Time to detection of positive BacT/Alert blood cultures and lack of need for routine subculture of 5- to 7-day negative cultures.

Consecutive BacT/Alert blood cultures which were instrument negative following a 7-day incubation were subcultured. Eighteen (0.2%) of 11,476 bottles had growth on subculture. Eleven of these eighteen isolates were considered contaminants on the basis of the identity of the organism and lack of other positive blood cultures from the same patient. In addition, analysis of time to instrument detection for approximately 2,900 positive blood cultures indicates that 5 or 6 days of incubation is sufficient for the routine detection of clinically significant organisms from BacT/Alert blood cultures. These data indicate that subculture of 5- to 7-day instrument-negative BacT/Alert blood culture bottles is not necessary.     

Growth and detection of filamentous fungi in the BacT/Alert blood culture system.

Little is known about the behaviour of filamentous fungi in most blood culture systems, despite their increasingly recognised role in infections of immunocompromised hosts. The ability of the BacT/Alert system (Organon Teknika, Durham, North Carolina, USA) to detect the growth of 19 such fungi was examined. Eleven species grew and were detected rapidly; two species did not grow. Six species grew slowly, and were generally only recovered with terminal subculture after prolonged incubation. The CO2 production graph for some of these fungi showed a slow but steady rise, insufficient to cause the apparatus to signal positive. These results show that the BacT/Alert system may miss some fungi, either because of no growth in the medium or undetected slow growth. The latter problem could be overcome by prolonged incubation and terminal subculture when fungal infection is considered likely. Alteration of the signalling mechanism might permit earlier detection of some slow growing fungi.     

Isolation of Kingella kingae from Synovial Fluids Using Four Commercial Blood Culture Bottles

 According to the literature, Kingella kingae may be an underdiagnosed cause of joint and bone infections in children. The use of the Bactec blood culture system for culture of joint fluids has dramatically improved the isolation of this fastidious bacterium. The aim of this study was to test the recovery rate and detection time of four commercial blood culture systems: three different BacT/Alert (Organon Teknika, USA) bottles and one Bactec (Becton Dickinson Microbiology Systems, USA) bottle, all inoculated with Kingella kingae strains mixed with pooled synovial fluids. For each strain the same inoculum and volume of synovial fluid was distributed into each of the four bottles. All 24 strains tested grew in the BacT/Alert Aerobic (100%) and the BacT/Alert Pedi-BacT (100%) bottles. Twenty-one strains grew in the BacT/Alert FAN aerobic (88%) bottle, and 15 strains grew in the Bactec Plus Aerobic F (63%) bottle, in both systems within 12 days (P<0.01). The Kingella kingae strains were first detected in the BacT/Alert Pedi-BacT bottles (P<0.001). The results were reproducible. The BacT/Alert blood culture bottles were superior to previously described blood culture systems in isolating Kingella kingae from synovial fluid, even with small inoculums and small volumes of synovial fluid.     

Detection of Brucella melitensis by the BacT/Alert automated system and Brucella broth culture

This study was conducted to evaluate the ability of the BacT/Alert automated blood culture system to detect Brucella spp. in comparison with traditional Brucella broth culture. Overall, 100 (50 bone marrow and 50 blood samples) paired cultures were obtained, and 59 were positive by at least one method. The Brucella broth culture method detected all 59 positive cultures (100%), and the BacT/Alert system detected 30 (50.8%) ( P  < 0.05). The mean detection times for B. melitensis were 4.5 days in the BacT/Alert system and 5 days in Brucella broth culture ( P  > 0.05). There is no significant difference between the two methods with respect to growth time of the microorganism, but Brucella broth culture is more sensitive than the BacT/Alert system.     

