Secnidazole for treatment of bacterial vaginosis: a systematic review

Background Bacterial vaginosis (BV) is one of the common vaginal infections among childbearing women. The usual treatment for BV is metronidazole; hence 30% of women have recurrence within 60 to 90 days after treatment. There are some studies which assessed the effect of secnidazole on BV. The aim of this systematic review was to investigate the effectiveness of secnidazole for treatment of BV. Methods The Cochrane Library, MEDLINE (PubMed), Scopus, and Web of Science (all databases from inception till October 28, 2018) were searched. Primary outcomes were clinical cure rate and microbiologic cure rate and the secondary outcomes were adverse events. Data was extracted from eligible studies by two review authors individually and analyzed by RevMan 5.3. Results Our search found six trials involving 1528 participants. Treatment with 2 g secnidazole could significantly reduce the risk of BV in patients with three or less episodes of BV in the last year by OR: 7.54 (95% CI, 3.89–14.60, p < 0.00001) and in patients with four or more episodes of BV in the last year (OR: 4.74, 95% CI: 1.51–14.84, p = 0.0.008). Secnidazole (2 g) could significantly increase the microbiologic cure rate in women with 3 or less episodes of BV in the last year (OR: 7.63, 95% CI: 2.30–25.33, p = 0.0009) but not in the women with 4 or more episodes of BV in the last year (OR: 20.17, 95% CI: 1.06–382.45, p = 0.05). The clinical cure rate, microbiological effect and the therapeutic cure rate of 2 g secnidazole was significantly more than that of 1 g secnidazole. The results showed that the clinical cure rate of 2 g secnidazole was not different from the following medications: metronidazole (500 mg bid for 5 days), secnidazole plus vaginal metronidazole, 2 g single dose of oral metronidazole and 2 g secnidazole plus vaginal ornidazole. Conclusion This review showed that 2 g and 1 g secnidazole were better than placebo, however, 2 g secnidazole was more effective than 1 g. Secnidazole 2 g was not different from metronidazole (500 mg bid for 5 days), or from secnidazole plus vaginal metronidazole, or 2 g single dose of oral metronidazole or from 2 g secnidazole plus vaginal ornidazole.


Background
Bacterial vaginosis (BV) is a type of vaginitis that results from change in bacterial vaginal flora due to the loss of hydrogen-peroxide generating lactobacilli and excessive growth of anaerobic bacteria and Gardnella vaginalis. BV is the most common gynecologic infection in women of childbearing age, affecting 40-50% of women within this age worldwide [1]. The vaginal flora contains various lactobacilli species that help maintain vaginal acidity (pH < 4.5) [2]. Vaginal microbiology is determined by factors that affect the strength of bacteria to survive, including vaginal pH, the presence of lactic acid produced by lactobacilli and hormonal agents such as estrogen, which fills the epithelial cells of the vagina with glycogen and is converted to lactic acid by lactobacilli [3]. The incidence of BV in the United States is 29.2%, which affects approximately 21 million women, and this disorder is the leading cause of 10 million annual visits due to the vaginitis [4]. BV's clinical symptoms are characterized by increased discharge which smells like fish and is uniform, dilute and grayish in color [5]. The known risk factors for this type of vaginitis include poor socioeconomic status, poor health level, early sexual activity, multiple sexual partners, psychological stress, and biogenetic factors [6]. Based on clinical criteria, the clinical diagnosis of BV is confirmed if there are at least three criteria. These criteria include: 1. vaginal discharge that is uniform and homogeneous, and is gray or white to yellowish in color, 2. PH of vaginal discharge equal or greater than 4.5. 3. positive whiff test (amine odor after adding 10% potassium hydroxide to vaginal discharge) and the presence of clue cells in the vaginal mucus smear sample (at least 1 in 5 cells of the vaginal epithelium) [7].

