- Email: [email protected]
Modified Granada Agar Medium for the detection of group B streptococcus carriage in pregnant women
ORIGINAL AR TICLE Modified Granada Agar Medium for the detection of group B streptococcus carriage in pregnant women G. Claeys1, G. Verschraegen1 and M. Temmerman2 Departments of 1Microbiology, and 2Obstetrics and Gynecology, Ghent University, Ghent, Belgium Objectives To improve the detection rate of group B streptococci (GBS) in pregnant women, aiming at the prevention of early-onset septicemia in the newborn. Methods The yield from culturing two sites, vaginal and anorectal, on a Modi®ed Granada Medium (MGM)
was compared with our standard approach of culturing a vaginal swab on blood agar (BA).
Results Samples were processed from 430 consecutive pregnant women. GBS was isolated from the vagina in
11.6% with BA, and in 13.7% with MGM. In 17.0% of anorectal samples, GBS was identi®ed with MGM. The combination of both sites and media had a yield of 20.0%. MGM identi®ed all but six (2%) of 310 GBS strains after aerobic incubation, with use of a cover slide, and missed only three strains (1%) after anaerobic incubation.
Conclusions Separate culture of vaginal and anorectal samples using the same MGM agar plate resulted in an
increase in detection rate for GBS of 76% as compared to BA alone. The technique is simple and results are available after overnight incubation. MGM was con®rmed as a speci®c medium for the identi®cation of GBS, with a sensitivity of 98±99%.
Keywords GBS, group B streptococcus, Streptococcus agalactiae, early-onset sepsis, prevention, perinatal infection, pregnancy Accepted 21 August 2000
Clin Microbiol Infect 2001; 7: 22±24
IN T R ODUC T ION Infections with group B streptococcus (GBS) or Streptococcus agalactiae are leading causes of neonatal morbidity and mortality. Some of these infections are preventable by intrapartum administration of antibiotics. Consensus guidelines have been formulated, and two strategies have been advocated [1,2], one being risk-based, and the other screening-based. With use of a culture-based approach, more women are treated with antibiotics and more infections can be prevented. In a recent analysis, a striking 65% decline in the number of early-onset disease cases was observed in a surveillance area in the USA. The decline coincided with an active policy of promotion of the above-mentioned guidelines [3]. For the culture-based approach, combined cultures from the distal vagina and the anorectum at 35±37 weeks of gestation are recommended, including a selective agar and a selective broth, to be subcultured on blood agar (BA) after 24 h of incubation [1,2].
Modi®ed Granada Agar Medium (MGM) is a selective and differential medium for the detection of GBS, producing orange carotenoid pigmented colonies under anaerobic conditions. The methotrexate added to the medium enhances the pigment production [4]. Only very few GBS strains are non-hemolytic and non-pigmented [5]. In a study of 800 vaginal and 450 vagino-anorectal samples of pregnant women, MGM and the selective broth were equally sensitive for detection of GBS in vaginal samples, and MGM was more sensitive in vaginoanorectal samples [6]. With the introduction of a screening-based prophylaxis strategy in our hospital, we decided to compare the culture of a vaginal and an anorectal sample on MGM with our hitherto used method of a vaginal swab solely plated on tryptic soy agar with 5% sheep blood. We report our ®ndings on the ®rst 430 patients. M AT ER I AL S AND ME T HOD S Specimens and patients
Corresponding author and reprint requests: G. Claeys, Laboratory for Microbiology, University Hospital, De Pintelaan 185, 9000 Ghent, Belgium Tel: 32 92 403645 Fax: 32 92 403659 E-mail: [email protected]
Separate samples from the vagina and the anorectum were taken from 430 pregnant women in gestational week 35±37. The samples were sent to the laboratory in Amies transport medium, and processed within 4 h.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases
Claeys et al Modified Granada Agar Medium for the detection of group B streptococcus
Processing of clinical samples The vaginal and anorectal samples were swabbed separately on both halves of an MGM plate, home-made as originally formulated [4], and incubated in an anaerobic jar for 18 h. From the vaginal sample a Gram stain was done; tryptic soy agar with 5% sheep blood (BA), enriched chocolate agar and gonococcal agar was inoculated, incubated in a 5% CO2 incubator, read after 1 and 2 days, and processed according to accepted methods [7]. After 1 day, b- and non-hemolytic colonies with the appearance of GBS were con®rmed with the CAMP test. When typical colonies were present on the BA, orange-pigmented colonies from the MGM were identi®ed as GBS; when no b-hemolytic colonies were present on BA, pigmented colonies were subjected to a CAMP test [7]. No enrichment broth was used. The yields of the three cultures (BA vagina, MGM vagina, MGM anorectal) were compared.
