Prevalence, risk factors, and adverse outcomes of bacterial vaginosis among pregnant women: a systematic review

Abstract

Introduction

Bacterial vaginosis (BV) is one of the most common genital tract infections among women of reproductive age. The existence of BV among pregnant women has momentously attracted the attention of both clinicians and the scientific community due to its potential link with adverse clinical outcomes in pregnancy.

Methods

To evaluate the prevalence, risk factors, and adverse outcomes of bacterial vaginosis among pregnant women, a comprehensive systematic review was conducted based on the preferred reporting items for systematic review and meta-analyses (PRISMA) criteria. PubMed, ScienceDirect, ClinicalTrials.gov and Cochrane database searches were conducted independently by two authors until May 13th, 2023.

Results

The search strategies yielded a total of 2237 records; among them, 12 studies met the inclusion criteria and were included in the qualitative synthesis. Majority of the included studies demonstrated a high prevalence of BV among African women. The risk of developing BV during pregnancy was highest among women with multiple sexual partners. Additionally, factors including age, socioeconomic status, unhygienic practices, ethnicity, 2nd trimester, spontaneous abortion, vaginal douching, symptoms, and history of sexually transmitted infections (STIs) were also associated with a higher prevalence of BV. Overall, 7 studies reported adverse outcomes during pregnancy which was directly associated with BV. Based on the review, it was found that PROM, PTB, and LBW were the most frequently reported adverse outcomes in pregnant women with BV.

Conclusion

In summary, the high prevalence of bacterial vaginosis necessitates a global surveillance approach to delineate the health risks imposed on both mother and child, and promote cost-effective strategic measures to alleviate the undesired consequences of BV during pregnancy.

Introduction

Globally, reproductive tract infection (ReTI) is a significant public health concern, especially in developing nations. The World Health Organization (WHO) reported the incidence of curable ReTI or sexually transmitted infections (STIs) among 357 million individuals every year on a global scale. These infections may occur either symptomatically or asymptomatically [1]. Bacterial vaginosis (BV) is one of the most frequently occurring genital tract infections among women of reproductive age [2]. BV is a dysbiosis of the vaginal microbiome, which increases the risk ReTIs transmission and leads to considerable physical and psychosocial discomfort [3]. In the United States, approximately one-third (29%) of women between 14 and 49 years have symptomatic or asymptomatic BV [4]. The typical clinical symptoms are smelly (fish-like) vaginal discharge along with itching, burning, and pain sensation [5].

Approximately one-third of the women in the world are BV positive, and this prevalence is reportedly higher among pregnant women from developing countries (Latin America, Asia, and most African countries). However, the highest prevalence of BV was reported in women of African descent [6]. BV occurs due to the perturbation of the vaginal ecosystem and the replacement of peroxidase-producing Lactobacilli as the normal vaginal flora with anaerobic bacteria. Decreased hydrogen peroxide-producing bacteria predispose to colonization of the causative organisms of BV [5]. Microbiologically, this vaginal infection is characterized by replacing lactobacillus-dominated vaginal microbiota through variable mixtures of strictly and facultative anaerobic bacteria, including Gardnerella vaginalis, Atopobium vaginae, and Mobiluncus spp [7].

The diagnosis of bacterial vaginosis involves the use of clinical (Amsel’s) and lab-based (Gram stain with an objective scoring system) methods [8]. Clinically, BV can be diagnosed through the presence of clue cells on microscopy or a vaginal pH ≥ 4.5, or based on the fishy odour of vaginal discharge. Nugent score is considered the gold standard diagnostic criteria for BV. This scoring system is a laboratory-based method that quantifies the number of Lactobacillus spp. present in vaginal swabs in comparison to other bacterial morphotypes such as Gardnerella vaginalis, Mobiluncus spp., and Prevotella/Bacteroides spp [9]. Amsel’s criteria are generally preferred in the clinical setting since Nugent criteria require considerable time and expert microscopic skills [10]. Amsel’s criteria include greyish white, thin, and homogeneous vaginal discharge with a pH value higher than 4.5, fishy or amine odour after the addition of 10% potassium hydroxide (KOH) and the presence of clue cells (> 20%) on microscopic examination. Using Amsel’s criteria, a BV positive diagnosis is attained if at least any three of the following four criteria are fulfilled [11]. The Nugent scoring system diagnoses BV based on a positive [7–10], intermediate [4–6], and a negative score (0–3) [12].

Symptomatic BV can be treated with oral metronidazole 500 mg twice daily for 7 days. Topical alternatives include vaginal metronidazole gel and clindamycin cream. In multiple recurrent cases of BV, longer therapeutic courses are often recommended [8]. Since BV treatment is based on the presence of abnormal vaginal discharge and the standard syndromic management of STIs, asymptomatic pregnant women may tend to remain untreated due to this therapeutic approach [9].