Treatment of bacterial vaginosis
Failure to properly treat bacterial vaginosis can lead to serious complications such as postpartum endometritis, pelvic inflammatory disease, premature rupture of the fetal membranes, preterm delivery, increased risk of post-hysterectomy infection, chorioamnionitis, spontaneous abortion, recurrent urinary tract infection and increased risk of intraepithelial cervical neoplasia [8,9].
Other complications of this infection include increased risk of sexually transmitted infections such as chlamydia, gonorrhea, HIV, and herpes simplex type 2, [10] so early diagnosis of BV and its treatment are important.
Usual treatments for BV include oral metronidazole 500 mg bid for 7 days, vaginal metronidazole gel 0.75% per day for 5 days, and vaginal clindamycin 2% per day for 7 days [11]. In spite of access to these regimens, there is 30% recurrence 60 to 90 days after treatment, which increases over time (50% in 12 months) [12]. Other treatments include actinazole from the nitroimidazole group [13] with a longer half-life than metronidazole (approximately 17 h compared to 8 h) for the treatment of vaginosis and strains of trichomoniasis are in usage in Asia and Europe [14]. In laboratory studies, the antimicrobial properties of acetyazol have been shown against many bacterial species involved in vaginosis.
Various clinical trials have reported the effects of one or two grams of oral secnidazole on the improvement of BV [15][16][17][18]. Despite the numerous clinical trials conducted in this area, there is no review study that compares the therapeutic effects of secnidazole with other treatments regarding the improvement of symptoms of BV. Therefore, the aim of the present systematic review is to examine the evidence from randomized clinical trials on the therapeutic effects of secnidazole on BV and compare it with metronidazole or placebo.

Types of studies
We recruited Randomized controlled trials (RCTs).

Types of participants
Women (of all ages) diagnosed with bacterial vaginosis using Amsel criteria were recruited for this study.

Types of interventions
Eligible trials compared single or combination treatment regimen (secnidazole) compared with conventional treatment (metronidazole) or placebo for bacterial vaginosis. We had no restriction regarding route of administration, dose, frequency or duration. We followed the Cochrane Collaboration reviewed methods for collection and analysis summary data. There was no limitation regarding to publication status, country, duration of follow-up or language. The search terms are presented in Additional file 1. We conducted the search on Cochrane Library (CENTRAL 2018), MEDLINE (PubMed), Scopus, and Web of Science (all databases from inception till October 28, 2018).

Types of outcomes
Two reviewers (FS, HJ) independently examined title/ abstracts of all studies according to our inclusion criteria. One of the review authors (MAA) checked for discrepancies which were resolved by discussion. One review author (MAA) entered the data into Review Manager 5.3. A second author independently checked the data (FS). Using a pre-designed data extraction sheet, two reviewers (MAA, FS) independently extracted the data from the included studies. The relevant data extracted from the included studies were study details (dates when the research was conducted, geographic location, participant inclusion criteria, funding sources, publication date), participant characteristics, interventions details (type, duration, route of administration, dose), outcome details (type of outcome, outcome assessment method), and bias assessment details (data necessary to assess the risk of bias, as described below).

Risk of bias assessment
We assessed the following risk of bias domains for each trial: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other bias as shown in Fig. 1 [19].

Measures of treatment effect
All our outcomes were binary outcomes, so we calculated the odds ratio (OR) with 95% confidence intervals (CI). The unit of analysis was the participant.
Meta-analysis Meta-analysis was only feasible for particular topics that at least two or more than two studies addressed those topics.
The included trials were conducted between 1996 and 2014. Two out of our six studies have conducted in the USA. The remaining studies were conducted in France, Venezuela, Turkey and India. One trial was phase II, five were phase III (Table 1).
We excluded four studies, because we were not able to find the full texts of those studies or they did not have a control group [23][24][25][26].

Risk of bias in included studies
Four out of six trials were rated as having a low risk of selection bias as they used appropriate random sequence generation method. However, only one trial had low risk for allocation concealment bias. One trial stated that the investigators were not blinded to the treatment. The remaining four trials reported no information for allocation concealment, thus they were rated having an unclear risk for selection bias (Fig. 1). For assessment the risk of bias and quality of studies, we used Grading of Recommendations, Assessment, Development and Evaluation (GRADEpro, Table 2).

Effects of interventions 2 g secnidazole versus placebo
The effect of intervention in terms of clinical cure of patients with three or less episodes of BV in the last year is illustrated in Fig. 3. Two studies [17,20] with total participants of 210 were recruited in the metaanalysis. As evident from this figure, treatment with 2 g secnidazole could significantly reduce the risk of BV in patients with three or less episodes of BV in the last year 7.54 (95% CI: 3.89-14.60, p < 0.00001) or in patients with four or more episodes of BV in the last year (OR: 4.74, 95% CI: 1.51-14.84, p = 0.0.008) (n = 78). The level of heterogeneity was high (I 2 = 76%) in the later analysis and did not change using random effect model. Because only two studies were in the meta-analysis, the sensitivity analysis was not applicable ( Fig. 3) [17,20].

