Objectives: The study objectives were to (1) determine the types and frequency of off-label (OL) or unlicensed (UL) medications used in a pediatric emergency department (PED) and before admission, (2) describe OL/UL-associated adverse drug events (ADEs) resulting in admission to the PED and those occurring during patient care in PED, and (3) determine the outcomes of these ADEs. Methods: Medical records of patients 18 years or younger admitted to the PED over a 5-month period were reviewed. Off-label/UL use of medications was determined based on Food and Drug Administration-approved labeling. The Adverse Drug Reaction Probability Scale was used to determine ADE causality. Data were analyzed using descriptive statistics. Results: A total of 2191 patients with 6675 medication orders were evaluated. About 26.2% (n = 1712) of medication orders were considered as OL/ UL use; 70.5% (n = 1208) of these medications were ordered as part of treatment in the PED, and the remaining 29.5% (n = 504) were home medications before their PED evaluation. Inhaled bronchodilators (30.4%), antimicrobials (14.8%), and antihistamines/antiemetics (9.1%) were the most common OL/UL medication classes. The frequency of ADEs among licensed medication use was greater compared with OL/UL use by 2-fold. Reported overall rate of ADEs was 0.6% (n = 40). Of these 40 ADEs, 5 resulted from the use of an OL/UL medication, 3 from home medication use, and 2 from PED-prescribed medications. Conclusions: The frequency of reported ADEs associated with OL/UL medications was less than the frequency of ADEs from licensed medication use, with overall ADE frequency of less than 1%.

Valproic acid (Depakote) is an antiepileptic drug approved for the treatment of bipolar disorder, migraine prophylaxis, and seizure disorders. While the exact mechanism is still unknown, thrombocytopenia, defined as platelet counts < 150,000/uL, has been reported secondary to Depakote treatment. The frequency of Depakote-induced thrombocytopenia varies greatly, with reported rates ranging from 5 to 54%. This adverse effect is dose-dependent and possible risk factors include lower baseline platelet counts, female gender, and high VPA serum concentrations.Our team came across two patient cases where thrombocytopenia during Depakote therapy was observed. Patient information was gathered through electronic medical records. The first patient was a 65-year-old male who was started on 500 mg Depakote ER three tablets at night for bipolar affective disorder. After several months on this dose, the patient’s platelets decreased to 59 X 103per microliter. One month after the drug was discontinued, the platelets recovered to 160 X 103per microliter. The second patient was a 57-year-old woman who had two occurrences of thrombocytopenia while on Depakote. The patient was started on Depakote for a seizure disorder. She was later admitted for symptomatic bradycardia, hypotension, and concern for thrombocytopenia. Her Depakote dose was decreased from 500 mg three times a day to twice a day. Approximately 5 weeks later, she presented to the emergency room for decreased arousal and hypotension. She was again found to have thrombocytopenia with a platelet count of 28 X 103per microliter with a Depakote level of 101 mcg/mL. The team discovered she had been receiving Depakote 500 mg three times a day following discharge from her last admission, not the reduced dose prescribed. On day four of admission, her platelets had not improved and the Depakote dose was decreased further to 250 mg twice daily. After Depakote was discontinued her platelets gradually improved and returned to normal after four days, the eighth day of admission. Utilizing the Naranjo adverse drug reaction probability scale, the first patient case had a probable reliability that this adverse reaction was due to Depakote, while the second patient case had a definite reliability.These cases illustrate the potential for thrombocytopenia secondary to Valproic acid use. Although this adverse event isn’t well understood, these cases add to the evidence that it can occur. Recognition of this reaction is important and clinicians should monitor hematologic labs, including platelets, for patients receiving Valproic acid.