GBS strains A series of 310 GBS strains, including isolates from non-study samples, was tested for pigment production. The strains were inoculated on two MGM plates, and incubated overnight. One MGM was incubated anaerobically, and the other was incubated in an ambient atmosphere after placement of a small (18 18 mm) cover slide upon the inoculum [6].
R E S ULT S Vaginal and anorectal samples were obtained from 430 pregnant women. GBS was detected in 86, which means that 20% of the women in the study were GBS carriers when using the new approach (Table 1). Of the 430 vaginal samples, 50 were GBS positive on BA (11.6%), compared to 59 with MGM (13.7%). Of the 86 GBS carriers, 58% were detected on BA and 68.5% were detected on MGM. No samples with GBS on BA were GBS negative on MGM.
23
Table 1 Comparison of GBS detected using one or two samples (vaginal, anorectal) cultured on non-selective (blood agar, BA) and/or selective medium (Modified Granada Medium, MGM)
Site
Medium
Vaginal
BA MGM MGM
Anorectal Both
Number (%) of pregnant women with a positive culture for GBS
Number (%) of carriers detected with sample/medium combination
50/430 (11.6%) 59/430 (13.7%) 73/430 (17%) 86/630 (20%)
50/86 (58%) 59/86 (68.5%) 73/86 (85%) 86/86 (100%)
Cultures of the 430 anorectal samples on MGM yielded GBS in 73 cases (17%), which implies a detection rate of 85%. In 13 women the vaginal sample was positive and the anorectal sample was negative, whereas 27 women had positive anorectal and negative vaginal cultures. As a consequence, the maximum number of GBS carriers was detected by combined culture of the two samples on MGM. A collection of 310 GBS strains, including isolates from nonstudy samples, was tested for hemolytic properties, reactions in the CAMP test, and pigment production under two different conditions. Three hundred and three strains were positive with all four tests. Abnormal reactions were found in seven isolates (Table 2). Six strains (2%) did not produce colored colonies when incubated aerobically; of these, three (1%) were pigmented when grown anaerobically. Six isolates (2%) were non-hemolytic, and this always corresponded to an absent or very weak pigmentation. Two isolates, one hemolytic and one non-hemolytic, were CAMP negative. All aberrant isolates had the group B Lance®eld antigen, as demonstrated by a coagglutination test. DI SC US S ION The purpose of our study was to compare the number of GBS carriers in pregnant women at 35±37 weeks of gestation
Table 2 Hemolytic and pigmentation characteristics of 310 GBS isolates
b-Hemolysis
CAMP reaction
Pigmentation on MGM incubated aerobicallya
^ ^ ^
^ ^
^ ^ ^
a
Coverslip placed upon the inoculum.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7, 22±24
Pigmentation on MGM incubated anaerobically
Number (%) of isolates
^ ^
303 (97%) 1 (< 1%) 1 (< 1%) 2 (1%) 3 (1%)
24 Clinical Microbiology and Infection, Volume 7 Number 1, January 2001
detected by plating a vaginal sample on BA to the use of MGM for a vaginal and an anorectal sample. We did not use enrichment culture in selective broth medium, as recommended in the CDC guidelines [1,2], because previous reports have shown equal or better sensitivity of MGM [4,6,8,9]. Furthermore, by using MGM, results are obtained faster and the workload is lower (no subcultures, and no identi®cation tests needed). The reason why enrichment culture does not have a better yield than MGM is probably that the selective broth, containing nalidixic acid and either gentamicin or colistin, also stimulates the growth of enterococci, competing with GBS. By using MGM, a selective differential medium, and the testing of both vaginal and anorectal specimens, we obtained an increase in yield of GBS carriers of 66.7% when compared with culture of only a vaginal swab on BA. These results are in agreement with previous reports [4,6,8,9]. The formation of a pigment on MGM is the crucial factor for improved yields of this medium; hence small quantities of GBS can sometimes be detected in an overgrowth of other bacteria. Only a very small number of GBS strains (1±2%) do not produce the typical pigment [10] and are thus overlooked on MGM. In our collection of 310 GBS isolates, only 2% were non-producers with the `coverglass' technique. If one adheres to the original technique (incubation in anaerobiosis), only 1% of GBS strains are non-producers, and thus false negative on MGM. Rare `Granada-negative' isolates can be detected by separate processing of the vaginal and the anorectal samples, as we did in this study. The vaginal sample can additionally be plated on a BA medium, which will reveal (non-hemolytic!) colonies suspected of being GBS. Furthermore, the vaginal specimen can be processed without fecal contamination, and the Gram stain can identify bacterial vaginosis, candidiasis etc. It has also been suggested that non-hemolytic, nonpigmented GBS strains are less virulent [10], which would be an additional point in favor of MGM. The approach that we evaluated in this study is important for the management of GBS disease prevention. It differs from the
