 |
|
 |
|
|
|
|
 |
We have had several months where our ER physicians seem to be ordering alot of blood cultures. Our infection control department is inquiring about what is practical. One particular practice common to most of the doctors is to order blood cultures x2 before any blood work has been processed. Is this typical? Acceptable? I know at one point in addition to fever of unknown origin was the CBC results. Is this still practiced? What if any, are the criteria to determine if a blood culture is warranted?
|
 |
Whether too many blood cultures are being ordered and whether they are clinically indicated, is, unfortunately, ultimately the decision of the clinician who orders them. There are some suggestions, however, that microbiologists can offer to their clinical colleagues. Of course, there is no correct answer. One of the best publications on the subject is a classic paper: Aronson, M. D. and D. H. Bor. 1987. Blood Cultures. Ann.Intern.Med 106:246-253.
Using information and ideas from that publication and others, I created a list of criteria for our physicians to consider a few years ago. It is printed here:
a. ONLY order blood cultures on patients with fever >101 F (38.3 C).
b. The following risk factors are associated with true bacteremia:
(1) IV drug abuse
(2) Shaking chills
(3) Major comorbidity; or rapidly ( <1 year) or ultimately ( <5 year) fatal underlying disease
(4) Acute abdomen
(5) WBC <1 x 10,000,000,000/L or >15 x 10,000,000,000/L, and Bands >10%.
Patients without these risk factors and temperature <101 F are likely to be not septic (98%). Observe such patients and watch for signs of septicemia before ordering blood cultures.
Laboratories can use internal quality assurance data to provide feedback to clinicians. Some strategies are outlined in the ASM Cumitech 1C, Blood Cultures IV and in the new CLSI document on Blood Cultures, M47-A. I urge you to have both references in your laboratory. One QA tool is to determine the number of blood cultures ordered per 1000 patient days. Somewhere around 120 is average. If your facility is receiving more than 150/1000 patient days, then you are on the very high side. Another QA measurement is percent of positive blood cultures. Most institutions run around 10-12% positive, with at least half of those being clinically important. If you are much below 8%, I would worry that they are drawing too many.
So I hope this helps to answer your question and that you are on the road to some QA activities.
Ellen Jo Baron, Ph.D., D(ABMM)
We hope that we were able to answer your question.
Please visit us again at the American Society for Microbiology.
DISCLAIMER
Askit is for general information purposes only and may not be relied upon by users for decisions or action in specific circumstances. Neither ASM nor the expert assures the accuracy or completeness of information provided. Questions are answered based on current information, which may change over time. Please note the date that a question was answered. If an answer is more than one year old, you may want to resubmit the question for an updated answer.
The advice and comments do not necessarily reflect the views of ASM. ASM and INDIVIDUAL experts expressly disclaim any and all liability for any direct, indirect, or special damages or loss of any nature incurred as a consequence of use of this site or any advice or comment by any expert. By using Askit, the user accepts the waiver of claims and all terms of use of ASM’s site, including the disclaimers set forth in the “Legal Rights” link on ASM’s home page.
|
| |
|
Return to Questions
|
|
|
|
 |
|
|