Os eventos adversos a medicamentos são comuns e provavelmente irão ocorrer em qualquer sistema, entretanto, é essencial identificar suas causas e tentar minimizar os riscos visando à segurança do paciente. O objetivo desse estudo foi analisar os eventos adversos a medicamentos em um hospital público do estado do Amazonas, com base nos relatos da equipe de enfermagem e propor estratégias à instituição para otimizar o sistema de medicamentos para a segurança do paciente. Um estudo do tipo descritivo, adotando-se a Técnica do Incidente Crítico como norteadora dos procedimentos metodológicos foi realizado com 15 enfermeiros, 09 auxiliares e 44 técnicos de enfermagem de um hospital de ensino de Manaus-AM. A análise dos dados constou da identificação e agrupamento das situações, comportamentos e consequências. A maioria dos profissionais entrevistados era do sexo feminino (89,7%), predominando a classe de idade de 31 a 35 anos (30,9%). Os resultados foram agrupados e categorizados em 47 situações, correspondentes a nove tipos de erros de medicação. Dos tipos de erros identificados pela equipe de enfermagem, os erros de diluição (23,4%), dose (14,9%), medicamentos não autorizados (14,9%), erro de via de administração (10,6%) e erro de paciente (8,5%) foram os mais frequentes. Com relação às categorias de comportamentos, estes foram distribuídos em função da percepção e comunicação do erro (36) e entre as consequências identificadas no estudo (103), 47 eram relacionadas aos pacientes e 56 aos profissionais de enfermagem. Os resultados apresentados indicam a necessidade de tomada imediata de iniciativas que minimizem essas situações e previna a ocorrência de futuros eventos adversos a medicamentos. / Adverse drug events are common and will probably occur in any system, however, it is essential to identify their causes and try to minimize the risks to patients\' safety. This descriptive study aimed to analyze adverse drug events at a public hospital in the state of Amazonas, based on the reports of the nursing team. The goal was to propose strategies for the institution to optimize the medication system for patient safety. The Critical Incident Technique guided the methodological procedures, carried out with 15 nurses, 09 nursing auxiliaries and 44 nursing technicians from a teaching hospital in Manaus, state of Amazonas. Data analysis consisted of identifying and grouping situations, behaviors and consequences. Most of the interviewed professionals were female (89.7%) from the age group of 31 to 35 (30.9%) years. Results of the studies were grouped and categorized into 47 situations, corresponding to nine types of medication errors. Of the types of errors identified by the nursing team, dilution errors (23.4%), dosing errors (14.9%), unauthorized medication (14.9%), wrong administration route (10.6%) and mistaken patient (8.5%) were the most frequent. The categories of behavior were distributed according to the perception and communication of error (36) and among the consequences identified in the study (103), of which 47 were related to patients and 56 to nursing professionals. The results indicate the need for immediate initiatives that minimize these situations and prevent the occurrence of future adverse drug events.

O estudo caracterizou o perfil biodemográfico e clínico dos pacientes internados na clínica médica do Hospital Universitário da USP e os fármacos prescritos, avaliou as prescrições médicas em relação à ocorrência ou não de interações medicamentosas potenciais (IMP) e os fatores relacionados a elas para elaboração de instrumento educativo para auxiliar a atuação da equipe multidisciplinar na detecção de eventos adversos evitáveis potenciais de relevância clínica. Foram extraídos dos prontuários médicos os dados referentes à internação, identificação e estado clínico dos pacientes internados na clínica médica, entre março e agosto de 2006. Coletou-se ainda o diagnóstico principal no resumo de alta e o tempo de internação. Os fármacos prescritos foram coletados das prescrições médicas para análise da ocorrência de interações medicamentosas potenciais e classificação, empregando-se o banco de monografias Micromedex&#174; DrugReax&#174;. Empregou-se estatística descritiva e regressão logística multivariada na análise dos dados. Analisaram-se 5.666 prescrições médicas, a média