Secnidazole versus metronidazole
One study [13] recruited 480 women to compare the effect of 2 g secnidazole versus metronidazole (500 mg bid for 5 days). Results showed that there was no significant difference between two groups in terms of clinical cure (OR: 0.87, 95% CI: 0.56-1.34, P = 0.53). Also the effect of 2 g oral secnidazole versus 2 g secnidazole plus vaginal metronidazole was compared in one study with 40 participants and the results showed that there was no significant difference between two groups (OR: 0.11, 95% CI: 0.01-2.08, p = 0.14) [22]. Another study with 172 participants for comparing the effect of 2 g oral secnidazole versus 2 g single dose oral metronidazole showed no significant difference between two groups regarding cure rate after one week using Amsel criteria (OR:1.28, 95% CI: 0.48-3.42, p = 0.62), or after 4 weeks (OR: 1.15, 95% CI: 0.55-2.40, p = 0.71) [21].
One study recruiting 39 women evaluated the effect of 2 g secnidazole versus 2 g secnidazole plus vaginal ornidazole and results showed no significant difference between two groups (OR: 0.29, 95% CI: 0.03-2.69, p = 0.28) [22]. Also the effect of 2 g oral secnidazole versus 2 g secnidazole plus vaginal metronidazole was compared in one study with 40 participants and the results showed that there was no significant difference between two groups (OR: 0.11, 95% CI: 0.01-2.08, p = 0.14).

Adverse effect
The adverse effects that reported in studies that 2 g secnidazole was used versus placebo were yeast infection, vulvovaginal pruritus, dyspareunia, nausea, increase the level of hepatic enzymes such as ALT and AST and headache. There was no significant difference between the two groups in any of the studies regarding adverse effects in 2 g secnidazole versus placebo (n = 233) (OR: 2.24, 95% CI: 0.85-5.94, P = 0.10) or versus 1 g secnidazole (n = 219) (OR: 1.25, 95% CI: 0.65-2.41, P = 0.50) [17,20].

Discussion
This systematic review designed to evaluate the effect of secnidazole on bacterial vaginosis in childbearing women.
BV is one of the most prevalent vaginitis in the United States and almost 30% of childbearing aged women are affected with this infection [27]. BV has some consequences such as; endometritis, postpartum fever, cellulitis in the hysterectomy cuff and infection after abortion [28] and also is a risk factor for acquisition of HIV and herpes simplex virus type 2 and other sexually transmitted diseases [29].
The main medication for treatment of BV is metronidazole that has 90% effectiveness, but the recurrence rate is high [30]. Secnidazole is a next-generation of 5nitroimidazole that has already approved in Europe and Asia as a single dose of 2 g for BV. This medication also was approved to use in the USA in 2004 [31]. The long half-life of secnidazole (17-28.8 h) makes it possible for a single dose to be effective [32]. The food and drug administration (FDA) of the United States recently approved the single 2 g dose of secnidazole for the treatment of BV according to two randomized controlled trials that conducted in the United States [33].   The results of this systematic review showed that 1 g secnidazole could significantly improve the clinical cure rate in women with BV compared to placebo, but the microbiological cure rate was not significantly different from placebo in women with 4 or more episodes of BV in the last year.
Our results also revealed that 2 g secnidazole significantly could treat BV in women with three or less or four and more episodes of BV in the last year in compare to placebo. Further, our results showed that 2 g secnidazole was significantly more effective in terms of clinical cure rate and microbiological impact than that of 1 g secnidazole.
We found only one study that compared the effect of 2 g secnidazole with metronidazole (500 mg bid for 5 days) or 2 g single dose of metronidazole. The results revealed no significant difference regarding clinical cure rate in both methods.
The diagnosis of BV is according to the Amsel criteria (at least three criteria should be present) including; homogenous grayish-white vaginal discharge, vaginal pH > 4.5, positive whiff test (fish odor with adding a drop of 10% potassium hydroxide to the vaginal discharge) and presence of more than 20% clue cells in the wet smear of vaginal discharge [34]. The clinical cure of BV are including; a negative test for amino odor after adding 10% potassium hydroxide solution to vaginal discharge, the number of clue cells less than 20% and pH of vaginal discharge < 4.5 [35]. In all studies recruited in this systematic review the Amsel criteria was used for clinical cure rate of BV.

Strengths and limitations of the study
This is the first time that we evaluated the effect of secnidazole on BV in a systematic review study. In this systematic review we only found six studies and the meta-analysis was only possible in some cases. High level of heterogeneity was observed in some studies that meta-analyses were performed. Because in most cases there were only two studies in the meta-analysis, the sensitivity analysis was not possible. Therefore the results should be considered with caution.