Risk factors and lethality of laboratory-confirmed bloodstream infection caused by non-skin contaminant pathogens in neonates

To evaluate risk factors and lethality of late onset laboratory-confirmed bloodstream infection (LCBI) in a Brazilian neonatal unit for progressive care (NUPC). This was a case-control study, performed from 2008 to 2012. Cases were defined as all newborns with late onset LCBI, excluding patients with isolated common skin contaminants. Controls were newborns who showed no evidence of late onset LCBI, matched by weight and time of permanence in the NUPC. Variables were obtained in the Hospital Infection Control Committee (HICC) database. Analysis was performed using the Statistical Package for the Social Sciences (SPSS). The chi-squared test was used, and statistical significance was defined as p < 0.05, followed by multivariate analysis. 50 patients with late onset LCBI were matched with 100 patients without late onset LCBI. In the group of patients with late onset LCBI, a significant higher proportion of patients who underwent surgical procedures (p = 0.001) and who used central venous catheter (CVC) (p = 0.012) and mechanical ventilation (p = 0.001) was identified. In multivariate analysis, previous surgery and the use of CVC remained significantly associated with infection (p = 0.006 and p = 0.047; OR: 4.47 and 8.99, respectively). Enterobacteriacea was identified in 14 cases, with three (21.4%) deaths, and Staphylococcus aureus was identified in 20 cases, with three (15%) deaths. Surgical procedures and CVC usage were significant risk factors for LCBI. Therefore, prevention practices for safe surgery and CVC insertion and manipulation are essential to reduce these infections, in addition to training and continuing education to surgical and assistance teams. Avaliar os fatores de risco e a letalidade da infecção da corrente sanguínea laboratorialmente confirmada (ICSLC) de início tardio em uma Unidade Neonatal de Cuidados Progressivos (UNCP) brasileira. Trata-se de um estudo caso-controle realizado de 2008 a 2012. Os casos foram definidos como todos os recém-nascidos com ICSLC de início tardio, excluindo pacien- tes isolados com contaminantes da pele comuns. Os controles foram recém-nascidos que não mostraram qualquer evidência de ICSLC de início tardio, sendo separados por peso e tempo de permanência na UNCP. As variáveis foram obtidas na base de dados da Comissão de Controle de Infecção Hospitalar (CCIH). A análise foi realizada utilizando o Pacote Estatístico para Ciências Sociais. O teste x2 foi utilizado e a relevância estatística foi definida como