Journal of Perinatology (2015) 35, 173–176 © 2015 Nature America, Inc. All rights reserved 0743-8346/15 www.nature.com/jp
ORIGINAL ARTICLE
Recurrence of group B streptococcus colonization in successive pregnancies LC Colicchia1,2,3, DS Lauderdale4, H Du5, M Adams5 and E Hirsch1,2 OBJECTIVE: To identify risk factors for group B streptococcus (GBS) colonization in a subsequent pregnancy using microbiological and clinical data from a prior pregnancy. STUDY DESIGN: A retrospective cohort study of women over a 10-year period, using laboratory records to identify women with GBS culture results available in two successive pregnancies. RESULT: One thousand eight hundred and ninety-four women met eligibility criteria. Of these, 1293 were not GBS-colonized in either pregnancy, 198 were colonized in both pregnancies and 403 had discordant colonization status. GBS colonization in the index pregnancy was positively associated with multiparity, premature delivery and lower maternal age and negatively associated with chorioamnionitis. The strongest predictor of colonization in a subsequent pregnancy was colonization in the index pregnancy (50% likelihood if colonized compared with 14% if not in the index pregnancy, relative risk 3.6, confidence interval (CI) = 3.1 to 4.3). GBS colonization in the subsequent pregnancy was independently associated with: GBS colonization in the index pregnancy (odds ratio (OR) = 6.28; CI = 4.91 to 8.05), preterm delivery in the index pregnancy (OR = 1.80; CI = 1.05 to 3.09) and prior early pregnancy loss (OR = 1.15; CI = 1.04 to 1.27). CONCLUSION: GBS colonization in a prior pregnancy is informative of colonization in a subsequent pregnancy. These data support providing antimicrobial prophylaxis in unscreened parous women with known prior GBS colonization. Journal of Perinatology (2015) 35, 173–176; doi:10.1038/jp.2014.185; published online 16 October 2014
INTRODUCTION Group B streptococcus (GBS) is the leading infectious cause of neonatal morbidity and mortality in the United States.1,2 The Centers for Disease Control and Prevention recommends screening all pregnant women for genital/rectal GBS colonization at 35 to 37 weeks’ gestation in every pregnancy regardless of prior GBS status. Prophylactic intrapartum antibiotics are recommended for all women with positive cultures (including GBS-positive urine cultures obtained at any time during pregnancy), while decisions regarding microbial prophylaxis in women with unknown GBS status are to be managed using a risk-based protocol.3,4 GBS colonization is known to be variable between pregnancies and transient during a given pregnancy. Between 13% and 54% of women with an initial positive culture will no longer be colonized at the time of delivery.5–8 Recurrent GBS colonization between pregnancies is reported at rates of 38% to 53%,9–12 but a clear understanding of the factors influencing GBS colonization over time is lacking. We posited that if microbiological and clinical risk factors for colonization could be identified from an index pregnancy, an opportunity may exist to use this information in the management of a subsequent pregnancy.
MATERIALS AND METHODS We performed a retrospective cohort study comparing parous women who were GBS-colonized to those who were GBS negative in their index
pregnancy. Approval to conduct this study was obtained from the Institutional Review Board (EH no.11-274). Derivation of the study subjects and their pregnancies is shown in Figure 1. All GBS vaginal–rectal cultures (both positive and negative) and all GBS-positive urine cultures processed by the NorthShore University HealthSystem clinical microbiology laboratory from 2003 to 2012 were collected. This data set was cross referenced with the electronic health record Data Warehouse (EPIC Clarity with Oracle and IBM COGNOS) to identify women with two successive pregnancies delivering in our center (Evanston Hospital) for whom appropriately timed GBS culture results were available (that is, either vaginal/rectal culture obtained between 35 and 37 weeks of pregnancy or o 5 weeks prior to a preterm delivery or positive urine culture at any time during pregnancy). Selected abstracted variables included in the analysis were maternal age, race, ethnicity, body mass index in kg m − 2, prenatal care provider, medical history, inter-pregnancy interval, tobacco use, peripartum antibiotic administration during the index pregnancy (to determine if there was any impact on colonization in the subsequent pregnancy), duration of membrane rupture and obstetric/neonatal outcomes (preterm delivery o37 weeks’ gestation, chorioamnionitis (defined as intrapartum temperature ⩾ 38 °C), mode of delivery and neonatal intensive care unit admission). The same microbiology laboratory processed all cultures during the study timeframe using standard methods via direct plating onto sheep blood agar and into LIM enrichment broth (Todd–Hewitt with CNA, no blood added).4 After overnight incubation, the enrichment broth was subcultured to a second sheep blood agar plate. Beta-hemolytic or nonbeta-hemolytic colonies resembling GBS on either the direct blood agar or the blood agar inoculated from the enrichment broth were identified using latex agglutination.
1 Department of Obstetrics and Gynecology, NorthShore University Health System, Evanston, IL, USA; 2Department of Obstetrics and Gynecology, University of Chicago, Chicago, IL, USA; 3Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh, Pittsburgh, PA, USA; 4Department of Health Studies, University of Chicago, Chicago, IL, USA and 5NorthShore University Health System Research Institute, Evanston, IL, USA. Correspondence: Dr LC Colicchia, Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh, Magee-Women’s Hospital, 300 Halket Street, Pittsburgh, PA 15213, USA. E-mail: [email protected] Received 19 May 2014; revised 22 August 2014; accepted 2 September 2014; published online 16 October 2014
Recurrence of group B streptococcus colonization LC Colicchia et al
174
Figure 1.