In the past, several risk factors have been associated with BV, including race and ethnicity, low socio-economic status, antibiotic therapy, multiple sex partners, smoking, and young or teenage age [5]. However, the risk factors for BV during pregnancy are inconsistent, underreported, and often overlooked [13]. Hence, BV risk factors are typically more prevalent among pregnant women and require further exploration. Clinicians need to understand the risk factors, symptoms, and potential sequelae of BV to monitor and treat women of reproductive age effectively [13].

The existence of BV among pregnant women has momentously attracted the attention of both clinicians and the scientific community due to its potential link with adverse clinical outcomes during pregnancy [7]. BV has been associated with a higher risk of multiple adverse outcomes, including preterm delivery, miscarriage, pelvic inflammatory disease, endometritis, and transmission or acquisition of human immunodeficiency virus type 1 (HIV-1) and STIs [3]. Increasing evidence indicates that vaginal microbiota (VMB) dysbiosis may be the causative reason behind premature rupture of membranes (PROM) and preterm birth (PTB) [14]. The mechanisms involved in pregnancy complications induced by BV are not fully understood, but it has been suggested that bacterial ascension from the vagina to the membranes and amniotic fluid results in preterm birth [7]. The practical implications of BV during pregnancy are not yet well defined. Further research is needed to formulate the necessary guidelines for prevention and address the associated adverse (maternal/fetal) outcomes [15]. To the best of the authors’ knowledge, no previous study to date has explored the current body of literature to collectively synthesize and report the prevalence, risk factors, and adverse outcomes of bacterial vaginosis among the pregnant population. Therefore, this is the first systematic review that: (i) delineates the global prevalence of BV among pregnant women receiving antenatal care; (ii) identifies the risk factors associated with BV and; (iii) explores the adverse outcomes during pregnancy due to BV.

Methods

To evaluate the prevalence, risk factors, and adverse outcomes of bacterial vaginosis among pregnant women, a comprehensive systematic review was conducted based on the preferred reporting items for systematic review and meta-analyses (PRISMA) criteria [16].

Literature search

PubMed, ScienceDirect, ClinicalTrials.gov and Cochrane database searches were conducted until May 13th, 2023. The comprehensive search was performed based on the following medical subject heading (MeSH) keywords: “bacterial vaginosis” OR “BV” AND “risk factors” AND “adverse outcomes” AND “pregnant” OR “pregnancy” and published between 2013 until 2023. Finally, the reference lists of articles were screened to determine any additional studies on BV among pregnant women in general. Searching and collecting the relevant papers were carried out independently by two reviewers.

Eligibility criteria

A specific screening process was followed to include or exclude studies in this systematic review, which was performed by two different authors on the basis of predefined eligibility criteria. The selection process of articles is presented in the PRISMA flowchart (Fig. 1). The inclusion criteria were: (1) observational, cross-sectional, and prospective cohort studies, systematic reviews, and meta-analyses on the prevalence, risk factors, and adverse outcomes of BV among pregnant women, and (2) articles published in the English language only. The exclusion criteria were: (1) editorials, basic literature reviews, and case report studies (2), conference abstracts, preprints, and duplicate publications (3), studies that assessed BV in non-pregnant women (4), studies that focused solely on the assessment of other conditions: Aerobic vaginosis, Candida albicans (vulvovaginal candidiasis) and Trichomonas vaginalis (trichomoniasis), or STIs, and (5) studies with missing information on BV prevalence, risk factors, and adverse outcomes.

Fig. 1.

PRISMA based flow diagram displaying the study selection

Study selection

Titles and abstracts screening was conducted independently by two reviewers (Fig. 1). Thereafter, the selected articles were fully read to determine if they met the inclusion criteria. If there were any uncertainties to include or exclude an article, a third reviewer was consulted to achieve mutual consensus. The bibliographies of articles were examined to retrieve further potential articles containing relevant data before reaching any final decision on the study selection. The studies selection ensured adequate sample size, sound methodology, identifying bias and confounding factors, appropriate data extraction and statistical analyses all added up to the quality assessment.

Data extraction

Data extraction was performed independently based on a data collection sheet designed by two reviewers, one person extracted the information and another person evaluated it. A list of included information was collected during the data extraction process which has been tabulated as follows: (1) first author’s name and study year (2), country/region of study (3) total number of study subjects (4), study setting (5), age range of participants (6), patients’ characteristics (7), prevalence of BV (7), name of microorganism involved (8), risk factors of BV (9), symptoms (10), adverse outcomes in pregnancy, and (11) summary of findings (Table 1). Although a formal quality assessment tool was not utilized in this systematic review, the methodological rigor of the included studies was considered during the review process. Key aspects, such as study design, sample size, clarity of outcome measures, and statistical methods, were critically reviewed to ensure the reliability and validity of the findings. This qualitative assessment provided a basis for interpreting the results and drawing conclusions.

