Use of the Roche LightCycler Strep B Assay for Detection of Group B Streptococcus from Vaginal and Rectal Swabs
http://www.100md.com
微生物临床杂志 2005年第8期
Division of Clinical Microbiology, Department of Laboratory Medicine and Pathology
Department of Obstetrics and Gynecology
Division of Infectious Diseases, Department of Internal Medicine, Mayo Clinic and Mayo Clinic College of Medicine, Rochester, Minnesota
Department of Laboratory Medicine and Pathology and Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota
Institute of Medical Microbiology and Hygiene, University of Regensburg, Regensburg, Germany
ABSTRACT
The results for a real-time PCR assay, using the LightCycler Strep B analyte-specific reagents (Roche Diagnostics Corporation, Indianapolis, Ind.), were compared to a direct plate method combined with a broth enrichment culture method for detection of group B streptococcus colonization in pregnant women. Two separate evaluations were conducted using two different automated nucleic extraction instruments, the MagNA Pure LC instrument (Roche Diagnostics Corporation) and the lower-capacity MagNA Pure Compact instrument (Roche Diagnostics Corporation). The sensitivities, specificities, and positive and negative predictive values for the different evaluation methods were as follow: for the LightCycler Strep B assay with MagNA Pure LC, 100, 97, 90, and 100%, respectively; for the LightCycler Strep B assay with MagNA Pure Compact, 92.5, 99, 97, and 97.5%, respectively. The LightCycler Strep B assay combined with either MagNA Pure LC or MagNA Pure Compact extraction is a suitable method for detecting group B streptococcus colonization in pregnant women. An advantage of the LightCycler assay over culture is the considerably reduced turnaround time for results.
INTRODUCTION
Guidelines from the Centers for Disease Control and Prevention recommend screening for vaginal or rectal carriage of group B streptococcus (GBS) (Streptococcus agalactiae) by culture in pregnant women at 35 to 37 weeks gestation (3). Both the vaginal and anal areas should be swabbed, and the culture method should include an enrichment broth. These recommendations are based on prior studies which have demonstrated that intrapartum surveillance for GBS and treatment of carriers significantly decreases early-onset (within 1 week after birth) GBS meningitis and sepsis in the newborn (2, 6, 7).
Recently, real-time PCR methods have been applied for direct detection of GBS from vaginal/rectal swabs (1, 5). These methods have been demonstrated to be as sensitive as culture but have the added advantage of being much faster. Because of their speed, real-time PCR tests have the potential for screening at the time of delivery. This could decrease the overutilization of antibiotics, especially in pregnant women who do not have prepartum screening for GBS.
The objective of the current study was to evaluate the ability of the LightCycler Strep B ptsI assay (Roche Diagnostics Corporation) to detect GBS carriage in pregnant women. This real-time PCR assay uses dual fluorescent resonance energy transfer (FRET) probes, also referred to as FRET hybridization probes. We compared the performance of this LightCycler assay to the results for a direct plate culture method combined with the results for a broth enrichment culture recommended by the Centers for Disease Control and Prevention. In addition, we evaluated two DNA extraction protocols, one using the MagNA Pure LC instrument (Roche Diagnostics Corporation) and the other using the lower-capacity MagNA Pure Compact instrument (Roche Diagnostics Corporation).
MATERIALS AND METHODS
Study design. The study was approved by the Institutional Review Board of the Mayo Foundation. Two separate evaluations were conducted using the LightCycler Strep B assay, with either the MagNA Pure LC instrument or the MagNA Pure Compact instrument. For each of these evaluations, a different set of vaginal/rectal swabs were collected from pregnant women.
For the MagNA Pure LC/LightCycler Strep B protocol, specimens were collected in June and July 2003. Eighteen specimens (10.3%) were excluded from the study because the patients from whom these swabs were collected declined to provide permission to use their specimens and medical histories for evaluation (Minnesota Statute 144.335). One hundred fifty-nine swabs from 154 patients were available for testing.
For the MagNA Pure Compact/LightCycler Strep B protocol, specimens were collected over the period July through September 2004. Twenty-five specimens (13.6%) were excluded from the study because the patients from whom these swabs were collected declined to provide permission to use their specimens and medical histories for evaluation (Minnesota Statute 144.335). One hundred fifty-nine swabs from 159 patients were available for testing.
Collection of specimens. All specimens were collected by physicians or midwives, by using a double-swab collection and transport system (CultureSwab and Liquid Stuart Double Swab; Becton Dickinson Diagnostic Systems [BDDS], Sparks, Md.). The double-swab specimen was collected by swabbing the vaginal introitus, followed by the rectum using the same swab. Alternatively, either the lower vagina or the rectum was swabbed. The double swab was placed into Stuart's transport medium for delivery to the microbiology laboratory. One swab was used for culture and the other for the real-time PCR assay.
