Abstract
We evaluated the Strep B optical immunoassay (OIA; ThermoBiostar, Inc.) for detecting light and heavy group B streptococcus colonization in 1,306 pregnant women. The women were examined at 20 to 32 weeks gestation and were from six countries. Compared to culture, the sensitivity and specificity of OIA were 13.3 and 98.4%, respectively, for light colonization and 41.5 and 97.7%, respectively, for heavy colonization.
Heavy colonization with group B streptococcus (GBS) may place women at high risk of delivering preterm, low-birth-weight infants (10, 11, 15). Earlier recognition of such high-risk women would be useful for intervention strategies. Detection of heavy colonization has traditionally relied on culture methods, which may not be feasible in some health care settings and require 1 to 3 days before results are available. Unfortunately, most rapid GBS tests have poor sensitivity (7, 13). The Strep B optical immunoassay (OIA; ThermoBiostar, Inc., Louisville, Colo.) is a rapid method which may have higher sensitivity for dense GBS colonization (14). We evaluated the performance of this test in an international, multicenter study.
We conducted the study at institutions participating in the Global Network for Perinatal and Reproductive Health (http://www.tju.edu/obgyn/rih/global.cfm). Centers were located in Khon Kaen and Bangkok, Thailand; Manila, The Philippines; Harare, Zimbabwe; Yangon, Myanmar; Dublin, Ireland; and Philadelphia, Pennsylvania. Participants were enrolled in the study between June 1999 and May 2001. Women attending prenatal care clinics and at 20 to 32 weeks gestation were eligible. Exclusion criteria included vaginitis symptoms, fever, vaginal bleeding, multiple pregnancy, use of antibiotics in the past 2 weeks, and active labor. Informed consent was obtained. The study was approved by institutional review boards at all study centers and supporting institutions.
At enrollment, study personnel conducted an interview and collected specimens. A sample for GBS rapid testing was collected from the cervix and lower vagina using a ThermoBiostar swab. OIAs were read by study site laboratory personnel and by the manufacturer's laboratory personnel. OIA kits and swabs were supplied by ThermoBiostar.
Methods for GBS culture and quantitation replicated those used in the Vaginal Infections and Prematurity Study (16). After collection of the OIA specimen, a Dacron swab (Copan swab) was used to collect a specimen from the cervix and lower vagina. Samples were streaked onto blood agar plates using the four-quadrant method and then placed into Lim broth (Todd-Hewitt broth containing selective antimicrobial agents). A clean catch urine specimen was collected for GBS culture on blood agar. Suspect colonies showing β-hemolysis were identified as GBS using latex agglutination. Heavy colonization was defined as isolation of GBS from either urine or direct plating on blood agar. Light colonization was defined as isolation of GBS from Lim broth alone. Dense GBS colonization was defined as growth in the third or fourth quadrant streaked on the blood agar plate (14).
Data were entered by using Epi-Info version 6 (9) and transmitted to the Centers for Disease Control and Prevention monthly. We calculated the sensitivity, specificity, positive predictive value, and negative predictive value of the OIA using culture results as the “gold standard.” The chi-square test was used to compare the results of OIA readings done by the manufacturer's staff versus the readings done by study center personnel.
We enrolled 1,306 women; their mean age was 26.6 years, and the mean gestational age was 26.3 weeks. GBS colonization was the highest in Philadelphia, Pa., (22.1%) and lowest in Yangon, Myanmar (7.1%). The rate of GBS isolation was higher from Lim broth culture than from other methods (Table 1).
TABLE 1.