de idade foi de 56,7±19,8 anos e o tempo médio de internação 10,7±9,4 dias. Os diagnósticos mais freqüentes foram broncopneumonia (138; 21,3%) e infarto agudo do miocárdio (57; 8,8%). Os grupos anatômicos mais freqüentemente prescritos, segundo o Anatomical Therapeutical Chemical, foram C (21,2%), A (17,5%), N (15,6%), B (15,0%) e J (13,6%). O número médio de fármacos por prescrição foi 5,7±2,9. Para a análise de IMP foram consideradas somente as prescrições com dois ou mais fármacos prescritos (5.336), das quais 3.097 (58,0%) apresentavam IMP. O número médio de fármacos entre essas prescrições foi 6,2±2,3. Observou-se associação significativa com as IMP identificadas no estudo: idade (p<0,001), tempo de internação (p<0,001), doenças cardiovasculares (p=0,0059) e número de fármacos (p<0,001). Tanto hipertensão arterial sistêmica quanto diabetes mellitus foram fatores de risco para ocorrência de IMP, com chances 4,93 e 2,79 vezes maiores de prescrição com IMP, respectivamente. Das 9.951 IMP observadas, 2.637 (26,5%) eram graves e bem documentadas. Observou-se associação entre o número de IMP graves e os pacientes com idade igual ou maior que 60 anos (p= 0,0007, chance 1,29 vezes maior). Diante do volume de informação sobre medicamentos e necessidade de consulta rápida para auxílio na tomada de decisões, os instrumentos educativos são de fundamental importância na garantia da qualidade e segurança da farmacoterapia. / The study characterized the demographic and clinical profile of patients admitted to the internal medicine unit of the USP University Hospital and the drugs prescribed, the prescriptions evaluated in relation to the occurrence of potential drug interactions (PDI) and factors related to them for developing educational tool to assist the performance of the multidisciplinary team in the detection of preventable adverse drug events of clinical relevance. Data were extracted from medical records regarding the identification and clinical status of patients from March to August 2006. It was still collected the main diagnosis in the discharge summary and the time of admission. The drugs prescribed were obtained from medical prescriptions for analysis of the occurrence of potential drug interactions and classification, using the database monographs Micromedex&#174; DrugReax&#174;. Descriptive statistics and logistic regression were used in data analysis. From 5.666 prescriptions evaluated, the average age was 56.7 ± 19.8 years and mean length of stay was 10.7 ± 9.4 days. The most common diagnoses were pneumonia (138, 21.3%) and heart infarct (57, 8.8%). The anatomical groups most frequently prescribed were C (21.2%), A (17.5%), N (15.6%), B (15.0%) and J (13.6%) according Anatomical Therapeutical Chemical. The average number of drugs per prescription was 5.7 ± 2.9. For the analysis of PDI were considered only the prescriptions with two or more drugs (5.336), and 3.097 (58.0%) presented PDI. The average number of drugs per prescription was 6.2 ± 2.3. There was a significant association the PDI identified in the study with: age (p <0.001), length of stay (p <0.001), cardiovascular diseases (p = 0.0059) and number of drugs prescribed (p <0.001). Both hypertension and diabetes mellitus were risk factors for the occurrence of PDI, with odds ratios 4.93 and 2.79, respectively. The frequency of major and well-documented drug interactions was 26.5%. From 9.951 PDI observed in 2.637 (26.5%) were considered major and well documented. There was association between the number of major PDI and elderly patients (p=0.0007, with odds ratio 1.29). Taking in consideration the excessive data available about drugs and the necessity of concise information to help clinical decisions, educational tools are essential to assure the quality and safety of pharmacotherapy.