Culture. One of the swabs was used to inoculate a Trypticase soy agar plate with 5% sheep blood (TSAII; BDDS) and a selective broth medium supplemented with colistin (10 μg/ml) and nalidixic acid (15 μg/ml) (LIM broth; BDDS). The plate and broth were incubated at 35 to 37°C with 5 to 7% CO2. After 16 to 24 h, colonies on plates suggestive of group B streptococci (catalase-negative gram-positive cocci with or without a narrow rim of beta-hemolysis) were identified using a latex agglutination test (BBL Streptocard acid latex test; BDDS). LIM broth was subcultured on a TSAII plate incubated under the same conditions. Identification of group B streptococcus was performed from whichever plate (primary TSAII or LIM broth subculture to TSAII) yielded growth of group B streptococcus first. Once group B streptococcus was identified, the culture was concluded. The primary TSAII plate and the LIM broth subculture plate were incubated for 2 days prior to being discarded as negative.
Nucleic acid extraction for LightCycler assay. Material from the swab was extracted and bacteria were lysed using S.E.T.S. II kit (catalog no. 03753158001; Roche Diagnostics Corporation). The swab was broken off in the combined inner and outer S.E.T.S. II tube (Fig. 1) using a BloodBlock protective wipe (Fisher Scientific) to avoid splashing, and the inner tube was capped. Material from the swab was extracted into the larger outer S.E.T.S. II tube containing silica beads by centrifuging at 20,800 x g for 1 min. Following centrifugation, the inner tube containing the swab was discarded. As a biosafety measure, the tubes containing the extracted swab material and beads were capped and heated at 95 to 100°C for 2 min. The tubes were cooled for 1 min at room temperature and processed in a MagNA Lyser (Roche Diagnostics Corporation) for 30 seconds at a speed setting of 6,000 or a Disruptor Genie (Scientific Industries, Bohemia, NY) for 60 seconds to lyse the streptococcus. We determined previously that lysis was equivalent for group B streptococci using either the MagNA Lyser or the Disruptor Genie at the settings described (unpublished data). The lysed material, some of which adhered to the sides and cap of the S.E.T.S. II outer tube after processing by the MagNA Lyser or Disruptor Genie, was collected in the bottom of the tube by centrifugation for 10 seconds at 1,000 x g to 5,000 x g.
Two different instruments were used for extracting DNA from the lysed specimen, the MagNA Pure LC and the MagNA Pure Compact. For extraction using the MagNAPure LC instrument, 200 μl of a mixture of 190 μl S.T.A.R. (stool transport and recovery) buffer (supplied separately [catalog no. 3335208; Roche Diagnostics Corporation [or as part of the S.E.T.S. II kit [catalog no. 03753158001; Roche Diagnostics Corporation]) and 10 μl extraction/PCR internal control (StrepB ptsI gene recovery template [catalog no. 03599463001; Roche Diagnostics Corporation]) were added to the lysate in the outer S.E.T.S. II tube and mixed gently, and 200 μl was removed for extraction. Nucleic acid was extracted using the MagNA Pure LC total nucleic acid isolation kit (Roche Diagnostics Corporation) with an elution volume of 100 μl.
For extraction using the MagNA Pure Compact instrument, 400 μl of a mixture of 395 μl S.T.A.R. buffer and 5 μl of internal control were added and mixed, and 400 μl was removed for extraction. The MagNA Pure Compact nucleic acid isolation kit I was used with an elution volume of 50 μl.
LightCycler Strep B assay. For each reaction mixture, 5 μl of the extracted swab material was added to 15 μl PCR reagent. The PCR reagent contains 2 μl LightCycler FastStart DNA master hybridization probe with enzyme (catalog no. 3003248; Roche Diagnostics Corporation), 2 μl LightCycler StrepB ptsI gene primer/hybridization probes or analyte-specific reagents (catalog no. 03595471001; Roche Diagnostics Corporation), and 11 μl of water. The LightCycler instrument with color compensation enabled was programmed as described in Table 1. LightCycler software version 3.5 was used for all experiments. Sterile water was used as a negative control. A positive control, LightCycler Strep B ptsI template DNA (catalog no. 0359463001; Roche Diagnostics Corporation) and a negative control were included in each run. A positive result was recorded by the technologist if a melting curve from the group B streptococcus hybridization probes was present in the 640-nm channel (F2/backF1) with a melting temperature (Tm) of 59°C ± 2.5°C (Fig. 2 and 3). Specimens were recorded as negative if there was not a melting curve present for group B streptococcus but a melting curve was present for the internal control in the 710-nm channel (F3/backF1) with a Tm of 59.3°C ± 2.5°C (indicating there was no significant extraction or PCR inhibition present in the specimen).
Analytical sensitivity of the LightCycler Strep B assay. Analytical sensitivity was determined using dilutions of group B streptococcus in Stuart's transport medium placed on swabs. The extraction protocols for both the MagNA Pure LC and MagNA Pure Compact instruments were used and the extracts analyzed in triplicate.
Group B streptococcus strain detection by the LightCycler Strep B assay. The ability of the LightCycler Strep B assay to detect a wide variety of well-characterized strains of group B streptococci (kindly provided by Patricia Ferrieri, University of Minnesota) was also studied. These strains are listed in Table 2. In addition, a panel of 105 well-characterized isolates of group B streptococci was tested using the same LightCycler assay in the laboratory of Udo Resichl in Regensburg, Germany.