Table 1.

Study characteristics and qualitative findings on the prevalence, risk factors, and adverse outcomes of bacterial vaginosis in pregnancy

First author, Year	Sample size (n)	Study setting	Age (years)	Patient characteristics	BV Prevalence % (n)	Microorganism	Symptoms	Risk factors	Adverse outcomes	Summary of findings
Krauss-Silva, 2014 [17]	1699	Women attending prenatal public services in Rio de Janeiro between 2006 and 2008, were evaluated and enrolled in a prospective study	-	Low socioeconomic and multiethnic South American population Asymptomatic pregnant women, < 20 weeks’ gestation, with no indication for elective PD and without risk factors of spontaneous PD	Prevalence of asymptomatic BV: 28.1% (1699) The prevalence of BV among white and black women was 28.1% (95% CI: 24.6–32.0%) and 32.5% (95% CI: 28.2–37.2%), respectively.	Mobiluncus species	-	Black ethnic background	Spontaneous preterm delivery The incidence of < 37 weeks’ spontaneous PDs among BV pregnant women with a pH = > 4.5 was 3.8%.	The prevalence of asymptomatic BV was higher in black women (p > 0.05). The RRs of spontaneous PD < 34 and < 37 weeks among women with BV, as compared with those with intermediate states were not statistically significant (p > 0.05).
Mengistie, 2014 [2]	252	57 symptomatic and 195 asymptomatic pregnant women attending antenatal care in Tikur Anbessa University Hospital, Addis Ababa, from November 2011 to April 2012 screened using Gram stain Nugent scoring system were randomly selected	18–40	Mean age: 27.6 (± 4.7) years; Residence: urban; Married: 241 (95.6%); Only one lifetime sexual partner: 201 (79.8%); Multigravida pregnant women: 161 (63.9%); Previous history of spontaneous Abortion: 24 (14.9%)	Overall BV prevalence: 19.4% (49) Symptomatic pregnant women: 31.6% (18) Asymptomatic pregnant women: 15.9% (31)	G. vaginalis Mobiluncus species	The presence of abnormal vaginal discharge (p = 0.01) and unpleasant smell (p = 0.005) were reported vaginal symptoms associated with bacterial vaginosis.	Multiple lifetime sexual partners and previous history of spontaneous abortion	-	The prevalence of bacterial vaginosis is higher among asymptomatic pregnant women and associated with a previous history of multiple lifetime sexual partners (OR: 8.6; 95% CI: 2.5, 29) and spontaneous abortion (OR: 5.9; 95% CI: 1.5, 23).
Afolabi, 2016 [18]	246	A prospective observational study was conducted between August 2012 and January 2013 with high vaginal swabs obtained from consecutive newly registered antenatal women between 14–36 weeks gestation	20–44	Mean age: 30.9 ± 4.5 years All Southern Nigerian women and the majority were married (98.8%) and working women (77.2%)	26% (64)	G vaginalis/ Gram-negative rods, and Mobiluncus species	Vaginal discharge, vulval itching, and offensive discharge	-	Of the 64 women who were positive for BV, 25% had preterm delivery, 14.1% had LBW, and 29.7% had PROM. Bacterial vaginosis was significantly associated with preterm delivery (risk ratio [RR], 2.68; 95% confidence interval [CI], 1.44–4.98), low birth weight (RR, 3.20; 95% CI, 1.29–7.94), and premature rupture of membranes (RR, 6.75; 95% CI, 3.11–14.67).	The prevalence rate of BV among pregnant women is high and is significantly associated with adverse pregnancy outcomes. Routine screening and treatment of women pre-conceptually may enable interventions to prevent these adverse outcomes.
Tachawatcharapunya, 2017 [19]	270	The prospective observational study was conducted at the Department of Obstetrics and Gynaecology, Faculty of Medicine Siriraj Hospital, Mahidol University between July 2015 and July 2016 among asymptomatic pregnant women during the early third trimester.	18–35	Mean age: 27.2 ± 5.3 years Mean gestational age at diagnosis: 29.2 ± 1.5 weeks Education above high school: 97.8% (44) Primiparity: 64.4% (29) SI during pregnancy: 73.3% (33)	19.3% (45)	Gardnerella/ Bacteroides spp	-	Primiparity was more common in the BV group (64.4%, p = 0.039)	PROM (IF 1.6, 95% CI 0.6–4.5; BV 2.2, 95% CI 0.8–6.2, p = 0.339), PB (IF 1.3, 95% CI 0.3–5.1; BV 2.3, 95% CI 0.6–9.4, p = 0.489) and LBW (IF 1.4, 95% CI 0.5–4.1; BV 1.5, 95% CI 0.5–4.9, p = 0.761).	BV during GA28-32 weeks in asymptomatic Thai pregnant women was prevalent at 19.3% and tended to increase pregnancy complications, including premature rupture of membrane, preterm birth, and low birth weight. However, no statistical significance was demonstrated