CDC guidelines, but GBS can be detected with at least equivalent sensitivity, at reduced cost in laboratory personnel and reagents and with results available after overnight incubation. If the MGM is available in the laboratory, it can also be used for the identi®cation of GBS from other clinical samples, avoiding quality problems of BA in the performance of the CAMP test. For bothpurposesMGMisausefulmedium,and wehopethatitwillbe widely commercially available in the near future. R EFER E NCE S 1. American Academy of Pediatrics Committee on Infectious Diseases and Committee on Fetus and New-born. Revised guidelines for prevention of early-onset group B streptococcal (GBS) Infection. Pediatrics 1997; 99: 489±96. 2. Centers for Disease Control and Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. MMWR 1996; 45: 1±24. 3. Schrag SJ, Zywicki S, Farley MM et al. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 2000; 342: 15±20. 4. Rosa M, Perez M, Carazo C et al. New Granada medium for detection and identification of group B streptococci. J Clin Microbiol 1992; 30: 1019±21. 5. Islam A. Rapid recognition of group B streptococci. Lancet 1977; i: 256±7. 6. Rosa-Fraile M, Rodriguez-Granger J, Cueto-Lopez M et al. Use of Granada medium to detect Group B streptococcal colonisation in pregnant women. J Clin Microbiol 1999; 37: 2674±7. 7. Isenbergh HD. Essential procedure for clinical microbiology. Washington DC: American Society for Microbiology, 1998. 8. Kelly VN, Garland SM. Evaluation of New Granada Medium (modified) for the antenatal detection of group B streptococcus. Pathology 1994; 26: 487±9. 9. Garcia G, Rodriguez MC, Bartolome B et al. Evaluation of the Granada agar plate for detection of vaginal and rectal Group B streptococci in pregnant women. J Clin Microbiol 1999; 37: 2648± 51. 10. Wennerstrom DE, Lee LN, Baseman AG et al. Genetics and characterisation of group B streptococcal pigment. In: Dunny GM, Cleary PP, McKay LL, eds. Genetics and molecular biology of streptococci, lactococci, and enterococci. Washington DC: American Society for Microbiology, 1991: 224±7.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7, 22±24
was compared with our standard approach of culturing a vaginal swab on blood agar (BA).
Results Samples were processed from 430 consecutive pregnant women. GBS was isolated from the vagina in
11.6% with BA, and in 13.7% with MGM. In 17.0% of anorectal samples, GBS was identi®ed with MGM. The combination of both sites and media had a yield of 20.0%. MGM identi®ed all but six (2%) of 310 GBS strains after aerobic incubation, with use of a cover slide, and missed only three strains (1%) after anaerobic incubation.
Conclusions Separate culture of vaginal and anorectal samples using the same MGM agar plate resulted in an
increase in detection rate for GBS of 76% as compared to BA alone. The technique is simple and results are available after overnight incubation. MGM was con®rmed as a speci®c medium for the identi®cation of GBS, with a sensitivity of 98±99%.
Keywords GBS, group B streptococcus, Streptococcus agalactiae, early-onset sepsis, prevention, perinatal infection, pregnancy Accepted 21 August 2000
Clin Microbiol Infect 2001; 7: 22±24
IN T R ODUC T ION Infections with group B streptococcus (GBS) or Streptococcus agalactiae are leading causes of neonatal morbidity and mortality. Some of these infections are preventable by intrapartum administration of antibiotics. Consensus guidelines have been formulated, and two strategies have been advocated [1,2], one being risk-based, and the other screening-based. With use of a culture-based approach, more women are treated with antibiotics and more infections can be prevented. In a recent analysis, a striking 65% decline in the number of early-onset disease cases was observed in a surveillance area in the USA. The decline coincided with an active policy of promotion of the above-mentioned guidelines [3]. For the culture-based approach, combined cultures from the distal vagina and the anorectum at 35±37 weeks of gestation are recommended, including a selective agar and a selective broth, to be subcultured on blood agar (BA) after 24 h of incubation [1,2].