Analytical specificity panel evaluation by the LightCycler Strep B assay. The analytical specificity of the LightCycler PCR assay was determined by evaluation of isolated colonies of bacteria. Colonies were suspended in water and lysed by boiling, and nucleic acid was extracted using the MagNA Pure LC instrument. To assure that bacterial nucleic acid was present in these extracts, PCR with broad-range 16S rRNA primers was used to determine whether amplifiable DNA was present. The bacteria tested are shown in Table 2 and included a wide range of gram-positive streptococci, as well as other gram-positive and gram-negative bacteria which can colonize or cause infection in the genitourinary or perirectal area.
Clinical evaluation and statistical analysis. Two separate comparative studies were conducted consecutively. The first study evaluated samples where DNA extracts were prepared using the MagNA Pure LC instrument, and the second study evaluated samples using the MagNA Pure Compact instrument. The results obtained using the LightCycler Strep B assay for each of these extraction methods were compared to the combined results obtained using the direct plate and broth enrichment culture methods. Therefore, if either or both of these culture methods were positive for group B streptococcus, the gold standard result was considered positive. Confidence intervals for sensitivity, specificity, and positive and negative predictive values were based on exact binomial probabilities (8).
Arbitration of discordant results for clinical study. In cases where the results for LightCycler Strep B assay were positive but the combined result for plate and broth enrichment culture was negative, a second PCR assay was performed on the archived extracted sample. This assay also used the LightCycler with FRET probe detection; however, the target DNA was the Christi-Atkins-Munch-Petersen (CAMP) factor (cfb) gene (5).
RESULTS
Analytical sensitivity of the LightCycler Strep B assay. Stuart's transport medium spiked to contain 5,000 CFU/ml was positive all three times, and Stuart's transport medium with 2,500 CFU/ml was positive two/three times for both the MagNA Pure LC and MagNA Pure Compact extractions. Stuart's transport medium spiked with 1,250 CFU/ml was positive one/three times with MagNA Pure LC extraction and two/three times with MagNA Pure Compact extraction.
Group B streptococcus strain detection by the LightCycler Strep B assay. All group B streptococcus strains or isolates tested at the Mayo Clinic (Table 2) or at the laboratory of Udo Reischl were detected by the LightCycler Strep B assay. No variation in the Tm of the FRET melting curve was observed in any of the specimens or strains tested.
Analytical specificity panel evaluation by the LightCycler Strep B assay. The LightCycler PCR assay was negative for all of the non-group B streptococcal organisms listed in Table 2.
Clinical evaluation using the MagNA Pure LC instrument. The LightCycler Strep B assay produced more positive results (n = 35) than the gold standard culture method (plate-broth enrichment) (n = 31). The sensitivity, specificity, and predictive values of the LightCycler Strep B assay compared with plate-broth culture are shown in Table 3. The internal control, LightCycler Strep B template DNA, was positive for all specimens tested, indicating the absence of extraction/PCR inhibition.
Four discordant results were observed; in all cases, the LightCycler Strep B assay was positive, but the plate-broth enrichment culture was negative. The crossing points for three of four of these discordant results were 30.5, 30.7, and 30.9; in contrast, the mean crossing point for all the concordant positive samples (LightCycler result agreed with culture result) was 28.8. Retesting the archived swab extracts for these three discordant samples, using the LightCycler assay for the CAMP factor gene, gave positive results, which suggests that the positive specimens were not contaminated with PCR product. The fourth discordant result was weakly positive, and retesting the archived swab extract for this specimen, using the LightCycler assay for the CAMP factor gene, gave a negative result.
Clinical evaluation using the MagNA Pure Compact instrument. The LightCycler Strep B assay produced 38 positive results compared with the gold standard method (plate-broth enrichment), which produced 40 positive results. The sensitivity, specificity, and predictive values compared with plate-broth culture are shown in Table 3. The internal control was positive for all specimens tested, indicating the absence of extraction/PCR inhibition.
Four discordant results were observed. The plate-broth culture method was positive in three cases when the LightCycler Strep B assay was negative. The LightCycler Strep B assay was positive in one case when the plate-broth culture was negative. For this LightCycler Strep B-positive, culture-negative discordant result, the LightCycler assay for the CAMP factor gene was positive.
DISCUSSION
Several previous studies have demonstrated the utility of real-time PCR for rapidly and reliably detecting group B streptococcus colonization in pregnant women (1, 4, 5). The sensitivities of the real-time PCR assays used in these studies compared with an enrichment broth culture method ranged from 90% to 100%. Additionally, in at least two of these studies, it was reported that group B streptococcal DNA was detected slightly more often than it was possible to isolate the organism by culture (1, 4). For one of these studies, the real-time PCR assay was a commercially available assay, the IDI-StrepB (Infectio Diagnostic [IDI], Quebec, Quebec, Canada) used on the SmartCycler instrument (4).
The results of the current study suggest that alternative commercially available reagents, the Roche LightCycler Strep B ptsI gene analyte-specific reagents, used in concert with either the Roche MagNA Pure LC or the Roche MagNA Pure Compact nucleic acid extraction instrument, is a suitable alternative to culture for the detection of group B streptococcus colonization in pregnant women.
