Printer Friendly

Optimal use of Myco/F lytic and standard BACTEC blood culture bottles for detection of yeast and mycobacteria.

Specific media for isolation of fungi and mycobacteria have been introduced on a number of automated blood culture systems; however, their optimal use in a clinical setting is not well defined. In the BACTEC 9240 system (BD Diagnostics, Sparks, Md), the Myco/F Lytic blood culture medium supports the growth of yeast, filamentous fungi, and mycobacteria through a formulation that includes brain-heart infusion broth, a rich media that supports fungal growth, and Middlebrook 7H9 base, a mycobacterial media (formulation listed at http://www. bd.com/ds/productCenter/BC-BactecMedia.asp, last accessed on November 7, 2007).

Myco/F Lytic culture offers advances over manual methods for detecting fungi and mycobacteria, such as the multistep Wampole ISOSTAT/ISOLATOR (Inverness Medical Professional Diagnostics, Princeton, NJ) lysis centrifugation system. The latter requires collection of blood culture sediment and plating on solid medium, both increasing labor and the risk of laboratory contamination during processing steps.1 In contrast, Myco/F Lytic culture does not require additional processing and, as a closed system, effectively eliminates the risk of laboratory-based contamination. Myco/F Lytic cultures are also monitored automatically and generally enable more rapid detection of mycobacteria and certain fungi. (2-4) However, Myco/F Lytic bottles are more expensive than manual methods. They also take a disproportionate amount of incubator space, since they require a 6-week incubation period, in contrast to 5 days for standard blood culture.

Mitigating against the need for optimal detection of all fungal pathogens by blood culture, alternative and more sensitive methods exist for detection of Cryptococcus neoformans, Aspergillus fumigatus, and other filamentous fungi; specifically, the cryptococcal antigen test, the Platelia galactomannan assay (Bio-Rad, Hercules, Calif) for invasive aspergillus, and the Histoplasma urinary antigen test. Furthermore, the (1-3)-[beta]-D-glucan assay detects a wide range of fungi, including Candida and certain molds, and provides a new, but as yet unproven, tool for identifying invasive fungal infection. (5) Additional serologic assays exist in either antibody- or antigen-detecting format for the diagnosis of dimorphic fungal infection.

Unlike for yeast, culture of most mycobacteria from blood requires specialized mycobacterial blood culture media (eg, Myco/F Lytic or analogous products made by other manufacturers) or less-automated alternatives (ie, collection in ISOLATOR tubes followed by plating on solid mycobacterial medium). Because our medical center serves a population with a low prevalence of disseminated active tuberculosis, most isolates represent atypical mycobacteria isolated from immunocompromised patients with very low CD4 cell counts. (6) Therefore, the use of Myco/F Lytic culture could potentially be targeted toward these patients.

Our hospital's computerized physician order entry system contains a number of screens for ordering laboratory tests grouped by specimen type. As the computerized physician order entry system blood test screen listed standard and Myco/F Lytic blood culture options consecutively, clinicians often ordered both types of blood cultures contemporaneously. This provided us with the opportunity to determine retrospectively the relative performance of standard and Myco/F Lytic blood culture in a clinical setting. Based on these data and a subgroup analysis of patients at risk for mycobacterial bloodstream infection, we developed an algorithm for use of the Myco/F Lytic bottle and describe its implementation and effectiveness.

MATERIALS AND METHODS

Patient data review was approved by our institutional review board. Positive standard and Myco/F Lytic blood cultures were identified using WHONET 5.3 software (7) (World Health Organization Collaborative Center for Surveillance of Antimicrobial Resistance, Boston, Mass).WHONET interrogates microbiology data downloaded from a laboratory information system based on user-defined parameters. For the purposes of this study, these parameters would include culture type (ie, blood culture), organism species or class (eg, fungi), and time period. Data were retrieved in Excel format (Microsoft, Redmond, Wash) to enable database analysis, and included specimen login date, culture bottle type, species, and patient medical record number to facilitate data parsing by episode of fungemia or mycobacteremia.