Modi®ed Granada Agar Medium (MGM) is a selective and differential medium for the detection of GBS, producing orange carotenoid pigmented colonies under anaerobic conditions. The methotrexate added to the medium enhances the pigment production [4]. Only very few GBS strains are non-hemolytic and non-pigmented [5]. In a study of 800 vaginal and 450 vagino-anorectal samples of pregnant women, MGM and the selective broth were equally sensitive for detection of GBS in vaginal samples, and MGM was more sensitive in vaginoanorectal samples [6]. With the introduction of a screening-based prophylaxis strategy in our hospital, we decided to compare the culture of a vaginal and an anorectal sample on MGM with our hitherto used method of a vaginal swab solely plated on tryptic soy agar with 5% sheep blood. We report our ®ndings on the ®rst 430 patients. M AT ER I AL S AND ME T HOD S Specimens and patients
Corresponding author and reprint requests: G. Claeys, Laboratory for Microbiology, University Hospital, De Pintelaan 185, 9000 Ghent, Belgium Tel: 32 92 403645 Fax: 32 92 403659 E-mail: [email protected]
Separate samples from the vagina and the anorectum were taken from 430 pregnant women in gestational week 35±37. The samples were sent to the laboratory in Amies transport medium, and processed within 4 h.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases
Claeys et al Modified Granada Agar Medium for the detection of group B streptococcus
Processing of clinical samples The vaginal and anorectal samples were swabbed separately on both halves of an MGM plate, home-made as originally formulated [4], and incubated in an anaerobic jar for 18 h. From the vaginal sample a Gram stain was done; tryptic soy agar with 5% sheep blood (BA), enriched chocolate agar and gonococcal agar was inoculated, incubated in a 5% CO2 incubator, read after 1 and 2 days, and processed according to accepted methods [7]. After 1 day, b- and non-hemolytic colonies with the appearance of GBS were con®rmed with the CAMP test. When typical colonies were present on the BA, orange-pigmented colonies from the MGM were identi®ed as GBS; when no b-hemolytic colonies were present on BA, pigmented colonies were subjected to a CAMP test [7]. No enrichment broth was used. The yields of the three cultures (BA vagina, MGM vagina, MGM anorectal) were compared.
GBS strains A series of 310 GBS strains, including isolates from non-study samples, was tested for pigment production. The strains were inoculated on two MGM plates, and incubated overnight. One MGM was incubated anaerobically, and the other was incubated in an ambient atmosphere after placement of a small (18 18 mm) cover slide upon the inoculum [6].
R E S ULT S Vaginal and anorectal samples were obtained from 430 pregnant women. GBS was detected in 86, which means that 20% of the women in the study were GBS carriers when using the new approach (Table 1). Of the 430 vaginal samples, 50 were GBS positive on BA (11.6%), compared to 59 with MGM (13.7%). Of the 86 GBS carriers, 58% were detected on BA and 68.5% were detected on MGM. No samples with GBS on BA were GBS negative on MGM.
23
Table 1 Comparison of GBS detected using one or two samples (vaginal, anorectal) cultured on non-selective (blood agar, BA) and/or selective medium (Modified Granada Medium, MGM)
Site
Medium
Vaginal
BA MGM MGM
Anorectal Both
Number (%) of pregnant women with a positive culture for GBS
Number (%) of carriers detected with sample/medium combination
50/430 (11.6%) 59/430 (13.7%) 73/430 (17%) 86/630 (20%)
50/86 (58%) 59/86 (68.5%) 73/86 (85%) 86/86 (100%)
Cultures of the 430 anorectal samples on MGM yielded GBS in 73 cases (17%), which implies a detection rate of 85%. In 13 women the vaginal sample was positive and the anorectal sample was negative, whereas 27 women had positive anorectal and negative vaginal cultures. As a consequence, the maximum number of GBS carriers was detected by combined culture of the two samples on MGM. A collection of 310 GBS strains, including isolates from nonstudy samples, was tested for hemolytic properties, reactions in the CAMP test, and pigment production under two different conditions. Three hundred and three strains were positive with all four tests. Abnormal reactions were found in seven isolates (Table 2). Six strains (2%) did not produce colored colonies when incubated aerobically; of these, three (1%) were pigmented when grown anaerobically. Six isolates (2%) were non-hemolytic, and this always corresponded to an absent or very weak pigmentation. Two isolates, one hemolytic and one non-hemolytic, were CAMP negative. All aberrant isolates had the group B Lance®eld antigen, as demonstrated by a coagglutination test. DI SC US S ION The purpose of our study was to compare the number of GBS carriers in pregnant women at 35±37 weeks of gestation