O estudo caracterizou o perfil biodemográfico e clínico dos pacientes internados na clínica médica do Hospital Universitário da USP e os fármacos prescritos, avaliou as prescrições médicas em relação à ocorrência ou não de interações medicamentosas potenciais (IMP) e os fatores relacionados a elas para elaboração de instrumento educativo para auxiliar a atuação da equipe multidisciplinar na detecção de eventos adversos evitáveis potenciais de relevância clínica. Foram extraídos dos prontuários médicos os dados referentes à internação, identificação e estado clínico dos pacientes internados na clínica médica, entre março e agosto de 2006. Coletou-se ainda o diagnóstico principal no resumo de alta e o tempo de internação. Os fármacos prescritos foram coletados das prescrições médicas para análise da ocorrência de interações medicamentosas potenciais e classificação, empregando-se o banco de monografias Micromedex&#174; DrugReax&#174;. Empregou-se estatística descritiva e regressão logística multivariada na análise dos dados. Analisaram-se 5.666 prescrições médicas, a média de idade foi de 56,7±19,8 anos e o tempo médio de internação 10,7±9,4 dias. Os diagnósticos mais freqüentes foram broncopneumonia (138; 21,3%) e infarto agudo do miocárdio (57; 8,8%). Os grupos anatômicos mais freqüentemente prescritos, segundo o Anatomical Therapeutical Chemical, foram C (21,2%), A (17,5%), N (15,6%), B (15,0%) e J (13,6%). O número médio de fármacos por prescrição foi 5,7±2,9. Para a análise de IMP foram consideradas somente as prescrições com dois ou mais fármacos prescritos (5.336), das quais 3.097 (58,0%) apresentavam IMP. O número médio de fármacos entre essas prescrições foi 6,2±2,3. Observou-se associação significativa com as IMP identificadas no estudo: idade (p<0,001), tempo de internação (p<0,001), doenças cardiovasculares (p=0,0059) e número de fármacos (p<0,001). Tanto hipertensão arterial sistêmica quanto diabetes mellitus foram fatores de risco para ocorrência de IMP, com chances 4,93 e 2,79 vezes maiores de prescrição com IMP, respectivamente. Das 9.951 IMP observadas, 2.637 (26,5%) eram graves e bem documentadas. Observou-se associação entre o número de IMP graves e os pacientes com idade igual ou maior que 60 anos (p= 0,0007, chance 1,29 vezes maior). Diante do volume de informação sobre medicamentos e necessidade de consulta rápida para auxílio na tomada de decisões, os instrumentos educativos são de fundamental importância na garantia da qualidade e segurança da farmacoterapia. / The study characterized the demographic and clinical profile of patients admitted to the internal medicine unit of the USP University Hospital and the drugs prescribed, the prescriptions evaluated in relation to the occurrence of potential drug interactions (PDI) and factors related to them for developing educational tool to assist the performance of the multidisciplinary team in the detection of preventable adverse drug events of clinical relevance. Data were extracted from medical records regarding the identification and clinical status of patients from March to August 2006. It was still collected the main diagnosis in the discharge summary and the time of admission. The drugs prescribed were obtained from medical prescriptions for analysis of the occurrence of potential drug interactions and classification, using the database monographs Micromedex&#174; DrugReax&#174;. Descriptive statistics and logistic regression were used in data analysis. From 5.666 prescriptions evaluated, the average age was 56.7 ± 19.8 years and mean length of stay was 10.7 ± 9.4 days. The most common diagnoses were pneumonia (138, 21.3%) and heart infarct (57, 8.8%). The anatomical groups most frequently prescribed were C (21.2%), A (17.5%), N (15.6%), B (15.0%) and J (13.6%) according Anatomical Therapeutical Chemical. The average number of drugs per prescription was 5.7 ± 2.9. For the analysis of PDI were considered only the prescriptions with two or more drugs (5.336), and 3.097 (58.0%) presented PDI. The average number of drugs per prescription was 6.2 ± 2.3. There was a significant association the PDI identified in the study with: age (p <0.001), length of stay (p <0.001), cardiovascular diseases (p = 0.0059) and number of drugs prescribed (p <0.001). Both hypertension and diabetes mellitus were risk factors for the occurrence of PDI, with odds ratios 4.93 and 2.79, respectively. The frequency of major and well-documented drug interactions was 26.5%. From 9.951 PDI observed in 2.637 (26.5%) were considered major and well documented. There was association between the number of major PDI and elderly patients (p=0.0007, with odds ratio 1.29). Taking in consideration the excessive data available about drugs and the necessity of concise information to help clinical decisions, educational tools are essential to assure the quality and safety of pharmacotherapy.