For the first clinical evaluation we conducted in which the high-capacity MagNA Pure LC instrument was used for extraction of DNA, the LightCycler Strep B assay had a sensitivity of 100% compared to the gold standard plate-broth enrichment culture. Additionally, four samples were positive by the LightCycler Strep B assay when the culture showed no growth of group B streptococci. A second in-house-developed LightCycler assay designed to detect a different gene in group B streptococcus, the CAMP factor gene, was used to arbitrate these discordant results. Three of the four samples that were positive by the LightCycler Strep B assay were also positive by the second LightCycler assay, which suggests that these results were true-positive results. It is possible that colonies of group B streptococcus grew on culture plates but were missed, as beta-hemolysis was not present. The MagNA Pure LC instrument can extract 32 samples with each run and is therefore suitable for laboratories with high numbers of specimens.
The MagNA Pure Compact instrument was used for extraction of DNA in the second clinical evaluation that we conducted. For this evaluation, the LightCycler Strep B assay had a sensitivity of 92.5% compared with the gold standard plate-broth enrichment culture. An additional specimen was positive by the LightCycler Strep B assay but negative by the plate-broth culture method. A positive result was also obtained for this sample using the in-house-developed LightCycler assay which targets the CAMP factor gene. It is therefore likely that this discordant LightCycler Strep B positive result was a true-positive result and again may have been missed on culture plates due to an absence of beta-hemolysis.
The time required for the LightCycler Strep B assay is approximately 50 min. The time requirement for processing the specimen prior to nucleic acid extraction, including the S.E.T.S. II tube centrifugation and bead beating by the MagNA Lyser was approximately 10 min for a batch of eight specimens. Extraction of DNA by the MagNA Pure LC instrument required 150 min for 32 samples and 40 min for 8 samples. Less time (25 min) was required when the fully loaded MagNA Pure Compact instrument was used (eight samples).
In cases where patients have a history of allergy to penicillin, susceptibility testing is required to discern suitable alternative antibiotics for prophylaxis. We have found that after the swab is removed from the collection tube for PCR, one can add broth to the swab collection tube and then incubate. This method recovers sufficient group B streptococci by culture so that susceptibility testing can be performed. Alternatively, a double swab could be used to collect all specimens, with one swab used for LightCycler PCR and the other used for culture-based susceptibility testing as required.
When testing is performed at 35 to 37 weeks gestation, batching of specimens may be suitable from the bedside perspective as well as from the personnel resource utilization requirements of the laboratory. The value of such testing for women who present in labor must be assessed individually for each institution. For example, if a laboratory cannot run tests on a stat basis, e.g., at nighttime, the most practical approach may still be to set up a plate-broth enrichment culture. On the following day if the plate culture is negative for group B streptococci, the turnaround time for results may be decreased if incubating broth is screened the following day by the LightCycler Strep B assay. However, studies are required to validate this approach. In general, the more quickly a result can be produced, the more quickly antibiotics can be discontinued if provided on an empirical basis. The MagNA Pure Compact instrument should permit a more rapid turnaround time for results than the higher-capacity MagNA Pure LC instrument. This may be particularly useful for samples obtained from women in labor.
In conclusion, the LightCycler Strep B assay, used either with MagNA Pure LC or MagNA Pure Compact instrumentation for nucleic acid extraction, is a suitable method for detecting group B streptococcus colonization in pregnant women. A distinct advantage of the LightCycler Strep B assay over traditional culture-based methods is the rapid turnaround time for results.
ACKNOWLEDGMENTS
JoAnn Brunette is thanked for her efforts in preparing the manuscript. We thank Roche Diagnostics Corporation for supplying the reagents for this study at no expense.
The conclusions of this study were solely those of the authors.
REFERENCES
Bergeron, M. G., D. Ke, C. Menard, F. J. Francois, M. Gagnon, M. Bernier, M. Ouellette, P. H. Roy, S. Marcoux, and W. D. Fraser. 2000. Rapid detection of group B streptococci in pregnant women at delivery. N. Engl. J. Med. 343:175-179.
Boyer, K. M., and S. P. Gotoff. 1986. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. N. Engl. J. Med. 314:1665-1669.
Centers for Disease Control and Prevention. 2002. Prevention of prenatal group B streptococcal disease. Morb. Mortal. Wkly. Rep. 51(RR-11):1-22.
Davies, H. D., M. A. Miller, S. Faro, D. Gregson, S. C. Kehl, and J. A. Jordan.2004 . Multicenter study of a rapid molecular-based assay for the diagnosis of group B Streptococcus colonization in pregnant women. Clin. Infect. Dis. 39:1129-1135.
Ke, D., C. Menard, F. J. Picard, M. Boissinot, M. Ouellette, P. H. Roy, and M. G. Bergeron.2000 . Development of conventional and real-time PCR assays for the rapid detection of group B streptococci. Clin. Chem. 46:324-331.
Mohle-Boetani, J. C., A. Schuchat, B. D. Plikaytis, J. D. Smith, and C. V. Broome. 1993. Comparison of prevention strategies for neonatal group B streptococcal infection. A population-based economic analysis. JAMA 270:1442-1448.
Rouse, D. J., R. L. Goldenberg, S. P. Cliver, G. R. Cutter, S. T. Mennemeyer, and C. A. Fargason, Jr. 1994. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis.Obstet. Gynecol. 83:483-494.
Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York, N.Y.(James R. Uhl, Emily A. Ve)
Department of Obstetrics and Gynecology
Division of Infectious Diseases, Department of Internal Medicine, Mayo Clinic and Mayo Clinic College of Medicine, Rochester, Minnesota
Department of Laboratory Medicine and Pathology and Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota
Institute of Medical Microbiology and Hygiene, University of Regensburg, Regensburg, Germany
ABSTRACT
The results for a real-time PCR assay, using the LightCycler Strep B analyte-specific reagents (Roche Diagnostics Corporation, Indianapolis, Ind.), were compared to a direct plate method combined with a broth enrichment culture method for detection of group B streptococcus colonization in pregnant women. Two separate evaluations were conducted using two different automated nucleic extraction instruments, the MagNA Pure LC instrument (Roche Diagnostics Corporation) and the lower-capacity MagNA Pure Compact instrument (Roche Diagnostics Corporation). The sensitivities, specificities, and positive and negative predictive values for the different evaluation methods were as follow: for the LightCycler Strep B assay with MagNA Pure LC, 100, 97, 90, and 100%, respectively; for the LightCycler Strep B assay with MagNA Pure Compact, 92.5, 99, 97, and 97.5%, respectively. The LightCycler Strep B assay combined with either MagNA Pure LC or MagNA Pure Compact extraction is a suitable method for detecting group B streptococcus colonization in pregnant women. An advantage of the LightCycler assay over culture is the considerably reduced turnaround time for results.
INTRODUCTION
Guidelines from the Centers for Disease Control and Prevention recommend screening for vaginal or rectal carriage of group B streptococcus (GBS) (Streptococcus agalactiae) by culture in pregnant women at 35 to 37 weeks gestation (3). Both the vaginal and anal areas should be swabbed, and the culture method should include an enrichment broth. These recommendations are based on prior studies which have demonstrated that intrapartum surveillance for GBS and treatment of carriers significantly decreases early-onset (within 1 week after birth) GBS meningitis and sepsis in the newborn (2, 6, 7).
Recently, real-time PCR methods have been applied for direct detection of GBS from vaginal/rectal swabs (1, 5). These methods have been demonstrated to be as sensitive as culture but have the added advantage of being much faster. Because of their speed, real-time PCR tests have the potential for screening at the time of delivery. This could decrease the overutilization of antibiotics, especially in pregnant women who do not have prepartum screening for GBS.
The objective of the current study was to evaluate the ability of the LightCycler Strep B ptsI assay (Roche Diagnostics Corporation) to detect GBS carriage in pregnant women. This real-time PCR assay uses dual fluorescent resonance energy transfer (FRET) probes, also referred to as FRET hybridization probes. We compared the performance of this LightCycler assay to the results for a direct plate culture method combined with the results for a broth enrichment culture recommended by the Centers for Disease Control and Prevention. In addition, we evaluated two DNA extraction protocols, one using the MagNA Pure LC instrument (Roche Diagnostics Corporation) and the other using the lower-capacity MagNA Pure Compact instrument (Roche Diagnostics Corporation).
MATERIALS AND METHODS
Study design. The study was approved by the Institutional Review Board of the Mayo Foundation. Two separate evaluations were conducted using the LightCycler Strep B assay, with either the MagNA Pure LC instrument or the MagNA Pure Compact instrument. For each of these evaluations, a different set of vaginal/rectal swabs were collected from pregnant women.
For the MagNA Pure LC/LightCycler Strep B protocol, specimens were collected in June and July 2003. Eighteen specimens (10.3%) were excluded from the study because the patients from whom these swabs were collected declined to provide permission to use their specimens and medical histories for evaluation (Minnesota Statute 144.335). One hundred fifty-nine swabs from 154 patients were available for testing.
For the MagNA Pure Compact/LightCycler Strep B protocol, specimens were collected over the period July through September 2004. Twenty-five specimens (13.6%) were excluded from the study because the patients from whom these swabs were collected declined to provide permission to use their specimens and medical histories for evaluation (Minnesota Statute 144.335). One hundred fifty-nine swabs from 159 patients were available for testing.
Collection of specimens. All specimens were collected by physicians or midwives, by using a double-swab collection and transport system (CultureSwab and Liquid Stuart Double Swab; Becton Dickinson Diagnostic Systems [BDDS], Sparks, Md.). The double-swab specimen was collected by swabbing the vaginal introitus, followed by the rectum using the same swab. Alternatively, either the lower vagina or the rectum was swabbed. The double swab was placed into Stuart's transport medium for delivery to the microbiology laboratory. One swab was used for culture and the other for the real-time PCR assay.
Culture. One of the swabs was used to inoculate a Trypticase soy agar plate with 5% sheep blood (TSAII; BDDS) and a selective broth medium supplemented with colistin (10 μg/ml) and nalidixic acid (15 μg/ml) (LIM broth; BDDS). The plate and broth were incubated at 35 to 37°C with 5 to 7% CO2. After 16 to 24 h, colonies on plates suggestive of group B streptococci (catalase-negative gram-positive cocci with or without a narrow rim of beta-hemolysis) were identified using a latex agglutination test (BBL Streptocard acid latex test; BDDS). LIM broth was subcultured on a TSAII plate incubated under the same conditions. Identification of group B streptococcus was performed from whichever plate (primary TSAII or LIM broth subculture to TSAII) yielded growth of group B streptococcus first. Once group B streptococcus was identified, the culture was concluded. The primary TSAII plate and the LIM broth subculture plate were incubated for 2 days prior to being discarded as negative.