Controlled comparison of BacT/Alert MB system, manual Myco/F lytic procedure, and isolator 10 system for diagnosis of Mycobacterium tuberculosis Bacteremia

We compared the performance of the BacT/Alert MB system, that of the manual Bactec Myco/F Lytic procedure, and that of the Isolator 10 lysis-centrifugation system in the detection of Mycobacterium tuberculosis bacteremia. Mean times to detection were 16.4 days for BacT/Alert MB versus 20.0 days for Myco/F Lytic, 16.5 days for BacT/Alert MB versus 23.8 days for Isolator 10, and 21.1 days for Bactec Myco/F Lytic versus 22.7 days for Isolator 10. There were no significant differences in yields. The mean (range) magnitude of mycobacteremia was 30.0 (0.4, 90.0) CFU/ml and was correlated with the time to positivity in the BacT/Alert MB system (r = -0.4920). M. tuberculosis bacteremia was detected more rapidly in a continuously monitored liquid blood culture system, but the mean time to positivity exceeded 3 weeks.     

Direct comparison of the BACTEC 9240 and BacT/ALERT 3D automated blood culture systems for candida growth detection

A direct comparison of two automated blood culture systems was conducted to compare their ability to detect Candida growth. The systems evaluated were the BACTEC 9240 (Bactec) and BacT/ALERT 3D (BacT). The aerobic, anaerobic, and mycology media for each system were evaluated: Bactec Plus Aerobic/F, Plus Anaerobic/F, and Myco/F Lytic bottles, respectively, and BacT FA, SN, and MB bottles, respectively. Each blood culture bottle was inoculated with fresh blood from healthy donors. Fifty isolates of Candida spp. were used. The six different blood culture bottles were each inoculated with 1000 yeasts per bottle and then incubated in the corresponding automated system. The BacT detected growth of 90% (135 of 150) of Candida pathogens, while Bactec detected 66% (100 of 150). Growth was detected in all BacT and Bactec mycology bottles, all BacT aerobic bottles, and by terminal subculture of all bottles. Sixty-five of 300 (22%) bottles had no growth detected; 50 from the Bactec (5 aerobic and 45 anaerobic) and 15 from the BacT (all anaerobic). Terminal subculture of "negative" bottles demonstrated viable yeast growth from all 65 bottles, representing 65 false-negatives. The mean time to growth detection in the BacT system was 25.62 h while the Bactec was 27.30 h (P < 0.01). Both automated blood culture systems detected all episodes of simulated candidemia when specialized mycology media were used. However, when only standard aerobic and anaerobic media were used, the BacT performed better than the Bactec in overall growth detection, time to growth detection, and number of false-negatives.     

Comparison of BacT/Alert with Signal blood culture system.

The BacT/Alert (Organon Teknika Corp., Durham, N.C.) is an automated blood culture system. It is based on the detection of CO2 by means of a colorimetric sensor internally attached to the bottom of culture bottles. The aerobic and anaerobic media of this system were compared with one bottle of the Signal system (Oxoid Ltd., Hampshire, United Kingdom). At bedside, 20 ml of blood was drawn from each adult patient. The two BacT/Alert bottles were inoculated with 5 ml of blood each; the Signal bottle was inoculated with 10 ml. A total of 5,284 sets (2,483 patients; 2.1 cultures per patient) consisting of three bottles each were evaluated, of which 781 sets (14.8%) revealed microorganisms (n = 892); 642 of these were considered to be pathogenic. Significantly more (P < 0.0001) pathogens were isolated from the two BacT/Alert bottles together (n = 584) than from the single Signal bottle (n = 515). Escherichia coli (P = 0.007), gram-negative bacteria other than members of the family Enterobacteriaceae or Pseudomonas spp. (P = 0.006), and yeasts (P = 0.02) were isolated more often from both or either BacT/Alert bottle. Comparing the systems in terms of 388 different organisms per septic episode, the difference between BacT/Alert and Signal was significant for the total number of septicemia cases (P = 0.003). More contaminants grew in the BacT/Alert system (173 versus 116; P = 0.0001). False-positive indications were more frequent in the BacT/Alert system, 198 (3.7%) aerobic bottles and 57 (1.1%) anaerobic bottles, than in the Signal bottles, 24 (0.5%) bottles.(ABSTRACT TRUNCATED AT 250 WORDS)     