Os eventos adversos a medicamentos são comuns e provavelmente irão ocorrer em qualquer sistema, entretanto, é essencial identificar suas causas e tentar minimizar os riscos visando à segurança do paciente. O objetivo desse estudo foi analisar os eventos adversos a medicamentos em um hospital público do estado do Amazonas, com base nos relatos da equipe de enfermagem e propor estratégias à instituição para otimizar o sistema de medicamentos para a segurança do paciente. Um estudo do tipo descritivo, adotando-se a Técnica do Incidente Crítico como norteadora dos procedimentos metodológicos foi realizado com 15 enfermeiros, 09 auxiliares e 44 técnicos de enfermagem de um hospital de ensino de Manaus-AM. A análise dos dados constou da identificação e agrupamento das situações, comportamentos e consequências. A maioria dos profissionais entrevistados era do sexo feminino (89,7%), predominando a classe de idade de 31 a 35 anos (30,9%). Os resultados foram agrupados e categorizados em 47 situações, correspondentes a nove tipos de erros de medicação. Dos tipos de erros identificados pela equipe de enfermagem, os erros de diluição (23,4%), dose (14,9%), medicamentos não autorizados (14,9%), erro de via de administração (10,6%) e erro de paciente (8,5%) foram os mais frequentes. Com relação às categorias de comportamentos, estes foram distribuídos em função da percepção e comunicação do erro (36) e entre as consequências identificadas no estudo (103), 47 eram relacionadas aos pacientes e 56 aos profissionais de enfermagem. Os resultados apresentados indicam a necessidade de tomada imediata de iniciativas que minimizem essas situações e previna a ocorrência de futuros eventos adversos a medicamentos. / Adverse drug events are common and will probably occur in any system, however, it is essential to identify their causes and try to minimize the risks to patients\' safety. This descriptive study aimed to analyze adverse drug events at a public hospital in the state of Amazonas, based on the reports of the nursing team. The goal was to propose strategies for the institution to optimize the medication system for patient safety. The Critical Incident Technique guided the methodological procedures, carried out with 15 nurses, 09 nursing auxiliaries and 44 nursing technicians from a teaching hospital in Manaus, state of Amazonas. Data analysis consisted of identifying and grouping situations, behaviors and consequences. Most of the interviewed professionals were female (89.7%) from the age group of 31 to 35 (30.9%) years. Results of the studies were grouped and categorized into 47 situations, corresponding to nine types of medication errors. Of the types of errors identified by the nursing team, dilution errors (23.4%), dosing errors (14.9%), unauthorized medication (14.9%), wrong administration route (10.6%) and mistaken patient (8.5%) were the most frequent. The categories of behavior were distributed according to the perception and communication of error (36) and among the consequences identified in the study (103), of which 47 were related to patients and 56 to nursing professionals. The results indicate the need for immediate initiatives that minimize these situations and prevent the occurrence of future adverse drug events.

京都大学 / 0048 / 新制・論文博士 / 博士(医学) / 乙第13376号 / 論医博第2210号 / 新制||医||1047(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 森田 智視, 教授 川上 浩司, 教授 佐藤 俊哉 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

Introduction :Les anticoagulants oraux soulèvent des problématiques majeures en termes de risque hémorragique et de bon usage. L’informatisation du dossier médical offre la possibilité d’accéder à de grandes bases de données que l’on peut exploiter de manière automatisée. L’objectif de ce travail est de montrer comment la réutilisation de données peut permettre d’étudier des problématiques liées aux anticoagulants et accompagner une démarche d’assurance de la qualité des soins. MéthodesCe travail a été réalisé sur les données informatisées (97 355 séjours) d’un centre hospitalier général. Pour chaque séjour nous disposons des données diagnostiques, biologiques, médicamenteuses, administratives et des courriers de sortie. Ce travail est organisé autour de 3 axes :Axe I. L’objectif est d’évaluer la qualité de la détection des facteurs pouvant majorer l’effet anticoagulant des antivitamines K (AVK), à l’aide de règles développées au cours de du projet européen PSIP (convention de subvention n° 216130). Une revue des cas sur une année a permis de calculer la valeur prédictive positive et la sensibilité des règles. Axe II. Nous avons réalisé une étude de cohorte historique sur les données de 2007 à 2012 pour déterminer les éléments majeurs impliqués dans l’élévation du risque hémorragique sous AVK dans la réalité clinique. Les cas étaient les séjours présentant une élévation de l’INR au-delà de 5, les témoins n’en présentaient pas. Axe III. Nous avons mis la réutilisation de données au service de l’étude de la qualité des prescriptions. D’une part nous avons évalué le suivi des recommandations de traitement du risque thromboembolique dans la fibrillation atriale (FA) chez la personne âgée, d’autre part nous avons étudié les modalités de prescription des