Nucleic acid extraction for LightCycler assay. Material from the swab was extracted and bacteria were lysed using S.E.T.S. II kit (catalog no. 03753158001; Roche Diagnostics Corporation). The swab was broken off in the combined inner and outer S.E.T.S. II tube (Fig. 1) using a BloodBlock protective wipe (Fisher Scientific) to avoid splashing, and the inner tube was capped. Material from the swab was extracted into the larger outer S.E.T.S. II tube containing silica beads by centrifuging at 20,800 x g for 1 min. Following centrifugation, the inner tube containing the swab was discarded. As a biosafety measure, the tubes containing the extracted swab material and beads were capped and heated at 95 to 100°C for 2 min. The tubes were cooled for 1 min at room temperature and processed in a MagNA Lyser (Roche Diagnostics Corporation) for 30 seconds at a speed setting of 6,000 or a Disruptor Genie (Scientific Industries, Bohemia, NY) for 60 seconds to lyse the streptococcus. We determined previously that lysis was equivalent for group B streptococci using either the MagNA Lyser or the Disruptor Genie at the settings described (unpublished data). The lysed material, some of which adhered to the sides and cap of the S.E.T.S. II outer tube after processing by the MagNA Lyser or Disruptor Genie, was collected in the bottom of the tube by centrifugation for 10 seconds at 1,000 x g to 5,000 x g.
Two different instruments were used for extracting DNA from the lysed specimen, the MagNA Pure LC and the MagNA Pure Compact. For extraction using the MagNAPure LC instrument, 200 μl of a mixture of 190 μl S.T.A.R. (stool transport and recovery) buffer (supplied separately [catalog no. 3335208; Roche Diagnostics Corporation [or as part of the S.E.T.S. II kit [catalog no. 03753158001; Roche Diagnostics Corporation]) and 10 μl extraction/PCR internal control (StrepB ptsI gene recovery template [catalog no. 03599463001; Roche Diagnostics Corporation]) were added to the lysate in the outer S.E.T.S. II tube and mixed gently, and 200 μl was removed for extraction. Nucleic acid was extracted using the MagNA Pure LC total nucleic acid isolation kit (Roche Diagnostics Corporation) with an elution volume of 100 μl.
For extraction using the MagNA Pure Compact instrument, 400 μl of a mixture of 395 μl S.T.A.R. buffer and 5 μl of internal control were added and mixed, and 400 μl was removed for extraction. The MagNA Pure Compact nucleic acid isolation kit I was used with an elution volume of 50 μl.
LightCycler Strep B assay. For each reaction mixture, 5 μl of the extracted swab material was added to 15 μl PCR reagent. The PCR reagent contains 2 μl LightCycler FastStart DNA master hybridization probe with enzyme (catalog no. 3003248; Roche Diagnostics Corporation), 2 μl LightCycler StrepB ptsI gene primer/hybridization probes or analyte-specific reagents (catalog no. 03595471001; Roche Diagnostics Corporation), and 11 μl of water. The LightCycler instrument with color compensation enabled was programmed as described in Table 1. LightCycler software version 3.5 was used for all experiments. Sterile water was used as a negative control. A positive control, LightCycler Strep B ptsI template DNA (catalog no. 0359463001; Roche Diagnostics Corporation) and a negative control were included in each run. A positive result was recorded by the technologist if a melting curve from the group B streptococcus hybridization probes was present in the 640-nm channel (F2/backF1) with a melting temperature (Tm) of 59°C ± 2.5°C (Fig. 2 and 3). Specimens were recorded as negative if there was not a melting curve present for group B streptococcus but a melting curve was present for the internal control in the 710-nm channel (F3/backF1) with a Tm of 59.3°C ± 2.5°C (indicating there was no significant extraction or PCR inhibition present in the specimen).
Analytical sensitivity of the LightCycler Strep B assay. Analytical sensitivity was determined using dilutions of group B streptococcus in Stuart's transport medium placed on swabs. The extraction protocols for both the MagNA Pure LC and MagNA Pure Compact instruments were used and the extracts analyzed in triplicate.
Group B streptococcus strain detection by the LightCycler Strep B assay. The ability of the LightCycler Strep B assay to detect a wide variety of well-characterized strains of group B streptococci (kindly provided by Patricia Ferrieri, University of Minnesota) was also studied. These strains are listed in Table 2. In addition, a panel of 105 well-characterized isolates of group B streptococci was tested using the same LightCycler assay in the laboratory of Udo Resichl in Regensburg, Germany.