Selection of the study population.
Chi square was used to analyze categorical variables (for example, GBS colonization status). Univariable logistic regression was conducted to identify additional clinical and demographic risk factors associated with colonization in the subsequent pregnancy. We then constructed three multivariable logistic regression models: one model included all women; a second included only women who were GBS positive in the index pregnancy; and the third included only women who were GBS negative in the index pregnancy. Potential predictors with Po0.25 in univariable analysis were included initially in the multiple regression models and removed in a backward fashion at a Po0.05 cutoff to identify significant predictors for the second pregnancy. Statistical analyses were conducted using the SAS 9.2 (SAS Inc., Durham, NC, USA) and STATA 12 (StataCorp., College Station, TX, USA) statistical and data analysis software.
RESULTS We identified 1894 women meeting inclusion criteria. A small subset of women (5.3%) had 42 deliveries, and for these women only the first two deliveries were included in the analysis. Patient and obstetric characteristics at the index pregnancy are summarized in Table 1. In comparison to GBS-negative patients, women who were GBS positive in the index pregnancy were more likely to be parous, more likely to be cared for in the resident clinic, younger and more likely to be African-American or Asian. Women who were GBS positive in the index pregnancy were also more likely to be exposed to antibiotics during the peripartum period, were more likely to deliver prematurely and were less likely to develop chorioamnionitis (P = 0.01) than women who were GBS negative. Body mass index, maternal diabetes, race and tobacco use did not significantly differ between the two groups. Among the 1894 eligible patients, 1293 were not GBS-colonized in either pregnancy, 198 were colonized in both pregnancies, 197 were colonized only in the index pregnancy and 206 were colonized only in the subsequent pregnancy (Table 2). The overall colonization rate was similar in index pregnancies (20.9%) and subsequent pregnancies (21.3%), but women who were GBScolonized in the index pregnancy had a 50% recurrent colonization rate in the subsequent pregnancy, while women who were GBS negative in the index pregnancy had a 14% colonization rate in the subsequent pregnancy (relative risk (RR) 3.6, 95% confidence interval (CI) 3.1 to 4.3, P o0.001). Among the subset of 31 patients with GBS-positive urine cultures in the index pregnancy, 16 (52%) were GBS positive in the subsequent pregnancy, and 15 (48%) were GBS negative, which was no different from recurrent Journal of Perinatology (2015), 173 – 176
colonization rates in women who were positive by vaginal/rectal culture. Multivariable logistic regression was performed for the entire cohort and separately for GBS-colonized and non-colonized women in the index pregnancy (Table 3). Among all patients, GBS colonization in the subsequent pregnancy was independently associated with GBS colonization status in the index pregnancy, preterm delivery in the index pregnancy and a history of early pregnancy loss. Among the subset of women who were GBS positive in the index pregnancy (n = 395), the risk of repeat colonization was higher with a history of preterm delivery or pregnancy loss o 20 weeks’ gestation in the index pregnancy. AfricanAmerican race was associated with diminished risk of recurrent colonization. Among women who were GBS negative in the index pregnancy (n = 1499), African-American race was associated with a higher risk of colonization in the subsequent pregnancy. DISCUSSION Our data indicate that GBS colonization status in an index pregnancy is the most significant determinant of colonization in a subsequent gestation (50% risk if culture positive in the index pregnancy versus 14% if culture negative, RR 3.65, 95% CI 3.1 to 4.3). By comparison, the impact of other variables affecting future colonization (prior preterm delivery, pregnancy loss and race) is small. Prior reports have demonstrated similar rates of recurrent GBS colonization in the United States and elsewhere. In a study of 251 GBS-positive Taiwanese women, Cheng et al.9 found a 38% risk of recurrent colonization compared with a 15% baseline colonization rate. Among American women, rates of recurrent GBS colonization are reported between 40% and 53%, while among women who tested GBS negative in a prior pregnancy, GBS colonization is 15% to 19%.10–12 Together with the present study, these results suggest that a 40% to 50% recurrent colonization risk is a reasonable estimate and also support the generalizability of our observations, in which cases were ascertained using unselected culture results from a single center. None of these reports, including the present study, allow differentiation between persistent colonization and new colonization, a distinction that would require molecular analysis not routinely performed in clinical labs. Identification of prior preterm birth and early pregnancy loss as independent risk factors for subsequent GBS colonization are © 2015 Nature America, Inc.