Sabour, 2018 [5]	43 studies	A comprehensive systematic review and meta-analysis approach was undertaken using the PRISMA criteria, and electronic databases were searched until December 01, 2017, for eligible studies	15–45	Iranian pregnant and non-pregnant women	Overall BV prevalence: 18.9% (95% CI 14–25) Pregnant women: 16.5% (95% CI 12.5–21.6).	Gardnerella vaginalis	-	Race and ethnicity, low socio-economic status, antibiotic therapy, multiple sex partners, smoking, and young or teenage age	Preterm premature rupture of membranes, preterm labour, abortion, and postpartum infections.	Prevalence of BV was higher than those reported in pregnant women from the USA (11–12%), Burkina Faso (6.4%), France (7.1%), Sweden (9.3%), Portugal (3.88%) and the United Kingdom (3.54%)
Aduloju, 2019 [20]	362	A descriptive cross-sectional study was conducted from April 2017 to June 2017 in the Department of Obstetrics and Gynaecology, EKSUTH, Ado-Ekiti, southwestern Nigeria.	19–40	Mean age: 26.24 ± 6.14 years Gestational age (2nd trimester): 71.7% (43) Tertiary education: 53.3% (32) Married: 95% (57) Employed: 86.7% (52) Parity (1–4): 75% (45)	16.6% (60)	-	Vaginal discharge, dysuria, dyspareunia, and lower abdominal pains	Age group 25–34 yrs, multiparity, higher education, and symptoms were significantly associated with BV (p < 0.05) Multiparity, higher education, marital relationship, and increased lifetime sexual partners	Women with BV had prelabour rupture of fetal membrane and had babies with prematurity, low birth weight, and Apgar score less than 5 at one minute	There should be screening for BV in pregnant women presenting with abnormal vaginal discharge so that they could be treated accordingly. This will mitigate the complications arising from bacterial vaginosis.
Kamga, 2019 [21]	309	A cross-sectional survey was administered to 309 women seeking antenatal care (ANC) in three health facilities in KHD between May to July 2016	18–37	Mean age 26.97 years With regards to place of residence, over half (63.4%) of the study participants resided in the urban area. Most (72.5%) participants were married and had secondary education (36.9%). Two hundred and twenty-eight (73.8%) women were multigravida with a mean gravidity of 2.86. Two hundred and seven (66.9%) of them had delivered and the mean parity was 1.48.	26.2 (81)	Gardnerella vaginalis Bacteroides fragilis Mobiluncus	-	BV was higher (29.5%) in rural area women (χ2 = 8.609. P = 0.014), those who did not use antibiotics (31.9%) prior to the study (χ2 = 12.893, P = 0.002), and women with no history of a genital tract infection (χ2 = 18.154, P = 0.001). There was a significant difference in prevalence with respect to gestation age (χ2 = 13.959, P = 0.007) with the highest occurring in women in the second trimester (31.7%). Women who practiced douching (χ2 = 23.935, P = 0.000) and did not wash pants with disinfectant (χ2 = 7.253, P = 0.027) had higher BV risk	-	BV prevalence was affected by some hygiene behaviors, sociodemographic and clinical factors. Screening and treatment of positive cases during antenatal visits to prevent adverse outcomes, as well as education of women on vaginal hygiene is highly recommended.
Tesfay, 2020 [22]	31 articles	A literature search was conducted in PubMed/ MEDLINE, science direct, PMC, and Free-text Web Searching using Google Scholar from January 2008 to December 2018.	-	BV among ANC pregnant women from the African region	Samples from a total of 19,354 pregnant women 4795 were positive for BV with an overall prevalence of 24.8% (4795/19,354) and the highest cumulative prevalence was reported from two studies conducted in more than two countries (1113/2349, 47.4%).	-	-	Fifteen out of 31 articles or studies (48.4%) are significantly associated with different risk factors. Out of the 15 articles 33.3% (5/15) of the associated risk factors are HIV followed by age and previous BV/STI (20%). Pregnant women who had a spontaneous history of abortion and the early stages of pregnancy were at high risk of being BV positive	-	The prevalence of BV among pregnant women was different in all regions of Africa with a pooled prevalence of 24.8%. HIV-infected pregnant women and sexually active women (mostly in the age group of 20–30 years) were at high risk for BV and some studies reported BV positive women are also at high risk of acquiring HIV infection.