Analytical specificity panel evaluation by the LightCycler Strep B assay. The analytical specificity of the LightCycler PCR assay was determined by evaluation of isolated colonies of bacteria. Colonies were suspended in water and lysed by boiling, and nucleic acid was extracted using the MagNA Pure LC instrument. To assure that bacterial nucleic acid was present in these extracts, PCR with broad-range 16S rRNA primers was used to determine whether amplifiable DNA was present. The bacteria tested are shown in Table 2 and included a wide range of gram-positive streptococci, as well as other gram-positive and gram-negative bacteria which can colonize or cause infection in the genitourinary or perirectal area.
Clinical evaluation and statistical analysis. Two separate comparative studies were conducted consecutively. The first study evaluated samples where DNA extracts were prepared using the MagNA Pure LC instrument, and the second study evaluated samples using the MagNA Pure Compact instrument. The results obtained using the LightCycler Strep B assay for each of these extraction methods were compared to the combined results obtained using the direct plate and broth enrichment culture methods. Therefore, if either or both of these culture methods were positive for group B streptococcus, the gold standard result was considered positive. Confidence intervals for sensitivity, specificity, and positive and negative predictive values were based on exact binomial probabilities (8).
Arbitration of discordant results for clinical study. In cases where the results for LightCycler Strep B assay were positive but the combined result for plate and broth enrichment culture was negative, a second PCR assay was performed on the archived extracted sample. This assay also used the LightCycler with FRET probe detection; however, the target DNA was the Christi-Atkins-Munch-Petersen (CAMP) factor (cfb) gene (5).
RESULTS
Analytical sensitivity of the LightCycler Strep B assay. Stuart's transport medium spiked to contain 5,000 CFU/ml was positive all three times, and Stuart's transport medium with 2,500 CFU/ml was positive two/three times for both the MagNA Pure LC and MagNA Pure Compact extractions. Stuart's transport medium spiked with 1,250 CFU/ml was positive one/three times with MagNA Pure LC extraction and two/three times with MagNA Pure Compact extraction.
Group B streptococcus strain detection by the LightCycler Strep B assay. All group B streptococcus strains or isolates tested at the Mayo Clinic (Table 2) or at the laboratory of Udo Reischl were detected by the LightCycler Strep B assay. No variation in the Tm of the FRET melting curve was observed in any of the specimens or strains tested.
Analytical specificity panel evaluation by the LightCycler Strep B assay. The LightCycler PCR assay was negative for all of the non-group B streptococcal organisms listed in Table 2.
Clinical evaluation using the MagNA Pure LC instrument. The LightCycler Strep B assay produced more positive results (n = 35) than the gold standard culture method (plate-broth enrichment) (n = 31). The sensitivity, specificity, and predictive values of the LightCycler Strep B assay compared with plate-broth culture are shown in Table 3. The internal control, LightCycler Strep B template DNA, was positive for all specimens tested, indicating the absence of extraction/PCR inhibition.
Four discordant results were observed; in all cases, the LightCycler Strep B assay was positive, but the plate-broth enrichment culture was negative. The crossing points for three of four of these discordant results were 30.5, 30.7, and 30.9; in contrast, the mean crossing point for all the concordant positive samples (LightCycler result agreed with culture result) was 28.8. Retesting the archived swab extracts for these three discordant samples, using the LightCycler assay for the CAMP factor gene, gave positive results, which suggests that the positive specimens were not contaminated with PCR product. The fourth discordant result was weakly positive, and retesting the archived swab extract for this specimen, using the LightCycler assay for the CAMP factor gene, gave a negative result.
Clinical evaluation using the MagNA Pure Compact instrument. The LightCycler Strep B assay produced 38 positive results compared with the gold standard method (plate-broth enrichment), which produced 40 positive results. The sensitivity, specificity, and predictive values compared with plate-broth culture are shown in Table 3. The internal control was positive for all specimens tested, indicating the absence of extraction/PCR inhibition.
Four discordant results were observed. The plate-broth culture method was positive in three cases when the LightCycler Strep B assay was negative. The LightCycler Strep B assay was positive in one case when the plate-broth culture was negative. For this LightCycler Strep B-positive, culture-negative discordant result, the LightCycler assay for the CAMP factor gene was positive.
DISCUSSION
Several previous studies have demonstrated the utility of real-time PCR for rapidly and reliably detecting group B streptococcus colonization in pregnant women (1, 4, 5). The sensitivities of the real-time PCR assays used in these studies compared with an enrichment broth culture method ranged from 90% to 100%. Additionally, in at least two of these studies, it was reported that group B streptococcal DNA was detected slightly more often than it was possible to isolate the organism by culture (1, 4). For one of these studies, the real-time PCR assay was a commercially available assay, the IDI-StrepB (Infectio Diagnostic [IDI], Quebec, Quebec, Canada) used on the SmartCycler instrument (4).
The results of the current study suggest that alternative commercially available reagents, the Roche LightCycler Strep B ptsI gene analyte-specific reagents, used in concert with either the Roche MagNA Pure LC or the Roche MagNA Pure Compact nucleic acid extraction instrument, is a suitable alternative to culture for the detection of group B streptococcus colonization in pregnant women.
