Pediatric Pulmonology 37:537–547 (2004)
Identification of Pseudomonas aeruginosa, Burkholderia cepacia Complex, and Stenotrophomonas maltophilia in Respiratory Samples From Cystic Fibrosis Patients Using Multiplex PCR Luiz V.F. da Silva Filho, MD, PhD,1,2* Adriana F. Tateno, BSc,2 Luciana de F. Velloso, MD,1 Jose´ E. Levi, PhD,2 Silvana Fernandes, BSc,2 Christina N.O. Bento, BSc,3 Joaquim C. Rodrigues, MD, PhD,1 and Sonia R.T.S. Ramos, MD, PhD1 Summary. A multiplex PCR method was developed to identify P. aeruginosa, B. cepacia complex, and S. maltophilia directly in sputum and oropharyngeal samples from CF patients. One hundred and six patients (53 male, and 53 female) attending our pulmonology clinic were studied from September 2000–April 2001. Two hundred and fifty-seven samples were cultured in selective media and submitted to multiplex PCR reactions, using three primer pairs targeting specific genomic sequences of each species, with an additional primer pair targeting a stretch of ribosomal 16S DNA, universal for bacteria, to act as a control. P. aeruginosa was isolated by culture in 56% of samples, B. cepacia complex in 4.3%, and S. maltophilia in 2.7%, while multiplex PCR identified P. aeruginosa in 78.7%, B. cepacia complex in 3.9%, and S. maltophilia in 3.1% of samples. Multiplex PCR results were verified by PCR reactions using different species-specific primers described in the literature and DNA sequencing of amplicons from a few samples. Comparing to culture results, the sensitivity and specificity values of multiplex PCR for bacterial identification were, respectively, 97.2% and 45.5% for P. aeruginosa, 45.5% and 97.9% for B. cepacia complex, and 40% and 97.6% for S. maltophilia. All 10 multiplex PCR-positive results for B. cepacia complex were confirmed using other species-specific primers described in the literature, while this approach confirmed results for S. maltophilia identification in 7/8 samples (87.5%). Sequencing of amplicons from samples culture-negative but multiplex PCR-positive for P. aeruginosa and B. cepacia complex confirmed their identity, while minor nucleotide differences among amplicons ruled out the hypothesis of PCR contamination. Pediatr Pulmonol. 2004; 37:537–547. ß 2004 Wiley-Liss, Inc. Key words: burkholderia cepacia; stenotrophomonas maltophilia; pseudomonas aeruginosa; cystic fibrosis; infections; diagnostic methods; polymerase chain reaction. 1
INTRODUCTION
Respiratory infections play a major role in lung disease of cystic fibrosis (CF) patients. CF patients usually present progressive deterioration of lung function associated with chronic and recurrent infections with specific bacteria, mainly Pseudomonas aeruginosa, Staphylococcus aureus, and Haemophilus influenzae. In recent years, Burkholderia cepacia and Stenotrophomonas maltophilia have also been increasingly recognized as infective agents for these patients.1,2 Pseudomonas aeruginosa is one of the most important respiratory pathogens for CF patients, and early colonization with this microorganism is associated with poor prognosis.3,4 Chronic colonization with mucoid strains of P. aeruginosa occurs in almost 80% of adult CF patients, and is associated with a significant immune response that results in progressive lung damage.5 Initial colonization with P. aeruginosa occurs in the first years of life, and several strategies to eradicate the microorganism or ß 2004 Wiley-Liss, Inc.
Instituto da Crianc¸a ‘‘Prof. Pedro de Alcaˆntara,’’ Hospital das Clı´nicas, University of Sa˜o Paulo Medical School, Sa˜o Paulo, Brazil.
2
Laboratory of Virology, Instituto de Medicina Tropical de Sa˜o Paulo, University of Sa˜o Paulo Medical School, Sa˜o Paulo, Brazil. 3
Microbiology Section of Central Laboratory, Hospital das Clı´nicas, University of Sa˜o Paulo Medical School, Sa˜o Paulo, Brazil. This work was presented at the 2002 North American CF Conference, New Orleans, LA. Grant sponsor: FAPESP ; Grant number: 1999/00067-9; Grant sponsor: CAPES, Brazil. *Correspondence to: Dr. Luiz Vicente Ferreira da Silva Filho, Instituto da Crianc¸a/HCFMUSP, Rua Grego´rio Paes de Almeida, 1148, Vica Madelena, Sa˜o Paulo, CEP 05404-001 Sa˜o Paulo, Brazil. E-mail:
[email protected] Received 6 February 2003; Revised 20 October 2003; Accepted 21 October 2003. DOI 10.1002/ppul.20016 Published online in Wiley InterScience (www.interscience.wiley.com).
538
da Silva Filho et al.
postpone chronic colonization on that occasion have been proposed in the last decade.6,7 However, these approaches rely on proper detection of the microorganism in the respiratory tract, which is usually attained by throat cultures, with low sensitivity.8,9 B. cepacia was first identified in sputum samples from cystic fibrosis patients in the early 1980s,10 and in some patients, it was associated with the occurrence of a fatal necrotizing pneumonia called B. cepacia syndrome. The inherent resistance to the majority of antimicrobial drugs, and further observations that B. cepacia strains can spread by social contact among patients, resulted in several recommendations for new approaches in microbiological surveillance11 and patient cohorting.12,13 As a result, isolation of B. cepacia in sputum from cystic fibrosis patients carries a significant medical and psychosocial burden. Strategies for B. cepacia infection control, however, depend on the accurate identification of this pathogen by the microbiology laboratory. This picture is complicated by the fact that B. cepacia is not currently recognized as a unique species, but a complex comprising, up to now, nine different genomovars or discrete genomic species. Determination of genomovar status is usually performed by genotypic testing,14–18 although phenotypic tests such as whole cell protein profile analysis have also been used.19 Stenotrophomonas maltophilia is a Gram-negative bacillus widely distributed in environmental habitats with inherent resistance to a variety of antibiotics, such as penicillins, carbapenems, and aminoglycosides.20 Originally considered as a harmless commensal, S. maltophilia has emerged as a significant pathogen for immunocompromised patients,21 and recent evidence suggests that it may also be an emergent pathogen for cystic fibrosis patients.1,22 Accurate identification of this microorganism is not always a straightforward procedure, and misidentifications have been reported.23 Previously, we described a PCR method for the identification of P. aeruginosa in sputum samples of CF patients.24 We decided to improve this method with the inclusion of additional primer pairs to identify B. cepacia complex and S. maltophilia directly in respiratory samples (sputum and throat swabs) from CF patients, resulting in a multiplex PCR approach for identification of these three bacterial species. MATERIALS AND METHODS Study Population
In total, 106 cystic fibrosis patients (53 male, 53 female; age range, 9 months–19 years; median age, 9.77 years) at the Pediatric Pulmonology Unit of the Instituto da Crianc¸a, University of Sa˜o Paulo, were studied from September 2000–April 2001. Diagnosis of CF was based on clinical symptoms and two positive sweat tests or identification of two mutations by genetic analysis, according to international guidelines.25