Recurrence of group B streptococcus colonization LC Colicchia et al
175 Table 1. Patient characteristics at index pregnancy by GBS colonization status
Maternal age at delivery (years)
Table 3. Factors affecting GBS colonization at the subsequent pregnancy: multivariable logistic regression
GBS positive, n = 395
GBS negative, n = 1499
P-value
29.7 ± 4.9
30.3 ± 4.7
0.03 o0.001
Race African-American Caucasian Asian Other
44 199 33 75
Hispanic/Latino ethnicity Parous (%)a BMI
27 (7.5) 41% 30.4 ± 5.5
111 (8.2) 34% 30.2 ± 5.3
Prenatal care provider Resident clinic Private clinic
59 (14.9) 330 (84.8)
146 (9.7) 1337 (90.1)
(13) (57) (9.4) (21)
92 815 68 349
(6.9) (62) (5.1) (26) 0.67 0.011 0.51
GBS colonized in the index pregnancy African-Americanc Preterm delivery of index pregnancya History of o20 week pregnancy lossd
95% CI
P-value
6.28
4.91–8.05
o 0.001
1.80
1.05–3.09
0.03
1.15
1.04–1.27
0.01
0.40 2.90
0.20-.079 1.07–7.85
0.009 0.035
1.27
1.04–1.54
0.017
2.09
1.28–3.44
0.003
0.003
GBS bacteriuria Prior early pregnancy lossb Maternal diabetes Tobacco use Gestational age at delivery (weeks) Preterm delivery ( o37 weeks) Cesarean delivery Chorioamnionitis Peripartum antibiotic administrationc Duration of membrane rupture (hours) NICU admission Inter-pregnancy interval (days)d
GBS negative in the index pregnancy African-Americanc
31 (7.8) 0.42 ± 0.90 36 (9.1) 28 (7.2) 39.2 ± 1.9 25 (6.3) 66 (17) 25 (6.3) 380 (96)
NA 0.49 ± 0.96 103 (6.9) 96 (6.5) 39.5 ± 1.3 48 (3.2) 249 (17) 160 (11) 416 (28)
0.23 0.13 0.62 o0.001 0.004 0.96 0.01 o0.001
8.3 ± 9.3
9.8 ± 19.0
0.15
9 (2.2) 675 ± 406
50 (3.3) 641 ± 369
0.28 0.12
Abbreviations: BMI, body mass index; GBS, group B Streptococcus; NICU, neonatal intensive care unit. Data are mean ± s.d. or n (%). When data were missing, percentages were calculated excluding cases from the denominator. P-values were calculated using t-tests for continuous variables and chi-square tests for categorical variables. aOne or more prior deliveries 420 weeks gestation. bIncludes terminations, miscarriages and ectopic pregnancies at o20 weeks’ gestation. cIncludes antibiotics given during antepartum, intrapartum and postpartum hospitalizations. dCalculated from the delivery of index pregnancy to the last menstrual period of subsequent pregnancy.
Table 2. GBS colonization status in a subsequent pregnancy by colonization status in the index pregnancy GBS colonization in subsequent pregnancy Positive GBS colonization in index pregnancy Positive (n = 395) 198 (50%) Negative (n = 1449) 206 (13.7%) a Po0.001 RR 3.6, 95% CI 3.1–4.3b
Negative 197 (50%) 1293 (86.2%)
Abbreviations: CI, confidence interval; GBS, group B Streptococcus; RR, relative risk. Percentages reported as percentage of women with initial colonization status to have subsequent pregnancy colonization status. a P-value is calculated using the chi-square test. bRelative risk of colonization in the subsequent pregnancy given colonization in the index pregnancy.
novel findings, possibly not previously identified due to the relatively large cohort studied here. The seemingly paradoxical observation that African-American race decreased the risk of colonization during the second pregnancy in patients who were GBS positive in the index pregnancy but increased this risk in women who were GBS negative in the index pregnancy may be © 2015 Nature America, Inc.
All patients GBS colonization at index pregnancy Preterm delivery of index pregnancya History of o20 week pregnancy lossb
Odds ratio
Abbreviations: CI, confidence interval; GBS, group B Streptococcus; OR, odds ratio. Multivariable logistic regression analysis of factors significantly affecting GBS colonization in a subsequent pregnancy. Variables with Po0.25 on univariable analysis were included in the multivariable analysis and removed in a backwards stepwise fashion to the Po0.05 significance level. aLess than 37 weeks gestation. bPrior miscarriage/ termination/ectopic pregnancy either before index pregnancy or between index and subsequent pregnancy. cCompared with Caucasian women. d Prior miscarriage/termination/ectopic pregnancy either before index pregnancy or between index and subsequent pregnancy.
related to the higher rate of colonization in African-American women than others in this study (32% compared with a baseline rate of 21%), with regression towards the mean. Among the strengths of the present study are its large patient population, with demographics and baseline GBS colonization rates comparable to the national average.1,4,13,14 The sample size (five times larger than the largest prior comparable report) provided power to identify differences between the two groups (that is, GBS-colonized and GBS non-colonized in the first pregnancy) as small as 15% for any single continuous variable. All cultures were performed according to the standard, commonly accepted protocol at a single institution. Limitations include the retrospective nature of the study and the lack of prospective follow-up. Also, identification of patients relied upon accurate linkage of microbiology results with the electronic medical record. Some cases may not have been identified properly. Current Centers for Disease Control and Prevention guidelines recommend management of laboring women with unknown GBS status according to a risk-based strategy, with antibiotics given for gestational age o 37 weeks, rupture of membranes 418 h in duration or intrapartum temperature 438 °C.4 Most GBS-colonized women delivering at term do not develop fever or prolonged rupture of membranes and thus would not receive antibiotics according to this risk-based protocol.15,16 However, even in the absence of risk factors, without treatment the infants in these cases have a 5 to 20 times greater risk of invasive neonatal GBS disease than infants born to non-colonized women.15–17 Given that 50% of women who were GBS positive in a prior pregnancy can be presumed to be colonized, including prior GBS-positivity as a risk-based indication for chemoprophylaxis seems reasonable, even in the absence of other risk factors. Such prophylaxis could decrease the incidence of early-onset neonatal disease from approximately 0.8% (that is, 1.6% of the 50% of colonized patients) to 0.05% (0.001% of colonized patients), a 16-fold decrease.4,15–19 Journal of Perinatology (2015), 173 – 176