For the first clinical evaluation we conducted in which the high-capacity MagNA Pure LC instrument was used for extraction of DNA, the LightCycler Strep B assay had a sensitivity of 100% compared to the gold standard plate-broth enrichment culture. Additionally, four samples were positive by the LightCycler Strep B assay when the culture showed no growth of group B streptococci. A second in-house-developed LightCycler assay designed to detect a different gene in group B streptococcus, the CAMP factor gene, was used to arbitrate these discordant results. Three of the four samples that were positive by the LightCycler Strep B assay were also positive by the second LightCycler assay, which suggests that these results were true-positive results. It is possible that colonies of group B streptococcus grew on culture plates but were missed, as beta-hemolysis was not present. The MagNA Pure LC instrument can extract 32 samples with each run and is therefore suitable for laboratories with high numbers of specimens.
The MagNA Pure Compact instrument was used for extraction of DNA in the second clinical evaluation that we conducted. For this evaluation, the LightCycler Strep B assay had a sensitivity of 92.5% compared with the gold standard plate-broth enrichment culture. An additional specimen was positive by the LightCycler Strep B assay but negative by the plate-broth culture method. A positive result was also obtained for this sample using the in-house-developed LightCycler assay which targets the CAMP factor gene. It is therefore likely that this discordant LightCycler Strep B positive result was a true-positive result and again may have been missed on culture plates due to an absence of beta-hemolysis.
The time required for the LightCycler Strep B assay is approximately 50 min. The time requirement for processing the specimen prior to nucleic acid extraction, including the S.E.T.S. II tube centrifugation and bead beating by the MagNA Lyser was approximately 10 min for a batch of eight specimens. Extraction of DNA by the MagNA Pure LC instrument required 150 min for 32 samples and 40 min for 8 samples. Less time (25 min) was required when the fully loaded MagNA Pure Compact instrument was used (eight samples).
In cases where patients have a history of allergy to penicillin, susceptibility testing is required to discern suitable alternative antibiotics for prophylaxis. We have found that after the swab is removed from the collection tube for PCR, one can add broth to the swab collection tube and then incubate. This method recovers sufficient group B streptococci by culture so that susceptibility testing can be performed. Alternatively, a double swab could be used to collect all specimens, with one swab used for LightCycler PCR and the other used for culture-based susceptibility testing as required.
When testing is performed at 35 to 37 weeks gestation, batching of specimens may be suitable from the bedside perspective as well as from the personnel resource utilization requirements of the laboratory. The value of such testing for women who present in labor must be assessed individually for each institution. For example, if a laboratory cannot run tests on a stat basis, e.g., at nighttime, the most practical approach may still be to set up a plate-broth enrichment culture. On the following day if the plate culture is negative for group B streptococci, the turnaround time for results may be decreased if incubating broth is screened the following day by the LightCycler Strep B assay. However, studies are required to validate this approach. In general, the more quickly a result can be produced, the more quickly antibiotics can be discontinued if provided on an empirical basis. The MagNA Pure Compact instrument should permit a more rapid turnaround time for results than the higher-capacity MagNA Pure LC instrument. This may be particularly useful for samples obtained from women in labor.
In conclusion, the LightCycler Strep B assay, used either with MagNA Pure LC or MagNA Pure Compact instrumentation for nucleic acid extraction, is a suitable method for detecting group B streptococcus colonization in pregnant women. A distinct advantage of the LightCycler Strep B assay over traditional culture-based methods is the rapid turnaround time for results.
ACKNOWLEDGMENTS
JoAnn Brunette is thanked for her efforts in preparing the manuscript. We thank Roche Diagnostics Corporation for supplying the reagents for this study at no expense.
The conclusions of this study were solely those of the authors.
REFERENCES
Bergeron, M. G., D. Ke, C. Menard, F. J. Francois, M. Gagnon, M. Bernier, M. Ouellette, P. H. Roy, S. Marcoux, and W. D. Fraser. 2000. Rapid detection of group B streptococci in pregnant women at delivery. N. Engl. J. Med. 343:175-179.
Boyer, K. M., and S. P. Gotoff. 1986. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. N. Engl. J. Med. 314:1665-1669.
Centers for Disease Control and Prevention. 2002. Prevention of prenatal group B streptococcal disease. Morb. Mortal. Wkly. Rep. 51(RR-11):1-22.
Davies, H. D., M. A. Miller, S. Faro, D. Gregson, S. C. Kehl, and J. A. Jordan.2004 . Multicenter study of a rapid molecular-based assay for the diagnosis of group B Streptococcus colonization in pregnant women. Clin. Infect. Dis. 39:1129-1135.
Ke, D., C. Menard, F. J. Picard, M. Boissinot, M. Ouellette, P. H. Roy, and M. G. Bergeron.2000 . Development of conventional and real-time PCR assays for the rapid detection of group B streptococci. Clin. Chem. 46:324-331.
Mohle-Boetani, J. C., A. Schuchat, B. D. Plikaytis, J. D. Smith, and C. V. Broome. 1993. Comparison of prevention strategies for neonatal group B streptococcal infection. A population-based economic analysis. JAMA 270:1442-1448.
Rouse, D. J., R. L. Goldenberg, S. P. Cliver, G. R. Cutter, S. T. Mennemeyer, and C. A. Fargason, Jr. 1994. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis.Obstet. Gynecol. 83:483-494.
Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York, N.Y.(James R. Uhl, Emily A. Ve)