Evaluation of the BacT/Alert and VITAL blood culture systems for the diagnosis of bacteremia

Objective: To evaluate the detection of bacterial growth in the BacT/Alert (Organon Teknika) and VITAL (bioMérieux) automated blood culture systems.
Methods: In accordance with the protocol of study, 1021 blood sample pairs for culture were obtained from adult patients admitted to the Emergency Room and Intensive Care Unit.
Results: In total, 139 (13.6%) clinically significant blood cultures were detected, of which 79 (56.8%) were detected by both systems, 48 (34.5%) only by BacT/Alert and 12 (8.6%) only by VITAL (P <0.0001). The BacT/Alert system detected positive blood cultures more rapidly for all groups of microorganisms. The VITAL system showed six false-negative blood cultures, while the BacT/Alert system showed none ( P =0.03). There was no significant difference between the number of false-positive blood cultures detected by the two systems.
Conclusions: In our study, overall the BacT/Alert system achieved a better recovery of microorganisms than the VITAL system.     

Potential use of BacT/Alert automated blood culture system for antifungal susceptibility testing.

A simple, rapid susceptibility test is needed to determine the possible resistance of yeasts to commonly used antifungal agents. The BacT/Alert automated blood culture system was evaluated as one technology for development of such a test. Yeast nitrogen base was used as the growth medium, and amphotericin B was the test antifungal agent. Isolates of various Candida species, Torulopsis glabrata, Saccharomyces cerevisiae, and Cryptococcus neoformans were evaluated. The results suggest that detection of amphotericin B resistance of yeast isolates within 12 to 14 h after inoculation of test medium is possible.     

Rapid detection of methicillin-resistant staphylococci from blood culture bottles by using a multiplex PCR assay

Rapid detection and accurate identification of methicillin-resistant staphylococci are critical for the effective management of infections caused by these organisms. We describe a multiplex PCR-based assay for the direct detection of methicillin-resistant staphylococci from blood culture bottles (BacT/Alert; Organon-Teknika, Durham, N.C.). A simple lysis method followed by a multiplex PCR assay designed to detect the nuc, mecA, and bacterial 16S rRNA genes was performed. A total of 306 blood culture specimens were collected over a period of 10 months from June 1998 to April 1999, consisting of 236 blood cultures growing staphylococci (including 124 methicillin-resistant Staphylococcus spp.), 50 positive blood cultures which grew organisms other than staphylococci, and 20 blood cultures that were negative for bacterial and fungal pathogens after 5 days of incubation and terminal subculture. DNA extraction, PCR, and detection could be completed in 2.5 h. Of the positive blood cultures with staphylococci, the multiplex PCR assay had a sensitivity and specificity of 99.2% and 100%, respectively. Our results show that rapid, direct detection of methicillin-resistant staphylococci is possible, allowing clinicians to make prompt and effective decisions for the management of patients with staphylococcal bacteremia.     

Three days of incubation may be sufficient for routine blood cultures with BacT/Alert FAN blood culture bottles

BacT/Alert FAN blood culture bottles have been shown to enhance the recovery of bacteria and yeast from blood compared with standard BacT/Alert bottles. It is well established that standard BacT/Alert blood culture bottles require no more than 5 days of incubation for the detection of routine bacteria and yeast. It is less clear, however, whether FAN bottles also routinely require 5 days of incubation. To address this question, we recently reviewed the results of 17,887 blood culture sets collected in FAN blood culture bottles at Geisinger Medical Center. Of these cultures, 1,780 were positive for bacteria or yeast, yielding a total of 1,242 clinically significant isolates. The numbers of isolates recovered on days 1, 2, 3, 4, and 5 were as follows: (values in parentheses are percentages of total significant isolates): 877 (71%), 269 (22%), 65 (5%), 18 (1%) and, 13 (1%), respectively. In total, 97.5% of all clinically significant isolates were detected in the first 3 days of incubation. Of the 31 significant isolates detected on day 4 or 5 of incubation, 17 were detected in concurrent blood cultures within the first 3 days of incubation. Chart reviews were conducted for the 13 patients with the remaining 14 isolates detected on day 4 or 5 to determine whether therapy was changed due to this blood culture result. Therapy was changed for only 1 patient. These results suggest that it may not be necessary to routinely incubate FAN blood culture bottles for more than 3 days.     