Recurrence of group B streptococcus colonization LC Colicchia et al
176 A strategy providing intrapartum chemoprophylaxis without prior screening to all women who were GBS-colonized in a previous pregnancy was evaluated hypothetically in a decision analysis, finding that culture avoidance and uniform treatment might be cost-effective.20 However, that study took into account only two prior small studies to determine the risk of recurrent colonization. The study was further limited by the use of manufacturer’s suggested price instead of actual drug prices for penicillin and by the assumption that all patients would be treated with penicillin instead of ampicillin or other antibiotics in the setting of penicillin allergy. Until rapid, accurate bedside GBS screening for women in labor is available and tested prospectively as part of a neonatal GBS disease avoidance protocol,4,21 we suggest that universal screening independent of prior pregnancy status remain the standard. We suggest that prior GBS colonization be considered as an additional indication for intrapartum treatment in women with unknown GBS status. Future studies should focus on maternal and microbial characteristics that determine persistence, transience and recurrence of GBS colonization. Data from such studies may be useful for further refining clinical protocols. CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Stoll BJ, Hansen NI, Sánchez PJ, Faix RG, Poindexter BB, Van Meurs KP et al. Early onset neonatal sepsis: the burden of group B Streptococcal and E. coli disease continues. Pediatrics 2011; 127: 817–826. 2 Yow MD, Leeds LJ, Thompson PK, Mason EO Jr, Clark DJ, Bleacher CW. The natural history of group B streptococcal colonization in the pregnant woman and her offspring. I. Colonization studies. Am J Obstet Gynecol 1980; 137: 34–38. 3 Melin P. Neonatal group B streptococcal disease: from pathogenesis to preventive strategies. Clin Microbiol Infect 2011; 17: 1294–1303. 4 Verani JR, McGee L, Schrag SJ, Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Center for Disease Control and Prevention (CDC). Prevention of perinatal group B streptococcal disease--revised guidelines from CDC, 2010. MMWR Recomm Rep 2010; 59: 1–36. 5 Allardice JG, Baskett TF, Seshia MM, Bowman N, Malazdrewicz R. Perinatal group B streptococcal colonization and infection. Am J Obstet Gynecol 1982; 142: 617–620.
Journal of Perinatology (2015), 173 – 176
6 Lin F-YC, Weisman LE, Azimi P, Young AE, Chang K, Cielo M et al. Assessment of intrapartum antibiotic prophylaxis for the prevention of early-onset group B Streptococcal disease. Pediatr Infect Dis J 2011; 30: 759–763. 7 Towers CV, Rumney PJ, Asrat T, Preslicka C, Ghamsary MG, Nageotte MP. The accuracy of late third-trimester antenatal screening for group B streptococcus in predicting colonization at delivery. Am J Perinatol 2010; 27: 785–790. 8 Yancey MK, Schuchat A, Brown LK, Ventura VL, Markenson GR. The accuracy of late antenatal screening cultures in predicting genital group B streptococcal colonization at delivery. Obstet Gynecol 1996; 88: 811–815. 9 Cheng PJ, Chueh HY, Liu CM, Hsu JJ, Hsieh TT, Soong YK. Risk factors for recurrence of group B streptococcus colonization in a subsequent pregnancy. Obstet Gynecol 2008; 111: 704–709. 10 Tam T, Bilinski E, Lombard E. Recolonization of group B Streptococcus (GBS) in women with prior GBS genital colonization in pregnancy. J Matern Fetal Neonatal Med 2012; 25: 1987–1989. 11 Turrentine MA, Ramirez MM. Recurrence of group B streptococci colonization in subsequent pregnancy. Obstet Gynecol 2008; 112: 259–264. 12 Page-Ramsey SM, Johnstone SK, Kim D, Ramsey PS. Prevalence of group B Streptococcus colonization in subsequent pregnancies of group B Streptococcuscolonized versus noncolonized women. Am J Perinatol 2013; 30: 383–388. 13 Regan JA, Klebanoff MA, Nugent RP. The epidemiology of group B streptococcal colonization in pregnancy. Vaginal Infections and Prematurity Study Group. Obstet Gynecol 1991; 77: 604–610. 14 Stapleton RD, Kahn JM, Evans LE, Critchlow CW, Gardella CM. Risk factors for group B streptococcal genitourinary tract colonization in pregnant women. Obstet Gynecol 2005; 106: 1246–1252. 15 Rouse DJ, Goldenberg RL, Cliver SP, Cutter GR, Mennemeyer ST, Fargason CA Jr. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis. Obstet Gynecol 1994; 83: 483–494. 16 Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS et al. A populationbased comparison of strategies to prevent early-onset group B streptococcal disease in neonates. N Engl J Med 2002; 347: 233–239. 17 Boyer KM, Gotoff SP. Strategies for chemoprophylaxis of GBS early-onset infections. Antibiot Chemother (1971) 1985; 35: 267–280. 18 Adair CE, Kowalsky L, Quon H, Ma D, Stoffman J, McGeer A et al. Risk factors for early-onset group B streptococcal disease in neonates: a population-based casecontrol study. CMAJ 2003; 169: 198–203. 19 Benitz WE, Gould JB, Druzin ML. Risk factors for early-onset group B streptococcal sepsis: estimation of odds ratios by critical literature review. Pediatrics 1999; 103: e77. 20 Turrentine MA, Ramirez MM, Mastrobattista JM. Cost-effectiveness of universal prophylaxis in pregnancy with prior group B streptococci colonization. Infect Dis Obstet Gynecol 2009; 2009: 934698. 21 Honest H, Sharma S, Khan KS. Rapid tests for group B Streptococcus colonization in laboring women: a systematic review. Pediatrics 2006; 117: 1055–1066.