Bacterial Strains
The strains used for the development and testing of the multiplex PCR method were American Type Culture Collection (ATCC) strains of P. aeruginosa ATCC 27853, B. cepacia ATCC 25416 (genomovar I), S. maltophilia ATCC 13637, E. coli ATCC 25922, and S. aureus ATCC 25923. Other organisms used included clinical isolates of Pseudomonas aeruginosa (n ¼ 6), B. cepacia complex (n ¼ 12), S. maltophilia (n ¼ 6), Pseudomonas stutzeri (n ¼ 2), Klebsiella pneumoniae (n ¼ 2), Haemophilus influenzae (n ¼ 4), Enterobacter aerogenes (n ¼ 2), Acynetobacter calcoaceticus (n ¼ 2), Alcaligenes xylosoxidans (n ¼ 4), and Staphylococcus epidermidis (n ¼ 2), provided by the Microbiology Section of Central Laboratory of the Hospital das Clı´nicas (Sa˜o Paulo, Brazil). Additionally, the following strains from the newly described genomovars of B. cepacia were tested: B. cepacia genomovar VI AU0645, B. ambifaria (genomovar VII) strains AMMD and ATCC 53266, B. anthina (genomovar VIII) strains W92 and J552, and B. pyrrocinia (genomovar IX) strains ATCC 15958 and ATCC 39277, all provided by Dr. Eshwar Mahenthiralingam (Cardiff School of Biosciences, Cardiff, UK). Clinical Samples
During the study period, samples of sputum or oropharyngeal swabs from CF patients were collected by one of the investigators (L.V.F.S.F. or L.F.V.) on each patient visit. Written informed consent was obtained from parents. Sputum samples were collected directly by expectoration in a sterilized plastic receptacle, and throat swabs were collected by direct friction of a sterile cotton swab in the posterior pharynx, if possible after coughing and with concomitant use of a tongue depressor. Sputum samples were initially processed as described by Wong et al.26 Briefly, an equal volume of a sterile solution of dithithreitol (DTT) 50 mg/ml in phosphate-buffered saline (PBS) with 0.1% of gelatin was added to the sputum samples, and after 30 min was homogenized by vortexing. Oropharyngeal swabs were kept on transport medium until delivery to the microbiology laboratory. All samples were taken to the microbiology laboratory within a short period of time (up to 4 hr), and immediately plated on selective media. After that, approximately 1 ml of each sputum sample was transferred to a criotube, and oropharyngeal swabs were put in criotubes containing 0.5 ml of PBS for 30 min. After this period, swabs were squeezed to tube walls and discarded. All samples were stored at 808C until DNA extraction. The Ethical Committee of our institution approved the study. Culture and Identification
Samples were cultivated on selective media, including blood agar (Columbia Agar, Oxoid), chocolate agar (GC
Multiplex PCR for Respiratory Pathogens in CF
Agar, Biobra´s, Sa˜o Paulo, Brazil), MacConkey agar (MacConkey Agar, Merck) and Burkholderia cepaciaselective medium (Burkholderia cepacia medium, Oxoid), incubated at 36 18C for a period of 18–48 hr. Bacterial identification was performed with the Vitek1 system (bioMe´rieux Vitek, Inc., St. Louis, MO), using Gramnegative (GNI) and Gram-positive (GPI) cards, and with additional biochemical tests for bacterial identification when necessary. Results were recorded after 24 hr of incubation at 378C. All isolates of Pseudomonas aeruginosa, Burkholderia cepacia, and Stenotrophomonas maltophilia were transferred to tryptic soy broth (Tryptic Soy Broth, Merck) with glycerol and stored in a 808C freezer. Preparation of Template DNA
Fresh cultures of the bacterial strains were suspended in 1 ml of sterile normal saline. Sputum and oropharyngeal samples or bacterial suspensions were then centrifuged at 10,000g for 10 min. Supernatants were removed, and the resultant pellets were resuspended in 250 ml of a solution containing 10 mM Tris HCl (pH 8.0), 0.1 M NaCl, 1 mM EDTA (pH 8.0), 5% Triton X-100, and 400 mg of lysozyme, and incubated for 30 min at 378C. This was followed by the addition of 250 ml of a solution consisting of 12 mM Tris HCl (pH 8.0), 6 mM EDTA (pH 8.0), 0.5% SDS, and 500 mg of proteinase K, with incubation at 568C for 1 hr. Samples were then boiled for 10 min. This was followed by two steps of organic extraction with phenol-chloroform (vol:vol) and DNA precipitation with 2.5 volumes of cold ethanol and 0.1 volume of sodium acetate (3 M, pH 5.2). After centrifugation, the pellet was washed with ethanol 70%, dried, and solubilized in sterile water, and DNA was quantified in an ultraviolet (UV) spectrophotometer at 260 nm (UltroSpec 3000 UV/Visible Spectrophotometer, Pharmacia, Uppsala, Sweden).
539
Development of Multiplex PCR
Multiplex PCR was optimized to identify three bacterial species (P. aeruginosa, B. cepacia complex, and S. maltophilia) and to include a universal bacterial primer pair, targeting 16S ribosomal DNA, to act as an internal control.27 Multiplex PCR development was based on a PCR protocol designed for P. aeruginosa identification,24 and additional oligonucleotide primers were added to identify B. cepacia complex and S. maltophilia. Oligonucleotide PCR primers included in the multiplex protocol are described in Table 1. Primers for S. maltophilia identification (Malt1/Malt2) were designed to amplify a 149-bp fragment of the chitinase A gene of the bacterium (GenBank access number AF014950), selected with the assistance of Primer31 software (http://www-genome. wi.mit.edu/cgi-bin/primer/primer3_www.cgi). All PCRs were performed using a Mastercycler gradient thermocycler (Eppendorf). Reactions were run on a final volume of 25 ml, containing 2.5 ml of clinical sample or 200 ng of bacterial DNA, 16 mM Tris HCl (pH 8.0), 80 mM KCl, 300 mM of dNTPs, 1.5 mM of MgSO4, 5 ml of PCR Enhancer Solution, 0.5 U of Platinum pfx DNA polymerase (GIBCO-BRL, Gaithersburg, MD), and the following concentrations of primers: 0.6 mM of VIC1/VIC2, 0.4 mM of Malt1/Malt2, 0.8 mM of Eub-16-1/CeMuVi-16-2457, and 0.1 mM of 11E/13B. The cycling parameters consisted of an initial denaturation at 968C for 5 min and 40 cycles at 968C for 1 min, 568C for 1 min, and 688C for 1 min, followed by a final extension step of 7 min at 688C. Following amplification, PCR products were visualized by electrophoresis in 1.5% agarose gel stained with ethidium bromide 0.5 mg/ml, with a UV transilluminator. A negative control (sample with no DNA added) was included in all PCR reactions, as well as a positive control consisting of a mixture of equal amounts (25 ng/ml) of DNA of P. aeruginosa ATCC 27853, B. cepacia
TABLE 1— Oligonucleotide Primers Utilized in Multiplex PCR Protocol Name
Sequence (50 ! 30 )1
VIC1
TTC CCT CGC AGA GAA AAC ATC
VIC2 Malt1 Malt2 Eub-16-1
CCT GGT TGA TCA GGT CGA TCT TAC CAC CCG TAC CTG GAC TT ATC GCA TCG TTG CTG TTG TA AGR GTT YGA TYM TGG CTC AG
CeMuVi-16-2457 11E 13B
CCG RCT GTA TTA GAG CCA GAG GAA GGT GGG GAT GAC GT AGG CCC GGG AAC GTA TTC AC
1
Gene target
PCR product
Pathogen
Reference no.