Controlled Clinical Laboratory Comparison of Two Supplemented Aerobic and Anaerobic Media Used in Automated Blood Culture Systems To Detect Bloodstream Infections

A 20-ml blood sample was collected from adult patients with suspected bloodstream infections and distributed equally into the four volume-controlled bottles of a blood culture set consisting of aerobic and anaerobic BACTEC Plus/F bottles and aerobic and anaerobic BacT/Alert FAN bottles. All bottles were incubated in their respective instruments for a standard 5-day protocol or until the instruments signalled positivity. Samples in all bottles with negative results by these instruments were terminally subcultured. A total of 8,390 blood culture sets were obtained during the study period, of which 4,402 (52.5%) met the study criteria. Of these, 946 (21.5%) were positive either by instrument signal or by additional terminal subculture of all negative bottles and yielded growth of microorganisms. Five hundred eighty-nine (13.4%) blood culture sets were considered to have recovered 663 clinically significant organisms. When both the BACTEC and the BacT/Alert systems were used, 465 positive sets were detected; BACTEC alone detected 52 positive sets and BacT/Alert alone detected 72 (P = 0.09). No differences were found between the two systems in microbial recovery rate from blood cultures obtained from patients on antibiotic therapy. Significantly more members of the family Enterobacteriaceae (P < 0.01) were detected from patients without antimicrobial therapy by BacT/Alert than by BACTEC. The false-negative rates were 0.20% for BACTEC and 0.32% for BacT/Alert. A significantly higher false-positive rate was found for BACTEC (P < 0.0001). Both systems were comparable for the time to detection of microorganisms. However, gram-positive bacteria were detected faster by BACTEC and Enterobacteriaceae were detected faster on average by BacT/Alert. We concluded that both systems are comparable in their abilities to recover aerobic and anaerobic organisms from blood cultures and a terminal subculture might not be necessary for either of the two systems. The increased positivity rate when using an anaerobic bottle in a two-bottle blood culture set is due to the additional blood volume rather than to the use of an anaerobic medium.     

Comparison of the BacT/Alert PF pediatric FAN blood culture bottle with the standard pediatric blood culture bottle, the Pedi-BacT

The performance of the BacT/Alert PF (Organon-Teknika Corp., Durham, N.C.), a new nonvented pediatric FAN blood culture bottle, was compared to that of the original pediatric bottle, the Pedi-BacT, with matched aerobic cultures obtained from two separate facilities. A total of 244 clinically significant isolates were recovered from 4,015 compliant pairs. Among the positive cultures, 170 (70%) isolates were detected in both the BacT/Alert PF and the Pedi-BacT bottles, while 47 (19%) isolates were recovered in the BacT/Alert PF bottle only and 27 (11%) isolates were recovered in the Pedi-BacT bottle only. Although isolation of specific microorganisms was comparable for the two bottles, the total number of organisms recovered by the BacT/Alert PF was greater than that by the Pedi-BacT (P = 0.0272). In addition, more organisms were recovered by the BacT/Alert PF bottle from the blood of patients receiving antimicrobial therapy (P = 0.0180). Overall time to detection was similar for the two bottles; however, a significantly decreased mean time to detection was recorded for yeast from the BacT/Alert PF bottle (22.9 h; P = 0.0001) and staphylococci from the Pedi-BacT bottle (22.5 h; P = 0.0056). One false-negative culture and five false-positive cultures occurred with the Pedi-BacT bottle, compared to one false-positive culture with the BacT/Alert PF bottle. The BacT/Alert PF bottle is a reliable blood culture bottle for pediatric blood culture specimens and may offer improved recovery of microbes from patients on antimicrobial therapy. The use of the nonvented bottle will both facilitate bottle processing and decrease expenditures for materials due to the elimination of the venting needles required for the original vented bottles.     