© 2015 Nature America, Inc.
Copyright of Journal of Perinatology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Recurrence of group B streptococcus colonization in successive pregnancies LC Colicchia1,2,3, DS Lauderdale4, H Du5, M Adams5 and E Hirsch1,2 OBJECTIVE: To identify risk factors for group B streptococcus (GBS) colonization in a subsequent pregnancy using microbiological and clinical data from a prior pregnancy. STUDY DESIGN: A retrospective cohort study of women over a 10-year period, using laboratory records to identify women with GBS culture results available in two successive pregnancies. RESULT: One thousand eight hundred and ninety-four women met eligibility criteria. Of these, 1293 were not GBS-colonized in either pregnancy, 198 were colonized in both pregnancies and 403 had discordant colonization status. GBS colonization in the index pregnancy was positively associated with multiparity, premature delivery and lower maternal age and negatively associated with chorioamnionitis. The strongest predictor of colonization in a subsequent pregnancy was colonization in the index pregnancy (50% likelihood if colonized compared with 14% if not in the index pregnancy, relative risk 3.6, confidence interval (CI) = 3.1 to 4.3). GBS colonization in the subsequent pregnancy was independently associated with: GBS colonization in the index pregnancy (odds ratio (OR) = 6.28; CI = 4.91 to 8.05), preterm delivery in the index pregnancy (OR = 1.80; CI = 1.05 to 3.09) and prior early pregnancy loss (OR = 1.15; CI = 1.04 to 1.27). CONCLUSION: GBS colonization in a prior pregnancy is informative of colonization in a subsequent pregnancy. These data support providing antimicrobial prophylaxis in unscreened parous women with known prior GBS colonization. Journal of Perinatology (2015) 35, 173–176; doi:10.1038/jp.2014.185; published online 16 October 2014
INTRODUCTION Group B streptococcus (GBS) is the leading infectious cause of neonatal morbidity and mortality in the United States.1,2 The Centers for Disease Control and Prevention recommends screening all pregnant women for genital/rectal GBS colonization at 35 to 37 weeks’ gestation in every pregnancy regardless of prior GBS status. Prophylactic intrapartum antibiotics are recommended for all women with positive cultures (including GBS-positive urine cultures obtained at any time during pregnancy), while decisions regarding microbial prophylaxis in women with unknown GBS status are to be managed using a risk-based protocol.3,4 GBS colonization is known to be variable between pregnancies and transient during a given pregnancy. Between 13% and 54% of women with an initial positive culture will no longer be colonized at the time of delivery.5–8 Recurrent GBS colonization between pregnancies is reported at rates of 38% to 53%,9–12 but a clear understanding of the factors influencing GBS colonization over time is lacking. We posited that if microbiological and clinical risk factors for colonization could be identified from an index pregnancy, an opportunity may exist to use this information in the management of a subsequent pregnancy.
MATERIALS AND METHODS We performed a retrospective cohort study comparing parous women who were GBS-colonized to those who were GBS negative in their index
pregnancy. Approval to conduct this study was obtained from the Institutional Review Board (EH no.11-274). Derivation of the study subjects and their pregnancies is shown in Figure 1. All GBS vaginal–rectal cultures (both positive and negative) and all GBS-positive urine cultures processed by the NorthShore University HealthSystem clinical microbiology laboratory from 2003 to 2012 were collected. This data set was cross referenced with the electronic health record Data Warehouse (EPIC Clarity with Oracle and IBM COGNOS) to identify women with two successive pregnancies delivering in our center (Evanston Hospital) for whom appropriately timed GBS culture results were available (that is, either vaginal/rectal culture obtained between 35 and 37 weeks of pregnancy or o 5 weeks prior to a preterm delivery or positive urine culture at any time during pregnancy). Selected abstracted variables included in the analysis were maternal age, race, ethnicity, body mass index in kg m − 2, prenatal care provider, medical history, inter-pregnancy interval, tobacco use, peripartum antibiotic administration during the index pregnancy (to determine if there was any impact on colonization in the subsequent pregnancy), duration of membrane rupture and obstetric/neonatal outcomes (preterm delivery o37 weeks’ gestation, chorioamnionitis (defined as intrapartum temperature ⩾ 38 °C), mode of delivery and neonatal intensive care unit admission). The same microbiology laboratory processed all cultures during the study timeframe using standard methods via direct plating onto sheep blood agar and into LIM enrichment broth (Todd–Hewitt with CNA, no blood added).4 After overnight incubation, the enrichment broth was subcultured to a second sheep blood agar plate. Beta-hemolytic or nonbeta-hemolytic colonies resembling GBS on either the direct blood agar or the blood agar inoculated from the enrichment broth were identified using latex agglutination.
1 Department of Obstetrics and Gynecology, NorthShore University Health System, Evanston, IL, USA; 2Department of Obstetrics and Gynecology, University of Chicago, Chicago, IL, USA; 3Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh, Pittsburgh, PA, USA; 4Department of Health Studies, University of Chicago, Chicago, IL, USA and 5NorthShore University Health System Research Institute, Evanston, IL, USA. Correspondence: Dr LC Colicchia, Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh, Magee-Women’s Hospital, 300 Halket Street, Pittsburgh, PA 15213, USA. E-mail: [email protected] Received 19 May 2014; revised 22 August 2014; accepted 2 September 2014; published online 16 October 2014
Recurrence of group B streptococcus colonization LC Colicchia et al
174
Figure 1.
Selection of the study population.
