High prevalence of Group B Streptococcus colonization among pregnant women in Amman, Jordan

Group B Streptococcus (S. agalactiae; “GBS”) is a gram-positive bacterium with a special capacity to cause perinatal infections of the mother, fetus, and/or newborn. This pathogen causes chorioamnionitis [1], preterm birth [2], stillbirth [3], meningitis [4] and is a leading cause of both early-onset (< 7 days of life) and late-onset (7–89 days of life) neonatal sepsis [5]. Globally, the burden of GBS disease is estimated to be 0.49–0.53 per 1000 livebirths, with a case fatality rate of 8.4–9.6% [6, 7]. The incidence of early-onset GBS disease is estimated to be 0.43 per 1000 livebirths, with a case fatality rate of 12.1%, twice that of late-onset disease [6].

In the United States, recommendations for routine screening for GBS in women between 35 and 37 weeks pregnant, followed by antibiotic prophylaxis 4h prior to delivery for colonized patients, led to a dramatic decrease in the incidence of early-onset GBS disease in neonates [8], with no change in the incidence of late-onset GBS. This screen-and-treat paradigm, however, has not been widely adopted outside of the United States, and policy decisions have been challenged by the absence of solid estimates of the number of at-risk mothers and babies in many parts of the world [9]. One barrier to obtaining reliable epidemiological data from low- and middle-income countries has been the absence of accurate point-of-care tests to determine GBS colonization without the need for time-consuming and expensive, laboratory-based cultivation practices [10].

A 2016 meta-analysis including data for over 70,000 women in 37 countries estimated global prevalence of maternal GBS colonization at 17.9% (95%CI: 16.2–19.7%) [11]. However, the authors noted substantial heterogeneity across and within regions, and reported few studies from Middle East and North African (MENA) countries, highlighting a paucity of maternal GBS prevalence data from the region. From our own review, we found many reports from the MENA region estimating the frequency of GBS in pregnant mothers, with estimates ranging from 1.6 to 32% (Table 1), however nearly half of these studies were conducted in only two countries – Iran and Israel – and over one-third of the studies were published more than 10 years ago. In Jordan, universal screening is not routine. Only one manuscript, published in 1991, estimated the GBS prevalence among 500 pregnant women in Jordan, finding nearly one-third (30.4%) of women were colonized by vaginal, rectal, and/or urine specimens [35]. Prior to launching future studies to assess novel diagnostics and vaccines, it is essential to establish the current burden of GBS colonization in the region. Given the scarcity of information about GBS burden of colonization in Jordan, we conducted a pilot study to determine the prevalence of GBS recto-vaginal colonization among pregnant women presenting for labor at one high-volume government hospital in Amman, Jordan, where GBS testing is not routine. Secondary objectives were to assess capsular serotypes of GBS specimens and to compare the performance of a GBS rapid diagnostic to culture.

Our surveillance study revealed nearly one in five women presenting for labor at Al-Bashir Hospital in Jordan were colonized with GBS. This represents a substantial burden of colonization within a context of no routine screening for GBS among pregnant women. Our findings are consistent with a worldwide-adjusted estimate for maternal GBS colonization of 18%, but a higher prevalence compared to Southern Asia (10%) and Eastern Asia (9.1%) [9]. However, our prevalence estimate is lower than the only other local data from Jordan, which reported 30.4% GBS colonization [35]. That study enrolled 500 women and included a positive GBS urine culture, rectal swab, and/or vaginal swab as part of their GBS colonization definition, but these data are nearly 30 years old [35]. The frequency of GBS colonization we found is average when compared to other countries in the MENA region (Table 1), where the average was 16.9% among 18,805 total subjects, with minimum of 1.6% and peak of 32% of subjects with GBS colonization. Thus, our study highlights that GBS colonization is common and potentially an important cause of neonatal disease. While our study did not include neonatal data, a 2017 study exploring neonatal sepsis at a tertiary hospital in Jordan found that GBS was isolated from 10% of neonatal sepsis cases; however, the authors did not explore overall prevalence of GBS among pregnant women [56]. Therefore, our study emphasizes that routine testing for GBS in mothers would be valuable in Jordan so IAP could be introduced to prevent early-onset GBS neonatal disease.