Controlled evaluation of 5 versus 10 milliliters of blood cultured in aerobic BacT/Alert blood culture bottles.

Bottles developed for use in the BacT/Alert automated blood culture system (Organon Teknika Corp., Durham, N.C.) can accept up to 10 ml of blood without falling below a 1:5 ratio of blood to broth. We compared the yield and speed of detection of microorganisms in 13,128 adequately filled, paired, aerobic bottles inoculated with 5 versus 10 ml of blood at three university hospitals. A total of 798 microorganisms causing sepsis grew in one or both bottles. The overall recovery of microorganisms from 10-ml samples exceeded that from 5-ml samples (P < 0.001); the increased yield attributed to the additional 5 ml in the samples was 7.2%. The increased yield from 10-ml inocula was most marked for Escherichia coli (P < 0.01) and other members of the family Enterobacteriaceae (P < 0.001). Ten-milliliter samples did not yield more gram-positive bacteria, nonfermentative gram-negative rods, or yeasts. When both bottles were positive, the bottles inoculated with 10 ml of blood showed growth sooner (P < 0.001). Earlier detection with 10-ml inocula was especially notable for coagulase-negative staphylococci (P < 0.001), streptococci (P < 0.001), E. coli (P < 0.025), and other members of the family Enterobacteriaceae (P < 0.025). We conclude that an increase in the volume of blood inoculated into BacT/Alert aerobic blood culture bottles from 5 to 10 ml will increase the overall yield and the speed of detection of clinically important blood pathogens.     

Comparison of BacT/Alert and BACTEC NR 860 blood culture systems in a laboratory not continuously staffed

Objective: To evaluate the performance of continuously working blood culture systems in a discontinuous laboratory system.
Method: The systems used were BacT/Alert (Organon Teknika Corp., Durham, NC) and BACTEC NR 860 (Becton Dickinson Diagnostic Instruments, Sparks, Md) in a comparison in a laboratory staffed 8 1/2 h on Mondays to Fridays and 4 1/2 h on Saturdays. Blood culture bottles (BacT/Alert aerobic and anaerobic, BACTEC NR 26 A and NR 27 A) were received thrice daily.
Results: From 1824 pairs of blood culture vials, 110 clinically significant microorganisms were recovered by both BACTEC and BacT/Alert, 43 by BACTEC alone, and 33 by BacT/Alert alone. The differences between the systems in total recovery and in recovery of individual species were not statistically significant. The average detection times were 13.36 h for BACTEC and 13.93 h for BacT/Alert ( P >0.1). These times represent only 35.6% (BACTEC) and 32.6% (BacT/Alert) of the total timespans from collection of blood to informing the ward of a positive result ( t _crd, clinically relevant detection time). If 24 h per day blood culture processing conditions and continuous transport of vials to the laboratory had been available, these percentages would have risen to 87% (BACTEC) and 87.5% (BacT/Alert). Under such 'ideal' conditions, t _trd could have been reduced by 22.16 h using BACTEC and by 26.81 h using BacT/Alert. The BacT/Alert system showed more false-positive results than the BACTEC system (80 (4.39%) versus 23 (1.26%), P <0.001).
Conclusions: No time benefit for detection of positive blood cultures is gained with continuously measuring systems, if loading and processing of vials is organized discontinuously, as in our laboratory.     

