Abstract
During the period 2006 and 2007, all blood cultures required by four units at high infective risk and most of those required by other units of the University Hospital of Palermo, Palermo, Italy were performed using a Bactec 9120 automated blood culture system with a complete set of Plus Aerobic/F, Plus Anaerobic/F, and Mycosis IC/F bottles. The aim of the study was to enable the authors to gain firsthand experience of the culture potentialities of the three different media, to obtain information regarding the overall and specific recovery of bacteria and yeasts from blood cultures in the hospital, and to reach a decision as to whether and when to utilize anaerobic and fungal bottles. Although very few bloodstream infections (1.8%) were associated with obligate anaerobes, the traditional routine use of anaerobic bottles was confirmed because of their usefulness, not only in the detection of anaerobes, but also in that of gram-positive cocci and fermentative gram-negative bacilli. In this study, Mycosis IC/F bottles detected 77.4% of all the yeast isolates, 87.0% of yeasts belonging to the species Candida albicans, and 45.7% of nonfermentative gram-negative bacilli resistant to chloramphenicol and tobramycin. In order to improve the diagnosis of fungemia in high-risk patients, the additional routine use of fungal bottles was suggested when, as occurred in the intensive-care unit and in the hematology unit of the University Hospital of Palermo, high percentages of bloodstream infections are associated with yeasts, and/or antibiotic-resistant bacteria and/or multiple bacterial isolates capable of inhibiting yeast growth in aerobic bottles.
Despite recent developments in microbiological biomolecular techniques, blood culture remains the most practical and reliable method in the diagnosis of bloodstream infections and one of the most important tools in the clinical microbiology laboratory, taking into account the role played by systemic infections as a major cause of mortality and morbidity, especially within the hospital environment. Various continuously monitoring blood culture systems are extensively used in order to increase the rapidity of isolation and identification of microorganisms in blood samples, and paired aerobic and anaerobic blood culture bottles are commonly used in order to obtain better overall recovery of microorganisms, as well as the recovery of specific types of bacteria. The traditional aerobic and anaerobic bacteriological media also detect episodes of candidemia, although the sensitivity in this case is estimated to be only 50% (12). The efficiency of yeast detection in blood samples can be increased by terminal subcultures, but the practice is time-consuming and the risk of underestimating yeast infections remains high. This is unacceptable if we take into account the expanding role of yeasts as agents of nosocomial infections, particularly in intensive-care units (ICUs) (4, 18, 23), and candidemia-associated mortality (40%), the highest of all hospital bloodstream infections (4). In order to facilitate faster and more efficient isolation of fungi from blood, the Bactec automated blood culture system (Becton Dickinson Diagnostic Systems, Sparks, MD) has media specifically formulated for this purpose. In addition to the traditional aerobic-anaerobic bacteriological set, their use has been recommended when blood culture is required for patients at risk of fungal infections (7, 16) and when a concomitant bloodstream bacterial infection is suspected of inhibiting yeast detection (16).
Nevertheless, fungal media are not routinely employed by many hospitals, as they require additional blood from patients, an increased amount of incubator space, and the additional costs of processing and purchasing bottles (10). Moreover, even the routine use of anaerobic bacteriological bottles in blood culture has been (21), and still is (13), a subject of discussion because of the reportedly declining rates of anaerobic bloodstream infections (2) and the possibility of recognizing them clinically and treating affected patients empirically for such infections (17, 19, 22).
The purpose of this study was to gain firsthand experience and, in particular, as suggested by Riley et al. (20), up-to-date knowledge of the overall recovery of bacteria and yeasts from blood cultures at our University Hospital of Palermo. For this purpose, we used a Bactec 9120 automated blood culture system with a complete set of aerobic and anaerobic bacteriological media and fungal medium: Plus Aerobic/F, Plus Anaerobic/F, and Mycosis IC/F bottles, respectively. With this set of bottles, we performed all the blood cultures required by four high-risk units—an ICU, a hematology unit (HU), an infectious-disease unit (IDU), and the cardio-surgery unit (CSU)— and most of those required by other units for the period 2006-2007. Only blood cultures performed with the whole set of bottles were considered in the study.
