Beth Israel Deaconess Medical Center Laboratory Manual

Back to Specimen Collection
Back to Main Page of Lab Manual

SUMMARY: A retrospective review of data at BIDMC (and also consistent with both the literature and practices at other comparable hospitals) indicate that yeasts (e.g., Candida, Cryptococcus) are adequately detected using standard blood culture bottles. At present, we offer two types of blood culture bottles that can detect yeast: standard blood culture bottles and a more resource intensive "fungal/mycobacterial" blood culture bottle. In our review, we found that these two systems are functionally equivalent for yeast. Therefore, for more effective use of resources, the more specialized bottle should be reserved for detection of organisms other than yeasts, i.e., Mycobacteria (specifically MAC) in patients with a CD4 count < 100 and dimorphic fungi, e.g., Histoplasma, Coccidiodes, or Blastomyces, (very rare in blood). Accordingly, the fungal/mycobacterial blood culture now will be given a more specific name ("blood culture for dimorphic fungi and mycobacteria") and will be taken off the Blood Test page menu of POE/OMR. However, the test will still be available via the Lookup function. More details on the appropriate use of different types of blood culture bottles are available both in the BIDMC lab manual and will appear as a comment when ordering "Blood Culture for Dimorphic Fungi and Mycobacteria. The new algorithm for blood culture bottle use was reviewed and approved by the BIDMC Core Clinical Services Committee on April 25, 2006.

A more detailed description of our review and resulting changes to ordering menus follows:

Our own data show that the mean and median time to detection for yeasts is essential the same for the standard blood culture bottle and the Myco/F Lytic bottles (i.e., fungal/mycobacterial blood culture bottles).

Time to Detection (in days)

	Myco/F Lytic	Standard Blood Cultures
average	1.9	2.0
median	1.5	1.8
mode	1.5	1.5

In 2005, 21 episodes of candidemia were detected by the Myco/F Lytic system during this period, and 52 patients were detected with standard aerobic/anaerobic blood culture sets. 20% of the isolates detected by the Myco/F Lytic bottle were missed by routine blood culture sets and 40% detected by standard blood culture bottles were missed by proximate Myco/F Lytic cultures. The organisms missed by the standard blood culture system are not known to be particularly fastidious with the exception of one Malassezia furfur isolate that should have been cultivated by a special test that we offer (culture under sterile olive oil) rather than either the Myco/F Lytic or standard blood culture system. One paper describes results of artificial inoculation with bottles, and that Candida glabrata appeared to grow more slowly in standard blood culture bottles. Of note, however, none of the "misses" in our data were Candida glabrata.

One interpretation of our data is that patients with low numbers of yeast in their blood might be detected variably by either blood culture system by chance, and a miss by one or the other might not reflect a superiority of either method.

There are no studies in the literature convincingly demonstrating the superiority of the MycoF-Lytic bottle in detecting Candida species in clinical specimens.

B. Negative Impact of Current Excessive Use of Myco/F Lytic Bottles

Operationally, Myco/F Lytic bottles now occupy an entire blood culture machine (capacity 240 bottles). We are now at or exceed our total blood culture capacity and need to pull bottles of early and manually plant them on media, creating significantly extra work for the laboratory. Without intervention, we will need to acquire another blood culture machine and do not have room for it.

Of note, only 1.8% of patients who had Myco/F Lytic cultures were positive for Candida species and 1.0% of patients drawn had a positive Mycobacterial culture. Many patients had multiple sets. Although the Myco/F Lytic bottle is supposed to draw 1-5mls of blood, its vacuum will draw 17mls (internal data) and many bottles are "overfilled". The blood loss to patients is not insignificant and could be eliminated by more streamlined use.

C. Practices at Other Hospitals

Other regional academic medical centers surveyed do not offer a "fungal" blood culture bottle, but instead offer a manual method known as an isolator tube. In these hospitals, this method is used for detection of molds (not yeasts) and Mycobacteria, and is not a primary method for identifying yeast. We also surveyed a number of other tertiary care hospitals out of the state, including those that offer the Myco/F Lytic blood culture system, and found similar findings, where use of methods other standard blood culture bottles to detect yeast were actively discouraged.

D. Specific Proposal

The following changes will be made in POE/OMR:

1. The current POE/OMR Test name ("Fungal/AFB Blood Culture") will be removed from blood test menu page and make available with a more specific name as a "lookup". Removing from the primary test menu will minimize indiscriminate checking off of boxes.
2. New POE test name

Blood Culture for Dimorphic Fungi and Mycobacteria

1. When the new test is order the following educational comment will appear in POE/OMR to guide appropriate usage:

Intercept Screen

Do not order this test for detection of Candida and related yeasts, as these are adequately detected in standard blood culture bottles.

For mycobacteria, this test should be used for detection of Mycobacterium avium complex (MAC) in HIV patients only with a CD4 count < 100 (typically < 50) and other similarly immunocompromised patients.

Order for dimorphic fungi, e.g., Histoplasma, Coccidiodes, or Blastomyces in someone with suspected systemic fungal infection & appropriate exposure history. NOTE: we have never had a positive Myco/F Lytic blood culture for these species at BIDMC. Consider using other methods for detection, e.g., antigen testing, serology, tissue biopsy and culture, or culture of respiratory fluids.

For invasive Aspergillosis detection, consider galactomannin assay and radiology findings, as fungal blood culture is insensitive. For systemic cryptococcosis, order a cryptococcal antigen test. For suspected Malassezia infection in patients on TPN, order a Malassezia blood culture.

A maximum of one test should be performed a week.

James Kirby, MD
Medical Director, Clinical Microbiology Laboratory

Howard Gold, MD
Medical Director, Antimicrobial Management

 Last reviewed/revised 2/15/2008