiques existent afin d’estimer précocement la durée du support ventilatoire. Objectifs. Déterminer l’incidence et les facteurs de risque précoces de ventilation mécanique invasive prolongée aux soins intensifs pédiatriques. Méthode. Nous avons conduit une étude descriptive rétroélective sur un an. Tous les épisodes de VI aux soins intensifs du Centre hospitalier universitaire Sainte Justine de Montréal ont été inclus. Les facteurs de risque de VI prolongée (≥ 96 heures) ont été déterminés par régression logistique. Résultats. Parmi les 360 épisodes de VI, 36% ont duré ≥ 96 heures. Les facteurs de risques de ventilation prolongée en analyse multivariée sont : âge <12 mois, score de PRISM ≥ 15 à l’admission, pression moyenne dans les voies aériennes ≥13 cm H2O au jour 1 de ventilation, utilisation de la sédation intraveineuse continue au jour 1 de ventilation et ventilation non invasive avant intubation. Conclusion. La VI prolongée survient chez environ un tiers des patients ventilés. Les patients de moins de 12 mois semblent être plus à risque que les enfants plus âgés et devraient bénéficier de stratégies différentes pour diminuer leur durée de ventilation mécanique. La sévérité de la maladie, l’agressivité du support ventilatoire, l’utilisation d’une sédation continue au premier jour de ventilation sont également des facteurs à considérer dans les études visant à diminuer la durée de support ventilatoire. / Rationale. Invasive mechanical ventilation is associated, if prolonged, with higher morbidity and mortality. Up to 64% of children hospitalized in pediatric intensive care units (PICU) require invasive ventilation but little epidemiological data are available on children requiring prolonged acute invasive mechanical ventilation. Objectives. To determine the incidence rate and early risk factors for prolonged acute invasive mechanical ventilation in children. Methods. We conducted a retrolective longitudinal cohort study over a one-year period. All consecutive episodes of invasive mechanical ventilation in the PICU of Sainte-Justine Hospital were included. Risk factors for prolonged (≥ 96 hours) versus short (< 96 hours) ventilation were determined by logistic regression. Results. Among the 360 episodes of invasive ventilation, 36% had a length ≥ 96 hours. Following multivariate analysis, significant risk factors for prolonged acute invasive mechanical ventilation were: age <12 months, PRISM score ≥ 15 at admission, mean airway pressure ≥13 cm H2O on day one, use of continuous intravenous sedation on day one and use of non-invasive ventilation prior to intubation. Conclusion. Prolonged acute mechanical ventilation occurs in approximately one third of ventilated children. Younger children (aged <12 months) may be different from older patients and may require different strategies to decrease the duration of mechanical ventilation. Severity of illness, the aggressiveness of ventilatory support required and the use of continuous intravenous sedation on the first day of ventilation are also risk factors to consider in trials aimed at reducing mechanical ventilation duration.

Le but de cette étude est d’adopter une intervention systémique visant à faciliter la communication entre les infirmières1 et les familles2 dans un contexte de soins intensifs (SI) en traumatologie. Cette recherche prend en compte la réalité et les besoins du milieu à l’étude grâce au devis participatif retenu. Le cadre de référence est composé de l’approche systémique familiale de Wright et Leahey (2005) ainsi que de la théorie du caring de Jean Watson (1979). La collecte des données a été inspirée du parcours de recherche en spirale de Lehoux, Levy et Rodrigue (1995). Cinq entrevues ont été menées auprès de membres de familles ainsi que trois groupes de discussion (focus groups) auprès d’infirmières. L’analyse des données a été effectuée selon le Modèle de Huberman et Miles (2003) qui préconise une démarche analytique continue et interactive. Les résultats obtenus révèlent un changement de pratique selon dix interventions systémiques infirmières spécifiques permettant de faciliter la communication avec les familles aux soins intensifs en traumatologie et soulignent l’importance d’agir sur le plan du contexte, de la relation et du contenu liés au processus de communication. De plus, ils permettent de démontrer un réel changement de pratique dans le milieu retenu pour l’étude. / Abstract The purpose on this study was to adopt systemic nursing interventions to enhance communication with families in a context of traumatology intensive cares. This participative study has the benefit of reflecting the real needs expressed by the persons concerned. The terms of reference for this study comprised the conceptual model as well as Wright and Leahey’s (2005) family systems nursing approach and the caring theory of Jean Watson (1979). The data collection was inspired by Lehoux, Levy and Rodrigue (1995) over five family interviews and three nurses’ meetings. The data collection was performed over a period of seven months. All of the interviews and meetings were recorded on audio tape and the verbatim’s were fully transcripted. The data analysis was done according to Huberman and Miles’ (2003) model, where analytical activities and data collection establish a continuous and interactive approach. The results obtained revealed ten specific systemic nursing interventions to enhance communication with families in this context and revealed the importance of the three levels of communication: cognitive, affective and behavioral. Moreover, the results are used to identify well-established interventions and recommend those who have to be improved.