MATERIALS AND METHODS
Plus Aerobic/F medium is an enriched soybean-casein digest broth with CO2 and an incorporated cation-exchange resin that is effective in removing β-lactam antibiotics, gentamicin/penicillin, and vancomycin (6); Plus Anaerobic/F medium is a prereduced enriched soybean-casein digest broth with added antibiotic-removing resin that is dispensed with CO2 and N2; and Mycosis IC/F medium is an enriched brain-heart infusion broth with CO2 that was specifically formulated for the isolation of yeasts and fungi from blood, containing chloramphenicol and tobramycin to inhibit bacterial growth and a lysis agent, saponin.
All bottles were inoculated, directly at the patient's bedside, with 8 to 10 ml each of fresh, whole blood; sent to our microbiology laboratory, where collected volumes for each bottle were verified and confirmed; and placed as soon as possible in a Bactec 9120 automated blood culture system. Specimens were incubated at 35°C with continuous agitation, and a fluorescence technology was used in order to evaluate the quantity and rate of CO2 production, as these are indicative of microbial growth. The protocol duration of incubation was 120 h for Plus Aerobic/F and Plus Anaerobic/F bottles and 160 h for Mycosis IC/F bottles.
Whenever there was any sign of microbial growth, 0.1-ml aliquots were withdrawn from each positive bottle, plated on solid medium, incubated at 35°C, and read after 24 to 48 h. Isolated organisms were identified with the BD Phoenix Automated Microbiology System (Becton Dickinson, Sparks, MD); in some cases, yeast identification was also performed by conventional microbiological procedures, in addition to the Mycotube method (BBL Diagnostics). Only bottles that proved to be positive with the use of the instrument were subcultured. No blind subcultures were performed on any bottles. The data recorded included which and how many microbial species were involved in each bloodstream infection, the numbers and the units of the infected patients, and whether microbial growth from blood cultures occurred in aerobic and anaerobic bacterial and/or fungal bottles.
No particular attempt was made to determine the clinical significance of any isolate, but isolates of coagulase-negative staphylococci, Bacillus spp., viridans group streptococci, Propionibacterium spp., and aerobic diphtheroids obtained from only one set of bottles were not considered in the comparative analyses, as they were judged to be clinically nonsignificant.
The role of microbial species in bloodstream infections in patients from different units and the positive rates of microbial growth in the different blood culture bottles were compared using the chi-square or Fisher's exact test when appropriate.
RESULTS
A total of 1,253 blood culture sets (3,759 bottles) were processed in our laboratory over the 2-year interval considered in this study. From 352 of these sets, 819 bacterial and yeast isolates were recovered, including 112 judged to be clinically nonsignificant.
The 707 clinically significant isolates were derived from frequently repeated blood cultures obtained from 221 patients with suspected bloodstream infections at the University Hospital of Palermo: 36 patients were from the ICU, 62 from the HU, 13 from the IDU, 28 from the CSU, and 82 from all other hospital units. Three hundred and two bacterial and yeast strains belonging to different species were considered potentially responsible for the bloodstream infections in the 221 patients: 52, 116, 13, 34, and 87 strains for patients from the ICU, HU, IDU, CSU, and other units, respectively. A great many patients from the HU were immunocompromised, and many polymicrobic growths were observed in the blood cultures from these patients.