Comparison of the two blood culture systems, Bactec 9240 and BacT/Alert 3D, in the detection of Candida spp. and bacteria with polymicrobial sepsis

The purpose of this investigation was to evaluate the performance of the Bactec 9240 and BacT/Alert 3D blood culture systems in the detection of Candida spp. and bacteria in simulated polymicrobial sepsis models. A total of 28 clinical isolates of Escherichia coli, Staphylococcus aureus, Candida albicans, and Candida glabrata were studied. Five polymicrobial models of C. albicans + S. aureus, C. albicans + E. coli, C. glabrata + S. aureus, C. glabrata + E. coli, and C. albicans + C. glabrata were prepared. Each combination was inoculated in five different blood culture vials. The two systems were compared for culture positivity and time to detection (TTD). Twenty-four mixed cultures with a yeast and a bacteria were tested. Bactec Mycosis vials could detect yeasts in all 24 cultures. The aerobic vials from both Bactec and BacT/Alert could detect both yeasts and bacteria in 22/24 (91.66?%) cultures. Bactec Plus Anaerobic/F and BacT/Alert FN vials could detect both microorganisms in 19/24 (79.16?%) and 4/24 (16.67?%) vials, respectively. Seven polymicrobial sepsis models with C. albicans + C. glabrata were also tested. Mycosis vials could detect both yeasts in 7/7 mixed cultures. The aerobic vials from Bactec and BacT/Alert could detect both yeasts in 3/7 and 2/7 mixed cultures, respectively. Bactec Plus Aerobic/F had a shorter TTD compared to BacT/Alert FA and Bactec Plus Anaerobic/F vials (p?<?0.0001 and p?<?0.01, respectively). The present study shows that the Bactec and BacT/Alert systems have different characteristics in the detection of yeasts and bacteria with polymicrobial sepsis.     

Isolation of Mycoplasma hominis using the BACTEC 9000 series blood culture system.

Mycoplasma hominis has been implicated as an important cause of septicaemia. There have been reported variances in the ability of blood culture systems to support the growth of this organism. In this study the ability of the BACTEC 9000 series automated system to grow and detect M hominis was assessed. Three of five wild M hominis strains grew in the BACTEC Anaerobic Plus/F medium but growth was not flagged by the detection mechanism of the system. It is recommended that users of the BACTEC 9000 series should use a seven day protocol and perform terminal subculture for suspected cases of M hominis septicaemia.     

Comparison of BACTEC PLUS blood culture media to BacT/Alert FA blood culture media for detection of bacterial pathogens in samples containing therapeutic levels of antibiotics

Blood culture bottles with antimicrobial removal systems are recommended for patients who develop fever while on antibiotics. This study compared the ability of Becton Dickinson (Sparks, MD) BACTEC PLUS bottles and bioMerieux (Durham, NC) BacT/Alert FA bottles to effectively remove vancomycin, cefoxitin, ceftriaxone, cefepime, piperacillin-tazobactam, ampicillin, oxacillin, gentamicin, and a combination of gentamicin/penicillin, thus allowing bacterial pathogens to grow. Each bottle was spiked with 10 ml of human blood, antibiotic, and strains of organisms susceptible to the antibiotic evaluated. The organisms used were type strains and clinical isolates of Staphylococcus aureus (methicillin susceptible and resistant), Streptococcus pneumoniae, a viridans streptococcus, Enterococcus faecalis, Enterococcus faecium, Streptococcus agalactiae, Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa. Testing was completed in triplicate, using 10 to 100 CFU/ml of organisms with various concentrations of each antibiotic. Two rounds of testing were completed per antibiotic/organism combination. Bottles were mixed and loaded onto their respective instruments as per the manufacturer's instructions. Antimicrobial removal was evaluated on the basis of time to detection of organism growth, for up to 5 days of incubation. Overall, the BacT/Alert FA system recovered 25.1% of strains from test bottles and 96.9% of strains from growth control bottles (no antibiotic added), and the BACTEC PLUS system recovered 95.1% of strains from test bottles and 100% of strains from growth control bottles. Both systems performed well in the detection of Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa in the presence of gentamicin. In the presence of ceftriaxone, neither system was able to recover Streptococcus pneumoniae. The ability to remove vancomycin and cefoxitin was also determined by measuring antibiotic levels remaining in bottles after 1 h of incubation. The results demonstrated remaining levels of 72 to 90% of vancomycin and 71 to 72% of cefoxitin in the BacT/Alert system. For the BACTEC system, remaining levels were 0 to 30% of vancomycin and 0% of cefoxitin. Under these simulated conditions, the BACTEC PLUS system was superior to the BacT/Alert FA system in recovering gram-positive and gram-negative bacterial pathogens in the presence of beta-lactam antibiotics, gentamicin/penicillin, and vancomycin.     