Rationnel : La pandémie de grippe A(H1N1)pdm09 a induit un grand nombre d’hospitalisation d’enfants en soins intensifs pédiatriques (SIP). L’objectif de cette étude a été de comparer l’incidence et la mortalité des enfants admis en SIP durant l’automne 2009 entre le Canada et la France, deux pays qui diffèrent essentiellement par l’immunisation de la population contre ce virus (première vague en été et taux de couverture vaccinale supérieur à 50% au Canada ; pas de vague estivale et couverture vaccinale de 18% en France). Méthodes : Nous avons comparé deux cohortes nationales qui ont inclue tous les patients avec une infection A(H1N1)pdm09 documentée, admis en SIP au Canada et en France entre le 1er Octobre 2009 et le 31 janvier 2010. Résultats : Au Canada, 160 enfants (incidence=2,63/100000 enfants) en 6 semaines ont été hospitalisés en SIP comparé aux 125 enfants (incidence=1,15/100000 enfants) en 11 semaines en France (p<0,001). Le taux de vaccination avant l’admission était inférieur à 25% parmi les enfants en situation critique dans les deux pays. La gravité à l’admission en SIP et le taux de mortalité ont été similaires au Canada et en France (4,4% en France vs 6,5% au Canada, p=0,45, respectivement). Au Canada, la vaccination contre le virus H1N1pdm09 a été associée avec une diminution du recours à la ventilation invasive (Odd Ratio 0.30, intervalle de confiance à 95% [0,11-0,83], p=0,02). Au Canada comparé à la France, les durées médianes de séjour en SIP et de ventilation invasive ont été plus courtes (2,9 vs 3 jours, p=0,03 et 4 vs 6 jours, p=0,02, respectivement). Conclusion : Les enfants canadiens et français critiquement malades ont été beaucoup moins nombreux à recevoir le vaccin contre le virus influenza A (H1N1)pdm09 en comparaison avec l’ensemble des enfants dans ces deux populations. Au Canada, où la couverture vaccinale a été élevée, le risque d’avoir une détresse respiratoire sévère était moins important parmi les enfants en situation critique ayant été vaccinés avant l’admission. / Background: The pandemic influenza A (H1N1)pdm09 resulted in a large number of admissions to pediatric intensive care units (PICUs). The objective of the study was to compare the incidence and mortality rate of children admitted to PICU in autumn 2009 between France and Canada, two countries that essentially differed by their population immunization to this virus (first pandemic wave in summer and vaccine coverage >50% in Canada; no wave in summer and vaccine coverage of 18% in France). Methods: We compared two national cohorts that included all patients with documented H1N1pdm09 infection, admitted to a PICU in Canada and in France between October 1st 2009 and January 31st 2010. Results: In Canada, 160 children (incidence=2.63/100,000 children) in 6 weeks were hospitalized in PICU compared to 125 children (incidence=1.15/100,000) in 11 weeks in France (p<0.001). Prior vaccination was under 25% among critically ill children in both countries. Severity of illness at PICU admission and mortality rates were similar in Canada and France (6.5%, vs 4.4 p=0.45, respectively). In Canada, H1N1pdm09 vaccination was associated with a decreased risk of requiring invasive ventilation (Odd Ratio 0.30, 95%Confidence Interval 0.11-0.83, p=0.02). In Canada as compared to France, median PICU length of stay and invasive ventilation durations were shorter (2.9 vs 3 days, p=0.03 and 4 vs 6 days, p=0.02, respectively). Conclusion: Critically ill Canadian and French children were much less likely to have received prior vaccination against influenza A (H1N1) pdm09 in comparison to all children in the populations. In Canada, where vaccination rate was higher, the risk of severe respiratory failure was less among those critically ill children receiving prior vaccination.