Bonneton, 2021 [23]	457	Nested within the BIRDY study, the present study consecutively recruited pregnant women in Guédiawaye (suburban neighborhood in Dakar) and Sokone (rural area near the Gambian border) primary health centers, from October 2013 to September 2018.	18–35	Most women were married (417/457, 91.2%) and 293/457 (64.1%) had a formal education (at least primary education). The median age was 28.1 years. Third semester of pregnancy	Overall, BV prevalence was 18.6% (85/457) [95% CI 15.4–22.6]) and was similar in suburban and rural areas (18.9% versus 18.1%, p = 0.843).	Gardnerella vaginalis Streptococcus and Candida spp	-	Multivariate analysis showed that primigravidity was the only factor independently associated with a lower risk of BV (aOR 0.35 [95% CI 0.17–0.72]).	-	Before authorities consider systematic BV screening for pregnant women, a larger study would be useful in documenting the prevalence, risk factors, and the impact of BV on pregnancy outcomes.
Mulinganya, 2021 [24]	533	The current study is part of the AVEONS study, whereof pregnant women seeking antenatal care were recruited in the Provincial Reference Hospital of Bukavu, DRC, between January and October 2017, and followed until delivery	20–35	The median age was 28.0 years (IQR 8.0 years). All participants lived in Bukavu. The majority lived in poverty (72.6%), completed primary school (88.4%) and nearly all were married (95.5%). Most participants were from the Shi tribe (66.7%) and almost all were Christians (93.7%)	The prevalence of BV was 26.3% (138) and approximately 47.1% (65) of the women with BV were asymptomatic	Gardnerella vaginalis/Bacteroides (Gram-variable coccobacilli), Mobiluncus (Gram-negative curved rods) cell types, and Candida cells and/or hyphae	-	Independent risk factors for BV were the use of alternatives to water for intravaginal washing, concurrent partners, unemployed status, the presence of vaginal Candida and clay consumption	BV was independently associated with both LBW and PTB of an infant with LBW.	BV was associated with adverse pregnancy outcomes. Hence, research on modifiable risk factor-based interventions to reduce the prevalence of BV, and on screening/ treatment of BV during antenatal care should be explored to reduce neonatal mortality and morbidity.
Yalew, 2022 [6]	422	This hospital-based descriptive cross-sectional study was conducted from February to June 2019 at Ayder Comprehensive Specialized Hospital (ACSH), Mekelle city	18–47	The mean age was 27.2 ± 5.0 years. Most of the study participants were married 78 (19.4%) and urban residents 90 (20.0%). In addition, the majority of the study participants had completed secondary school 42 (24.6%) and were housewives 47 (21.1%) Multigravida: 57 (21.6) Gestational age (2nd trimester: 49 (23.7)	The total prevalence of BV was 20.1% (85/422) which included BV alone at 12.3% (52/422), BV with AV at 5.5% (23/422), BV with candidiasis 1.7% (7/422), and BV + AV + trichomoniasis 0.7% (3/422).	Enterobacteriaceae Staphylococcus	White homogenous discharge	Symptoms of BV [2.672 (1.547, 4.615), (P < 0.001)] and second trimester [0.563 (0.324, 0.979), (P = 0.042)] were found significantly associated with BV (P < 0.05).	-	BV was more common in symptomatic vs. asymptomatic people (P < 0.001), and in second-trimester vs. first-trimester samples (P = 0.042).
Ng, 2023 [15]	237	A prospective cohort study over one-year duration was conducted from December 2014 until December 2015, involving 237 women who presented with abnormal vaginal discharge, preterm labour, or preterm prelabour rupture of membrane between 22- and 34-weeks period of gestation	26–32	The median age for the study sample was 30.0 years. Almost all the pregnant women were married (98.3%). The majority were Malay (67.9%), followed by Chinese (23.2%), Indian (4.2%), and others (4.6%). More than half of the pregnant women were multiparous (58.4%) with a median monthly income of 3,500 Malaysian Ringgit	10.1% (24)	Gardnerella vaginalis	Abnormal vaginal discharge	-	There was a significantly higher preterm birth rate, below 34 weeks (22.7% vs. 6.2%, p = 0.019) in women with BV. Placental pathology revealed more than half (55.6%) of women with BV had histologic chorioamnionitis. Neonatal morbidity was significantly higher with exposure to BV, with lower median birth weight, higher rate of neonatal intensive care unit admission (41.7% vs. 19.0%, p = 0.010), increased intubation for respiratory support (29.2% vs. 7.6%, p = 0.004) and respiratory distress syndrome (33.3% vs. 9.0%, p = 0.002).	More research is needed to formulate guidelines for prevention, early detection and treatment of BV during pregnancy to reduce intrauterine inflammation and the associated adverse fetal outcomes