To assess whether either the Myco/F Lytic or standard bottles were superior for isolation of Candida species, we examined whether isolates were uniquely identified by either culture method. For this analysis, we reasoned that a standard and a Myco/F Lytic blood culture must be drawn within 24 hours of one another for a meaningful comparison (referred to as contemporaneous culture in "Results"). Over a greater time, fluctuating levels of organisms related to treatment or disease course would presumably make such a comparison inappropriate.

Mean time to detection was calculated for all positive blood cultures during the indicated period. Time to detection was determined by subtracting blood culture login time (ie, just before bottles are put into incubators) from the time positive bottles were flagged and Gram stain results entered into the laboratory information system. As the first positive bottle within a standard blood culture set represents the clinically actionable data point, only its time point was used in time to detection analysis, except for the comparison of anaerobic and aerobic detection times for Candida glabrata, as noted.

During this study, blood cultures were collected as part of normal clinical practice and incubated either for 5 days (Standard/ 10 Aerobic/F and Lytic/10 Anaerobic/F bottles) or 6 weeks (Myco/F Lytic bottle) on the BACTEC 9240 system. Yeasts were identified using the Vitek 2 YST card (NC) from bioMerieux (Durham, NC), the Vitek 2 ID-YST card (OC) or, rarely, by a reference laboratory (Mayo Clinic, Rochester, Minn, or Fungus Testing Laboratory, San Antonio, Tex).

JMP statistical software (SAS Institute, Cary, NC) was used to analyze data. Determination of statistical significance was performed using the nonparametric Wilcoxon rank sum test for time to detection determinations and the [chi square](2) test for categorical data. P < .05 was considered statistically significant.

RESULTS

During a 2-year period (2005-2006), our institution performed approximately 71 000 standard and 3400 Myco/F Lytic blood cultures. Approximately 0.2% and 1.2%, respectively, were positive for Candida species, representing 85 unique patient episodes of candidemia. Of these, 53 episodes were excluded from analysis, since they did not have standard and Myco/F Lytic cultures drawn contemporaneously. Of the 32 remaining patient episodes, 19 were identified by both standard and Myco/F Lytic cultures, 9 were positive by standard culture alone, and 4 were positive by Myco/F Lytic culture alone (Table). A potential trend toward enhanced detection by standard blood culture was not found to be statistically significant (P = .12). The average time to detection for the 2 types of bottles was also not statistically different (P = .24), with a mean of 1.8 [+ or -] 0.9 days for Myco/F Lytic and 2.0 [+ or -] 1.0 days for standard culture.

An extended analysis was also specifically performed for C glabrata from 2003 to 2006 because of reports of its decreased detection by standard blood culture media in simulated blood culture experiments. (8,9) Seventeen unique patient episodes of C glabrata bloodstream infection were identified, during which standard and Myco/F Lytic bottles were drawn contemporaneously. Of these, 7 patients' episodes were identified by both types of blood culture, 7 were identified by standard culture alone, and 3 were identified by Myco/F Lytic culture alone--a nonstatistically significant difference in detection rate between bottle types. To rule out the possibility that larger numbers of cultures drawn contemporaneously by either method might have biased sensitivity in its favor, the total number of positive and negative bottles drawn within 24 hours of positive cultures was added together. By this measure, there were a total of 19 positive and 15 negative standard culture sets, and 11 positive and 9 negative Myco/F Lytic cultures. Here again, neither method showed proportionally better detection (P = .85) within the power of this study. However, when examining time to detection for C glabrata, we did observe a statistically significant decrease (P = .01) in mean time to detection in Myco/F Lytic (1.8 [+ or -] 0.8 days) versus standard (2.6 [+ or -] 1.1 days) blood culture.