algD GDP mannose dehydrogenase (GenBank Y00337)
520 bp
P. aeruginosa
chitA (GenBank AF014950)
149 bp
S. maltophilia
Ribossomal 16S (GenBank M22518)
463 bp
B. cepacia complex
15
Ribossomal 16S
233 bp
Eubacteria
27
Degenerate base code is: R ¼ A or G; Y ¼ C or T; M ¼ A or C.
24
540
da Silva Filho et al.
ATCC 25416 (genomovar I), and S. maltophilia ATCC 13637. All PCR reactions were set up in a laminar flow UV hood, according to standard precautions to avoid PCR carry-over.28 Confirmation of Multiplex PCR and Culture Results
All positive results for B. cepacia complex and S. maltophilia (culture and multiplex PCR) were confirmed by two approaches: verification of identification of the bacterial strain with specific PCR, and verification of the result in clinical samples with specific PCR, using different primer pairs described in the literature.17,29 Samples with identification of P. aeruginosa by culture and negative multiplex PCR results were also verified for the identification of isolated strain, using primers VIC1/ VIC2 as described by Silva Filho et al.24 Additionally, 8 samples with identification of P. aeruginosa by multiplex PCR and negative culture results were submitted to DNA sequencing of amplicons, in order to confirm their identity and exclude laboratorial contamination. This procedure was also performed in 5 samples with identification of B. cepacia complex by multiplex PCR but negative culture results. Primers BCR1/BCR217 were used for identification of B. cepacia complex, as described by McDowell et al.,30 while primers SM1/SM4 were selected for S. maltophilia identification, and were used as described by Whitby et al.29 For DNA sequencing, amplicons of 520 bp of P. aeruginosa or 463 bp from B. cepacia complex were cut from agarose gels and purified with a Geneclean II kit (Bio101, Vista, CA). Sequencing was performed using primers VIC1/VIC2 or Eub-16-1/CeMuVi-16-2457 and a Big Dye Terminator kit (Applied Biosystems, Inc., Foster City, CA), followed by automated sequencing on ABI PRISM 377 (Applied Biosystems, Inc.).
primer pairs required several adjustments to the PCR reaction conditions, such as raising buffer ionic strength, changing primers concentrations, using a hot-start polymerase (Platinum pfx) and modifications of cycling conditions. Since Malt1/Malt2 was the only ‘‘new’’ primer pair (never tested before), it was validated by testing against a panel of 12 different bacterial species and an additional 7 samples of S. maltophilia DNA (ATCC 13637 and 6 strains provided by the Microbiology Laboratory) before inclusion in the multiplex PCR protocol, resulting in the appearance of the expected fragment of 149 bp only in the samples containing S. maltophilia DNA (data not shown). The other primer pair included in the reaction (Eub-16-1/CeMuVi-16-2457), selected to identify B. cepacia complex strains, was chosen for its ability to identify the five main genomovars of the B. cepacia complex, its suitable amplicon size, and melting temperature compatible to the multiplex PCR protocol.15 This primer pair was further shown to amplify 16S DNA fragments of the newly described B. cepacia genomovars (B. cepacia genomovar VI, B. ambifaria, B. anthina, and B. pyrrocinia; data not shown). The multiplex PCR method was further tested against a panel of bacteria including P. aeruginosa, B. cepacia complex, S. maltophilia, S. aureus, P. stutzeri, K. pneumoniae, H. influenzae, E. aerogenes, S. epidermidis, A. calcoaceticus, A. xylosoxidans, and E. coli. All samples resulted in the amplification of the 233-bp fragment from the 16S ribosomal DNA, universal for bacteria. Only samples of DNA from P. aeruginosa, S. maltophilia, and B. cepacia complex resulted in additional fragments of 520, 149, and 463 bp, respectively. Additionally, the positive control sample containing a mixture of DNA obtained from these three bacterial species (P. aeruginosa, S. maltophilia, and B. cepacia complex) resulted in the
Statistical Analysis
Sensitivity and specificity of the multiplex PCR method, compared with culture, were calculated using 2 2 contingency tables.31 RESULTS Development of Multiplex PCR
A previously described PCR method for the identification of P. aeruginosa in sputum samples from CF patients was the starting point for the development of the multiplex PCR protocol. This method included primers targeting a 520-bp fragment of the algD GDP mannose dehydrogenase gene of P. aeruginosa and another primer pair targeting a portion of 16S ribosomal DNA, universal for bacteria, to act as control.24 The addition of two new
Fig. 1. Multiplex PCR reaction of DNA samples obtained from: lane 1, S. maltophilia ATCC13637; lane 2, P. aeruginosa ATCC27853; lane 3, B. cepacia ATCC25416 (genomovar I); lane 4, S. aureus ATCC25923; lane 5, H. influenzae (clinical sample); lane 6, E. coli ATCC25922; lane 7, A. xylosoxidans (clinical sample). Lanes 8–10, sputum samples with positive cultures for P. aeruginosa, B. cepacia complex, and S. maltophilia, respectively. Lane 11, Mix of DNA obtained from S. maltophilia, P. aeruginosa, and B. cepacia genomovar I strains, 100 ng each. Lane 12, negative control (no DNA added). Lane M, molecular size markers (100-bp ladder).