Chi square was used to analyze categorical variables (for example, GBS colonization status). Univariable logistic regression was conducted to identify additional clinical and demographic risk factors associated with colonization in the subsequent pregnancy. We then constructed three multivariable logistic regression models: one model included all women; a second included only women who were GBS positive in the index pregnancy; and the third included only women who were GBS negative in the index pregnancy. Potential predictors with Po0.25 in univariable analysis were included initially in the multiple regression models and removed in a backward fashion at a Po0.05 cutoff to identify significant predictors for the second pregnancy. Statistical analyses were conducted using the SAS 9.2 (SAS Inc., Durham, NC, USA) and STATA 12 (StataCorp., College Station, TX, USA) statistical and data analysis software.
RESULTS We identified 1894 women meeting inclusion criteria. A small subset of women (5.3%) had 42 deliveries, and for these women only the first two deliveries were included in the analysis. Patient and obstetric characteristics at the index pregnancy are summarized in Table 1. In comparison to GBS-negative patients, women who were GBS positive in the index pregnancy were more likely to be parous, more likely to be cared for in the resident clinic, younger and more likely to be African-American or Asian. Women who were GBS positive in the index pregnancy were also more likely to be exposed to antibiotics during the peripartum period, were more likely to deliver prematurely and were less likely to develop chorioamnionitis (P = 0.01) than women who were GBS negative. Body mass index, maternal diabetes, race and tobacco use did not significantly differ between the two groups. Among the 1894 eligible patients, 1293 were not GBS-colonized in either pregnancy, 198 were colonized in both pregnancies, 197 were colonized only in the index pregnancy and 206 were colonized only in the subsequent pregnancy (Table 2). The overall colonization rate was similar in index pregnancies (20.9%) and subsequent pregnancies (21.3%), but women who were GBScolonized in the index pregnancy had a 50% recurrent colonization rate in the subsequent pregnancy, while women who were GBS negative in the index pregnancy had a 14% colonization rate in the subsequent pregnancy (relative risk (RR) 3.6, 95% confidence interval (CI) 3.1 to 4.3, P o0.001). Among the subset of 31 patients with GBS-positive urine cultures in the index pregnancy, 16 (52%) were GBS positive in the subsequent pregnancy, and 15 (48%) were GBS negative, which was no different from recurrent Journal of Perinatology (2015), 173 – 176
colonization rates in women who were positive by vaginal/rectal culture. Multivariable logistic regression was performed for the entire cohort and separately for GBS-colonized and non-colonized women in the index pregnancy (Table 3). Among all patients, GBS colonization in the subsequent pregnancy was independently associated with GBS colonization status in the index pregnancy, preterm delivery in the index pregnancy and a history of early pregnancy loss. Among the subset of women who were GBS positive in the index pregnancy (n = 395), the risk of repeat colonization was higher with a history of preterm delivery or pregnancy loss o 20 weeks’ gestation in the index pregnancy. AfricanAmerican race was associated with diminished risk of recurrent colonization. Among women who were GBS negative in the index pregnancy (n = 1499), African-American race was associated with a higher risk of colonization in the subsequent pregnancy. DISCUSSION Our data indicate that GBS colonization status in an index pregnancy is the most significant determinant of colonization in a subsequent gestation (50% risk if culture positive in the index pregnancy versus 14% if culture negative, RR 3.65, 95% CI 3.1 to 4.3). By comparison, the impact of other variables affecting future colonization (prior preterm delivery, pregnancy loss and race) is small. Prior reports have demonstrated similar rates of recurrent GBS colonization in the United States and elsewhere. In a study of 251 GBS-positive Taiwanese women, Cheng et al.9 found a 38% risk of recurrent colonization compared with a 15% baseline colonization rate. Among American women, rates of recurrent GBS colonization are reported between 40% and 53%, while among women who tested GBS negative in a prior pregnancy, GBS colonization is 15% to 19%.10–12 Together with the present study, these results suggest that a 40% to 50% recurrent colonization risk is a reasonable estimate and also support the generalizability of our observations, in which cases were ascertained using unselected culture results from a single center. None of these reports, including the present study, allow differentiation between persistent colonization and new colonization, a distinction that would require molecular analysis not routinely performed in clinical labs. Identification of prior preterm birth and early pregnancy loss as independent risk factors for subsequent GBS colonization are © 2015 Nature America, Inc.