Table 2 Hemolytic and pigmentation characteristics of 310 GBS isolates
b-Hemolysis
CAMP reaction
Pigmentation on MGM incubated aerobicallya
^ ^ ^
^ ^
^ ^ ^
a
Coverslip placed upon the inoculum.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7, 22±24
Pigmentation on MGM incubated anaerobically
Number (%) of isolates
^ ^
303 (97%) 1 (< 1%) 1 (< 1%) 2 (1%) 3 (1%)
24 Clinical Microbiology and Infection, Volume 7 Number 1, January 2001
detected by plating a vaginal sample on BA to the use of MGM for a vaginal and an anorectal sample. We did not use enrichment culture in selective broth medium, as recommended in the CDC guidelines [1,2], because previous reports have shown equal or better sensitivity of MGM [4,6,8,9]. Furthermore, by using MGM, results are obtained faster and the workload is lower (no subcultures, and no identi®cation tests needed). The reason why enrichment culture does not have a better yield than MGM is probably that the selective broth, containing nalidixic acid and either gentamicin or colistin, also stimulates the growth of enterococci, competing with GBS. By using MGM, a selective differential medium, and the testing of both vaginal and anorectal specimens, we obtained an increase in yield of GBS carriers of 66.7% when compared with culture of only a vaginal swab on BA. These results are in agreement with previous reports [4,6,8,9]. The formation of a pigment on MGM is the crucial factor for improved yields of this medium; hence small quantities of GBS can sometimes be detected in an overgrowth of other bacteria. Only a very small number of GBS strains (1±2%) do not produce the typical pigment [10] and are thus overlooked on MGM. In our collection of 310 GBS isolates, only 2% were non-producers with the `coverglass' technique. If one adheres to the original technique (incubation in anaerobiosis), only 1% of GBS strains are non-producers, and thus false negative on MGM. Rare `Granada-negative' isolates can be detected by separate processing of the vaginal and the anorectal samples, as we did in this study. The vaginal sample can additionally be plated on a BA medium, which will reveal (non-hemolytic!) colonies suspected of being GBS. Furthermore, the vaginal specimen can be processed without fecal contamination, and the Gram stain can identify bacterial vaginosis, candidiasis etc. It has also been suggested that non-hemolytic, nonpigmented GBS strains are less virulent [10], which would be an additional point in favor of MGM. The approach that we evaluated in this study is important for the management of GBS disease prevention. It differs from the
CDC guidelines, but GBS can be detected with at least equivalent sensitivity, at reduced cost in laboratory personnel and reagents and with results available after overnight incubation. If the MGM is available in the laboratory, it can also be used for the identi®cation of GBS from other clinical samples, avoiding quality problems of BA in the performance of the CAMP test. For bothpurposesMGMisausefulmedium,and wehopethatitwillbe widely commercially available in the near future. R EFER E NCE S 1. American Academy of Pediatrics Committee on Infectious Diseases and Committee on Fetus and New-born. Revised guidelines for prevention of early-onset group B streptococcal (GBS) Infection. Pediatrics 1997; 99: 489±96. 2. Centers for Disease Control and Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. MMWR 1996; 45: 1±24. 3. Schrag SJ, Zywicki S, Farley MM et al. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 2000; 342: 15±20. 4. Rosa M, Perez M, Carazo C et al. New Granada medium for detection and identification of group B streptococci. J Clin Microbiol 1992; 30: 1019±21. 5. Islam A. Rapid recognition of group B streptococci. Lancet 1977; i: 256±7. 6. Rosa-Fraile M, Rodriguez-Granger J, Cueto-Lopez M et al. Use of Granada medium to detect Group B streptococcal colonisation in pregnant women. J Clin Microbiol 1999; 37: 2674±7. 7. Isenbergh HD. Essential procedure for clinical microbiology. Washington DC: American Society for Microbiology, 1998. 8. Kelly VN, Garland SM. Evaluation of New Granada Medium (modified) for the antenatal detection of group B streptococcus. Pathology 1994; 26: 487±9. 9. Garcia G, Rodriguez MC, Bartolome B et al. Evaluation of the Granada agar plate for detection of vaginal and rectal Group B streptococci in pregnant women. J Clin Microbiol 1999; 37: 2648± 51. 10. Wennerstrom DE, Lee LN, Baseman AG et al. Genetics and characterisation of group B streptococcal pigment. In: Dunny GM, Cleary PP, McKay LL, eds. Genetics and molecular biology of streptococci, lactococci, and enterococci. Washington DC: American Society for Microbiology, 1991: 224±7.
ß 2001 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 7, 22±24