Multiplex PCR for Respiratory Pathogens in CF
541
TABLE 2— Bacterial Isolates From 257 Respiratory Tract Cultures From CF Patients Followed at Pediatric Pulmonology Unit of Instituto da Crianc¸a, Sa˜o Paulo, Brazil, From September 2000–April 2001 Pathogen1
Fig. 2. Multiplex PCR reaction of serially diluted DNA of Pseudomonas aeruginosa ATCC27853. Lanes 1–6, P. aeruginosa DNA 500 ng, 50 ng, 5 ng, 0.5 ng, 0.05 ng, and 0.005 ng, respectively. Lane 7, negative control (no DNA added). Lane M, molecular size markers (100-bp ladder).
appearance of four amplicons (520, 463, 233, and 149 bp) (Fig. 1). Sensitivity of the multiplex PCR method was assessed by serial dilutions of DNA of type strains of P. aeruginosa (ATCC 27853), B. cepacia genomovar I (ATCC 25416), and S. maltophilia (ATCC 13637). The method was able to detect 5 pg of P. aeruginosa DNA (Fig. 2), 50 pg of S. maltophilia, and 250 pg of B. cepacia genomovar I DNA, corresponding to approximately 103 colony-forming units (CFU) of P. aeruginosa and 105 CFU of B. cepacia genomovar I. Since the S. maltophilia genome size is unknown, it was not possible to estimate how many CFU of the bacterium correspond to 50 pg of DNA. Given that samples with quite different compositions were tested (sputum and oropharyngeal swabs), the influence of high total DNA concentration in the PCR reaction was also assessed. It was observed that, for sputum samples, total DNA amounts as high as 15 mg were obtained with the selected DNA extraction method, and a 50 or 100 dilution was adequate to produce reliable PCR results (data not shown). Oropharyngeal samples produced significantly less DNA amounts than sputum, and so it was not necessary to dilute samples prior to PCR. Multiplex PCR of Clinical Samples
During the 6-month study period, 257 samples were obtained from 106 CF patients. Culture results of these
Number of samples
%
144 118 62 26 11 9 7 7 7 2 2 1 1 1 1 1 1 1
56.0 45.9 24.1 10.1 4.3 3.5 2.7 2.7 2.7 0.8 0.8 0.4 0.4 0.4 0.4 0.4 0.4 0.4
P. aeruginosa S. aureus H. influenzae H. parainfluenzae B. cepacia complex E. coli S. maltophilia E. cloacae K. pneumoniae Moraxella spp. P. stutzeri E. aerogenes Acinetobacter spp. A. calcoaceticus Proteus mirabilis P. fluorescens Flavimonas oryzihabitans Citrobacter freundii 1
More than one bacterial isolate per sample.
samples identified P. aeruginosa strains in 144/257 samples (56.0%), B. cepacia complex in 11/257 (4.3%), and S. maltophilia in 7/257 (2.7%). The complete list of microorganisms isolated during the study is presented in Table 2. Multiplex PCR was performed in 254 samples (three samples were lost during transport). Pseudomonas aeruginosa was identified in 200/254 samples (78.7%), while B. cepacia complex in 10/254 (3.9%), and S. maltophilia in 8/254 (3.1%) (Table 3). Multiplex PCR identified P. aeruginosa in several samples with concomitant negative culture, and this was observed mainly in samples from throat swabs (53% of PCR-positive samples were negative by culture). P. aeruginosa was identified in 24 patients during the study period only by multiplex PCR. Nine of these patients had never shown positive cultures before. Comparison of culture and multiplex PCR results is displayed in Table 4. Only 4/144 samples with positive culture for P. aeruginosa were negative with multiplex
TABLE 3— Multiplex PCR Results of Respiratory Samples From CF Patients1
Multiplex PCR result
Sputum (n ¼ 141)
Oropharyngeal swabs (n ¼ 113)
Total (n ¼ 254)
Universal for bacteria P. aeruginosa B. cepacia complex S. maltophilia
141 (100) 106 (75.2) 8 (5.7) 5 (3.5)
113 (100) 94 (83.2) 2 (1.8) 3 (2.6)
254 (100) 200 (78.7) 10 (3.9) 8 (3.1)
1
Numbers in parentheses, percentages.
542
da Silva Filho et al. TABLE 4— Comparison of Multiplex PCR and Culture Results of Respiratory Samples From CF Patients Number of samples
P. aeruginosa B. cepacia complex S. maltophilia
Cultureþ, PCRþ
Cultureþ, PCR
Culture, PCRþ
Culture, PCR
Total
140 5 2
4 6 3
60 5 6
50 238 243
254 254 254
PCR, but 60/110 samples with negative culture for P. aeruginosa were positive with the multiplex PCR method. Multiplex PCR was negative in 6/11 samples with positive culture for B. cepacia complex; however, an additional 5 samples among culture-negative samples were positive for B. cepacia complex on multiplex PCR. S. maltophilia was identified by multiplex PCR in 2/5 culture-positive samples, but another 6 samples were identified by multiplex PCR among culture-negative samples. Sensitivity, specificity, and negative and positive predictive values of the multiplex PCR method are shown in Table 5. Confirmation of Multiplex PCR and Culture Results
The identity of bacterial strains obtained from samples with negative multiplex PCR results was confirmed for 2/4 P. aeruginosa strains and for all B. cepacia complex strains; only 1/3 S. maltophilia strains isolated from samples with negative multiplex PCR results had its identity confirmed; one strain was lost in the microbiology laboratory, and the remaining strain was negative in a specific PCR using primers Malt1/Malt2. The use of primers BCR1/BCR2 in a specific PCR protocol described by McDowell et al.30 confirmed all B. cepacia complex-positive results obtained with the multiplex PCR method, even those observed on culturenegative samples. One B. cepacia complex culturepositive sample that was negative on multiplex PCR, however, was positive with primers BCR1/BCR2, although all other B. cepacia complex culture-positive samples (negative in the multiplex PCR method) were negative with primers BCR1/BCR2 (Table 6). A similar ap-
proach to verify culture and multiplex PCR results for S. maltophilia identification, using primers SM1/SM4,29 confirmed 7/8 positive multiplex PCR results. One sample (culture-negative) that was positive in the multiplex PCR was negative when PCR with primers SM1/SM4 was performed, indicating a false-positive result of the multiplex PCR method (Table 7). Besides that, two culturepositive samples that were negative on multiplex PCR were positive in a PCR reaction with primers SM1/SM4. One S. maltophilia strain isolated from one of these samples, furthermore, had its identity confirmed by PCR, using the same primers as in the multiplex PCR protocol (Malt1/Malt2) (Table 7). DNA sequencing was performed on amplicons of 520 bp (corresponding to amplification of the algD GDP mannose dehydrogenase gene of P. aeruginosa) from 8 samples, with identification of P. aeruginosa by multiplex PCR and negative culture results. An additional 5 samples with identification of B. cepacia complex by multiplex PCR but culture-negative results were selected for sequencing of the 463-bp amplicon of the 16S DNA. Sequences were submitted to the BLAST program (Basic Local Alignmennt Search Tool, National Center of Bioinformatics, NCBI),32 and high similarity scores (95–98%) were found with two P. aeruginosa entries contained in GenBank: the P. aeruginosa PAO1 whole genome (AE004774) and the algD GDP mannose gene of P. aeruginosa (Y00337). Similarly, sequences of 16S DNA of B. cepacia complex also showed high similarity scores (98–100%) with several other B. cepacia complex entries contained in GenBank, also including sequences of the newly described genomovars B. anthina, B. ambifaria, and B. pyrrocinia. Alignment of sequences was performed with the ClustalW program (http://www.clustalw.