En réanimation, il n’existe pas de gold standard pour estimer le débit de filtration glomérulaire (DFG). Nous avons mesuré la clairance du iohexol chez 20 patients en insuffisance circulatoire aiguë (injection de 5 mL de iohexol et cinétique riche sur 24h). Les clairances urinaire et plasmatique étaient équivalentes ; la clairance plasmatique n’était pas influencée par le remplissage. Nous avons étudié la distribution de la clairance du iohexol chez 85 patients en insuffisance circulatoire aiguë. Quarante-et-un patients (48%) avaient un DFG < 30 mL.min-1, 29 (34%) entre 30 et 60mL.min-1, 10 (12%) entre 60 et 90mL.min-1, 4 (5%) entre 90 et 130 mL.min-1 et 1 (1%) > 130 mL.min-1. Nous avons mesuré les biomarqueurs lésionnels [TIMP-2].[IGFBP-7] juste avant, 6h et 24 h après un scanner injecté en réanimation; il n’y a pas eu d’augmentation significative des biomarqueurs, confortant l’hypothèse d’une toxicité négligeable des produits de contraste iodés en réanimation. En conclusion, le iohexol peut être considéré comme un gold standard pour l’estimation du DFG chez des patients en insuffisance circulatoire aiguë en termes de faisabilité, fiabilité et sécurité. / There is no gold standard for glomerular filtration rate (GFR) estimation in intensive care unit. We measured iohexol clearance in 20 patients experiencing acute circulatory failure (5 mL iohexol bolus, urine and blood-sample collections over 24h). Urinary and plasma clearances were equivalent; rapid fluid infusion did not influence plasma clearance. We studied iohexol clearance repartition in 85 patients experiencing acute circulatory failure. Forty-one (48%) had a GFR < 30 mL.min-1, 29 (34%) between 30 and 60mL.min-1, 10 (12%) between 60 and 90mL.min-1, 4 (5%) between 90 and 130 mL.min-1 and 1 (1%) > 130 mL.min-1. We measured lesion biomarkers [TIMP-2].[IGFBP-7], before, 6h and 24h after an injected computed tomography scan; there was no significant raise in the biomarkers. This result supports the hypothesis that contrast media are armless in intensive care units. To conclude, iohexol can be considered as a gold standard for GFR estimation in acute-circulatory-failure patients regarding feasibility, reliability and safety.