anticoagulants oraux directs (AOD).Résultats : Axe I : La valeur prédictive positive des règles de détection des facteurs favorisant l’élévation de l’INR sous AVK est de 22,4%, leur sensibilité est de 84,6%. Les règles les plus contributives sont les règles de détection d’un syndrome infectieux et de l’administration d’amiodarone. Axe II : Les facteurs majeurs d’élévation du risque hémorragique sous AVK mis en évidence par l’étude de cohorte sont le syndrome infectieux, le cancer, l’hyprotidémie et l’insuffisance hépatique. Axe III : Le taux de suivi des recommandations dans la fibrillation atriale chez le sujet âgé est de 47.8%. Seuls 45% des patients reçoivent des anticoagulants oraux, 22,9% ne reçoivent aucun traitement antithrombotique et 32,1% reçoivent des antiagrégants plaquettaires. Les AOD sont quant à eux prescrits à des posologies inadaptées chez 15 à 31,4% des patients, respectivement pour le dabigatran et le rivaroxaban. Ces erreurs sont principalement des sous-dosages en AOD dans la FA de la personne âgée (82.6%). Discussion : L’informatisation des dossiers médicaux a permis la constitution de grandes bases de données médico-administratives, qui peuvent être utilisées à des fins variées comme nous le montrons dans ce travail. Dans le premier axe nous avons montré que des systèmes d’aide à la décision à base de règles permettent de caractériser les facteurs impliqués dans les surdosages en AVK avec une bonne sensibilité mais avec une faible valeur prédictive positive. Le second axe a montré que l’on pouvait utiliser ces données à des fins exploratoires pour identifier les facteurs liés à l’élévation de l’INR chez les patients recevant des AVK en pratique réelle. Le troisième axe montre que les systèmes à base de règles peuvent aussi être utilisés pour identifier des prescriptions inappropriées à des fins d’amélioration de la qualité des soins. Dans le domaine de l’anticoagulation ce travail ouvre des perspectives innovantes en vue de l’amélioration de la qualité des soins. / Introduction :Oral anticoagulants raise major issues in terms of bleeding risk and appropriate use. The computerization of medical records offers the ability to access large databases that can be explored automatically. The objective of this work is to show how routinely collected data can be reused to study issues related to anticoagulants in a supportive approach to quality of care.MethodsThis work was carried out on the electronic data (97,355 records) of a community hospital. For each inpatient stay we have diagnostic, biological, drug and administrative data, and the discharge letters. This work is organized around three axes:Axis I. The objective is to evaluate the accuracy of the detection of factors that may increase the anticoagulant effect of vitamin K antagonists (VKA), using rules developed in the PSIP european project (grant agreement N° 216130). A case review on one year enabled the calculation of the positive predictive value and sensitivity of the rules. Axis II. We conducted a cohort study on data from 2007 to 2012 to determine the major elements involved in raising the risk of bleeding related to VKA in clinical reality. Cases were the stays with an elevation of the INR beyond 5, the controls did not have.Axis III. We made data reuse serve a study of the quality of the prescriptions. On the one hand we assessed treatment of the thromboembolic risk recommendations in atrial fibrillation (AF) in the elderly, on the other hand we investigated the prescription of direct oral anticoagulants.Results : Axis I : The positive predictive value of the rules intended to detect the factors favoring the elevation of INR in case of treatment with VKA is 22.4%, the sensitivity is 84.6%. The main contributive rules are the ones intended to detect an infectious syndrome and amiodarone.Axis II : The major factor increasing the INR with VKA treatment highlighted by the cohort study are infectious syndrome, cancer, hepatic insufficiency and hypoprotidemia. The recommendations compliance rate in atrial fibrillation in the elderly is 47.8%. Only 45% of patients receive oral anticoagulants, 22.9% do not receive antithrombotic treatment at all and 32.1% received platelet aggregation inhibitors. Direct oral anticoagulants are prescribed at inadequate dosages in 15 to 31.4% of patients, respectively for dabigatran and rivaroxaban. These errors are mainly underdosages in the elderly with atrial fibrillation (82.6%).Discussion : The computerization of medical records has led to the creation of large medical databases, which can be used for various purposes as we show in this work. In the first work axis we have shown that rule-based decision support systems detect the contributing factors for VKA overdose with a good sensitivity but a low positive predictive value. The second line shows that we could use the data for exploratory purposes to identify factors associated with increased INR in patients receiving VKA in “real life practice”. The third line shows that the rule-based systems can also be used to identify inappropriate prescribing for the purpose of improving the quality of care. In the field of anticoagulation this work opens up innovative perspectives for improving the quality of care.