TABLE 5— Sensitivity, Specificity, and Positive (PPV) and Negative Predictive (NPV) Values of Multiplex PCR Method When Compared to Culture Results, for Each Bacterial Species Studied
P. aeruginosa B. cepacia complex S. maltophilia
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
97.2 45.5 40.0
45.5 97.9 97.6
70.0 50.0 25.0
92.6 97.5 98.8
Multiplex PCR for Respiratory Pathogens in CF
543
TABLE 6— Verifications of Multiplex PCR and Culture Results for B. cepacia Complex
Sample 06 19 21 29 55 68 80 106 130 137 157 176 177 196 199 220 Number of positives
Origin of sample Sputum Sputum Sputum Sputum Oropharynx Oropharynx Sputum Sputum Sputum Sputum Sputum Sputum Oropharynx Sputum Oropharynx Sputum
Culture
Multiplex PCR
Species-specific PCR using primers BCR1/BCR21
þ þ þ þ þ þ þ þ þ þ þ 11
þ þ þ þ þ þ þ þ þ þ 10
þ þ þ þ þ þ þ þ þ þ þ 11
Specific PCR of DNA from strains isolated by culture þ NA þ þ þ NA NA þ þ þ þ þ þ NA NA þ 11
Genomovar status of isolate2 B. vietnamiensis NA B. multivorans Genomovar III (I) Genomovar III (I) NA NA B. multivorans Genomovar III (I) Genomovar III (I) Genomovar III (I) Genomovar III (I) ND NA NA B. vietnamiensis
1
Species-specific PCR performed as described by McDowell et al.30 Determination of genomovar status of 11 B. cepacia complex strains isolated during study,35 performed as described by Whitby et al.;16 method does not differentiate between genomovars I and III. NA, not applicable; ND, genomovar status not determined.
2
genome.ad.jp), and high similarity among fragments of the P. aeruginosa algD GDP mannose dehydrogenase gene was found, with minor nucleotide differences indicating diverse origins instead of PCR contamination (data not shown). Alignment of B. cepacia complex sequences also showed minor nucleotide differences within our samples and between ours and others contained in GenBank, but to a lower extent (1 or 2 positions only). DISCUSSION
We report on a multiplex PCR method suitable for the identification of P. aeruginosa, B. cepacia complex, and S. maltophilia in respiratory samples from CF patients.
Two oligonucleotide primers from the chiA gene were designed to specifically amplify the DNA of S. maltophilia and were incorporated in a multiplex PCR reaction that included an additional three primer pairs, targeting sequences from P. aeruginosa, B. cepacia complex, and a well-conserved region of ribosomal 16S DNA. The method was applied to 254 respiratory samples obtained from CF patients and showed high negative predictive values for the identification of the three pathogens (>90%), and high sensitivity for P. aeruginosa identification, but low sensitivity for B. cepacia complex and S. maltophilia detection, when compared to culture. Although culture of sputum is a reliable and inexpensive method to detect respiratory pathogens in CF patients,
TABLE 7— Verifications of Multiplex PCR and Culture Results for S. Maltophilia1
Sample
Origin of the sample
39 56 65 73 111 113 165 167 189 216 252 Number of positives
Sputum Oropharynx Sputum Oropharynx Sputum Sputum Sputum Sputum Sputum Sputum Oropharynx
1
Culture
Multiplex PCR
Species-specific PCR using primers SM1/SM4
Specific PCR of DNA from strains isolated by culture
þ þ þ þ þ 6
þ þ þ þ þ þ þ þ 8
þ þ þ þ þ þ þ þ þ 9
NA NA NA NA Na þ þ NA þ NA 3
Species-specific PCR performed as described by Whitby et al.29 NA, not applicable; Na, not available.
544
da Silva Filho et al.
the identification of bacterial species such as B. cepacia complex and S. maltophilia is not always a straightforward procedure, and misidentifications have been reported.23,33 McMenamin et al., studying 1,051 bacterial isolates recovered from 608 patients at 115 CF treatment centers in 91 US cities, reported as much as 11% misidentification of B. cepacia complex strains (wrongly assigned as B. cepacia by referring laboratories) and 36% of B. cepacia complex strains among isolates not specifically identified or identified as a species other than B. cepacia complex.33 Accurate identification of these bacteria, however, is essential for adequate care and to avoid cross-infections among CF patients. There are specific recommendations for microbiological approaches when culturing respiratory specimens from CF patients,34 and they include the use of a B. cepacia complex-selective medium. Nevertheless, the use of selective media for B. cepacia complex isolation is not a universally adopted practice among microbiology laboratories involved in CF care. Shreve et al.11 performed a survey to determine microbiology laboratories practices with regard to respiratory specimens from CF patients in 142 North American sites. The use of B. cepacia complex-selective media was reported by 73% of these sites, but over half of the laboratories did not carry out long-term incubation at reduced temperature, conditions which may increase the yield of B. cepacia. Additionally, sites performing partial rather than complete protocols for B. cepacia complex isolation had significantly lower 2-year cumulative prevalence rates for this microorganism.11 During the present study, we introduced the use of selective medium for B. cepacia complex isolation, and this resulted in a significant increase in the isolation of the organism when compared to the previous 2 years,35 although the incubation period was only 48 hr. Isolation of P. aeruginosa in the respiratory tract is almost a rule for adult CF patients, but its initial appearance has significant prognostic implications concerning morbidity and mortality.3,36 Towards that, monitoring airway colonization is a very important aspect of clinical follow-up of CF patients, using sputum or oropharyngeal cultures. Previous studies demonstrated good correlation between sputum and lung tissue cultures.37,38 The accuracy of oropharyngeal cultures in detecting lower respiratory pathogens, however, was addressed by several studies and remains a matter for debate. Ramsey et al. suggested that positive oropharyngeal cultures are highly predictive, but negative cultures do not rule out the presence of pathogens in the lower airway.8 Further studies comparing oropharyngeal and bronchoalveolar lavage cultures concluded that oropharyngeal cultures do not reliably predict the presence of bacterial pathogens in the lower airways of young CF children, but suggested that a negative culture for P. aeruginosa is helpful in discarding lower airway infection with this bacterium.39,40 A recent prospective study of P. aeruginosa infections in young CF children,
however, suggested that infection may occur much earlier than previously found, and a serological response indicating infection rather than colonization can occur very early in life in CF patients.9 The authors also showed genotypical correlation among isolates recovered from oropharyngeal and bronchoalveolar lavage samples. We hypothesized that the use of a method with a higher sensitivity to detect P. aeruginosa would improve its early detection in oropharyngeal samples from CF patients. In fact, multiplex PCR identified 60 P. aeruginosa-positive samples that were negative by culture, and 50/60 of them were oropharyngeal swabs, suggesting that this discrepancy may be related to sampling rather to culture identification failures, although this hypothesis cannot be excluded. Multiplex PCR was also able to identify 24 P. aeruginosa-colonized patients who would not be identified by standard culture, and this was the first evidence of P. aeruginosa colonization for 9 of these patients. Unfortunately, serological response was not assessed to ensure whether or not the multiplex PCR method was the first real evidence of P. aeruginosa colonization. A prospective trial using molecular methods for early P. aeruginosa