Importantly, we found among standard blood culture sets that C glabrata was isolated almost 2-fold more frequently (P < .001) in anaerobic than in aerobic bottles. Specifically, 35 cultures were positive in the anaerobic bottle alone, 18 sets were positive in both anaerobic and aerobic bottles, and 8 were positive in the aerobic bottle alone. Furthermore, when limiting time to detection comparisons to sets in which both aerobic and anaerobic bottles were positive, C glabrata grew significantly faster (P = .004) in anaerobic (2.3 [+ or -] 1.2 days) than aerobic (3.3 [+ or -] 0.9 days) bottles.

The incidence of non-Candida fungal infection in our study was limited to Cryptococcus and 2 isolates of Malassezia detected after subculture on olive oil. Because of the limited number of Candida neoformans isolates in any given year, this analysis was extended for an 8-year period from 1999 through 2006. During this 8-year period, there were 29 positive standard and 9 positive Myco/F Lytic cultures representing 10 patient episodes. A total of 7 episodes were detected in both bottle types; 1 episode was detected in a single Myco/F Lytic culture, but not in a second Myco/F Lytic and 2 standard blood culture sets drawn contemporaneously; and 2 episodes were detected by standard culture in the absence of Myco/F Lytic culture. Here again, evidence supporting superiority in detection by Myco/F Lytic culture was not found. In addition, with the exclusion of 1 outlier (11 days to positivity by Myco/F Lytic culture), the average time to detection was not statistically different (P = .34), with a mean of 3.0 [+ or -] 1.0 days for Myco/F Lytic and 3.2 [+ or -] 0.9 days for standard culture. Of note, all 10 patients also had positive cryptococcal antigen tests in blood and/or cerebral spinal fluid at the time of their positive blood cultures.

The next step in developing an algorithm to optimize use of Myco/F Lytic culture was to examine whether patients at risk for mycobacterial bloodstream infection could be reliably identified by laboratory and clinical criteria and thereby selectively cultured. Among 68 patient episodes of mycobacteremia identified by Myco/F Lytic culture during a 10-year period from 1997 through 2006, there were 59 patients with Mycobacterium avium complex; 6 with Mycobacterium tuberculosis complex; and 3 with Mycobacterium abscessus, Mycobacterium simiae, and coinfection with M avium complex and M kansasii, respectively. For the 59 patients with M avium complex who had a CD4 count determination within 1 month of a positive blood culture sample, the mean and median CD4 counts were 22/[mm.sup.3] and 8/[mm.sup.3], respectively, which are consistent with previous observations. (6) The 6 patients with M tuberculosis bloodstream infection had highly suppressed CD4 counts, leukemia/lymphoma, and/or M tuberculosis also isolated from other sites. As expected, the small number of rapidly growing mycobacteria isolates was also adequately detected using standard blood culture. Mycobacterium abscessus was detected by standard blood culture initially and 1 month later in 2 of 3 standard blood cultures drawn contemporaneously with a positive Myco/F Lytic culture. Additionally, an isolate of M mucogenicum was detected initially by standard blood culture, whereas follow-up Myco/F Lytic and standard cultures were both negative. Based on these observations, restriction ofMyco/ F Lytic culture to those with depressed CD4 counts, severe immunosuppression from hematologic malignancies, and risk factors for tuberculosis should provide the most targeted use of this culture modality.

