Justificativa â A pneumonia hospitalar (PH) à causa de morbidade e mortalidade elevadas em pacientes hospitalizados. A terapia antimicrobiana empÃrica adequada e precoce pode salvar a vida de mais da metade dos pacientes com PH e deve ser baseada em padrÃes locais de sensibilidade a antimicrobianos. Praticamente todo tratamento de PH Ã, inicialmente, empÃrico. O objetivo deste trabalho à contribuir com o conhecimento do padrÃo regional de resistÃncia de microrganismos nesse contexto. Metodologia â Foram estudados 588 resultados de culturas de lavado bronco-alveolar (LBA) de pacientes internados em Fortaleza, processados na rotina de um laboratÃrio privado, no perÃodo de janeiro de 1996 a dezembro de 2001. Como resultado de pesquisa aos prontuÃrios mÃdicos desses pacientes, dois grupos foram criados: Grupo 1, com n=147, composto por pacientes com pneumonia hospitalar (PH) notificada segundo os critÃrios do Center for Disease Control and Prevention (CDC); Grupo 2, com n=382, pacientes com pneumonia nÃo-hospitalar (PNH). Utilizou-se a tÃcnica quantitativa de cultivo descrita nos trabalhos de Kahn e Jones (1987), Salata et al. (1987) e Carvalho (1997). IdentificaÃÃo e antibiogramas foram realizados no equipamento VITEK BioMerrieux, exceto para Streptococcus pneumoniae e Stenotrophomonas maltophilia cujos antibiogramas foram realizados pelo mÃtodo Kirby-Bauer e E-test respectivamente. Resultados â No Grupo 1, os cinco microrganismos mais freqÃentes foram Pseudomonas aeruginosa [56 casos (38,1%)], Staphylococcus aureus [24 casos (16,3%)], Klebsiella pneumoniae [12 casos (8,2%)], Acinetobacter spp [12 casos (8,2%)] e Serratia marcescens [10 casos (6,8%)]. No Grupo 2, encontraram-se, mais freqÃentemente, Pseudomonas aeruginosa [113 casos (29,6%)], Staphylococcus aureus [89 casos (23,3%)], Klebsiella pneumoniae [32 casos (8,4%)] e Acinetobacter spp [31 casos (8,1%)]. NÃo foi observada diferenÃa significativa entre os dois grupos para a etiologia. O mesmo ocorreu com o perfil de resistÃncia dos organismos, exceto para o Grupo 1 com S. aureus/oxacilina (p=0,027) e P. aeruginosa/piperacilina/tazobactam (p=0,003). No perfil de resistÃncia do conjunto total de amostras (n=751), destaca-se a de P. aeruginosa ao imipenem, de 40,8%; de Acinetobacter spp ao imipenem, de 10,0%; de Acinetobacter spp a Ampicilina/Sulbactam, de 44,3%; e de S. aureus a oxacilina, de 67,3%. O intervalo de tempo entre a data de internaÃÃo e a realizaÃÃo da cultura foi maior do que 7 dias em 60,18% dos casos. ConclusÃo - no trato respiratÃrio, o problema da multiresistÃncia bacteriana à evidente e preocupante com alta prevalÃncia de bacilos gram-negativos multiresistentes, marcadamente P. aeruginosa e Acinetobacter spp., assim como elevada resistÃncia a oxacilina nas amostras de Staphylococcus aureus. O Grupo 2 nÃo possui caracterÃsticas de pneumonia comunitÃria (PC), provavelmente, porque o tempo entre a internaÃÃo e a realizaÃÃo da cultura foi longo o suficiente para permitir a colonizaÃÃo do trato respiratÃrio superior pela microflora do ambiente hospitalar. à possÃvel que o Grupo 2 seja constituÃdo por pacientes com pneumonia comunitÃria severa refratÃria à antibioticoterapia que necessita internaÃÃo, ou que tiveram vÃrias internaÃÃes anteriores, com conseqÃente colonizaÃÃo por microflora hospitalar. InvestigaÃÃes subseqÃentes devem conferir atenÃÃo especial a esse aspecto. Pode ser Ãtil, neste contexto, o emprego de tÃcnicas de Biologia Molecular para melhor caracterizaÃÃo dos microrganismos isolados / Hospital acquired pneumonia (HAP) is associated with high morbidity and mortality in hospitalized patients. Early, appropriate, and adequate empiric antibiotic therapy can save lives of more than half of all HAP patients and must be based on local data. This study will provide local patterns of antibiotic resistance. Practically all primary therapy of HAP is empiric and information from surveillance program of a given hospital is very important. We studied 588 Bronchoalveolar lavage cultures results from hospitalized patients performed in a private lab during a period of 6 years from 1996 to 2001. As a result of searching patientâs records, two groups were assigned: Group 1, n=147, patients with HAP notified by Nosocomial Infection Commission according to Center for Disease Control and Prevention-CDC; Group 2, n=382, patients with No-Nosocomial Pneumonia. Bacteriologic cultures were done quantitatively with a threshold of >= 105 according to Kahn and Jones (1987), Salataet al. (1987) and Carvalho (1997). Identification and susceptibility tests were performed on VITEK BioMerrieux except for Streptococcus pneumoniae and Stenotrophomonas maltophilia. In patients from Group 1, the five most frequent agents were: P. aeruginosa 56 cases (38,1%), S.aureus 24 (16,3%), K. pneumonia 12 (8,2%), Acinetobacter spp 12 (8,2%) and S. marcescens 10 (6,8%). Group 2 shows: P. aeruginosa 113 (29,6%), Staphylococcus aureus 89 (23,3%), Klebsiella pneumoniae 32 (8,4%), Acinetobacter spp 31 (8,1%) and Candida spp 20 (5,2%). There was no significant difference between resistance profile of isolates when distributed in two groups except S. aureus/Oxacilina (p=0,027), P.aeruginosa/Piperacilina/Tazobactam (p=0,003). The resistance profile in total (n=751) was: P. aeruginosa/Imipenem 40,8%, Acinetobacter spp/Imipenem 10,0%, Acinetobacter spp/AmpicilinaSulbactam 44,3% e S. aureus/Oxacilina 67,3%. The time between admission date and culture request was more than 7days in 60,18% in both groups. Conclusion: a) drug-resistance in lower respiratory tract infection(LRTI) is a serious concern mainly with high prevalence of multi-R gram-negative like P. aeruginosa and Acinetobacter with high resistance for Imipenem and other β- IactÃmic and S. aureus with high resistance to Oxacilina. There was no significant difference between the two groups. Group 2 did not show characteristics of Community-Acquired Pneumoniae (CAP) maybe because of large intervals of time between admission and request of culture, or this kind of patient had either severe CAP with no response to prior multiple antimicrobial therapy or previous hospitalizations or even had recent hospitalization and consequent colonization. Forwards molecular studies should be performed on isolates to provide better characterization of lower respiratory tract pathogens.

Antibiotic resistant bacteria, a growing health crisis, arise due to antibiotic overuse and misuse. Resistant infections endanger the lives of patients and are financially burdensome. Aggregate antimicrobial susceptibility reports, called antibiograms, are critical for tracking antibiotic susceptibility and evaluating the likelihood of the effectiveness of different antibiotics to treat an infection prior to the availability of patient specific susceptibility data. This research leverages the Massachusetts Statewide Antibiogram database, a rich dataset composed of antibiograms for $754$ antibiotic-bacteria pairs collected by the Massachusetts Department of Public Health from $2002$ to $2016$. However, these antibiograms are at least a year old, meaning antibiotics are prescribed based on outdated data which unnecessarily furthers resistance. Our objective is to employ data science techniques on these antibiograms to assist in developing more responsible antibiotic prescription practices. First, we use model selectors with regression-based techniques to forecast the current antimicrobial resistance. Next, we develop an assistant to immediately identify clinically and statistically significant changes in antimicrobial resistance between years once the most recent year of antibiograms are collected. Lastly, we use k-means clustering on resistance trends to detect antibiotic-bacteria pairs with resistance trends for which forecasting will not be effective. These three strategies can be implemented to guide more responsible antibiotic prescription practices and thus reduce unnecessary increases in antibiotic resistance.

Antibiotic resistant bacteria, a growing health crisis, arise due to antibiotic overuse and misuse. Resistant infections endanger the lives of patients and are financially burdensome. Aggregate antimicrobial susceptibility reports, called antibiograms, are critical for tracking antibiotic susceptibility and evaluating the likelihood of the effectiveness of different antibiotics to treat an infection prior to the availability of patient specific susceptibility data. This research leverages the Massachusetts Statewide Antibiogram database, a rich dataset composed of antibiograms for $754$ antibiotic-bacteria pairs collected by the Massachusetts Department of Public Health from $2002$ to $2016$. However, these antibiograms are at least a year old, meaning antibiotics are prescribed based on outdated data which unnecessarily furthers resistance. Our objective is to employ data science techniques on these antibiograms to assist in developing more responsible antibiotic prescription practices. First, we use model selectors with regression-based techniques to forecast the current antimicrobial resistance. Next, we develop an assistant to immediately identify clinically and statistically significant changes in antimicrobial resistance between years once the most recent year of antibiograms are collected. Lastly, we use k-means clustering on resistance trends to detect antibiotic-bacteria pairs with resistance trends for which forecasting will not be effective. These three strategies can be implemented to guide more responsible antibiotic prescription practices and thus reduce unnecessary increases in antibiotic resistance.