When patients from all the units were taken into account, gram-positive cocci, gram-negative bacilli, and yeasts were involved in the infections of 58.4%, 62.40%, and 12.7%, respectively. Staphylococcus aureus (18.1%), Escherichia coli (15.4%), Pseudomonas aeruginosa (10.4), and Staphylococcus epidermidis, among the coagulase-negative staphylococci (8.1%), were the most frequently identified species. In comparison with results from all other units, significantly (P < 0.01) higher percentages of involvement of gram-positive cocci were observed in patients from the HU and CSU, of fermentative gram-negative bacilli in patients from the HU, of nonfermentative gram-negative bacilli in patients from the ICU and HU, and of yeasts in patients from the ICU. Candida albicans, prevailing in the ICU, was isolated from 25% of patients in the unit; other Candida species and Tricosporon cutaneum were the only yeasts isolated in the HU. Anaerobes were involved in only four infections: Bacteroides fragilis in one patient from one of the other units (a medical unit) and Clostridium perfringens in one patient from the HU and in two patients from two other units (one a medical and one a general surgery unit). Probably clinically significant Corynebacterium spp. were isolated from two patients in the HU.
With regard to detection by the Bactec bottles of the 707 bacterial and yeast strains obtained from the 221 patients with suspected bloodstream infections, it was observed that 72 strains (10.2%) grew in all three bottles, 299 (42.3%) grew in both the Plus Aerobic/F and Plus Anaerobic/F bottles, 75 (10.6%) grew in both the Plus Aerobic/F and Mycosis IC/F bottles, 5 (0.7%) grew in both in the Plus Anaerobic/F and Mycosis IC/F bottles, 154 (21.8%) grew only in the Plus Aerobic/F bottle, 65 (9.2%) grew only in the Plus Anaerobic/F bottle, and 37 (5.2%) grew only in the Mycosis IC/F bottle. If the last two types of bottles had not been used in the assays, 17 and 12 patients, respectively, would have been classified as blood culture negative.
Table 1 shows the whole numbers and percentages of strains belonging to each microbial species detected by the single bacterial aerobic, bacterial anaerobic, and fungal media. Table 2 shows the numbers and percentages of strains belonging to each microbial species that grew only in the Plus Aerobic/F bottles or that were instead detected either in the Plus Anaerobic/F or Mycosis IC/F bottles; the percentages were always calculated by comparison with the total number of isolates of each species.
TABLE 1.
Growth detection in the Bactec 9120 automated blood culture system
| Microorganism(s) | No. of isolates | No. (%) of isolates from Bactec bottles
|
||
|---|---|---|---|---|
| Plus Aerobic/F | Plus Anaerobic/F | Mycosis IC/F | ||
| Gram-positive cocci | 335 | 299 (89.3) | 256 (76.4) | 51 (15.2) |
| S. aureus | 66 | 63 (95.5) | 59 (89.4) | 1 (1.5) |
| Coagulase-negative staphylococcia | 170 | 154 (90.6) | 111 (65.3) | 23 (13.5) |
| Enterococcib | 88 | 75 (85.2) | 76 (86.4) | 27 (30.7) |
| Other streptococcic | 11 | 7 (63.6) | 10 (90.9) | 0 (0.0) |
| Corynebacterium spp. | 4 | 4 (100) | 4 (100) | 0 (0.0) |
| Gram-negative bacilli | 301 | 253 (84.1) | 171 (56.8) | 90 (29.9) |
| Fermenters | 174 | 144 (82.8) | 137 (78.7) | 32 (18.4) |
| E. coli | 70 | 61 (87.1) | 53 (75.7) | 9 (12.9) |
| Enterobacter spp.d | 53 | 40 (75.5) | 40 (75.5) | 19 (35.8) |
| Other Enterobacteriaceaee | 51 | 43 (84.3) | 44 (86.3) | 4 (7.8) |
| Nonfermenters | 127 | 109 (85.8) | 34 (26.8) | 58 (45.7) |
| Acinetobacter spp.f | 24 | 20 (83.3) | 10 (41.7) | 9 (37.5) |
| P. aeruginosa | 54 | 47 (87.0) | 9 (16.7) | 34 (63.0) |
| Stenotrophomonas maltophilia | 21 | 17 (81.0) | 5 (23.8) | 6 (28.6) |
| Other gram-negative bacillig | 28 | 25 (89.3) | 10 (35.7) | 9 (32.1) |
| Anaerobesh | 5 | 0 (0.0) | 5 (100) | 0 (0.0) |
| Yeasts | 62 | 44 (71.0) | 5 (8.1) | 48 (77.4) |
| C. albicans | 23 | 13 (56.5) | 2 (8.7) | 20 (87.0) |
| Other Candida spp.i | 29 | 23 (79.3) | 3 (10.3) | 24 (82.8) |
| T. cutaneum | 10 | 8 (80.0) | 0 (0.0) | 4 (40.0) |
| Total | 707 | 600 (84.9) | 441 (62.4) | 189 (25.7) |
Includes 58 S.epidermidis, 24 Staphylococcus hominis, 19 Staphylococcus haemolyticus, and 69 coaglulase-negative Staphylococcus sp. isolates.