detection would probably help clarify these questions. Sputum samples that were positive for P. aeruginosa by multiplex PCR but negative by culture, on the other hand, may reflect mainly misidentifications by the Vitek1 system, a situation previously observed.24 The PCR approach for P. aeruginosa detection in respiratory samples from CF patients was previously described by our group,24 and we therefore decided to improve it by adding primers for concomitant detection of B. cepacia complex and S. maltophilia. The development of the multiplex PCR method proved to be the most difficult part of the study, since the addition of new primer pairs demanded several adjustments to the reaction conditions. Toward that, the excellent review by Henegariu et al.41 on critical steps in multiplex PCR reactions provided invaluable help for the development of such a PCR strategy. Increasing the concentration of salt in buffer solution (80 mM KCl, 32 mM Tris-HCl) in this setting was one of the suggestions that resulted in significant improvement of reaction performance (data not shown). The use of PCR adjuvants such as dimethyl sulfoxide (DMSO) and bovine serum albumin (BSA), however, did not result in any improvement and therefore was not adopted. PCR identification of respiratory pathogens in CF patients was described by several authors in the past decade,24,29,30,42–46 but is still used mainly for research purposes. A multiplex PCR method for identification of respiratory pathogens in CF patients was described in 1996 by Karpati et al.,45 targeting P. aeruginosa, S. maltophilia, and B. cepacia, using a set of three primers based on 16S rRNA sequences. The method proved to be highly reproducible and sensitive, and was also applied to direct detection in 90 sputum samples. Sensitivity for
Multiplex PCR for Respiratory Pathogens in CF
P. aeruginosa detection was similar to ours (93%), but sensitivity was poor for S. maltophilia identification, and only one sample was found to be B. cepacia-positive. New approaches for the molecular detection of B. cepacia complex and S. maltophilia were recently described. Identification of B. cepacia complex by amplification of the recA gene17,30 offers the advantage of genomovar determination by further analysis of the amplicon by digestion with restriction enzymes, and this was also shown to be feasible using sputum as the DNA source.30,46 Species-specific PCR detection of S. maltophilia was described by Whitby et al.,29 using 23S rRNA gene as the target, and was also shown to be able to detect the bacterium directly in sputum samples. In the present study, we observed very high negative predictive values for identification of the three bacteria using multiplex PCR (P. aeruginosa, B. cepacia complex, and S. maltophilia), while positive predictive values were relatively low (Table 5). This means that the multiplex PCR method was able to rule out the presence of these bacteria with great confidence, while it was not adequate to establish respiratory colonization by these organisms with the same conviction. The observed low positive predictive values, however, may be influenced by the fact that culture was the ‘‘gold standard’’ method to be compared with the multiplex PCR. The increasing use of molecular methods for pathogen detection has resulted in some uncertainty as to whether culture is the adequate gold standard to evaluate new techniques, since molecular methods are usually able to detect microbial genomes in smaller amounts than those required by traditional microbiology methods,47 and do not rely on the viability of organisms. We therefore decided to use molecular methods of bacterial detection to verify the discordant results of culture and multiplex PCR. Identification of B. cepacia complex was verified by recA gene detection17,30 and S. maltophilia by the PCR strategy described by Whitby et al.,29 since these molecular approaches were tested on a large panel of bacterial strains of each relevant species and showed excellent sensitivity and specificity. Excellent correlation was found between the multiplex PCR method and PCR reactions with primers BCR1/ BCR2 for identification of B. cepacia complex, and only one sample with positive culture but a negative result on multiplex PCR was positive with these primers. This difference may be attributed to the lower sensitivity of multiplex PCR, approximately 105 CFU of B. cepacia complex per PCR tube, while sensitivity of the recA PCR using primers BCR1/BCR2, although defined as the amount of bacteria per gram of sputum (106 CFU g sputum1), was significantly higher.30 Although recA PCR showed a higher sensitivity and better correlation with culture results than the multiplex PCR method, the former was tested in adult CF patients, who had been colonized for a long period and who may have higher
545
levels of B. cepacia complex bacteria in the sputum. The pediatric patients of the present study, in contrast, were likely to be in earlier stages of colonization and therefore may have a lower carriage of the organism. All samples with B. cepacia complex identified only by 16S DNA amplification in multiplex PCR, however, were also positive by recA gene amplification using primers BCR1/ BCR2. Additionally, primers Eub-16-1/CeMuVi-16-2457 were shown to amplify the 16S DNA gene of the newly described B. cepacia genomovars (genomovars VI–IX), and two of these genomovars in particular (B. anthina and B. pyrrocinia) may confound PCR-based diagnostic tools.18 Confirmation of multiplex PCR results for S. maltophilia identification using PCR with primers SM1/ SM4, in contrast, was concordant in only 7/8 samples. One sample multiplex PCR-positive and culture-negative for S. maltophilia was also negative on PCR with primers SM1/SM4 (sample 167, Table 6), indicating the possibility of a false-positive result with the multiplex PCR method. Additionally, two other samples with positive culture for S. maltophilia and negative multiplex PCR results were positive using primers SM1/SM4. Although the use of the primers Malt1/Malt2 resulted in amplification of the expected 149-bp fragment in one S. maltophilia strain isolated from one of these samples, these results suggest that these primers may not be adequately sensitive and/or specific to identify S. maltophilia, and therefore should be tested against a larger panel of S. maltophilia strains from different origins. In this respect, a study by Coenye et al., describing significant genomic heterogeneity within the S. maltophilia species, may explain the low sensitivity observed with the primers Malt1/ Malt2.48 Discordant results of P. aeruginosa identification, on the other hand, were verified by using the same primers employed in the multiplex PCR, since they were extensively tested previously.24 Sequencing of a small subset of amplicons of P. aeruginosa (8 samples) and B. cepacia complex (5 samples) with discordant multiplex PCR/ culture results was additionally performed in order to exclude carry-over contamination. This approach relies on the fact that microbial organisms have defined levels of sequence variation within the amplified region, whereas contaminants tend to be derived from a single source.47 Alignment of sequences obtained from the P. aeruginosa algD GDP mannose dehydrogenase gene showed significant variability, while alignment of B. cepacia complex 16S DNA gene sequences showed minor differences, probably reflecting a higher conservancy of ribosomal genes. The results for P. aeruginosa ruled out the hypothesis of amplicon contamination, whereas for the B. cepacia complex, recA gene amplification of samples with discordant results, in addition to the observed minor nucleotide differences, makes it very unlikely.
