A recent survey conducted in the Clinical Microbiology Laboratory indicates that blood culture practices at UIHC are suboptimal. Between July and September of this year, an average of only 1.54 blood culture sets were obtained per septic episode in adult patients. The optimum approach to detecting bacteremia and fungemia is to perform three separate blood cultures (each consisting of 20 mls of total blood) per septic episode. Two sets should be drawn initially from different sites and a third drawn sometime thereafter during the next 12 hours. The exact timing of the third culture is unimportant. With each of the three cultures, the 20 ml blood sample should be split evenly between an aerobic and anaerobic bottle. The only circumstance that justifies fewer than three blood cultures is in patients with limited access to sufficient blood volume, i.e. infants and young children. In such cases, whenever possible collect three blood cultures but limit the volume to 5-10 mls per draw and inoculate the entire volume into just an aerobic bottle.

Two other issues deserve comment. Blood culture contamination rates at UIHC are extremely high (i.e. approximately 10%). Before collecting blood specimens, disinfect the collection site, whether it is a catheter entry port or skin, with both 70% alcohol (first) and 10% povidone iodine (second). The rubber septae of blood culture bottles should be prepped with 70% alcohol.

Secondly, it is always preferable to draw blood for culture by traditional venipuncture techniques. In patients who do not have venipuncture access, however, it is satisfactory to draw blood from indwelling vascular catheters (e.g. A-lines, central venous catheters or peripheral lines) when performing blood cultures. You must, however, first aspirate at least a 10 ml aliquot of blood off the line and discard it before collecting the specimen that will be cultured. The drawing site should be reprepped with both alcohol and iodine immediately before collecting this specimen.

Thank you for your cooperation on these issues. If you have any questions, please do not hesitate to contact Dr. Gary Doern (x6-8616), Director, Clinical Microbiology Laboratory or Dr. Bradley Brittigan (x6-3674), Chief, Infectious Diseases.