Recurrence of group B streptococcus colonization LC Colicchia et al
175 Table 1. Patient characteristics at index pregnancy by GBS colonization status
Maternal age at delivery (years)
Table 3. Factors affecting GBS colonization at the subsequent pregnancy: multivariable logistic regression
GBS positive, n = 395
GBS negative, n = 1499
P-value
29.7 ± 4.9
30.3 ± 4.7
0.03 o0.001
Race African-American Caucasian Asian Other
44 199 33 75
Hispanic/Latino ethnicity Parous (%)a BMI
27 (7.5) 41% 30.4 ± 5.5
111 (8.2) 34% 30.2 ± 5.3
Prenatal care provider Resident clinic Private clinic
59 (14.9) 330 (84.8)
146 (9.7) 1337 (90.1)
(13) (57) (9.4) (21)
92 815 68 349
(6.9) (62) (5.1) (26) 0.67 0.011 0.51
GBS colonized in the index pregnancy African-Americanc Preterm delivery of index pregnancya History of o20 week pregnancy lossd
95% CI
P-value
6.28
4.91–8.05
o 0.001
1.80
1.05–3.09
0.03
1.15
1.04–1.27
0.01
0.40 2.90
0.20-.079 1.07–7.85
0.009 0.035
1.27
1.04–1.54
0.017
2.09
1.28–3.44
0.003
0.003
GBS bacteriuria Prior early pregnancy lossb Maternal diabetes Tobacco use Gestational age at delivery (weeks) Preterm delivery ( o37 weeks) Cesarean delivery Chorioamnionitis Peripartum antibiotic administrationc Duration of membrane rupture (hours) NICU admission Inter-pregnancy interval (days)d
GBS negative in the index pregnancy African-Americanc
31 (7.8) 0.42 ± 0.90 36 (9.1) 28 (7.2) 39.2 ± 1.9 25 (6.3) 66 (17) 25 (6.3) 380 (96)
NA 0.49 ± 0.96 103 (6.9) 96 (6.5) 39.5 ± 1.3 48 (3.2) 249 (17) 160 (11) 416 (28)
0.23 0.13 0.62 o0.001 0.004 0.96 0.01 o0.001
8.3 ± 9.3
9.8 ± 19.0
0.15
9 (2.2) 675 ± 406
50 (3.3) 641 ± 369
0.28 0.12
Abbreviations: BMI, body mass index; GBS, group B Streptococcus; NICU, neonatal intensive care unit. Data are mean ± s.d. or n (%). When data were missing, percentages were calculated excluding cases from the denominator. P-values were calculated using t-tests for continuous variables and chi-square tests for categorical variables. aOne or more prior deliveries 420 weeks gestation. bIncludes terminations, miscarriages and ectopic pregnancies at o20 weeks’ gestation. cIncludes antibiotics given during antepartum, intrapartum and postpartum hospitalizations. dCalculated from the delivery of index pregnancy to the last menstrual period of subsequent pregnancy.
Table 2. GBS colonization status in a subsequent pregnancy by colonization status in the index pregnancy GBS colonization in subsequent pregnancy Positive GBS colonization in index pregnancy Positive (n = 395) 198 (50%) Negative (n = 1449) 206 (13.7%) a Po0.001 RR 3.6, 95% CI 3.1–4.3b
Negative 197 (50%) 1293 (86.2%)
Abbreviations: CI, confidence interval; GBS, group B Streptococcus; RR, relative risk. Percentages reported as percentage of women with initial colonization status to have subsequent pregnancy colonization status. a P-value is calculated using the chi-square test. bRelative risk of colonization in the subsequent pregnancy given colonization in the index pregnancy.
novel findings, possibly not previously identified due to the relatively large cohort studied here. The seemingly paradoxical observation that African-American race decreased the risk of colonization during the second pregnancy in patients who were GBS positive in the index pregnancy but increased this risk in women who were GBS negative in the index pregnancy may be © 2015 Nature America, Inc.
All patients GBS colonization at index pregnancy Preterm delivery of index pregnancya History of o20 week pregnancy lossb
Odds ratio
Abbreviations: CI, confidence interval; GBS, group B Streptococcus; OR, odds ratio. Multivariable logistic regression analysis of factors significantly affecting GBS colonization in a subsequent pregnancy. Variables with Po0.25 on univariable analysis were included in the multivariable analysis and removed in a backwards stepwise fashion to the Po0.05 significance level. aLess than 37 weeks gestation. bPrior miscarriage/ termination/ectopic pregnancy either before index pregnancy or between index and subsequent pregnancy. cCompared with Caucasian women. d Prior miscarriage/termination/ectopic pregnancy either before index pregnancy or between index and subsequent pregnancy.
related to the higher rate of colonization in African-American women than others in this study (32% compared with a baseline rate of 21%), with regression towards the mean. Among the strengths of the present study are its large patient population, with demographics and baseline GBS colonization rates comparable to the national average.1,4,13,14 The sample size (five times larger than the largest prior comparable report) provided power to identify differences between the two groups (that is, GBS-colonized and GBS non-colonized in the first pregnancy) as small as 15% for any single continuous variable. All cultures were performed according to the standard, commonly accepted protocol at a single institution. Limitations include the retrospective nature of the study and the lack of prospective follow-up. Also, identification of patients relied upon accurate linkage of microbiology results with the electronic medical record. Some cases may not have been identified properly. Current Centers for Disease Control and Prevention guidelines recommend management of laboring women with unknown GBS status according to a risk-based strategy, with antibiotics given for gestational age o 37 weeks, rupture of membranes 418 h in duration or intrapartum temperature 438 °C.4 Most GBS-colonized women delivering at term do not develop fever or prolonged rupture of membranes and thus would not receive antibiotics according to this risk-based protocol.15,16 However, even in the absence of risk factors, without treatment the infants in these cases have a 5 to 20 times greater risk of invasive neonatal GBS disease than infants born to non-colonized women.15–17 Given that 50% of women who were GBS positive in a prior pregnancy can be presumed to be colonized, including prior GBS-positivity as a risk-based indication for chemoprophylaxis seems reasonable, even in the absence of other risk factors. Such prophylaxis could decrease the incidence of early-onset neonatal disease from approximately 0.8% (that is, 1.6% of the 50% of colonized patients) to 0.05% (0.001% of colonized patients), a 16-fold decrease.4,15–19 Journal of Perinatology (2015), 173 – 176