Outcome measures

Various adverse outcomes in relation to BV were investigated among the pregnant women receiving antenatal care. These outcomes included PTD, PTB, PROM, LBW, preterm labor (PTL), spontaneous abortion, premature baby, intrauterine growth restriction (IUGR), chorioamnionitis, and intrauterine fetal death or stillbirth.

Results

Study characteristics

The search strategies yielded a total of 2237 records; among them, 12 studies [2, 5, 6, 15, 17–24] met the inclusion criteria and were included in the qualitative synthesis. The article selection process is shown in the PRISMA flow diagram (Fig. 1). The majority of the studies were conducted in Africa including Ethiopia (n = 2), Nigeria (n = 2), Cameroon (n = 1), Senegal (n = 1), and the Democratic Republic of Congo (n = 1). The remaining studies on pregnant women with BV were conducted in Brazil (n = 1), Bangkok (n = 1), Malaysia (n = 1), and Iran (n = 1). Most of the included studies were observational prospective cohorts (n = 7), followed by descriptive cross-sectional studies (n = 3), and systematic reviews (n = 2). The age range of the patients was between 18 and 47 years. Both symptomatic and asymptomatic pregnant women with BV were reported in this review. The majority of the BV-positive women were in the 2nd trimester of pregnancy, multigravida, educated, married, working, and belonged to an African race. The study characteristics and qualitative findings on the prevalence, risk factors, and adverse outcomes of bacterial vaginosis among pregnant women are synthesized in a tabular form (Table 1).

Prevalence and risk factors of BV

Among the included studies, the highest BV prevalence was noted among women attending prenatal public services in Rio de Janeiro, Brazil, between 2006 and 2008 [17]. The results showed that the prevalence of BV among white and black women was 28.1% (95% CI: 24.6–32.0%) and 32.5% (95% CI: 28.2–37.2%), respectively [17]. The second highest prevalence of BV (26.3%) was reported among women from the Democratic Republic of Congo, along with the highest number of asymptomatic BV cases (47.1%) diagnosed during pregnancy [24]. While the lowest prevalence of BV (10.1%) recorded in this review was in Malaysia [15].

In the present study, Gardnerella vaginalis was the most frequently identified bacteria in pregnant women with BV. The presence of abnormal vaginal discharge was the most prominent symptom of bacterial vaginosis, and the majority of the women complained about it during antenatal check-ups. The risk of developing BV during pregnancy was higher among women with multiple sexual partners based on the cumulative evidence from the included studies [2, 5, 20, 24]. Additionally, factors including age [5, 20, 22], socioeconomic status [5, 21, 24], unhygienic practices [21, 24], ethnicity [5, 17], gestational age of 14–26 weeks (2nd trimester) [21, 24], history of spontaneous abortion [2, 22], vaginal douching [21, 24], symptoms [6, 20], and history of STIs [22] were also associated with a higher prevalence of BV.

Adverse outcomes in pregnant BV women

Overall, 58.3% (n = 7) of studies reported adverse outcomes during pregnancy which were directly associated with bacterial vaginosis. In Brazil [17], the incidence of < 37 weeks’ spontaneous preterm deliveries (PDs) among BV pregnant women with a pH = > 4.5 was 3.8%. Affolabi et al. found that out of 26% of women (n = 64) who were diagnosed with BV, 29.7% had PROM, 25% had a preterm delivery, and 14.1% had LBW [18]. It was further noted that BV had a significant relationship with PD (RR, 2.68; 95% CI, 1.44–4.98), LBW (RR, 3.20; 95% CI, 1.29–7.94), and PROM (RR, 6.75; 95% CI, 3.11–14.67). The study in Bangkok [19] also reported similar BV adverse outcomes among Thai asymptomatic pregnant women during the gestational age of 28–32 weeks. Although the findings were not statistically significant it was revealed that BV tends to increase the risk of PROM, PTB, premature baby, and LBW. The findings were in line with a systematic review in Iran [5], a cross-sectional study in Nigeria [20], and a cohort study [24] that highlighted PROM, PTB, and LBW as some of the important adverse effects of BV in pregnant women. The Malaysian study found a significantly higher preterm birth rate, below 34 weeks (22.7% vs. 6.2%, p = 0.019) in women with BV. Placental pathology revealed more than 55.6% of women with chorioamnionitis. Neonatal morbidity was significantly higher with lower median birth weight, higher rate of neonatal intensive care unit admission (41.7% vs. 19.0%, p = 0.010), increased intubation for respiratory support (29.2% vs. 7.6%, p = 0.004) and respiratory distress syndrome (33.3% vs. 9.0%, p = 0.002) due to BV exposure [15]. Based on the review of the included studies, it was found that PROM (n = 4), PTB (n = 4), and LBW (n = 4) were the most frequently reported adverse outcomes in pregnant women with BV. The adverse outcomes reported in the included studies have been presented below (Fig. 2).

Fig. 2.