546
da Silva Filho et al.
In conclusion, we describe a multiplex PCR protocol for concomitant detection of P. aeruginosa, B. cepacia complex, and S. maltophilia directly in respiratory samples from CF patients. The method showed high negative predictive values for the identification of the three pathogens (>90%), which means that it is suitable for ruling out the presence of these bacteria with great confidence and may be used as an additional tool in infection control practices for CF patients.49 The method also showed high sensitivity for P. aeruginosa identification but low sensitivity for B. cepacia complex and S. maltophilia detection, when compared to culture. Confirmation of the obtained results with other molecular methods described in the literature, however, suggests that the multiplex PCR method is indeed a reliable method for the detection of these three bacterial species in respiratory samples from CF patients. During the study period, multiplex PCR results were not available for the clinical team, and therefore the 9 patients with identification of P. aeruginosa who had never been culture-positive were not submitted to eradication treatment protocols. Unfortunately, all these patients were shown to be colonized by P. aeruginosa in the following 2 years. The clinical performance of this test, concerning early P. aeruginosa detection in oropharyngeal swabs, will be better characterized by prospective trials comparing standard culture and serology. Moreover, the described multiplex method may be the basis for a modified PCR approach presenting higher sensitivity, such as real-time PCR. ACKNOWLEDGMENTS
We thank all the medical staff of the Pediatric Pulmonology Unit of Instituto da Crianc¸a (Sa˜o Paulo, Brazil) for their assistance during the study period. We also thank Dr. Eshwar Mahenthiralingan for supplying strains of the newly described genomovars of the Burkholderia cepacia complex. L.V.F.S.F. was sponsored by CAPES, Brazil. REFERENCES 1. Demko CA, Stern RC, Doershuk CF. Stenotrophomonas maltophilia in cystic fibrosis: incidence and prevalence. Pediatr Pulmonol 1998;25:304–308. 2. Beringer PM, Appleman MD. Unusual respiratory bacterial flora in cystic fibrosis: microbiologic and clinical features. Curr Opin Pulm Med 2000;6:545–550. 3. Hudson VL, Wielinski CL, Regelmann WE. Prognostic implications of initial oropharyngeal bacterial flora in patients with cystic fibrosis diagnosed before the age of two years. J Pediatr 1993;122: 854–860. 4. Nixon GM, Armstrong DS, Carzino R, Carlin JB, Olinsky A, Robertson CF, Grimwood K. Clinical outcome after early Pseudomonas aeruginosa infection in cystic fibrosis. J Pediatr 2001;138:699–704. 5. Govan JR, Deretic V. Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev 1996;60:539–574.
6. Valerius NH, Koch C, Hoiby N. Prevention of chronic Pseudomonas aeruginosa colonisation in cystic fibrosis by early treatment. Lancet 1991;338:725–726. 7. Frederiksen B, Koch C, Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997;23:330–335. 8. Ramsey BW, Wentz KR, Smith AL, Richardson M, WilliamsWarren J, Hedges DL, Gibson R, Redding GJ, Lent K, Harris K. Predictive value of oropharyngeal cultures for identifying lower airway bacteria in cystic fibrosis patients. Am Rev Respir Dis 1991;144:331–337. 9. Burns JL, Gibson RL, McNamara S, Yim D, Emerson J, Rosenfeld M, Hiatt P, McCoy K, Castile R, Smith AL, Ramsey BW. Longitudinal assessment of Pseudomonas aeruginosa in young children with cystic fibrosis. J Infect Dis 2001;183:444– 452. 10. Isles A, MacLusky I, Corey M, Gold R, Prober C, Fleming P, Levison H. Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J Pediatr 1984;104:206–210. 11. Shreve MR, Butler S, Kaplowitz HJ, Rabin HR, Stokes D, Light M, Regelmann WE. Impact of microbiology practice on cumulative prevalence of respiratory tract bacteria in patients with cystic fibrosis. J Clin Microbiol 1999;37:753–757. 12. LiPuma JJ. Burkholderia cepacia. Management issues and new insights. Clin Chest Med 1998;19:473–486. 13. Govan JR. Infection control in cystic fibrosis: methicillinresistant Staphylococcus aureus, Pseudomonas aeruginosa and the Burkholderia cepacia complex. J R Soc Med [Suppl] 2000;93: 40–45. 14. Vandamme P, Holmes B, Vancanneyt M, Coenye T, Hoste B, Coopman R, Revets H, Lauwers S, Gillis M, Kersters K, Govan JR. Occurrence of multiple genomovars of Burkholderia cepacia in cystic fibrosis patients and proposal of Burkholderia multivorans sp. nov. Int J Syst Bacteriol 1997;47:1188–1200. 15. Bauernfeind A, Schneider I, Jungwirth R, Roller C. Discrimination of Burkholderia multivorans and Burkholderia vietnamiensis from Burkholderia cepacia genomovars I, III, and IV by PCR. J Clin Microbiol 1999;37:1335–1339. 16. Whitby PW, Carter KB, Hatter KL, LiPuma JJ, Stull TL. Identification of members of the Burkholderia cepacia complex by species-specific PCR. J Clin Microbiol 2000;38:2962–2965. 17. Mahenthiralingam E, Bischof J, Byrne SK, Radomski C, Davies JE, Av-Gay Y, Vandamme P. DNA-Based diagnostic approaches for identification of Burkholderia cepacia complex, Burkholderia vietnamiensis, Burkholderia multivorans, Burkholderia stabilis, and Burkholderia cepacia genomovars I and III. J Clin Microbiol 2000;38:3165–3173. 18. Vandamme P, Henry D, Coenye T, Nzula S, Vancanneyt M, LiPuma JJ, Speert DP, Govan JR, Mahenthiralingam E. Burkholderia anthina sp. nov. and Burkholderia pyrrocinia, two additional Burkholderia cepacia complex bacteria, may confound results of new molecular diagnostic tools. FEMS Immunol Med Microbiol 2002;33:143–149. 19. Henry DA, Mahenthiralingam E, Vandamme P, Coenye T, Speert DP. Phenotypic methods for determining genomovar status of the Burkholderia cepacia complex. J Clin Microbiol 2001;39:1073–1078. 20. Denton M, Kerr KG. Microbiological and clinical aspects of infection associated with Stenotrophomonas maltophilia. Clin Microbiol Rev 1998;11:57–80. 21. Spencer RC. The emergence of epidemic, multiple-antibioticresistant Stenotrophomonas (Xanthomonas) maltophilia and Burkholderia (Pseudomonas) cepacia. J Hosp Infect [Suppl] 1995;30:453–464.
