Abstract
The purpose of this study was to evaluate the clinical utility of bacterial rRNA-targeted reverse transcription-quantitative PCR (BrRNA RT-qPCR) assays for identifying the bacterial pathogens that cause fever with neutropenia in pediatric cancer patients, by comparing the bacterial detection rate of this technique with that of blood culture. One milliliter of blood was collected from pediatric patients who developed fever with neutropenia following cancer chemotherapy. BrRNA RT-qPCR was performed using 16 primer sets, each designed for a specific type of bacteria. The entire BrRNA RT-qPCR procedure took less than 5 h. Blood culture was performed at the same time, following the standard institutional procedure. Blood from 13 patients was collected during 23 febrile neutropenic episodes. Of these samples, bacteria were identified in 16 by BrRNA RT-qPCR (69.6%) and in 4 by blood culture (17.4%, P < 0.001). In all 4 blood culture-positive samples, BrRNA RT-qPCR detected the same type of bacteria as that identified by culture. In 9 samples, more than 4 types of bacteria were identified simultaneously by BrRNA RT-qPCR, most of which were anaerobic bacteria known to be part of the gut flora. We conclude that BrRNA RT-qPCR could be useful in the diagnosis of fever with neutropenia, given its high bacterial detection rate, short turnaround time, and the small blood sample required compared with the standard blood culture techniques. Our findings also indicate that anaerobic intestinal bacteria, which are difficult to detect by standard culture techniques, may be responsible for some cases of febrile neutropenia.
Fever with neutropenia is a frequent complication in patients undergoing cancer chemotherapy. Because bacterial infection is considered to be the most common cause of this complication, which can be life threatening if untreated, it is recommended that antibiotic treatment be started immediately when neutropenic patients develop fever (10, 17, 18). However, blood culture, a current standard microbiological assessment for febrile neutropenic patients, can identify pathogens in only 10 to 37% of these patients (4, 6, 7, 19) and can take more than 24 h to yield a report. Therefore, the majority of cases are treated empirically with broad-spectrum antibiotics and without microbiological evidence to support the treatment. More sensitive techniques would help physicians optimize the therapy for individual cases of fever with neutropenia.
We recently developed a technique, bacterial rRNA-targeted reverse transcription-quantitative PCR (BrRNA RT-qPCR), to detect subdominant bacterial populations in feces (12, 13, 23). In an in vitro study, we also showed that BrRNA RT-qPCR can be used to detect bacteria in blood samples (12), but we did not seek to identify bacterial pathogens in a clinical setting.
The objective of this study was to evaluate the clinical utility of BrRNA RT-qPCR for identifying the bacterial pathogens responsible for fever in neutropenic patients. We conducted BrRNA RT-qPCR on blood samples from pediatric cancer patients with fever and neutropenia and compared the bacterial detection rates with the results obtained from blood culture.
MATERIALS AND METHODS
Participants.
This study enrolled patients who were hospitalized in the pediatric ward of Juntendo University Hospital, were 1 month to 18 years of age, and developed fever higher than 38.0°C during a neutropenic period following cancer chemotherapy for solid tumors, lymphoma, or leukemia. Neutropenia was defined as white blood cell (WBC) counts of less than 1,000/mm3 or absolute neutrophil counts of less than 500/mm3. Patients remained enrolled if they underwent more than one cycle of cancer chemotherapy during the study period, which means that if a patient developed fever more than once but in different neutropenic periods, samples were collected during each febrile episode. However, if a patient developed fever that subsided in a few days and then developed fever again during the same neutropenic period, we did not collect a sample during the second wave of fever, because it was considered a recurrence or treatment failure of the preceding episode of febrile neutropenia. A cycle of chemotherapy was defined as the duration from the start of one set of treatments to the start of the next set, which was sufficient to allow patients to recover from acute adverse effects of the therapy, including neutropenia.