Since standard blood culture appeared to adequately detect yeast, we formulated a new clinical algorithm for targeted use of Myco/F Lytic culture. Here, Myco/F Lytic culture was recommended primarily for detection of mycobacterial bloodstream infection in patients with CD4 counts lower than 100/[mm.sup.3], based on prior studies (4,6) and validated in our own patient population. We took a number of steps to facilitate adoption of this new algorithm. First, in the computerized physician order entry system, Myco/F Lytic culture was given a more specific name--that is, "Blood Culture for Dimorphic Fungi and Mycobacteria"--replacing the previous name, "Fungal/Mycobacterial Blood Culture." Second, we removed the Myco/F Lytic test from the blood test ordering screen. Instead, it was made available through a lookup function, where typing the names of mycobacteria or dimorphic fungi would pull up this test option. When a care provider then selects this test, a pop-up comment appears in bold, red letters explaining the new algorithm, providing a link to more detailed data, and listing preferred alternative testing options--that is, "For invasive Aspergillosis detection, consider galactomannan 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 total parenteral nutrition, order a Malassezia blood culture" for plating on olive oil-supplemented fungal media. Although we include the name "dimorphic fungi" in the new test name based on reported ability of the Myco/F Lytic medium to isolate Histoplasma capsulatum and potentially other dimorphic fungi as well, (10) note is made in the computerized physician order entry system comment that dimorphic fungi have never been isolated from Myco/F Lytic bottles in our hospital system. After reading the pop-up comment, providers then have the option of ordering or canceling the test. Lastly, education on the algorithm was provided to hospital staff by e-mail.

The Figure shows the dramatic decrease in ordering of Myco/F Lytic bottles after implementation of these changes. Normalizing data to the total number of standard blood cultures drawn during each month, and excluding data from April and May 2006 during implementation, there was an almost 3-fold reduction in usage (P = .002).

COMMENT

A retrospective review of culture data over an extended period suggests that standard blood culture was equivalent to Myco/F blood culture for detection of yeast. A potential trend toward enhanced detection of yeasts by standard blood culture was thought to reflect a larger blood volume generally inoculated into standard (20 mL per set) than Myco/F Lytic (5 mL) bottles, rather than a true superiority of standard media. Based on these data, an algorithm was implemented targeting use of the more resource intensive Myco/F Lytic culture to patients at risk for mycobactemia. The resulting sustained reduction in Myco/F Lytic usage led to a projected savings of approximately $12 000 to $14 000 per year based on cost of supplies and phlebotomist time, and enhanced capacity by freeing half a blood culture incubator.

Our ability to adequately detect C glabrata in standard culture contrasts with suboptimal detection reported previously in simulated blood culture experiments. (8,9) There are a number of potential reasons for this. First, and most likely, different standard media were used here (BACTEC Standard/10 Aerobic/F and Lytic/10 Anaerobic/F media) than in simulated blood culture studies (BACTEC PLUS Aerobic/F and Anaerobic/F). (8,9) The presence of the lytic agent saponin in both the anaerobic lytic media used here and in Myco/F Lytic media, where this yeast grows well, might explain the enhanced growth of C glabrata; alternatively, resin found in PLUS bottles used in simulated studies might be inhibitory. Second, laboratory-adapted organisms used in simulated studies may have lost their capacity for optimal growth in standard blood culture media. Lastly, candidemic patient blood, as a result of lower levels of inhibitory factors, such as complement (11) or immune cells, may foster growth of yeast more than healthy patient blood inoculated into simulated cultures.

More generally, the noninferiority of standard blood culture for detection of yeast may depend on the specific combination of aerobic and anaerobic media used, and in our study could only be achieved through use of both aerobic and anaerobic media. The optimal choice of blood culture media will clearly need to evolve with future advances in blood culture technology and an understanding of their impact on clinical outcome. Our study highlights the need for ongoing evaluation of new culture media in clinical practice, so that they can be used in the most effective way.

Accepted for publication June 11, 2008.

[GRAPHIC OMITTED]

References

(1.) Creger RJ, Weeman KE, Jacobs MR, et al. Lack of utility of the lysis-centrifugation blood culture method for detection of fungemia in immunocompromised cancer patients. J Clin Microbiol. 1998;36:290-293.

(2.) Martinez-Sanchez L, Ruiz-Serrano J, Bouza E, et al. Utility of the BACTEC Myco/F lytic medium for the detection of mycobacteria in blood. Diagn Microbiol Infect Dis. 2000;38:223-226.