Multiplex PCR for Respiratory Pathogens in CF 22. Burns JL, Emerson J, Stapp JR, Yim DL, Krzewinski J, Louden L, Ramsey BW, Clausen CR. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin Infect Dis 1998;27:158–163. 23. Burdge DR, Noble MA, Campbell ME, Krell VL, Speert DP. Xanthomonas maltophilia misidentified as Pseudomonas cepacia in cultures of sputum from patients with cystic fibrosis: a diagnostic pitfall with major clinical implications. Clin Infect Dis 1995; 20:445–448. 24. Silva Filho LV, Levi JE, Oda Bento CN, da Silva Ramos SR, Rozov T. PCR identification of Pseudomonas aeruginosa and direct detection in clinical samples from cystic fibrosis patients. J Med Microbiol 1999;48:357–361. 25. Rosenstein BJ, Cutting GR. The diagnosis of cystic fibrosis: a consensus statement. Cystic Fibrosis Foundation Consensus Panel. J Pediatr 1998;132:589–595. 26. Wong K, Roberts MC, Owens L, Fife M, Smith AL. Selective media for the quantitation of bacteria in cystic fibrosis sputum. J Med Microbiol 1984;17:113–119. 27. Relman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent of bacillary angiomatosis. An approach to the identification of uncultured pathogens. N Engl J Med 1990;323:1573– 1580. 28. Kwok S, Higuchi R. Avoiding false positives with PCR. Nature 1989;339:237–238. 29. Whitby PW, Carter KB, Burns JL, Royall JA, LiPuma JJ, Stull TL. Identification and detection of Stenotrophomonas maltophilia by rRNA-directed PCR. J Clin Microbiol 2000;38:4305–4309. 30. McDowell A, Mahenthiralingam E, Moore JE, Dunbar KE, Webb AK, Dodd ME, Martin SL, Millar BC, Scott CJ, Crowe M, Elborn JS. PCR-based detection and identification of Burkholderia cepacia complex pathogens in sputum from cystic fibrosis patients. J Clin Microbiol 2001;39:4247–4255. 31. Dawson-Saunders B, Trapp RG. Basic and clinical biostatistics. Norwalk, CT: Appleton and Lange; 1994. p 234. 32. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol 1990;215:403–410. 33. McMenamin JD, Zaccone TM, Coenye T, Vandamme P, LiPuma JJ. Misidentification of Burkholderia cepacia in US cystic fibrosis treatment centers: an analysis of 1,051 recent sputum isolates. Chest 2000;117:1661–1665. 34. Cystic Fibrosis Foundation. Microbiology and infectious disease in cystic fibrosis. In: Cystic Fibrosis Foundation, editor. Concepts of care. Rockville, MD: Cystic Fibrosis Foundation; 1994. p 1–26. 35. Silva Filho LV, Velloso LF, Bento CN, Gytin E, Tateno AF, Levi JE, Rodrigues JC, Ramos SR. Use of selective medium for Burkholderia cepacia isolation in respiratory samples from cystic fibrosis patients. Rev Inst Med Trop Sao Paulo 2002;44:203–208. 36. Nixon GM, Armstrong DS, Carzino R, Carlin JB, Olinsky A, Robertson CF, Grimwood K. Early airway infection, inflam-
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
547
mation, and lung function in cystic fibrosis. Arch Dis Child 2002;87:306–311. Thomassen MJ, Klinger JD, Badger SJ, van Heeckeren DW, Stern RC. Cultures of thoracotomy specimens confirm usefulness of sputum cultures in cystic fibrosis. J Pediatr 1984;104:352–356. Moller LV, Timens W, van der BW, Kooi K, de Wever B, Dankert J, van Alphen L. Haemophilus influenzae in lung explants of patients with end-stage pulmonary disease. Am J Respir Crit Care Med 1998;157:950–956. Armstrong DS, Grimwood K, Carlin JB, Carzino R, Olinsky A, Phelan PD. Bronchoalveolar lavage or oropharyngeal cultures to identify lower respiratory pathogens in infants with cystic fibrosis. Pediatr Pulmonol 1996;21:267–275. Rosenfeld M, Emerson J, Accurso F, Armstrong D, Castile R, Grimwood K, Hiatt P, McCoy K, McNamara S, Ramsey B, Wagener J. Diagnostic accuracy of oropharyngeal cultures in infants and young children with cystic fibrosis. Pediatr Pulmonol 1999;28:321–328. Henegariu O, Heerema NA, Dlouhy SR, Vance GH, Vogt PH. Multiplex PCR: critical parameters and step-by-step protocol. Biotechniques 1997;23:504–511. McIntosh I, Govan JR, Brock DJ. Detection of Pseudomonas aeruginosa in sputum from cystic fibrosis patients by the polymerase chain reaction. Mol Cell Probes 1992;6:299–304. O’Callaghan EM, Tanner MS, Boulnois GJ. Development of a PCR probe test for identifying Pseudomonas aeruginosa and Pseudomonas (Burkholderia) cepacia. J Clin Pathol 1994;47: 222–226. Campbell PW III, Phillips JA III, Heidecker GJ, Krishnamani MR, Zahorchak R, Stull TL. Detection of Pseudomonas (Burkholderia) cepacia using PCR. Pediatr Pulmonol 1995;20: 44–49. Karpati F, Jonasson J. Polymerase chain reaction for the detection of Pseudomonas aeruginosa, Stenotrophomonas maltophilia and Burkholderia cepacia in sputum of patients with cystic fibrosis. Mol Cell Probes 1996;10:397–403. Drevinek P, Hrbackova H, Cinek O, Bartosova J, Nyc O, Nemec A, Pohunek P. Direct PCR detection of Burkholderia cepacia complex and identification of its genomovars by using sputum as source of DNA. J Clin Microbiol 2002;40:3485–3488. Yolken RH. Nucleic acid amplification assays for microbial diagnosis: challenges and opportunities. J Pediatr 2002;140:290– 292. Coenye T, Vanlaere E, LiPuma J, Vandamme P. Diversity of Stenotrophomonas maltophilia isolates recovered from CF patients. J Cystic Fibrosis [Suppl] 2003;2:42. Saiman L, Siegel J. Infection control recommendations for patients with cystic fibrosis: microbiology, important pathogens, and infection control practices to prevent patient-to-patient transmission. Infect Control Hosp Epidemiol [Suppl] 2003; 24:6–52.