Includes 32 Enterococcus faecium, 30 Enterococcus. faecalis, and 26 Enterococcus sp. isolates.
Includes two Streptococcus pyogenes, two Streptococcus pneumoniae, two Streptococcus bovis, two Streptococcus intermedius, and three Streptococcus group C isolates.
Includes 48 Enterobacter cloacae, 3 Enterobacter aerogenes, and 2 Enterobacter spp.
Includes 8 Salmonella enterica serovar Typhi, 3 S. enterica serovar Paratyphi A, 8 S. enterica serovar Enteritidis, 17 Klebsiella pneumoniae, 3 Klebsiella oxytoca, 6 Proteus vulgaris, 5 Serratia marcescens, and 1 Serratiaa plymuthica isolates.
Includes 8 Acinetobacter baumannii, 3 Acinetobacterlwoffii, 1 Acinetobacter haemolyticus, and 12 Acinetobacter spp.
Includes 1 Pseudomonas putida, 14 Pseudomonas spp., 21 Stenotrophomonas maltophilia, 9 Achromobacter spp., and 4 Ochrobactrum anthropi isolates.
Includes two Bacteroides fragilis and three Clostridium perfrigens isolates.
Includes 4 C. tropicalis, 3 Candida guilliermondi, 1 Candida glabrata, 1 Candida lusitaniae, 1 Candida parapsilosis, and 19 Candida sp. isolates.
TABLE 2.
Microorganisms grown only in Plus Aerobic/F bottles or detected only by other Bactec media
| Microorganism(s) | No. of isolates | No. (%) of isolates from Bactec bottles
|
||
|---|---|---|---|---|
| Plus Aerobic/F alone | Plus Anaerobic/F alone | Mycosis IC/F alone | ||
| Gram-positive cocci | 335 | 65 (19.4) | 30 (9.0) | 5 (1.5) |
| S. aureus | 66 | 7 (10.6) | 3 (4.5) | 0 (0.0) |
| Coagulase-negative staphylococci | 170 | 49 (28.8) | 11 (6.5) | 4a (2.4) |
| Enterococci | 88 | 8 (9.1) | 12 (13.6) | 1b (1.1) |
| Other streptococci | 11 | 1 (9.1) | 4 (36.4) | 0 (0.0) |
| Corynebacterium spp. | 4 | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Gram-negative bacilli | 301 | 78 (25.9) | 29 (9.6) | 15 (5.0) |
| Fermenters | 174 | 30 (17.2) | 23 (13.2) | 4 (2.3) |
| E. coli | 70 | 16 (22.9) | 7 (10.0) | 1 (1.4) |
| Enterobacter spp. | 53 | 7 (13.2) | 8 (15.1) | 3c (5.7) |
| Other Enterobacteriaceae | 51 | 7 (13.7) | 8 (15.7) | 0 (0.0) |
| Nonfermenters | 127 | 48 (37.8) | 6 (4.7) | 11 (8.7) |
| Acinetobacter spp. | 24 | 7 (29.2) | 2 (8.3) | 2 (8.3) |
| P. aeruginosa | 54 | 16 (29.6) | 2 (3.7) | 4 (7.4) |
| Stenotrophomonas maltophilia | 21 | 13 (61.9) | 1 (4.8) | 3 (14.3) |
| Other gram-negative bacilli | 28 | 12 (42.9) | 1 (3.6) | 2d (7.1) |
| Anaerobes | 5 | 0 (0.0) | 5 (100) | 0 (0.0) |
| Yeasts | 62 | 11 (17.7) | 1 (1.6) | 17 (27.4) |
| C. albicans | 23 | 2 (8.7) | 0 (0.0) | 10 (43.5) |
| Other Candida spp. | 29 | 3e (10.3) | 1f (3.4) | 5g (17.2) |
| T. cutaneum | 10 | 6 (60.0) | 0 (0.0) | 2 (20.0) |
| Total | 707 | 154 (21.8) | 65 (9.2) | 37 (5.2) |
Includes two taphylococcus haemolyticus, one Staphylococcus epidermidis, and one coagulase-negative Staphylococcus sp. isolates.