Recurrence of group B streptococcus colonization LC Colicchia et al
176 A strategy providing intrapartum chemoprophylaxis without prior screening to all women who were GBS-colonized in a previous pregnancy was evaluated hypothetically in a decision analysis, finding that culture avoidance and uniform treatment might be cost-effective.20 However, that study took into account only two prior small studies to determine the risk of recurrent colonization. The study was further limited by the use of manufacturer’s suggested price instead of actual drug prices for penicillin and by the assumption that all patients would be treated with penicillin instead of ampicillin or other antibiotics in the setting of penicillin allergy. Until rapid, accurate bedside GBS screening for women in labor is available and tested prospectively as part of a neonatal GBS disease avoidance protocol,4,21 we suggest that universal screening independent of prior pregnancy status remain the standard. We suggest that prior GBS colonization be considered as an additional indication for intrapartum treatment in women with unknown GBS status. Future studies should focus on maternal and microbial characteristics that determine persistence, transience and recurrence of GBS colonization. Data from such studies may be useful for further refining clinical protocols. CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Stoll BJ, Hansen NI, Sánchez PJ, Faix RG, Poindexter BB, Van Meurs KP et al. Early onset neonatal sepsis: the burden of group B Streptococcal and E. coli disease continues. Pediatrics 2011; 127: 817–826. 2 Yow MD, Leeds LJ, Thompson PK, Mason EO Jr, Clark DJ, Bleacher CW. The natural history of group B streptococcal colonization in the pregnant woman and her offspring. I. Colonization studies. Am J Obstet Gynecol 1980; 137: 34–38. 3 Melin P. Neonatal group B streptococcal disease: from pathogenesis to preventive strategies. Clin Microbiol Infect 2011; 17: 1294–1303. 4 Verani JR, McGee L, Schrag SJ, Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Center for Disease Control and Prevention (CDC). Prevention of perinatal group B streptococcal disease--revised guidelines from CDC, 2010. MMWR Recomm Rep 2010; 59: 1–36. 5 Allardice JG, Baskett TF, Seshia MM, Bowman N, Malazdrewicz R. Perinatal group B streptococcal colonization and infection. Am J Obstet Gynecol 1982; 142: 617–620.
Journal of Perinatology (2015), 173 – 176
6 Lin F-YC, Weisman LE, Azimi P, Young AE, Chang K, Cielo M et al. Assessment of intrapartum antibiotic prophylaxis for the prevention of early-onset group B Streptococcal disease. Pediatr Infect Dis J 2011; 30: 759–763. 7 Towers CV, Rumney PJ, Asrat T, Preslicka C, Ghamsary MG, Nageotte MP. The accuracy of late third-trimester antenatal screening for group B streptococcus in predicting colonization at delivery. Am J Perinatol 2010; 27: 785–790. 8 Yancey MK, Schuchat A, Brown LK, Ventura VL, Markenson GR. The accuracy of late antenatal screening cultures in predicting genital group B streptococcal colonization at delivery. Obstet Gynecol 1996; 88: 811–815. 9 Cheng PJ, Chueh HY, Liu CM, Hsu JJ, Hsieh TT, Soong YK. Risk factors for recurrence of group B streptococcus colonization in a subsequent pregnancy. Obstet Gynecol 2008; 111: 704–709. 10 Tam T, Bilinski E, Lombard E. Recolonization of group B Streptococcus (GBS) in women with prior GBS genital colonization in pregnancy. J Matern Fetal Neonatal Med 2012; 25: 1987–1989. 11 Turrentine MA, Ramirez MM. Recurrence of group B streptococci colonization in subsequent pregnancy. Obstet Gynecol 2008; 112: 259–264. 12 Page-Ramsey SM, Johnstone SK, Kim D, Ramsey PS. Prevalence of group B Streptococcus colonization in subsequent pregnancies of group B Streptococcuscolonized versus noncolonized women. Am J Perinatol 2013; 30: 383–388. 13 Regan JA, Klebanoff MA, Nugent RP. The epidemiology of group B streptococcal colonization in pregnancy. Vaginal Infections and Prematurity Study Group. Obstet Gynecol 1991; 77: 604–610. 14 Stapleton RD, Kahn JM, Evans LE, Critchlow CW, Gardella CM. Risk factors for group B streptococcal genitourinary tract colonization in pregnant women. Obstet Gynecol 2005; 106: 1246–1252. 15 Rouse DJ, Goldenberg RL, Cliver SP, Cutter GR, Mennemeyer ST, Fargason CA Jr. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis. Obstet Gynecol 1994; 83: 483–494. 16 Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS et al. A populationbased comparison of strategies to prevent early-onset group B streptococcal disease in neonates. N Engl J Med 2002; 347: 233–239. 17 Boyer KM, Gotoff SP. Strategies for chemoprophylaxis of GBS early-onset infections. Antibiot Chemother (1971) 1985; 35: 267–280. 18 Adair CE, Kowalsky L, Quon H, Ma D, Stoffman J, McGeer A et al. Risk factors for early-onset group B streptococcal disease in neonates: a population-based casecontrol study. CMAJ 2003; 169: 198–203. 19 Benitz WE, Gould JB, Druzin ML. Risk factors for early-onset group B streptococcal sepsis: estimation of odds ratios by critical literature review. Pediatrics 1999; 103: e77. 20 Turrentine MA, Ramirez MM, Mastrobattista JM. Cost-effectiveness of universal prophylaxis in pregnancy with prior group B streptococci colonization. Infect Dis Obstet Gynecol 2009; 2009: 934698. 21 Honest H, Sharma S, Khan KS. Rapid tests for group B Streptococcus colonization in laboring women: a systematic review. Pediatrics 2006; 117: 1055–1066.
© 2015 Nature America, Inc.
Copyright of Journal of Perinatology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