Because this was our first application of this technique in a clinical setting, we additionally collected blood samples from 6 of the enrolled patients and 7 healthy adult volunteers to assess whether false positives could occur due to contamination associated with the procedure of blood collection. Blood was collected when these participants were afebrile, defined as a temperature lower than 37.5°C, and had no clinical or laboratory findings that suggested infection for at least the previous 3 days. To eliminate the possibility of subclinical bacterial infection, samples were excluded from analysis if the participant developed a fever higher than 38.0°C within 3 days after the sampling or if the patient had neutropenia or a C-reactive protein (CRP) level higher than 0.3 mg/dl at the time of the sample collection.
Written informed consent was obtained from the guardians of pediatric patients and directly from the adult volunteers. The procedures were approved by our institutional review board and were in accordance with the Helsinki Declaration.
Sample collection and study design.
As soon as the neutropenic patients developed fever, blood was drawn from indwelling vascular devices or peripheral veins. From each sample, 1.0 ml of blood was allocated for BrRNA RT-qPCR, and 10 to 15 ml was used for blood culture. A complete blood cell count (CBC) was obtained at the same time to confirm that the patients were neutropenic. Antibiotic therapy was started empirically immediately following the blood sampling, in accordance with the standard institutional policy for treating febrile neutropenia.
As infection-free control samples, 1.0 ml of blood was drawn from afebrile patients through indwelling vascular devices during scheduled routine blood tests, which included CBC and CRP measurement. One milliliter of blood was drawn from healthy adult volunteers via a peripheral vein.
The primary outcome was the detection rate of bacteria by BrRNA RT-qPCR and by blood culture. The types and amounts of bacteria detected by BrRNA RT-qPCR were the secondary outcomes.
BrRNA RT-qPCR and target bacteria.
One milliliter of blood was added to 2.0 ml of RNAprotect bacterial reagent (Qiagen GmbH, Hilden, Germany) immediately after collection, and the sample was stored at −30°C until used for analysis. The methods for total RNA extraction and RT-qPCR were described previously (12). Briefly, RNA was isolated using a modified acidic guanidinium thiocyanate-phenol-chloroform extraction method. RT-qPCR was conducted using a Qiagen OneStep RT-PCR kit (Qiagen GmbH).
The target bacteria and sequences of the corresponding primers used in the current study are listed in Table 1. Twelve primer sets had been developed for BrRNA RT-qPCR in previous studies, for detection of Staphylococcus species, Enterococcus species, Enterobacteriaceae, Pseudomonas species, Bacillus cereus, the Bacteroides fragilis group, the Atopobium cluster, the Clostridium coccoides group, the Clostridium leptum subgroup, Clostridium perfringens, Bifidobacterium species, and Prevotella species (9, 11-16, 23). Although some of these bacteria are usually found in feces as subdominant bacterial populations but are not frequently detected in the blood as pathogens of systemic infection, we decided to use all of them as target bacteria in the current study, since this was the first assessment of BrRNA RT-qPCR in a clinical setting. In addition, 4 primer sets for the genus Streptococcus were newly developed. Of these, 3 sets were species specific for S. pyogenes, S. agalactiae, and S. pneumoniae/S. mitis. The fourth was a genus-specific primer set that targeted all Streptococcus species, including but not limited to the above 3 species. These primers were developed following a previously described method (12). In brief, to develop rRNA-targeted primers, we constructed a multiple alignment of the target bacteria using the Clustal X program with 16S and 23S rRNA gene sequences obtained from the DDBJ/GenBank/EMBL databases (22). We checked their specificities using the Probe Match program of the Ribosomal Database Project (RDP-II), release 9 (5).
TABLE 1.