Prevalence of GBS according to various culture methods and OIA by study center
| Study site (no. of samples) | Prevalence (%) of GBS by method
|
|||
|---|---|---|---|---|
| Lim broth culture | Blood agar culture | Urine culture | OIA | |
| Dublin, Ireland (203) | 11.8 | 3.5 | 3.5 | 3.5 |
| Yangon, Myanmar (226) | 7.1 | 0.4 | 0.0 | 0.4 |
| Manila, The Philippines (200) | 6.0 | 0.0 | 2.5 | 5.0 |
| Bangkok, Thailand (200) | 12.0 | 3.0 | 3.5 | 1.5 |
| Khon Kaen, Thailand (200) | 14.5 | 8.0 | 4.0 | 4.5 |
| Philadelphia, Pa. (68) | 19.1 | 16.2 | 14.0 | 14.7 |
| Harare, Zimbabwe (209) | 11.7 | 5.3 | 4.9 | 7.7 |
| All study sites (1,306) | 10.9 | 4.0 | 3.5 | 4.3 |
Compared to culture, the sensitivity of OIA for detecting any GBS colonization was 25.7% (95% confidence interval [CI], 19.0 to 33.6%) and ranged from 6.3% (95% CI, 0.3 to 32.3%) in Yangon, Myanmar, to 53.3% (95% CI, 27.4 to 77.7%) in Philadelphia, Pa. OIA sensitivity was 13.3% for lightly colonized women, 41.5% for heavily colonized women, and 68.4% for densely colonized women (Table 2). OIA sensitivity and specificity were 50.0 and 97.6%, respectively, compared with culture on blood agar and 26.8 and 98.4% compared with Lim broth. The OIA was positive in 1.4% of women who were negative by all culture methods. OIA test readings performed by the manufacturer and by study site personnel were similar (kappa of 0.701); the sensitivities from study site personnel and manufacturer readings was 29.7% (95% CI, 22.5 to 37.9%) and 25.7% (95% CI, 19.0 to 33.6%), respectively.
TABLE 2.
Diagnostic performance of OIA for detection of vaginal carriage of GBS compared to culture in women with light, heavy, and dense GBS colonization (n = 1,306)
| GBS colonization levela | Performance characteristicb
|
|||
|---|---|---|---|---|
| Sensitivity | Specificity | Positive predictive value | Negative predictive value | |
| Light | 13.3 (7.1, 22.9) | 98.4 (97.5, 99.0) | 37.9 (21.3, 57.6) | 94.1 (92.5, 95.3) |
| Heavy | 41.5 (29.7, 54.4) | 97.7 (96.6, 98.4) | 48.2 (34.8, 61.8) | 97.0 (95.8, 97.8) |
| Dense | 68.4 (43.5, 86.4) | 96.5 (95.2, 97.5) | 24.5 (14.2, 38.6) | 99.5 (98.8, 99.8) |
The GBS colonization levels were defined as follows: light, isolation of GBS from Lim broth alone; heavy, isolation of GBS from either urine or direct plating on blood agar; dense, growth in the third or fourth quadrant streaked on the blood agar plate (14).
The four performance characteristics, sensitivity, specificity, and positive and negative predictive values, are all given as percentages. The 95% CIs are given in parentheses.
Our study demonstrated that the sensitivity of OIA to detect GBS colonization compared with that of culture was low, although the sensitivity of OIA in detecting heavy or dense colonization was higher than its sensitivity for detecting light colonization. Our findings confirm earlier work that found reliance on blood agar alone was insensitive for detecting light colonization and that comparing rapid detection methods with blood agar alone falsely elevated the sensitivity of these tests. Enrichment broth media, such as Lim broth, enhance GBS growth while suppressing growth of other bacteria, thus improving identification in specimens with low bacterial inoculum (3, 6).
In previous studies, the performance of the OIA has been compared with selective broth media. The specificity was uniformly high, 85 to 98%, similar to the 98.4% in our study. Conversely, previously reported test sensitivities varied widely but were generally higher (37 to 81%) (2, 5, 7, 12-14, 17-19) than the value we found (26.8%). The most important reason for this variability may be the extent to which the subjects were colonized. The sensitivity of the OIA has been higher (61 to 100%) when compared with third- and fourth-quadrant positive blood agar cultures (2, 5, 7, 12-14). Our finding was within this range (68.4%).
Our study had limitations, as it did not evaluate the performance of the OIA 4 to 5 weeks before delivery or at delivery, when it would have the most clinical value (8). However, the average gestational age for our participants was 26.3 weeks, a time when the fetus is considered viable in many centers, making our data potentially useful for evaluating the role of GBS in preterm birth. We did not collect specimens from the rectum; collecting both rectal and vaginal cultures increases the detection of GBS carriage in pregnant women (1). If we had collected rectal specimens, we would have identified more GBS carriers, and the OIA performance would have been worse.
In conclusion, the low sensitivity of the Strep B OIA may limit its usefulness for detecting GBS in pregnant women, although the best performance of the test is in women who are heavily colonized and who therefore may be at highest risk for complications due to GBS infection. New tests, such as the newly licensed PCR test, may provide other options (4). For now, culture of specimens from the rectum and lower vagina remain the gold standard for detecting GBS colonization in pregnant women (8).
