Predictive Value of Sialidase in the Diagnosis of Sexually Transmitted Infections: a Cross-section Study Based on Vaginal Microecology Evaluation System

Background: The purpose of this study was to determine whether any indexes of vaginal microecology evaluation system (VMES) could predict sexually transmitted infections (STIs). Methods: A total of 1032 women, who presented to Gynecology Outpatient Clinic of the First Aliated Hospital of Xi'an Jiaotong University between November 2016 and November 2020, were included in the study. Incident STIs was dened as any case of Chlamydia trachomatis (CT), Mycoplasma genitalium (MG) and Neisseria gonorrhoeae (NG), and were conrmed using the RNA-based simultaneous amplication and testing (SAT) assay. VEMS is structured in two parts, one is the morphological characterizations encompassing bacterial density, ora diversity, predominant ora, pathogenic microorganisms and indicators of inammation, the other is the functional indexes, including vaginal pH, cleanliness, hydrogen peroxide (H2O2), sialidase, β-glucuronidase, leukocyte esterase, and acetylglucosaminidase. Bacterial vaginosis (BV) was diagnosed by Gram stain (Nugent score). Associations were mainly assessed using logistic regression (LR). Results: SAT assay detected STIs in 64 (6.2%) of the 1032 samples tested and 136 (13.2%) women had a clinical BV diagnosis using Nugent score. Multivariate logistic regression analysis revealed that women with sialidase-positive were more likely to test positive for STIs (aOR=3.081, 95% CI=1.586-5.984, P=0.001). Of 896 women without clinical BV, signicant association was also found for sialidase and STIs (aOR =4.133, 95% CI= 1.140-14.978, P=0.031). Conclusions: Sialidase may be a useful indicator to help clinicians identify these women who are risk for


Background
Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG) and Mycoplasma genitalium (MG) are common sexually transmitted infections (STIs) and treat mostly in the outpatient clinic in China. In 2016, the estimated number of incident cases was 124.3 million for CT, 30.6 million for NG worldwide [1]. Besides, the prevalence of MG was estimated to be 1.3% among countries with higher levels of development and 3.9% among countries with lower levels of development [2]. STIs is known to cause various complications including infertility, choronic pelvic in ammatory disease, premature delivery, neonatal encephalitis and others [3][4][5]. However, most carriers are asymptomatic and still transmissible, making the identi cation and diagnosis of patients with STIs a public health issue [6][7][8].
Vaginal microecology evaluation system (VMES) has gained popularity in China recently due to its effective, inexpensive and quick to conduct. Wet preparations of vaginal swab samples that are examined microscopically and stained by Gram staining are studied using the tool. It consists of two parts, one is the morphological characterizations encompassing bacterial density, ora diversity, predominant ora, pathogenic microorganisms and indicators of in ammation, the other is the functional indexes, including vaginal pH, cleanliness, hydrogen peroxide (H2O2) and activity of sialidase, β-glucuronidase, leukocyte esterase, and acetylglucosaminidase [9]. In 2016, infectious Disease collaborative group of Obstetrics and Gynecology Branch in Chinese Medical Association proposed the standard VMES and published the expert consensus (Table 1) [10]. The evaluation of the vaginal microecosystem can help the clinician to make a prompt diagnosis and provide new opportunities for the comprehensive management of dysbacteriosis. In the resource constrained settings of many developing countries, screening tests for STIs using culture, ELISA, and polymerase chain reaction are not practical options because of their costs or procedures [11][12][13]. Given this, there is a clinical need for a cheap, rapid, and simple assay that would allow clinicians to make an early predictive diagnosis during the initial patient visit. Therefore, we evaluated whether any indexes of VMES that could predict the presence of STIs.  The results were also divided into four levels (denoted as + to ++++): grade I (+): 1-3 kinds of ora; grade II (++): 4-6 kinds of ora; grade III (+++): 7-10 kinds of ora; and grade IV (++++): more than 10 kinds of ora. "++" and "+++" were viewed as normal.
(C) Predominant ora: it refers to the maximum number of microorganism observed in the microscopic eld under oil immersion lens. Only Large Gram-positive rods (Lactobacillus) was determined as a normal ora.
(D) Vaginal cleanliness evaluation: It was graded by the density of Lactobacillus, epithelial cells, white blood cells, other ora into 4 levels, with Grade I-II as normal and grades III-IV as abnormal usually accompanied by in ammation. Speci cally, the judgment of vaginal cleanliness were as follows (Table   S1) [14]:  (Table S1).
(F) Pathogen: TV (Trichomonas vaginosis) was diagnosed by motile agellates under a microscope and VVC (vulvovaginal candidiasis) was con rmed when the yeast and(or) pseudohyphae were observed through a microscope.
All these enzymes were classi ed as positive or negative result.
(B) Vaginal pH: pH paper strips ranging from 3.8 to 5.4 were used to measure the vaginal pH. The normal vaginal pH values range from 3.8 to 4.5.

Detection of STIs
CT, NG and MG were detected utilizing the RNA-based SAT (simultaneous ampli cation and testing) assay and STIs was considered positive if any of CT, MG, NG test was positive.

Statistical analyses
Differences in study population characteristics were tested using the χ2, Fisher exact or Mann-Whitney U test, accordingly. The unadjusted odds ratios (ORs) were obtained from an unconditional, binary logistic model with only one dichotomous covariate of interest in the model. All variables associated with the outcome (STIs) at P < 0.1 were entered into the logistic regression model. Then stepwise backward elimination logistic regression was performed to determine which factors were independently associated with STIs, and the nal reduced model was constructed. P values < 0.05 were considered signi cant.
Finally, we also presented the results in the subset of women without BV using the stepwise backward elimination logistic regression. We obtained the unadjusted and adjusted log odds using computer software (IBM SPSS version 25).