Includes one Enterococcus faecalis isolate.
Includes three Enterobacter cloacae isolates.
Includes one Pseudomonas sp. and one Achromobacter sp. isolates.
Includes three Candida sp. isolates.
Includes one C. tropicalis isolate.
Includes one Candida lusitaniae, one C. tropicalis, and three Candida sp. isolates.
It is clear that Plus Anaerobic/F and, in particular, Mycosis IC/F bottles detected significantly (P < 0.01) lower numbers of strains than Plus Aerobic/F bottles. Plus Aerobic/F bottles detected high percentages of strains belonging to all species except obligate anaerobes. Plus Anaerobic/F bottles detected obligate anaerobes and other strains mainly belonging to gram-positive cocci, Corynebacterium spp., and fermentative gram-negative bacilli. Finally, Mycosis IC/F bottles mainly detected yeasts and antibiotic-resistant bacteria like P. aeruginosa, other Pseudomonadaceae, Acinetobacter spp., Enterobacter spp., or enterococci.
The strains growing only in Plus Aerobic/F bottles, Plus Anaerobic/F bottles, or Mycosis IC/F bottles mainly belonged to coagulase-negative staphylococci and nonfermentative gram-negative bacilli; to fermentative gram-negative bacilli, enterococci, and other streptococci; and to yeasts and nonfermentative gram-negative bacilli, respectively. Yeasts, detected by Mycosis IC/F and Plus Aerobic/F bottles at the nonsignificantly different percentages of 77.4% and 71.0% were detected by Plus Anaerobic/F bottles at a very low percentage of 8.1%. Eleven yeast strains grew only in Plus Aerobic/F bottles; 1 (a Candida tropicalis strain) grew in a Plus Anaerobic F bottle; and 17, mainly C. albicans strains, grew in Mycosis IC/F bottles. Six of the last 17 strains (4 C. albicans, 1 C. tropicalis, and 1 T. cutaneum) were obtained in the course of blood cultures that also exhibited bacterial growth in the paired Plus Aerobic/F bottles. When considered alone, C. albicans strains were isolated more frequently (P = 0.047) from Mycosis IC/F bottles than from Plus Aerobic/F bottles. Mean detection times for yeasts were significantly shorter with the Mycosis IC/F medium than with the Plus Aerobic/F medium (29.19 ± 16.90 h versus 38.01 ± 16.61 h; P < 0.001).
DISCUSSION
According to the characteristics of the different culture media, Plus Aerobic/F bottles detected aerobic, facultatively anaerobic, and aerotolerant bacteria and yeasts; Plus Anaerobic/F bottles detected obligate anaerobes and strains mainly belonging to facultatively anaerobic and aerotolerant species; and Mycosis IC/F bottles detected mainly yeasts and bacteria resistant to chloramphenicol and tobramycin. Moreover, due to the differences in the medium compositions and the additional amounts of inoculated blood, Plus Anaerobic/F and Mycosis IC/F bottles increased the number of clinically relevant isolates by about 10% and 6%, respectively, compared to Plus Aerobic/F bottles. The whole detection rates were 84.9% by Plus Aerobic/F bottles, 62.4% by Plus Anaerobic/F bottles, and 26.7% by Mycosis IC/F bottles. Obligate anaerobes grew only in Plus Anaerobic bottles.
Yeasts were detected by Mycosis IC/F bottles at significantly higher rates than by Plus Aerobic/F bottles if they belonged to the C. albicans species or at more comparable rates if they belonged to all other species. A few yeasts, less than 10%, were also detected by Plus Anaerobic/F bottles. Although with some differences in the detection rates for C. albicans and non-albicans Candida isolates, analogous evidence of the advantageous use of specialized mycology media for the diagnosis of fungemia were obtained by Horvath et al. (11) experimentally and by Meyer et al. (16) during retrospective analyses of numerous blood culture pairs composed of one Mycosis IC/F bottle and one Plus Aerobic/F bottle.
In some blood cultures, yeasts grew only in Plus Aerobic/F bottles or in Mycosis IC/F bottles, probably due to low fungemia levels; in others, they only grew in Mycosis IC/F bottles, but bacterial growth was also observed in the paired Plus Aerobic/F bottles. In these cases, yeast detection by Mycosis IC/F, but not by Plus Aerobic/F, bottles might be due to the bacterial growth having been prevented in the Mycosis IC/F medium, which also prevented the bacterial inhibition of yeast growth. According to Meyer et al. (16), this selective effect of antibiotics in the Mycosis IC/F medium might well be the main reason for its advantage in yeast isolation. However, we have observed that Mycosis IC/F medium detects bacterial strains belonging to different species, including P. aeruginosa, a microorganism known for its ability to inhibit the growth of several yeast species both in vitro (8) and in vivo (9). Therefore, the possibility that bacterial inhibition of yeast isolation can also occur in the Mycosis IC/F bottles when yeasts are present concomitantly with antibiotic-resistant bacteria should be considered.
By deploying the entire set of Bactec aerobic, anaerobic, and fungal media, 707 clinically relevant microbial isolates were detected in a 2-year study in the University Hospital of Palermo, and 302 unrepeated isolates belonging to various microbial species were shown to be associated with bacteremic episodes in 221 patients from different units. Of the microorganisms involved in all of the infections, 58.4% were gram-positive cocci, 33.5% were fermentative gram-negative bacilli, 29.0% were nonfermentative gram-negative bacilli, 0.9% were corynebacteria, 1.8% were anaerobes, and 13.1% were yeasts. When these results are compared with those obtained in other epidemiological studies of nosocomial bloodstream infections recently carried out in Italy (15) and other parts of the world (1, 3, 5, 14, 17, 20), our hospital presents a very low incidence of anaerobic infections and, conversely, relatively high rates of isolation of nonfermentative gram-negative bacilli and yeasts. Nonfermentative gram-negative bacilli were more frequently isolated during bloodstream infections in patients from the ICU and HU than in those from all other units, and yeasts (C. albicans) were more frequently isolated in patients from the ICU.
According to Riley et al. (20), “the decision of whether to routinely utilize an anaerobic blood culture bottle should be influenced by the overall recovery of bacteria and yeasts as well as the recovery of specific types of bacteria and yeasts.” In this regard, the isolation of very few obligate anaerobes and numerous aerobic nonfermentative gram-negative bacilli from patients with nosocomial bloodstream infections suggests the advantageous use in our hospital of two aerobic bottles with only selective culturing for anaerobes instead of one aerobic bottle and one anaerobic bottle inoculated with equal volumes of blood. However, anaerobes grew only in anaerobic bottles, and their presence in blood, even though rare, is considered to be significantly and independently associated with increased mortality (1). Aerotolerant and facultatively anaerobic bacteria were efficiently detected with anaerobic bottles, and even although we have not verified this, anaerobic bottles are considered by some authors (20) to be more effective than aerobic bottles in the isolation of S. aureus, an important pathogen responsible for many hospital bloodstream infections. Nevertheless, we do not consider our own data to be significant enough to justify a change in the traditionally and still largely recommended use of paired aerobic and anaerobic bacteriological media in routine blood cultures.
On the contrary, we believe it is appropriate to suggest that Mycosis IC/F bottles, currently reserved for single patients with risk factors related to fungemia (16), should be routinely used when blood cultures are required for patients in units with epidemiological histories of bloodstream infections similar to those observed in our ICU and HU. Our results, in fact, revealed high percentages of yeast infections in the ICU and high percentages of polymicrobic infections in the HU that could most certainly interfere with yeast isolation if Mycosis IC/F bottles were not used. Moreover, many infections in the ICU and HU were caused by aerobic nonfermentative gram-negative bacilli, which, because they were multiresistant to antibiotics in many cases, were also detected by Mycosis IC/F bottles. Finally, many patients in the ICU and HU presented highly compromised defense systems and were therefore exposed to the risk of fungal infections. As analogous conditions were not observed in any other unit in the hospital, the use of Mycosis IC/F bottles in blood cultures, required for patients in these units, will be recommended in the future only when the risk factors for fungemia are present.
Footnotes
Published ahead of print on 15 October 2008.
REFERENCES
- 1.Diekema, D. J., S. E. Beekmann, K. C. Chapin, K. A. Morel, E. Munson, and G. V. Doern. 2003. Epidemiology and outcome of nosocomial and community-onset bloodstream infection. J. Clin. Microbiol. 413655-3660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dorsher, C. W., J. E. Rosemblatt, W. R. Wilson, and D. M. Ilstrup. 1991. Anaerobic bacteremia: decreasing rate over a 15-year period. Rev. Infect. Dis. 13633-636. [DOI] [PubMed] [Google Scholar]
- 3.Durmaz, G., T. Us, A. Aydinli, A. Kiremitci, N. Kiraz, and Y. Akgün. 2003. Optimum detection times for bacteria and yeast species with the BACTEC 9120 aerobic blood culture system: evaluation for a 5-year period in a Turkish University Hospital. J. Clin. Microbiol. 41819-821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Edmond, M. B., S. E. Wallace, D. K. McClish, M. A. Pfaller, R. N. Jones, and R. P. Wenzel. 1999. Nosocomial bloodstream infections in United States hospitals: a three year analysis. Clin. Infect. Dis. 29239-244. [DOI] [PubMed] [Google Scholar]
- 5.Elouennass, M., I. Sahnoun, A. Zrara, T. Bajjou, and S. Elhamzaoui. 2008. Epidemiology and susceptibility profile of blood culture isolates in an intensive care unit (2002-2005). Med. Mal. Infect. 3818-24. [DOI] [PubMed] [Google Scholar]
- 6.Flayhart, D., A. P. Borek, T. Wakefield, J. Dick, and K. C. Carroll. 2007. 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. J. Clin. Microbiol. 45816-821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fricker-Hidalgo, H., B. Lebeau, H. Pelloux, and R. Grillot. 2004. Use of the BACTEC 9240 system with Mycosis-IC/F blood culture bottles for detection of fungemia. J. Clin. Microbiol. 421855-1856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Grillot, R., V. Portmann-Coffin, and P. Ambroise-Thomas. 1994. Growth inhibition of pathogenic yeasts by Pseudomonas aeruginosa in vitro: clinical implications in blood cultures. Mycoses 37343-347. [DOI] [PubMed] [Google Scholar]
- 9.Hockey L. J., N. K Fujita, T. R. Gibson, D. Rotrosen, J. Z. Montgomerie, and J. E. Edwards, Jr. 1982. Detection of fungemia obscured by concomitant bacteremia: in vitro and in vivo studies. J. Clin. Microbiol. 161080-1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Horvath, L. L., and D. R. Hospenthal. 2004. Letter. J. Clin. Microbiol. 421855-1856.15071071 [Google Scholar]
- 11.Horvath, L. L., B. J. George, C. K. Murray, L. S. Harrison, and D. R. Hospenthal. 2004. Direct comparison of the BACTEC 9240 and BacT/ALERT 3D automated blood culture systems for Candida growth detection. J. Clin. Microbiol. 42115-118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jones, J. M. 1990. Laboratory diagnosis of invasive candidiasis. Clin. Microbiol. Rev. 332-45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Karunakaran, R., N. S. Raja, K. F. Quek, V. C. Hoe, and P. Navaratnam. 2007. Evaluation of the routine use of the anaerobic bottle when using the BACTEC blood culture system. J. Microbiol. Immunol. Infect. 40445-449. [PubMed] [Google Scholar]
- 14.Koh, E. M., S. G. Lee, C. K. Kim, M. Kim, D. Yong, K. Lee, J. M. Kim, D. S. Kim, and Y. Chong. 2007. Microorganisms isolated from blood cultures and their antimicrobial susceptibility patterns at a university hospital during 1994-2003. Korean J. Lab. Med. 27265-275. [DOI] [PubMed] [Google Scholar]
- 15.Luzzaro, F., E. F. Viganò, D. Fossati, A. Grossi, A. Sala, C. Sturia, M. Sbudelli, and A. Toniolo 2002. Prevalence and drug susceptibility of pathogens causing bloodstream infections in northern Italy: a two-year study in 16 hospitals. Eur. J. Clin. Microbiol. Infect. Dis. 21849-855. [DOI] [PubMed] [Google Scholar]
- 16.Meyer, M. H., V. Letscher-Bru, B. Jaulhac, J. Waller, and E. Candolfi. 2004. Comparison of Mycosis IC/F and Plus Aerobic/F media for diagnosis of fungemia by the Bactec 9240 system. J. Clin. Microbiol. 42773-777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Morris, A. J., M. L. Wilson, S. Mirrett, and L. B. Reller. 1993. Rationale for selective use of anaerobic blood cultures. J. Clin. Microbiol. 312110-2113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Rangel-Frausto, M. S., T. Wiblin, H. M. Blumberg, L. Saiman, J. Patterson, M. Rinaldi, M. Pfaller, J. E. Edwards, Jr., W. Jarwis, J. Dawson, R. P. Wenzel, and the NEMIS Study Group. 1999. National epidemiology of mycoses survey (NEMIS): variations in rates of bloodstream infections due to Candida species in seven surgical intensive care units and six neonatal intensive care units. Clin. Infect. Dis. 29253-258. [DOI] [PubMed] [Google Scholar]
- 19.Reimer, L. G., M. L. Wilson, and M. P. Weinstein. 1997. Update on detection of bacteremia and fungemia. Clin. Microbiol. Rev. 10444-465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Riley, J. A., B. J. Heiter, and P. P. Bourbeau. 2003. Comparison of recovery of blood culture isolates from two BacT/ALERT FAN aerobic blood culture bottles with recovery from one FAN aerobic bottle and one FAN anaerobic bottle. J. Clin. Microbiol. 41213-217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sharp, S. E. 1991. Routine anaerobic blood cultures: still appropriate today? Clin. Microbiol. Newsl. 13179-181. [Google Scholar]
- 22.Sharp, S. E., J. C. McLaughlin, J. M. Goodman, J. Moore, S. M. Spanos, D. W. Keller III, and R. J. Poppiti, Jr. 1993. Clinical assessment of anaerobic isolates from blood cultures. Diagn. Microbiol. Infect. Dis. 1719-22. [DOI] [PubMed] [Google Scholar]
- 23.Vincent, J. L., D. J. Bihari, P. M. Suter, H. A. Bruining, J. White, M. H. Nicolas-Chanoin, M. Wolff, R. C. Spencer, and M. Hemmer. 1995. The prevalence of nosocomial infection in intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care (EPIC) study. JAMA. 274639-644. [PubMed] [Google Scholar]