Les objectifs de la thèse sont l'a nalyse approfondie d'un système de supervision automatique de la ventilation artificielle des patients hospitalisés en soins intensifs et l'élaboration de solutions pour améliorer et étendre son fonctionnement. Ce système adapte l'assistance en pression de la ventilation spontanée avec aide inspiratoire (AI) par un rétrocontrôle basé sur la fréquence respiratoire du patient et, comme variables de "sécurité", le volume courant et le CO2 de fin d'expiration (etCO2). Il établit ainsi une classification ventilatoire et règle un niveau de pression d'AI.Sur la base d'études publiées rapportant des limitations, d'analyses d'une base de données rétrospectives acquises sur patient, d'études sur banc-test et d'études observationnelles prospectives réalisées chez les patients, nous avons étudié précisément le fonctionnement du système. Pour la plupart des limitations, une solution a été proposée et évaluée sur banc. A partir d'une étude clinique, nous avons proposé une amélioration de la procédure de traitement du signal etCO2 par le système. En nous basant sur les observations de la base de données, nous avons décrit une procédure automatisée de sevrage de la PEP dont un niveau supérieur à 5 mbar entrave le sevrage par le système. Sur le même principe, nous avons souhaité, en amont de l'AI, tenter d'automatiser un changement de mode depuis la ventilation assistée contrôlée. Au travers d'une étude clinique, nous avons déterminé des critères ventilatoires qui pourraient permettre d'automatiser cette procédure. L'ensemble a permis la définition d'une méthodologie d'évaluation et d'amélioration d'un système automatisé de ventilation artificielle / The objectives of the thesis are the detailed analysis of an automated system of management of artificial ventilation for patients in intensive care unit and the elaboration of solutions to improve and upgrade its functioning. This system automatically adapts the level of pressure support ventilation (PSV) through a feedback based on the respiratory rate of the patient and, as “safety” variables, the tidal volume and the end tidal of expired CO2 (etCO2). The system, called “SmartCare”, establishes a ventilatory classification and sets a level of inspiratory pressure.We precisely studied the system functioning on the basis of published studies reporting limitations, on a database analysis compound with retrospective data acquired on patients, of bench-test studies and observational prospective studies performed in patients. For most of the limitations, a solution was proposed and evaluated on bench. Using the results of a clinical study, we proposed an improvement of etCO2 signal processing. From observations of the database, we described a procedure of automated weaning of positive end expiratory pressure level which hinders pressure support weaning if above 5 mbar. On the same principle, upstream to PSV, we wished to automate a change of ventilation mode from assist control ventilation. Through a clinical study, we determined ventilatory criteria which could allow automation of this procedure. The whole project allowed to define a methodology of evaluation and improvement of an automated mechanical ventilation system

A prevenção de úlcera por pressão representa grande desafio no ambiente hospitalar, especialmente na Unidade de Terapia Intensiva, mediante a diversidade de fatores de risco apresentados pelos pacientes. A pesquisa teve como objetivo avaliar o efeito da construção e utilização de um protocolo embasado em evidências para prevenção de úlcera por pressão em Unidade de Terapia Intensiva, nas ações de enfermagem e na incidência do evento. Foi realizada na Unidade de Terapia Intensiva de hospital universitário, de João Pessoa, Paraíba, após aprovação do Comitê de Ética em Pesquisa. O percurso metodológico foi construído a partir das etapas do processo da adoção de uma inovação conforme Rogers, utilizando-se as abordagens quanti e qualitativa, em três fases: pré-intervenção, intervenção e pós- intervenção. Nas fases pré e pós-intervenção, por meio de dois estudos, investigou- se a realidade caracterizando-se as ações dos profissionais para prevenção durante a higiene corporal e a incidência de úlcera por pressão. A fase de intervenção possibilitou a construção e utilização do protocolo de prevenção de úlcera por pressão com a participação dos profissionais de saúde da Unidade de Terapia Intensiva, utilizando-se como estratégias o grupo focal, a avaliação da concordância pelos profissionais e o desenvolvimento de ações educativas, visando à persuasão para a adoção das recomendações. Os resultados das fases pré e pós-intervenção foram comparados utilizando testes estatísticos para verificar o impacto da utilização do protocolo. No estudo qualitativo, desenvolvido durante a fase de intervenção, a análise e interpretação dos depoimentos dos profissionais foram realizadas pela técnica de análise de conteúdo conforme Bardin. A comparação dos resultados dos estudos das fases pré e