Based on clinical trials and observational studies, early-onset GBS disease can be prevented by the administration of IAP during labor to GBS colonized women, with potential efficacy of 80% [57, 58]. Therefore, two screening methods have been implemented to identify GBS antenatally: risk-based or universal culture-based screening between 35 and 37 weeks. In 2002, CDC recommended switching to culture-based universal screening. Subsequently, an uptake of prenatal screening and IAP was rapid and widespread [8]. In addition, the incidence of invasive early-onset GBS disease decreased by more than 80% from 1.8 cases/1000 live births in the early 1990s to 0.26 cases/1000 live births in 2010, estimating that over 70,000 cases of early-onset GBS invasive disease were prevented in the United States [8]^. However, globally, the majority of countries that routinely screen for GBS are risk-based, and unfortunately very few of these countries are from the MENA area, including Jordan which does not routinely test for GBS colonization. In our study, only 9/39 (23.1%) women with confirmed GBS received an intrapartum antibiotic; five of these were also diagnosed by rapid test.

There are no established international standards for sampling GBS colonization; however, CDC recommends rectovaginal swabs at 35–37 weeks with selective enrichment broth culture, but this approach is not always feasible for low and middle-income settings [9]. Additional barriers to routine testing in these low resource settings may include the lack of microbiology capacity to perform routine testing for GBS, lack of timely knowledge of the results being reported back to the clinicians because of separate settings where prenatal care is performed and where lab testing would be performed, and lack of electronic medical records for the obstetricians to obtain antenatal results. One way to address these issues is the introduction of point-of-care testing when women present in labor. Unfortunately, the rapid diagnostic used in this pilot study was substantially less sensitive and specific than culture. Furthermore, over one-quarter of women who received an intrapartum antibiotic due to a positive rapid GBS antigen test (11/43, 25.6%) were found later to be GBS culture-negative, leading to unnecessary use of antibiotics. Our GBS antigen test results were poorer than other studies using other antigen-based tests that yielded 57.3% sensitivity and 99.5% specificity in Canada [59] and 100% sensitivity and 92.9% specificity in India [60]. Other point-of-care testing methods such as PCR testing have been reported to have high sensitivity (92.9–100%) and specificity (81.1–97.5%) [19, 61, 62] and may be a better solution. However, these higher-performing tests typically are cost-prohibitive for lower-resource settings [12, 63], and further development of affordable point-of-care GBS tests are urgently needed, particularly those that also simultaneously provide information about antibiotic sensitivities.

GBS has 10 known serotypes; however, five (Ia, Ib, II, III, and V) colonize the rectovaginal tracts of women in all regions, accounting for 98% of serotypes globally [9]. Moreover, these five serotypes represent 97% of invasive isolates in all regions with serotype data [7], with serotype III the most prevalent serotype across the United Nations sub-regions. Our study also identified serotype III as the most prevalent, followed by Ia and II serotypes. This is similar to other studies that report serotype III as the first- or second-most common isolate colonizing women [64]. Knowing which serotypes are most widespread will be important when GBS conjugate vaccines become available [65]. Therefore, further studies are needed to know which serotypes are responsible for both colonization and invasive neonatal disease in the MENA region, including Jordan.

Our results should be viewed in light of a number of limitations. This study was conducted at one facility – a government hospital – over 2 months and may not be generalizable to other settings in Jordan or the MENA region. However, the frequency of GBS colonization is similar to the average reported for all MENA regions. Furthermore, we are unable to report neonatal outcome information and our use of one senior microbiologist may have affected our results. In addition, we did not test for antibiotic susceptibility, which is important considering reports of increasing erythromycin and clindamycin resistance in GBS [66]. Still, given the dearth of information about GBS in Jordan and the wider MENA region, our results help to shed light on the high prevalence of GBS colonization among pregnant women.

We found high prevalence of GBS and both under- and over-treatment of GBS among pregnant women in Jordan. These results highlight the unmet need for routine GBS testing during pregnancy and support expanded research in the region, including defining the GBS resistance patterns, serotyping information, risk factors, and neonatal outcomes. They also emphasize the need for improved rapid GBS diagnostics for developing world settings.