Acknowledgments
This project was performed with the financial support of the Rockefeller Foundation; the Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases; the Prevention Committee and the Office of Minority and Women's Health, Centers for Disease Control and Prevention, Atlanta, Ga.; USAID; INCLEN; and Thomas Jefferson University.
We thank the Rockefeller Foundation; the Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases; the Prevention Committee and the Office of Minority and Women's Health, Centers for Disease Control and Prevention, Atlanta, Ga.; USAID; INCLEN; and Thomas Jefferson University for technical support on this project. In addition, we thank ThermoBiostar, Inc., for supplying the OIA kits and for technical support for this study.
J. Thinkhamrop, S. Limpongsanurak, M. Festin, S. Daly, A. Schuchat, P. Lumbiganon, E. Zell, T. Chipato, A. A. Win, M. J. Perilla, J. E. Tolosa, and C. G. Whitney are members of the Global Network for Perinatal and Reproductive Health (GNPRH), Department of Obstetrics and Gynecology, Thomas Jefferson University, Philadelphia, Pa.. GNPRH study centers and coinvestigators follow: Khon Kaen University (Khon Kaen, Thailand), Unchalee Tatttawasart, Kaewjai Khamsuk, and Ladda Sriboonreung; Chulalongkorn University (Bangkok, Thailand), Sompop Limpongsanurak, Surasith Chaithonhgwattana, Pongpun Nunthapisud, Sumanee Nin-gate, and Wanwadee Sripitak; Coombe Women's Hospital (Dublin, Ireland), Amanda Cotter, Julie Grantham, and James Fagan; Institute of Medicine, Women's Central Hospital (Yangon, Myanmar), Kyi Kyi Thinn, Win Win Maw, Cho Cho Oo, MarLar Win, Thuzar Han, Mya Mya Aye, Tin Ohn Myat, and Katherine Ba-Thike; University of Zimbabwe (Harare, Zimbabwe), Alexio Mashu, Laina Chidede, Marshall Munjoma, Travor Nyamurera, and Evelyn Mudzviti; University of the Philippines (Manila, The Philippines), Guadalupe N. Villanueva, Maybelle Cagayan, Concepcion Ang, Francis Sarmiento, Daniel Morales, and Celeridad Padrinao; Population Council (Bangkok, Thailand), Josephine Sauvarin, Christopher Elias, and Washiraporn Suramaythangkoon; Centers for Disease Control and Prevention, Falgunee Parekh, Carolyn Wright, Richard Facklam, and John Elliott; Department of Obstetrics and Gynecology, University of Alabama at Birmingham, William Andrews, Martha Lyon, Sue Cliver, and Robert Goldenberg; Thomas Jefferson University, Catherine Farrell, Babu Cheku, Tracey James, Letitia Lee, Michelle Di Vito, Michelle Pollino, Donald Jungkind, Jim Bondi, Vincenzo Berghella, Jay Goldberg, Ratana Komwilaisak, and Jennifer Culhane.
REFERENCES
- 1.Badri, M. S., S. Zawaneh, A. C. Cruz, G. Mantilla, H. Baer, W. N. Spellacy, and E. M. Ayoub. 1977. Rectal colonization with group B streptococcus: relation to vaginal colonization of pregnant women. J. Infect. Dis. 135:308-312. [DOI] [PubMed] [Google Scholar]
- 2.Baker, C. J. 1996. Inadequacy of rapid immunoassays for intrapartum detection of group B streptococcal carriers. Obstet. Gynecol. 88:51-55. [DOI] [PubMed] [Google Scholar]
- 3.Baker, C. J., J. C. Clark, and F. F. Barrett. 1973. Selective broth medium for isolation of group B streptococci. Appl. Microbiol. 26:884-885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bergeron, M. G., D. Ke, C. Ménard, F. J. François, 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. [DOI] [PubMed] [Google Scholar]
- 5.Boyer, K. M., C. A. Gadzala, P. D. Kelly, D. E. Fisher, J. B. Paton, and S. P. Gotoff. 1983. Selective intrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. I. Epidemiologic rationale. J. Infect. Dis. 148:795-801. [DOI] [PubMed] [Google Scholar]
- 6.Boyer, K. M., C. A. Gadzala, P. D. Kelly, L. I. Burd, and S. P. Gotoff. 1983. Selective intrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. II. Predictive value of prenatal cultures. J. Infect. Dis. 148:802-808. [DOI] [PubMed] [Google Scholar]
- 7.Carroll, K. C., D. Ballou, M. Varner, H. Chun, R. Traver, and J. Salyer. 1996. Rapid detection of group B streptococcal colonization of the genital tract by a commercial optical immunoassay. Eur. J. Clin. Microbiol. Infect. Dis. 15:206-210. [DOI] [PubMed] [Google Scholar]
- 8.Centers for Disease Control and Prevention. 2002. Prevention of perinatal group B streptococcal disease—revised guidelines from CDC. Morb. Mortal. Wkly. Rep. 51(RR11):1-22. [Google Scholar]
- 9.Dean, A. G., J. A. Dean, D. Coulombier, et al. 1995. Epi-Info version 6: a word processing, database, and statistics program for epidemiology on microcomputers. Centers for Disease Control and Prevention, Atlanta, Ga.