Study Population
Of the 1032 women included, all belonged to Han Chinese ethnicity from northwestern China, 563 attended with clinical symptoms of urogenital symptoms such as "itching" "abnormal vaginal discharge" or "cervicitis" and others visited for pre-pregnancy examination, abortion history or pelvic in ammation, etc. They ranged in age from 15 to 68 years (mean age = 30) and 18.4% of subjects under 25 years old. 64 (6.2%) participates were diagnosed with STIs and there were 32 cases of CT,17 cases of MG, and 16 cases of NG.

STIs and Other reproductive tract infections
Nugent score con rmed that 136 of the 1032 patients had BV and these subjects were more likely to be STIs-positive than subjects without BV. In univariable analyses, we found that BV(Nugent score ≥ 7) was associated with a 1.9-fold increase (OR = 1.942, 95% CI = 1.042-3.618, P = 0.037) in the odds of incident STIs. In addition, the STIs-positive and STIs-negative groups were similar in TV, VVC or AV (Table 2).

Multivariate logistic regression analysis
Because the individual markers (bacterial density and predominant ora) that connected to the Nugent score had independent associations with STIs, the individual markers rather than the composite variable "BV" were used in the nal logistic regression model. After adjustment for predominant ora, bacterial density, vaginal PH, Leukocyte esterase and acetylglucosaminidase, the multivariate logistic regression model revealed that sialidase-positive and younger age (15-25y) were independently associated with STIs. Sialidase-positive was signi cantly associated with an increased odds of STIs (aOR = 3.08; 95% CI = 1.586-5.984, P = 0.001). Further, younger subjects(15-25y) were 5.1 times more likely to be infected with STIs (aOR = 5.042, 95% CI, 2.890-8.798, P = 0.001) ( Table 3).

Discussion
The study focused on the relationship between STIs and indexes of vaginal microecology evaluation system in Xi'an district. We found that sialidase is strongly associated with STIs which suggests that information from VMES can be used to predict the presence of STIs. Vaginal microecology tests are inexpensive and widely available tools especially in China that may potentially help clinicians estimate which patients are at risk for STIs.
Our study demonstrates that the rates of STIs acquisition are increased in the BV affected patient and younger women, which is consistent with some previous ndings about STIs [16][17][18]. BV is characterized by a reduced number of Lactobacillus spp. and an increased diversity of anaerobic bacterial species, including Gardnerella vaginalis, Atopobium vaginae, Prevotella spp., etc. There are strong biological arguments in favor of this hypothesis that BV is risk factor for STI acquisition. Firstly, Lactobacillus species protect against STIs by producing antimicrobial compounds, including organic acid to lowers the vaginal pH, H2O2, target-speci c bacteriocins and possibly biosurfactants [19]. Thus, the lack of Lactobacillus species would result in greater susceptibility to infections. Secondly, vaginal strict and facultative anaerobes appear to produce cytokines, nitrosamines and in ammatory mediators, which are supposed to be related to cervical in ammation [20], and mucin-degrading enzymes that reduce the infection barrier [21,22].
We identi ed the main factor associated with STIs was sialidase. Sialidase is an enzyme involved in the pathogenesis of several diseases by cleaving sialic acid from terminal glycans of glycoproteins and thus leading to impairment of the local immune response [23][24][25]. Not surprising that elevated sialidase represents an abnormal state and associated with higher risk, should these women be exposed to an STI pathogen. In former study, Briselden et studied 50 vaginal uid specimens from women with BV and found that elevated levels of sialidase activity were detected in those with BV (84%) than without BV (P 0.001) [26]. Meanwhile, the results of our study also showed a strong correlation between BV and sialidase. Furthermore, increased sialidase activity has been detected in the vaginal uid of BV patients [26,27], and is the basis of a marketed quick test for diagnosis of BV [28]. Therefore, then we also focused on the subjects of BV-negative group. We nd that women with sialidase-positive have higher prevalence of STIs compared with those without sialidase positivity even in the BV negativity group, and 30% of those with an elevated sialidase were infected with STIs in subset of women without BV, which demonstrating sialidase-positive associated with the vulnerability to STIs regardless of the BV. Therefore, sialidase may be a better predictor of STIs than BV.
However, these ndings may be somewhat limited by lacking some other possible determinants and confounders for the acquisition of STIs, such as sexual risk behavior, previous STIs, condom use. Another limitation of this study is that the cross-sectional design limits our ability to interpret the associations we found and we cannot determine whether STIs cause an elevated vaginal sialidase or if women with high vaginal sialidase are more susceptible to STIs. Besides, the low prevalence of STIs in this study came as a surprise, compared to the prevalence in other parts of the world [1]. It might be explained that more than four in ve of participants over age 25. Thus, the study should be repeated with a larger group of women to con rm our results due to the lack of positive samples.

Conclusions
Overall, our study supports that sialidase is a feasible option to predict the people who need to be further NAATs examined, especially in the absence of BV. In view of this, the VMES may offer great potential for prediction in diagnosis of infectious disease. It is noteworthy that the quick prediction and diagnosis of STIs is key to propose a comprehensive approach to sexual and reproductive health.