Target bacteria and primer sequences
| Target | Primer sequence and directiona (5′-3′) | Reference(s) or source |
|---|---|---|
| Streptococcus | F: AGCTTAGAAGCAGCTATTCATTC | This study |
| R: GGATACACCTTTCGGTCTCTC | ||
| Streptococcus pyogenes | F: AAGAGAGACTAACGCATGTTAGTAATTT | This study |
| R: AATGCCTTTAACTTCAGACTTAAAAA | ||
| Streptococcus agalactiae | F: GTTATTTAAAAGGAGCAATTGCTT | This study |
| R: TTGGTAGATTTTCCACTCCTACCA | ||
| Streptococcus pneumoniae/S. mitis | F: CAATGTGGACTCAAAGATTATAGAAGAATG | This study |
| R: GTCATGATACTAAGGCGCCCTA | ||
| Staphylococcus | F: ACGGTCTTGCTGTCACTTATA | 12 |
| R: TACACATATGTTCTTCCCTAATAA | ||
| Enterococcus | F: ATCAGAGGGGGATAACACTT | 13 |
| R: ACTCTCATCCTTGTTCTTCTC | ||
| Enterobacteriaceae | F: TGCCGTAACTTCGGGAGAAGGCA | 12 |
| R: TCAAGGACCAGTGTTCAGTGTC | ||
| Pseudomonas | F: CAAAACTACTGAGCTAGAGTACG | 12 |
| R: TAAGATCTCAAGGATCCCAACGGCT | ||
| Bacillus cereus | F: TCGAAATTGAAAGGCGGC | 9, 23 |
| R: CCAGCTTATTCAACTAGCACTT | ||
| Bacteroides fragilis group | F: AYAGCCTTTCGAAAGRAAGAT | 14 |
| R: CCAGTATCAACTGCAATTTTA | ||
| Atopobium cluster | F: GGGTTGAGAGACCGACC | 16 |
| R: CGGRGCTTCTTCTGCAGG | ||
| Clostridium coccoides group | F: AAATGACGGTACCTGACTAA | 15 |
| R: CTTTGAGTTTCATTCTTGCGAA | ||
| Clostridium leptum subgroup | F: GCACAAGCAGTGGAGT | 16 |
| R: CTTCCTCCGTTTTGTCAA | ||
| Clostridium perfringens | F: GGGGGTTTCAACACCTCC | 11, 14 |
| R: GCAAGGGATGTCAAGTGT | ||
| Bifidobacterium | F: CTCCTGGAAACGGGTGG | 15 |
| R: GGTGTTCTTCCCGATATCTACA | ||
| Prevotella | F: CACRGTAAACGATGGATGCC | 15 |
| R: GGTCGGGTTGCAGACC |
F, forward; R, reverse.
The method for determining the number of bacteria in a sample was described previously (12, 13, 23) and was used here essentially as described. Briefly, standard curves for the RT-qPCR were generated with a threshold cycle (CT) value and the corresponding bacterial cell count of a dilution series of reference strains, which was determined microscopically. Next, the CT values of three serial dilutions of RNA extracted from the blood sample were applied to the standard curve to obtain the corresponding number of bacterial cells in a sample. As reference strains for the 4 new primers, we used Streptococcus mutans IFO13955T, S. pyogenes ATCC 12344T, S. agalactiae JCM 5671T, and S. pneumoniae DSM 20566T.
The specificity of the 12 previously developed primer sets in the BrRNA RT-qPCR was determined by assessing RNA extracted from 50 species of bacteria (12); the 4 newly developed primer sets were tested with 44 species of bacteria. All the primers gave positive results for only the target species and did not cross-react with any of the nontarget species tested. The minimum detectable number of all the target bacteria by BrRNA RT-qPCR was 1 bacterial cell per 1-ml sample (12, 13, 23; this study).
The entire BrRNA RT-qPCR assay, including the RNA extraction step, could be completed in 5 h.
Blood culture.
Each blood sample was subdivided into a two-bottle set (92F aerobic and 93F anaerobic resin blood culture bottles; Becton and Dickinson Company, Japan) and processed in an automatic culture device (Bactec 9240; Becton and Dickinson Company, Japan) at the institutional laboratory. Susceptibility testing was performed on agar plates using the dilution method.
Statistical analysis.
Bacterial detection rates are expressed as whole numbers and percentages. Differences in the detection rates of BrRNA RT-qPCR and blood culture were assessed by McNemar's test. The types and numbers of bacteria detected by BrRNA RT-qPCR are presented descriptively.
RESULTS
Participants.