Adverse outcomes reported in BV-positive pregnant women

Discussion

This study aimed to review the factors associated with BV and adverse pregnancy outcomes to highlight the gravity of this issue among pregnant women. Studies on the BV prevalence, risks, and outcomes were systematically identified and reviewed to synthesise all relevant literature on the topic. Bacterial vaginosis has emerged as a global women’s health issue due to its association with pelvic inflammatory infections, cervicitis, endometritis, urinary tract infections, postoperative infections, spontaneous abortions, preterm delivery, increased susceptibility to HIV, and abnormal Pap smears [25]. One of the most prevalent lower genital tract infections among women of reproductive age is BV; however, its diagnosis can be problematic. Around 50% of women with no signs or symptoms of BV have been identified with Gardnerella [25]. It is crucial to determine the prevalence, risk factors, and adverse impact of BV so that early screening, diagnosis, and effective management can be implemented among pregnant women.

Worldwide, the percentage of pregnant or non-pregnant women affected by BV at any given point of time varies between 5% and 70% from country to country, and it may also differ within the same county e.g., the differences in prevalence in certain parts of Africa (highest in southern and eastern Africa). BV is least prevalent in Asia, Europe, and America, and the prevalence is much higher in developing countries compared to developed countries [22]. The reported prevalence of BV among pregnant women globally ranges between 4.9 and 49% [13]. The estimated BV prevalence rates in pregnant women as cited by Sabour et al. [5] include the following: Kenya (37%), Zimbabwe (32.5%), India (20.5%), %), Ethiopia (19.4%), Nigeria (17.3%), Denmark (17%), USA (11–12%), Sweden (9.3%), France (7.1%), Burkina Faso (6.4%), UK (3.54%), and Portugal (3.88%). Various factors including the clinical setting, sociodemographic factors, diagnostic criteria, gestational age, etc., may influence the prevalence of BV among pregnant and non-pregnant populations [13]. Among the included studies, a high prevalence (32.5%) was noted among South-American black women attending prenatal public services in Rio de Janeiro, Brazil, followed by 26.3% in residents of Central Africa. But since this study analysis in Brazil was based on old data collected (2006–2008), there may be certain discrepancies in the current BV statistics. Based on the inclusion criteria of this review, majority of the studies on BV published within the 10 years’ time frame were conducted in certain parts of Africa. Therefore, our study sample is biased towards African population. However, a high prevalence of BV among African pregnant women is also evident in the current literature. A study on pregnant low-income African-American women reported a prevalence of 25% [13]. Similarly, another study in the sub-Saharan African countries suggested variations in the BV prevalence from 25 to 50% and from 9 to 23% [23].

BV is common in both symptomatic and asymptomatic pregnant women; however, the majority of the studies supported a higher BV prevalence among asymptomatic women [2, 19, 22]. According to a systematic review published in Africa, approximately 50 − 80% of BV-positive patients were asymptomatic [22]. In contrast, the study conducted on Ethiopian pregnant women found that symptomatic pregnant women with symptoms of white homogenous discharge were significantly (P = 0.001) 2.7 times higher to be BV positive than asymptomatic women [6]. However, their study population included women with mixed vaginal infections (aerobic vaginitis, candidiasis, and trichomoniasis) and not BV alone. Another study in India reported about 48.4% of asymptomatic BV cases and was conducted in non-pregnant women [26]. Overall, there appear to be some differences in the findings in relevance to the symptomatic vs. asymptomatic criteria which is most likely related to the different study populations and characteristics.

The majority of the included studies indicated that women with concurrent sexual partners have a significantly higher risk of developing BV. Although there could be additional risk factors such as age, socioeconomic status, unhygienic behaviours, ethnic background, gestational age, vaginal douching, and history of spontaneous abortion. According to a recently published cross-sectional study in Nigeria, the prevalence of BV was significantly higher among pregnant women (45.5%, p = 0.045), with low education level (66.7%, p = 0.001), unemployment status (36.1%, p = 0.0013), and nulliparity [36.4%, p = 0.0274), while no significant association was found between BV prevalence and the pregnancy trimester (p = 0.253) [27]. While a study in India found that BV was significantly associated with multipara status, and the risk was much higher among women from lower socioeconomic classes and poor educational backgrounds [25]. Literature also indicates that vaginal douching and frequent intercourse may increase the risk of developing BV during pregnancy [13].

This review strongly supports the association between adverse pregnancy outcomes (e.g., premature rupture of membrane, preterm birth or delivery, and premature/small for gestational age / low birth weight baby) and bacterial vaginosis. In addition, BV-positive women are more susceptible to STIs and HIV transmission [22]. It has been proposed that the increased risk of adverse reproductive and pregnancy outcomes might be directly attributable to BV-related bacterial species rather than BV itself [14]. The vaginal microbiota dominated by lactobacilli, helps to protect pregnant women against RTIs through the production of lactic acid. Pregnancy causes an increased vaginal discharge due to the rise of serum oestrogen levels. This discharge contains higher cervical mucus, and as the pregnancy progresses, it predisposes women to RTIs/STIs, such as bacterial vaginosis [3]. Due to this vaginal dysbiosis, the presence of other RTIs and poor obstetric outcomes may predominantly occur among pregnant women and make them highly vulnerable. The presence of BV-related bacteria and/or sexually transmissible microorganisms in BV-positive women can lead to opportunistic infections including HIV acquisition and proliferation [28] because the loss of lactobacilli and alteration in pH develops a more permissive milieu for endogenous aerobes and anaerobes in high concentrations [3]. As a result of this imbalanced vaginal flora, globally, 10–30% of the BV-positive pregnant women give premature birth, and approximately 70% of them face preterm delivery accompanied by perinatal mortality [28].