A farmacoterapia complexa com múltiplos medicamentos aumenta a probabilidade de interações medicamentosas (IM) e eventos adversos a medicamentos(EAM) em pacientes de unidade de terapia intensiva (UTI). Este estudo teve como objetivos: analisar e classificar as IM potenciais dos tipos fármaco-fármaco e fármaco-nutrição enteral e os EAM detectados, durante a internação de pacientes, na UTI de um hospital de ensino, identificar os EAM relacionados a IM e investigar os fatores associados com IM e EAM. O estudo transversal, correlacional, retrospectivo investigou 299 prontuários de pacientes internados por cinco ou mais dias na UTI. As IM potenciais foram identificadas e classificadas empregando o software Drug Reax. Investigou-se IM em prescrições de três momentos: 24 horas de internação, mediana da internação e alta. Os EAM foram identificados empregando o método de monitorização intensiva e também critérios de rastreabilidade de EAM. Os EAM foram classificados em relação à causalidade e à gravidade Os dados foram analisados empregando análise descritiva, univariada e regressão logística múltipla. A maioria da casuística era do sexo feminino (50,5%) e com mediana de idade de 57 anos. Em 24 horas a prevalência de pacientes com IM potenciais foi 68,6%. A maior frequência de pacientes com IM potenciais foi 73,9% na mediana da internação. Na alta detectou-se que 69,6% dos pacientes apresentaram IM potenciais. Cerca de 99% das interações era do tipo fármaco-fármaco. As interações fármaco-nutrição enteral foram mais frequentes em prescrições de pacientes em alta. Em relação a gravidade a maioria das IM eram graves e moderadas. O número de medicamentos administrados e o uso de medicamentos do sistema cardiovascular foram os fatores associados com a ocorrência de IM, de forma independente, nos três momentos da internação. Detectou-se que 34,1% dos pacientes apresentaram EAM. Os EAM mais frequentes foram reações adversas a medicamentos (RAM) do Tipo A, de causalidade possível e gravidade moderada. Os pacientes com EAM relacionados a IM corresponderam a 7% da casuística. O tempo de internação e a administração de medicamentos cardiovasculares apresentaram associação com a ocorrência de EAM na regressão logística . Os resultados desta investigação mostraram que apesar da frequência elevada de IM potenciais, o número de EAM relacionados a IM foi reduzido, entretanto envolveu RAM de significância clínica como nefrotoxicidade e sedação excessiva que podem comprometer a segurança do paciente. / Complex pharmacotherapy with multiple medications increases the chances of drug interactions (DI) and adverse drug events (ADE) in patients at intensive care units (ICU). This study aimed to analyze and classify potential drug interactions of the following types: drug-drug and drug-enteral nutrition and also ADE detected during patients\' hospitalization in the ICU of a teaching hospital; to identify drug adverse events related to DI; and investigate factors associated with DI and ADE. This crosssectional, correlational and retrospective study investigated 299 medical records of patients hospitalized for five or more days in the ICU. Potential DI were identified and classified through the software Drug-Reax. Interactions in prescriptions were identified in three occasions: at 24 hours of hospitalization, at median of stay and at discharge of ICU stay . ADE were identified through the method of intensive monitoring and also adverse drug event trigger. ADE were classified regarding causality and severity.Data were analyzed through descriptive analysis, univariate and multiple logistic regression. The majority was female (50.5%) and median age was 57 years. The prevalence of patients with potential DI at 24h was 68.6%. The highest frequency of patients with potential DI was 73.9% in the median of hospitalization. We found that 69.6% of patients presented potential DI at discharge. About 99% were drug-drug interactions. The drug-enteral nutrition interactions were more frequent in patients\' prescriptions at discharge. Most IM were severe or moderate.The number of administered medication and the use of medication for the cardiovascular system were the factors independently associated with the occurrence of DI in the three occasions of hospitalization. The study also detected that 34.1% of patients presented ADE The most frequent ADE were adverse drug reactions (ADR) type A, of possible causality and moderate severity. Patients with ADE related to DI corresponded to 7% of the sample. The multiple logistic regression revealed that time of hospitalization and administration of cardiovascular medication were associated with the occurrence of ADE. This study\'s results revealed that despite the high frequency of potential DI, the number of ADE related to DI was small, however, these involved ADR of clinical significance such as nephrotoxicity and oversedation, which might compromise patient safety.