pós-intervenção evidenciou que a utilização do protocolo influenciou a prática clínica dos profissionais de enfermagem, observando-se mudanças significativas na adoção de medidas de prevenção como: avaliação do risco para úlcera por pressão utilizando a Escala de Braden durante a internação do paciente (p<0,001), incremento na utilização do hidratante corporal (p<0,001), inspeção da pele nas proeminências ósseas em todas as regiões corporais (p<0,001), aumento da utilização do lençol móvel para elevação do paciente do leito durante a movimentação (p<0,001), utilização de travesseiros para proteger as proeminências ósseas do joelho (p=0,015) e sob as panturrilhas para manter os calcâneos flutuantes (p<0,005). Observou-se ainda aumento na utilização de coberturas para proteção da pele nas áreas de proeminências ósseas (p=0,005). Quanto à incidência de úlcera por pressão observou-se diferença estatisticamente significante (p=0,0069) entre as duas fases da pesquisa com redução do índice de 35,7% (pré-intervenção) para 8,3% (pós-intervenção). Identificou-se que as variáveis associadas à ocorrência de úlcera por pressão foram: uso de antibióticos, vasoconstrictores, ventilação mecânica; tempo prolongado de internação; menores escores na Escala de Braden e de Glasgow. Os resultados denotam a importância da utilização do protocolo de prevenção de úlcera por pressão no serviço e ressaltam a necessidade do envolvimento da instituição na manutenção de um programa de educação permanente que envolva a equipe multiprofissional, no provimento de recursos humanos e materiais para garantir a continuidade na adoção das boas práticas para prevenção de úlcera por pressão e no monitoramento contínuo do problema / Preventing pressure ulcers poses a great challenge in the hospital context, especially in Intensive Care Unit, because of the range of risk factors the patients present. The objective in this research was to assess the effect of the construction and use of an evidence-based protocol to prevent pressure ulcers in Intensive Care Unit on nursing actions and on the incidence rates of the event. The study was undertaken at the Intensive Care Unit of a university hospital in João Pessoa/Paraíba, after receiving ethical clearance. The method was constructed based on Rogers\' steps in the innovation adoption process, using the quantitative and qualitative approaches, in three phases: pre-intervention, intervention and post-intervention. In the pre- and post-intervention phases, the reality was investigated through two studies, characterizing the professionals\' preventive actions during bodily hygiene and the incidence of pressure ulcers. The intervention phase permitted the construction and use of the pressure ulcers prevention protocol, involving health professionals from the Intensive Care Unit, through the adoption of the following strategies: focus group, assessment of the professionals\' agreement and development of educative actions to persuade the professionals to adopt the recommendations. The results of the pre and post-intervention phases were compared, using statistical tests to verify the impact of using the protocol. In the qualitative study, developed during the intervention phase, Bardin\'s content analysis technique was applied for the analysis and interpretation of the professionals\' statements. The comparison between the study results of the pre and post-intervention phases evidenced that the use of the protocol significantly influenced the nursing professionals\' clinical practice, revealing significant changes in the adoption of prevention measures, including: pressure ulcers risk assessment using the Braden scale during the patient\'s hospitalization (p<0.001), increased use of body hydrating lotion (p<0.001), skin inspection on bony prominences in all body regionso (p<0.001), increased use of blankets to raise the patient from the bed while moving (p<0.001); use of pillows to protect bony prominences on the knee (p=0.015) and under the calves to maintain the heels suspended (p<0.005). In addition, an increase was observed in the use of covers to protect areas of bony prominences (p=0.005). Concerning the incidence of pressure ulcers, a statistically significant difference (p=0.0069) was observed between the two research phases, with a reduction from 35.7 % (pre-intervention) to 8.3% (post-intervention). It was identified that the variables associated with the occurrence of pressure ulcers were: use of antibiotics, vasoconstrictors, mechanical ventilation; extended hospitalization time; lower scores on Braden and Glasgow Scale. The results indicate the importance of using the pressure ulcers prevention protocol in the service and highlight the need for the institution to engage in the maintenance of a continuing education program for the multiprofessional team, in the provision of human and material resources to guarantee continuity in the adoption of best practices for pressure ulcers prevention and in the continuous monitoring of the problem