Comparative analysis of simulated candidemia using two different blood culture systems and the rapid identification of Candida albicans

The goal of this study was to determine the time to detection of Candida species isolates using the two most commonly used automated blood culture systems, and to evaluate rapid, widely available methods for the presumptive identification of C. albicans. Candidemia models of eight commonly detected Candida species were prepared using ATCC standards. The times to detection were evaluated using the BACTEC 9240 (Becton Dickinson) and BacT/Alert 3D (bioMerieux) automated blood culture systems. The presence of pseudohyphae clusters was examined by Gram staining and wet preparation. Germ tube tests were performed directly from blood culture bottles. All samples were cultured on blood agar plates and macroscopically examined for the presence of an irregular margin (spiking). Most Candida species (6/8) except C. glabrata and C. krusei grew more rapidly in aerobic than in anaerobic conditions. Clusters of pseudohyphae were observed in cultures of C. albicans and C. tropicalis. All culture bottles positive for C. albicans were positive by the germ tube test and macroscopically showed 'spiking.' Aerobic and anaerobic blood culture systems can effectively detect candidemia. Furthermore, the direct germ tube test may be the most useful available morphological presumptive identification method for C. albicans.     

Assessment of routine use of an anaerobic bottle in a three-component, high-volume blood culture system.

The relative value of routine anaerobic blood culture for recovery of organisms and identification of episodes of bloodstream infection was assessed in a three-component, high-volume blood culture system which employs aerobic and anaerobic bottles of BacT/Alert (Organon-Teknika, Durham, N.C.) and aerobic cultures of Isolator (Wampole Laboratories, Cranbury, N.J.). The results of 5,595 blood culture sets from patients with suspected bloodstream infection were analyzed. Compared with either the aerobic BacT/Alert bottle or aerobic culture of Isolator, the BacT/Alert anaerobic bottle recovered significantly fewer isolates (242 versus 294, P < 0.05; 242 versus 298, P < 0.05) but did not detect significantly fewer episodes of bloodstream infection (141 versus 157, P > 0.05; 141 versus 147, P > 0.05). The BacT/Alert anaerobic bottle recovered significantly more isolates of obligately anaerobic bacteria (16 versus 4, P < 0.05; 16 versus 0, P < 0.05) and detected significantly more episodes of bloodstream infection caused by obligately anaerobic bacteria (10 versus 3, P < 0.05; 10 versus 0, P < 0.05) than either the aerobic bottle of BacT/Alert or the aerobic culture of Isolator. The combination of the BacT/Alert anaerobic bottle and the aerobic culture of Isolator recovered as may isolates (374 versus 377) and detected as many episodes of bloodstream infection (194 versus 191) as the combination of the aerobic bottle of BacT/Alert and the aerobic culture of Isolator, and both of these combinations identified at least 8% more isolates and detected at least 3% more bloodstream infections than the combination of the BacT/Alert aerobic and anaerobic bottles. Further analysis of the data revealed that the utility of the BacT/Alert anaerobic bottle, especially when combined with the aerobic culture of Isolator, resulted from not only enhanced recovery of obligately anaerobic bacteria but also effective recovery of facultatively anaerobic bacteria. These results demonstrate the utility of the anaerobic BacT/Alert bottle for detecting bloodstream infection caused by either facultatively anaerobic bacteria or obligately anaerobic bacteria and support the routine inclusion of anaerobic blood culture in the three-component blood culture system used in our hospital.