A farmacoterapia complexa com múltiplos medicamentos aumenta a probabilidade de interações medicamentosas (IM) e eventos adversos a medicamentos(EAM) em pacientes de unidade de terapia intensiva (UTI). Este estudo teve como objetivos: analisar e classificar as IM potenciais dos tipos fármaco-fármaco e fármaco-nutrição enteral e os EAM detectados, durante a internação de pacientes, na UTI de um hospital de ensino, identificar os EAM relacionados a IM e investigar os fatores associados com IM e EAM. O estudo transversal, correlacional, retrospectivo investigou 299 prontuários de pacientes internados por cinco ou mais dias na UTI. As IM potenciais foram identificadas e classificadas empregando o software Drug Reax. Investigou-se IM em prescrições de três momentos: 24 horas de internação, mediana da internação e alta. Os EAM foram identificados empregando o método de monitorização intensiva e também critérios de rastreabilidade de EAM. Os EAM foram classificados em relação à causalidade e à gravidade Os dados foram analisados empregando análise descritiva, univariada e regressão logística múltipla. A maioria da casuística era do sexo feminino (50,5%) e com mediana de idade de 57 anos. Em 24 horas a prevalência de pacientes com IM potenciais foi 68,6%. A maior frequência de pacientes com IM potenciais foi 73,9% na mediana da internação. Na alta detectou-se que 69,6% dos pacientes apresentaram IM potenciais. Cerca de 99% das interações era do tipo fármaco-fármaco. As interações fármaco-nutrição enteral foram mais frequentes em prescrições de pacientes em alta. Em relação a gravidade a maioria das IM eram graves e moderadas. O número de medicamentos administrados e o uso de medicamentos do sistema cardiovascular foram os fatores associados com a ocorrência de IM, de forma independente, nos três momentos da internação. Detectou-se que 34,1% dos pacientes apresentaram EAM. Os EAM mais frequentes foram reações adversas a medicamentos (RAM) do Tipo A, de causalidade possível e gravidade moderada. Os pacientes com EAM relacionados a IM corresponderam a 7% da casuística. O tempo de internação e a administração de medicamentos cardiovasculares apresentaram associação com a ocorrência de EAM na regressão logística . Os resultados desta investigação mostraram que apesar da frequência elevada de IM potenciais, o número de EAM relacionados a IM foi reduzido, entretanto envolveu RAM de significância clínica como nefrotoxicidade e sedação excessiva que podem comprometer a segurança do paciente. / Complex pharmacotherapy with multiple medications increases the chances of drug interactions (DI) and adverse drug events (ADE) in patients at intensive care units (ICU). This study aimed to analyze and classify potential drug interactions of the following types: drug-drug and drug-enteral nutrition and also ADE detected during patients\' hospitalization in the ICU of a teaching hospital; to identify drug adverse events related to DI; and investigate factors associated with DI and ADE. This crosssectional, correlational and retrospective study investigated 299 medical records of patients hospitalized for five or more days in the ICU. Potential DI were identified and classified through the software Drug-Reax. Interactions in prescriptions were identified in three occasions: at 24 hours of hospitalization, at median of stay and at discharge of ICU stay . ADE were identified through the method of intensive monitoring and also adverse drug event trigger. ADE were classified regarding causality and severity.Data were analyzed through descriptive analysis, univariate and multiple logistic regression. The majority was female (50.5%) and median age was 57 years. The prevalence of patients with potential DI at 24h was 68.6%. The highest frequency of patients with potential DI was 73.9% in the median of hospitalization. We found that 69.6% of patients presented potential DI at discharge. About 99% were drug-drug interactions. The drug-enteral nutrition interactions were more frequent in patients\' prescriptions at discharge. Most IM were severe or moderate.The number of administered medication and the use of medication for the cardiovascular system were the factors independently associated with the occurrence of DI in the three occasions of hospitalization. The study also detected that 34.1% of patients presented ADE The most frequent ADE were adverse drug reactions (ADR) type A, of possible causality and moderate severity. Patients with ADE related to DI corresponded to 7% of the sample. The multiple logistic regression revealed that time of hospitalization and administration of cardiovascular medication were associated with the occurrence of ADE. This study\'s results revealed that despite the high frequency of potential DI, the number of ADE related to DI was small, however, these involved ADR of clinical significance such as nephrotoxicity and oversedation, which might compromise patient safety.