- 10.McKenzie, H., M. L. Donnet, P. W. Howie, N. B. Patel, and D. T. Benvie. 1994. Risk of preterm delivery in pregnant women with group B streptococcal urinary infections or urinary antibodies to group B streptococcal and E. coli antigens. Br. J. Obstet. Gynecol. 101:107-113. [DOI] [PubMed] [Google Scholar]
- 11.Moller, M., A. C. Thomsen, K. Borch, K. Dinesen, and M. Zdravkovic. 1984. Rupture of fetal membranes and premature delivery associated with group B streptococci in urine of pregnant women. Lancet 2:69-70. [DOI] [PubMed] [Google Scholar]
- 12.Nguyen, T. M., D. W. Gauthier, T. D. Mules, B. S. Nuwayhid, M. A. Viana, and P. C. Schreckenberger. 1998. Detection of group B streptococcus: comparison of an optical immunoassay with direct plating and broth-enhanced culture methods. J. Matern. Fetal Med. 7:172-176. [DOI] [PubMed] [Google Scholar]
- 13.Ostroff, R. M., and J. W. Steaffens. 1995. Effect of specimen storage, antibiotics, and feminine hygiene products on the detection of group B streptococcus by culture and the STREP B OIA test. Diagn. Microbiol. Infect. Dis. 22:253-259. [DOI] [PubMed] [Google Scholar]
- 14.Park, C. H., D. Ruprai, N. M. Vandel, D. L. Hixon, and F. E. Mecklenburg. 1996. Rapid detection of group B streptococcal antigen from vaginal specimens using a new optical immunoassay technique. Diagn. Microbiol. Infect. Dis. 24:125-128. [DOI] [PubMed] [Google Scholar]
- 15.Regan, J. A., S. Chao, and L. S. James. 1981. Premature rupture of membranes, preterm delivery, and group B streptococcal colonization of mothers. Am. J. Obstet. Gynecol. 141:184-186. [DOI] [PubMed] [Google Scholar]
- 16.Regan, J. A., M. A. Klebanoff, R. P. Nugent, D. A. Eschenbach, W. C. Blackwelder, Y. Lou, R. S. Gibbs, P. J. Rettig, D. H. Martin, and R. Edelman. 1996. Colonization with group B streptococci in pregnancy and adverse outcome. Am. J. Obstet. Gynecol. 174:1354-1360. [DOI] [PubMed] [Google Scholar]
- 17.Reisner, D. P., M. J. Haas, R. W. Zingheim, M. A. Williams, and D. A. Luthy. 2000. Performance of a group B streptococcal prophylaxis protocol combining high-risk treatment and low-risk screening. Am. J. Obstet. Gynecol. 182:1335-1343. [DOI] [PubMed] [Google Scholar]
- 18.Samadi, R., A. Stek, and J. S. Greenspoon. 2001. Evaluation of a rapid optical immunoassay-based test for group B streptococcus colonization in intrapartum patients. J. Matern. Fetal Med. 10:203-208. [DOI] [PubMed] [Google Scholar]
- 19.Song, J. Y., L. L. Lin, S. Shott, N. Kimber, J. Tangora, A. Cohen, A. Wells, M. Maezes, A. Aroutcheva, and S. Faro. 1999. Evaluation of the Strep B OIA test compared to standard culture methods for detection of group B streptococci. Infect. Dis. Obstet. Gynecol. 7:202-205. [DOI] [PMC free article] [PubMed] [Google Scholar]