Fourteen patients with 25 episodes of fever with neutropenia were enrolled. Of these, 2 episodes were excluded because they did not meet the criteria (WBC count of more than 1,000/mm3 at the time of sample collection), and samples taken from 13 patients during 23 episodes of fever with neutropenia were assessed. The patients' baseline characteristics and underlying diseases are shown in Table 2.
TABLE 2.
Baseline characteristics of patients
| Characteristic | Value for group |
|---|---|
| Age [yrs; median (range)] | 5.8 (1.6-14.8) |
| Male [no. (%)] | 6 (43) |
| Underlying disease [no. (%) of neutropenic episodes; no. (%) of patients with disease] | |
| Acute lymphoblastic leukemia | 8 (34.8); 5 (38.5) |
| Acute myeloid leukemia | 1 (4.3); 1 (7.7) |
| Lymphoma | 3 (13.0); 2 (15.4) |
| Neuroblastoma | 8 (34.8); 3 (23.1) |
| Clear cell sarcoma of kidney | 2 (8.7); 1 (7.7) |
| Medulloblastoma | 1 (4.3); 1 (7.7) |
| Total no. | 23; 13 |
During the same period, 20 blood samples were collected as infection-free controls for the BrRNA RT-qPCR. Of these, 13 were from 6 afebrile patients and 7 were from 7 healthy adult volunteers. Four of the samples collected from patients were excluded (CRP levels at the time of sample collection were higher than 0.3 mg/dl), so a total of 16 control samples were assessed.
Identification of bacteria in samples from patients with fever and neutropenia.
Of the 23 samples from patients with fever and neutropenia, bacteria were detected in 16 by BrRNA RT-qPCR (69.6%) and in 4 by blood culture (17.4%, P < 0.001). The types and amounts of bacteria detected by BrRNA RT-qPCR are shown in Table 3.
TABLE 3.
Results for BrRNA RT-qPCR-positive samples
 |
—, not detected or negative; Pos, positive.
Blood culture positive for Staphylococcus epidermidis.
Blood culture positive for S. mitis.
Blood culture positive for S. oralis.
In all 4 blood culture-positive samples, BrRNA RT-qPCR detected the same type of bacterium as was identified by the culture method: Staphylococcus epidermidis in 1 sample, S. mitis in 2 samples, and S. oralis in 1 sample. (Staphylococcus epidermidis and S. oralis were identified at the genus level by BrRNA RT-qPCR.) More than one type of bacteria were identified simultaneously in 10 of the 16 BrRNA RT-qPCR-positive samples (62.5%).
Results from infection-free control samples.
The BrRNA RT-qPCR assay did not detect bacteria in any of the 16 infection-free control samples.
DISCUSSION
This study demonstrated that BrRNA RT-qPCR can be used to identify bacterial pathogens responsible for fever with neutropenia in pediatric cancer patients. To our knowledge, this is the first study in which BrRNA RT-qPCR was used in a clinical setting to detect bacterial pathogens. The detection rate obtained by the BrRNA RT-qPCR assay was markedly higher than that obtained from the standard blood culture in this study, which was comparable to the detection rate reported in the literature (4, 6, 7, 19). BrRNA RT-qPCR may be a more sensitive technique because it can detect bacteria that grow poorly in culture media. In addition, rRNA is a universal constituent of the bacterial ribosome, and high copy numbers are present in a single bacterial cell as housekeeping genes. We previously found that the RT-qPCR assay targeting rRNA is 64- to 1,024-fold more sensitive than conventional qPCR that targets DNA (12). Conventional PCR has been used to identify the bacterial pathogens in patients who have a clinical diagnosis of systemic bacterial infection. However, the bacterial detection rates by conventional PCR were the same as those obtained by blood culture or even lower, which is its major limitation for this clinical application (24, 25). Our findings indicate that BrRNA RT-qPCR overcomes this issue and enables the detection of bacteria with high sensitivity compared to that of conventional PCR or standard blood culture.