(3.) Vetter E, Torgerson C, Feuker A, et al. Comparison of the BACTEC MYCO/ F Lytic bottle to the isolator tube, BACTEC Plus Aerobic F/bottle, and BACTEC Anaerobic Lytic/10 bottle and comparison of the BACTEC Plus Aerobic F/bottle to the Isolator tube for recovery of bacteria, mycobacteria, and fungi from blood. J Clin Microbiol. 2001;39:4380-4386.

(4.) Waite RT, Woods GL. Evaluation of BACTEC MYCO/F lytic medium for recovery of mycobacteria and fungi from blood. J Clin Microbiol. 1998;36:1176-1179.

(5.) Pickering JW, Sant HW, Bowles CAP, Roberts WL, Woods GL. Evaluation of a (1 [right arrow] 3)-[beta]-D-glucan assay for diagnosis of invasive fungal infections. J Clin Microbiol. 2005;43:5957-5962.

(6.) Nightingale SD, Byrd LT, Southern PM, Jockusch JD, Cal SX, Wynne BA. Incidence of Mycobacterium avium-intracellulare complex bacteremia in human immunodeficiency virus-positive patients. J Infect Dis. 1992;165:1082-1085.

(7.) O'Brien TF, Stelling JM. WHONET: removing obstacles to the full use of information about antimicrobial resistance. Diagn Microbiol Infect Dis. 1996;25: 162-168.

(8.) Horvath LL, Hospenthal DR, Murray CK, Dooley DP. Detection of simulated candidemia by the BACTEC 9240 system with plus aerobic/F and anaerobic/F blood culture bottles. J Clin Microbiol. 2003;41:4714-4717.

(9.) Horvath LL, George BJ, Murray CK, Harrison LS, Hospenthal DR. Direct comparison of the BACTEC 9240 and BacT/ALERT 3D automated blood culture systems for candida growth detection. J Clin Microbiol. 2004;42:115-118.

(10.) Fuller DD, Davis TE Jr, Denys GA, York MK. Evaluation of BACTEC MYCO/ F Lytic medium for recovery of mycobacteria, fungi, and bacteria from blood. J Clin Microbiol. 2001;39:2933-2936.

(11.) Tuite A, Mullick A, Gros P. Genetic analysis of innate immunity in resistance to Candida albicans. Genes Immun. 2004;5:576-587.

James E. Kirby, MD; Meghan Delaney, DO; Qinfang Qian, MD, PhD; Howard S. Gold, MD

From the Department of Pathology (Drs Kirby and Qian) and the Division of Infectious Diseases and Silverman Institute for Health Care Quality and Safety (Dr Gold), Beth Israel Deaconess Medical Center, Boston, Mass; and the Puget Sound Blood Center, Seattle, Wash (Dr Delaney).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: James E. Kirby, MD, Department of Pathology, Beth Israel Deaconess Medical Center, 330 Brookline Ave--YA309, Boston, MA 02215 (e-mail: jekirby@bidmc.harvard.edu).
Contemporaneous Cultures Positive for Candida
Species in Standard and/or Myco/F Lytic Bottles

 Standard Myco/F Lytic Both

C albicans 4 1 6
C parapsilosis 1 2 5
C glabrata 2 1 4
C tropicalis 1 0 3
C pararugosa 0 0 1
C kefyr 1 0 0
Total 9 4 19
COPYRIGHT 2009 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kirby, James E.; Delaney, Meghan; Qian, Qinfang; Gold, Howard S.
Publication:Archives of Pathology & Laboratory Medicine
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2009
Words:3181
Previous Article:Urine Mercury excretion following meso-dimercaptosuccinic acid challenge in fish eaters.
Next Article:In situ detection of inflammatory cytokines and apoptosis in pemphigus foliaceus patients.
Topics:

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters