RESUMO
BACKGROUND: Reducing the blood-culture contamination rate is a constant challenge for health services. This study aimed to analyze whether blood-culture (BC) collection using sterile gloves reduces the contamination rate when compared to the non-sterile gloves, and to compare baseline and intervention periods. METHODS: A randomized clinical trial, performed in an intensive care unit in Brazil and paired in two groups: sterile (BCs obtained with modified sterile technique: only sterile gloves; no fenestrated drape or dedicated sterile collection kit) and clean (clean technique: usual care with non-sterile gloves). Two paired blood samples were obtained from each patient by trained and calibrated nurses. BCs were processed by conventional microbiological methods and the results were issued by blinded microbiologists. RESULTS: There was no difference (P = 1.00) in the contamination rate of BC between the sterile (1%) and clean (1%) groups. However, there was a significant difference (P = 0.05; relative risk: 0.17; 95% confidence interval: 0.04-0.70) in the contamination rate between baseline (6.1%; 20/330) and intervention (1%; 2/200). CONCLUSIONS: This study suggests that the aseptic care provided in obtaining samples is more important than the sterile technique itself, and highlights the value of standardizing the practices, qualification and calibration of phlebotomists.