All the samples that were positive for bacteria by blood culture were also positive by BrRNA RT-qPCR, with the same bacteria detected in both techniques; in contrast, only 4 of the 16 positive BrRNA RT-qPCR samples were positive by blood culture. Interestingly, for Staphylococcus, in the samples that were positive by both techniques, the bacterial counts (sample 1 in Table 3; 25 cells) were higher than in those detected only by BrRNA RT-qPCR (samples 2 and 8 to 11; 2 to 8 cells). A similar pattern was observed for the detection of S. mitis. These findings support our assertion that BrRNA RT-qPCR is more sensitive than blood culture for detecting small amounts of bacteria in the blood. On the other hand, some organisms were not detected by culture, even though BrRNA RT-qPCR detected relatively large amounts of them, such as the Enterobacteriaceae and Pseudomonas bacteria in sample 9. The patient from whom sample 9 was obtained developed hypotension and multiple organ dysfunctions immediately after the sample collection, which was clinically diagnosed as septic shock. Although no organism was identified even by subsequent repeated cultures, the results of the BrRNA RT-qPCR analysis fit the patient's clinical course. To assess the association between the results of BrRNA RT-qPCR and culture for Enterobacteriaceae and Pseudomonas bacteria, a further study that includes patients whose blood culture is positive for these organisms is required.
In the present study, BrRNA RT-qPCR simultaneously detected more than one type of bacteria in 10 of 23 samples. Surprisingly, in 9 of them, more than 4 types of bacteria were detected, including Enterococcus species, Enterobacteriaceae, and obligate anaerobes, i.e., the C. coccoides group, C. leptum subgroup, B. fragilis group, and Atopobium cluster. Because these organisms are part of the gut flora, their simultaneous detection in the bloodstream may indicate bacterial translocation, which is defined as the invasion of extraintestinal organs, including the bloodstream, by indigenous intestinal bacteria that cross the gut mucosa. Bacterial translocation occurs when the intestinal barrier function is damaged by chemotherapy, and therefore, it is a likely etiology of infection in cancer patients undergoing chemotherapy (1, 3, 21). More than half of the organisms that make up the gut flora cannot be grown in culture media, which may explain the low detection rate of these pathogens by culturing the blood from patients with fever with neutropenia (8, 20). The finding that the infection-free controls were all negative by the BrRNA RT-qPCR method indicates that the detection of these bacteria in the current study was probably not false-positive results.
The entire BrRNA RT-qPCR process takes less than 5 h, while standard blood culture can take more than 24 h to identify pathogens. This marked reduction in turnaround time could allow physicians to start the management of fever with neutropenia with a pathogen-specific treatment rather than with an empirical treatment. In addition, BrRNA RT-qPCR requires only 1.0 ml of blood, while considerably more is required for an adequate blood culture (2). This method may benefit pediatric patients more greatly than adults, because it is often difficult to obtain enough blood from children for a culture assay and because their condition can deteriorate more rapidly than that of adults, which makes the early initiation of appropriate treatment even more critical.
The main limitation of this study is the small sample size. Although the sample size was sufficient to show a significant difference in the bacterial detection rate, which was our primary outcome, only 4 cases were positive by blood culture, and some pathogens that are typical in febrile neutropenic patients, such as Gram-negative rods, were not detected by the culture method. We also did not assess the clinical course and outcome of the febrile neutropenia. In a future study, these clinical data could help distinguish whether the bacteria detected by BrRNA RT-qPCR but not by blood culture are truly the pathogens responsible for the fever. Regarding the limitations of the BrRNA RT-qPCR itself, it cannot evaluate bacterial susceptibility to antibiotics, and it identifies only bacteria for which there are specific primers. Therefore, the combination of culture techniques with BrRNA RT-qPCR and regular surveillance of the antibiotic susceptibility patterns of emerging bacterial species may be important for the effective clinical use of BrRNA RT-qPCR.
In conclusion, our findings indicate that BrRNA RT-qPCR can be used to identify bacteria from the blood of pediatric cancer patients with fever and neutropenia with a higher detection rate, shorter turnaround time, and smaller blood sample than standard blood culture. Our results also suggest that anaerobic intestinal bacteria, which are difficult to detect by the current culture techniques, may be responsible for some cases of febrile neutropenia. We conclude that BrRNA RT-qPCR is a convenient and sensitive tool for detecting bacteria in the blood and that it may be useful as an adjunct to culture techniques for diagnosing fever with neutropenia.
