Background: Although blood cultures are commonly used to diagnose sepsis, the accuracy of the results is highly dependent on proper technique, especially the collection of a sufficient volume of blood. Current evidence-based guidelines recommend that at least 10 mL of blood be collected for each culture (5 mL for an aerobic collection bottle and 5 mL for an anaerobic collection bottle). In 1993, insufficient blood volume was recognized as a widespread quality-control problem that limited the accuracy of blood culture results. It is unknown whether this problem has improved. Donnino and colleagues examined whether hospital-based health care professionals follow evidence-based recommendations for blood culture collection.
The Study: Participants were anonymously surveyed at a tertiary care center in Detroit, Mich. Health care professionals in the emergency department, medical and surgical general wards, and intensive care unit who were qualified to collect blood for culture were included in the study. Participants completed short questionnaires about their job titles, years of experience, and specimen collection, including the appropriate blood volume needed for a culture.
Results: Of the 360 participants surveyed, 355 were included in the analysis: 116 physicians, 192 nurses, 32 technicians, and 15 health care professionals from other or unspecified disciplines. Overall, 44 percent of respondents reported that a blood volume of less than 10 mL per culture (less than 5 mL per bottle) was acceptable; this included higher percentages of nurses and technicians than physicians. Twenty-one percent of all respondents reported that less than 1 mL per bottle was sufficient; this included 11 percent of physicians, 25 percent of nurses, and 38 percent of technicians. When divided by work location, 31 percent of respondents who worked in the intensive care unit, 24 percent in the emergency department, and 12 percent in a general ward reported that 1 mL or less was sufficient.
Conclusion: The authors conclude that there is a widespread lack of awareness about the appropriate blood volume required for a culture. This lack of knowledge is greatest among nurses and technicians, who are responsible for collecting blood for culture at the institution where the study was conducted. Assuming they can be generalized to include other hospitals, these findings could have significant implications for hospital-based care. Further studies and local quality-improvement projects could help ensure proper blood collection for culture.