Strengths and limitations

The extensive database searches and cross-referencing ensured no duplication of data with previously published systematic reviews. Any potentially overlapping studies were analyzed for relevance and then categorized appropriately based on the scope of this review. The findings of this study may help to strengthen the current practices and offer valuable insights that could potentially reform future guidelines and policies in the field of obstetrics and gynaecology. This comprehensive review serves as a valuable resource for obstetricians and other healthcare providers responsible for the care of pregnant individuals. It emphasizes the necessity of routine screening and treatment of bacterial vaginosis during pregnancy, and underscores the importance of further research to develop effective strategies for the prevention of BV and its associated adverse outcomes in this target population.

Although the present study contributed to some meaningful insights, it had some limitations as well. The study’s selection criteria led to the inclusion of a higher number of studies targeting African women which affected the overall pooled data on BV prevalence, risk factors, and pregnancy outcomes. Hence, this has caused potential biases in our research. Furthermore, the inclusion of different study designs led to some degree of data heterogeneity in this systematic review. Since most of the included studies had an observational design, their analysis needs to be interpreted with caution. This study focused on BV prevalence irrespective of the test method used. Most of the studies included in this review have used the Nugent scoring system which is considered the gold standard for BV diagnosis. Other factors that contributed to the heterogeneity of included studies were the differences in patient characteristics (age range, parity, symptomatic or asymptomatic BV, gestational age, presence of mixed infections, and sexual relationship with partner). The scarcity of relevant studies did not allow us to adjust the confounders which resulted in heterogeneity in the study findings. Future reviews on this topic may benefit from incorporating standardized quality assessment frameworks to enhance the methodological transparency.

Recommendations

Worldwide, the high prevalence of BV imposes a concomitant increase in economic burden. Consistent prevention and sustainable treatment strategies are urgently needed to address this burden [29]. Holistic screening programs should be administered in all clinics as part of routine antenatal care or services to identify potential symptomatic or asymptomatic BV-positive cases. The majority of the BV-positive women in this review were educated and employed, which suggests that a higher literacy and working status among pregnant women may motivate them towards obtaining better healthcare services. Hence educational programs might be an effective strategy to increase BV awareness among unemployed women from rural areas with low socioeconomic and educational backgrounds. Simultaneously, educational interventions may also help to control the modifiable or behavioural risk factors such as lack of hygiene, douching, and risky sexual behaviours during pregnancy. Since the study noted some marked disparities in the BV prevalence based on race, multiparity, and gestational age (second trimester), it is advisable that multiparous females, women from an African origin, and women who are 14–26 weeks pregnant, should be actively screened for BV as an at-risk population. Due to the presence of an increasing number of asymptomatic cases, it is highly recommended to run diagnostic tests for BV during the early stages of pregnancy and prevent any possibility of preterm delivery and miscarriage. Accordingly, an immediate treatment plan should be executed by the healthcare providers to mitigate any other adverse events and/or the acquisition of severe STIs. Future studies should focus on identifying causal linkages between BV and adverse health outcomes during pregnancy.

Conclusion

This review strongly supports the association between adverse pregnancy outcomes (e.g., premature rupture of membrane, preterm birth or delivery, and premature/small for gestational age / low birth weight baby) and BV. In summary, the high prevalence of bacterial vaginosis necessitates a global surveillance approach to delineate the health risks imposed on both mother and child and promote cost-effective strategic measures to alleviate the undesired consequences of BV during pregnancy.

Authors’ contributions

NS, CSJT, MH, and STN: conceived and designed the study. NS, VN, RS, and ASAA: conducted the study and acquired data. NS, VN, RS, and ASAA analyzed the data. NS and STN: drafted the manuscript. CSJT, MH, DSWN, SNSJHZ, KKT, SNENZ, AR, and SKO: critically reviewed and revised the manuscript. All authors contributed to the article and approved the final version for publication.

Funding

This work was supported by the Universiti Malaya Impact-Oriented Interdisciplinary Research Grant (IIRG) (IIRG001A-2021HWB).

Data availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Medical Research Ethics Committee of the University Malaya Medical Centre on 30th March 2022 (MRECID.NO: 2022129-10953). All research activities conformed to the principles embodied in the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.