Acknowledgments
This study was performed as a project of the Probiotics Research Course of Juntendo University, Graduate School of Medicine. This course is funded by Yakult Honsha Co., Ltd.
Footnotes
Published ahead of print on 29 March 2010.
REFERENCES
- 1.Balzan, S., C. de Almeidaq Uadros, R. de Cleva, B. Zilberstein, and I. Cecconello. 2007. Bacterial translocation: overview of mechanisms and clinical impact. J. Gastroenterol. Hepatol. 22:464-471. [DOI] [PubMed] [Google Scholar]
- 2.Baron, E. J., M. P. Weinstein, W. M. Dunne, Jr., P. Yagupsky, D. F. Welch, and D. M. Wilson. 2005. Cumitech 1C, Blood cultures IV. Coordinating ed., E. J. Baron. ASM Press, Washington, DC.
- 3.Berg, R. D., and A. W. Garlington. 1979. Translocation of certain indigenous bacteria from the gastrointestinal tract to the mesenteric lymph nodes and other organs in a gnotobiotic mouse model. Infect. Immun. 23:403-411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Castagnola, E., V. Fontana, I. Caviglia, S. Caruso, M. Faraci, F. Fioredda, M. L. Garrè, C. Moroni, M. Conte, G. Losurdo, F. Scuderi, R. Bandettini, P. Tomà, C. Viscoli, and R. Haupt. 2007. A prospective study on the epidemiology of febrile episodes during chemotherapy-induced neutropenia in children with cancer or after hemopoietic stem cell transplantation. Clin. Infect. Dis. 45:1296-1304. [DOI] [PubMed] [Google Scholar]
- 5.Cole, J. R., B. Chai, T. L. Marsh, R. J. Farris, Q. Wang, S. A. Kulam, S. Chandra, D. M. McGarrell, T. M. Schmidt, G. M. Garrity, and J. M. Tiedje. 2003. The Ribosomal Database Project (RDP-II): previewing a new autoaligner that allows regular updates and the new prokaryotic taxonomy. Nucleic Acids Res. 31:442-443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gençer, S., T. Salepçi, and S. Ozer. 2003. Evaluation of infectious etiology and prognostic risk factors of febrile episodes in neutropenic cancer patients. J. Infect. 47:65-72. [DOI] [PubMed] [Google Scholar]
- 7.Greenberg, D., A. Moser, P. Yagupsky, N. Peled, Y. Hofman, J. Kapelushnik, and E. Leibovitz. 2005. Microbiological spectrum and susceptibility patterns of pathogens causing bacteraemia in paediatric febrile neutropenic oncology patients: comparison between two consecutive time periods with use of different antibiotic treatment protocols. Int. J. Antimicrob. Agents 25:469-473. [DOI] [PubMed] [Google Scholar]
- 8.Guarner, F., and J. R. Malagelada. 2003. Gut flora in health and disease. Lancet 361:512-519. [DOI] [PubMed] [Google Scholar]
- 9.Hansen, B. M., T. D. Leser, and N. B. Hendriksen. 2001. Polymerase chain reaction assay for the detection of Bacillus cereus group cells. FEMS Microbiol. Lett. 202:209-213. [DOI] [PubMed] [Google Scholar]
- 10.Jun, H. X., S. Zhixiang, W. Chun, A. H. Reksodiputro, D. Ranuhardy, K. Tamura, T. Matsumoto, D. G. Lee, S. V. Purushotaman, V. Lim, A. Ahmed, Y. Hussain, M. Chua, A. Ong, C. Y. Liu, P. R. Hsueh, S. F. Lin, Y. C. Liu, P. Suwangool, S. Jootar, and J. J. Picazo. 2005. Clinical guidelines for the management of cancer patients with neutropenia and unexplained fever. Int. J. Antimicrob. Agents 26(Suppl. 2):S128-S132. [DOI] [PubMed] [Google Scholar]
- 11.Kikuchi, E., Y. Miyamoto, S. Narushima, and K. Itoh. 2002. Design of species-specific primers to identify 13 species of Clostridium harbored in human intestinal tracts. Microbiol. Immunol. 46:353-358. [DOI] [PubMed] [Google Scholar]
- 12.Matsuda, K., H. Tsuji, T. Asahara, Y. Kado, and K. Nomoto. 2007. Sensitive quantitative detection of commensal bacteria by rRNA-targeted reverse transcription-PCR. Appl. Environ. Microbiol. 73:32-39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Matsuda, K., H. Tsuji, T. Asahara, K. Matsumoto, T. Takada, and K. Nomoto. 2009. Establishment of an analytical system for the human fecal microbiota, based on reverse transcription-quantitative PCR targeting of multicopy rRNA molecules. Appl. Environ. Microbiol. 75:1961-1969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Matsuki, T. 2007. Development of quantitative PCR detection method with 16S rRNA gene-targeted genus- and species-specific primers for the analysis of human intestinal microflora and its application. Nippon Saikingaku Zasshi 62:255-261. (In Japanese.) [DOI] [PubMed] [Google Scholar]
- 15.Matsuki, T., K. Watanabe, J. Fujimoto, Y. Miyamoto, T. Takada, K. Matsumoto, H. Oyaizu, and R. Tanaka. 2002. Development of 16S rRNA-gene-targeted group-specific primers for the detection and identification of predominant bacteria in human feces. Appl. Environ. Microbiol. 68:5445-5451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Matsuki, T., K. Watanabe, J. Fujimoto, T. Takada, and R. Tanaka. 2004. Use of 16S rRNA gene-targeted group-specific primers for real-time PCR analysis of predominant bacteria in human feces. Appl. Environ. Microbiol. 70:7220-7228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Pizzo, P. A. 1999. Fever in immunocompromised patients. N. Engl. J. Med. 341:893-900. [DOI] [PubMed] [Google Scholar]
- 18.Pizzo, P. A. 1993. Management of fever in patients with cancer and treatment-induced neutropenia. N. Engl. J. Med. 328:1323-1332. [DOI] [PubMed] [Google Scholar]
- 19.Ritchie, S., S. Palmer, and R. Ellis-Pegler. 2007. High-risk febrile neutropenia in Auckland 2003-2004: the influence of the microbiology laboratory on patient treatment and the use of pathogen-specific therapy. Intern. Med. J. 37:26-31. [DOI] [PubMed] [Google Scholar]
- 20.Sears, C. L. 2005. A dynamic partnership: celebrating our gut flora. Anaerobe 11:247-251. [DOI] [PubMed] [Google Scholar]
- 21.Song, D., B. Shi, H. Xue, Y. Li, X. Yang, B. Yu, Z. Xu, F. Liu, and J. Li. 2006. Confirmation and prevention of intestinal barrier dysfunction and bacterial translocation caused by methotrexate. Dig. Dis. Sci. 51:1549-1556. [DOI] [PubMed] [Google Scholar]
- 22.Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tsuji, H., K. Matsuda, T. Asahara, M. Kiwaki, and K. Nomoto. August 2006. Quantitative analysis of microbes with rRNA as target. Japanese patent, PCT no. 301467.
- 24.Varani, S., M. Stanzani, M. Paolucci, F. Melchionda, G. Castellani, L. Nardi, M. P. Landini, M. Baccarani, A. Pession, and V. Sambri. 2009. Diagnosis of blood stream infection in immunocompromised patients by real-time PCR. J. Infect. 58:346-351. [DOI] [PubMed] [Google Scholar]
- 25.Westh, H., G. Lisby, F. Breysse, B. Boddinghaus, M. Chomarat, V. Gant, A. Goglio, A. Raglio, H. Schuster, F. Stuber, H. Wissing, and A. Hoeft. 2009. Multiplex real-time PCR and blood culture for identification of bloodstream pathogens in patients with suspected sepsis. Clin. Microbiol. Infect. 15:544-551. [DOI] [PubMed] [Google Scholar]