Asymptomatic Bacterial Vaginosis in Cervical Cytology: A Preliminary Report of Incidental Finding in Routine Cervical Cancer Screening

Background: Bacterial vaginosis is a complex pathogenic group of bacteria which can cause a wide range of symptoms in affected women in Obstetric and Gynaecological practice. Its occurrence usually indicates alteration in the normal vaginal flora expected in majority of the women of reproductive age. Amongst the maintenance agents of the vaginal health, lactobacilli occupy a prominent place. Asymptomatic bacterial vaginosis is postulated to be linked to the development of cervical intraepithelial neoplasia by some researchers while some differ in this regard. Objective: To determine the rate of occurrence of asymptomatic bacterial vaginosis and associated factors in women who underwent cervical cytology for cervical cancer screening. Study Design: A cross sectional study involving microscopic examination of cervical smears of eligible clients presenting for routine Pap smear. Setting: The study was carried out in a well woman’s facility established and manned by female medical doctors in Enugu state with the assistance of trained nursing staff. Various female cancers and other communicable non-communicable disease screening cervical pre cancer stages and infections, more elaborate studies are needed to determine the usefulness of treatment or otherwise of bacterial vaginosis which constitutes significant incidental findings.


Introduction
A woman's vagina is known to harbor loads of bacteria regarded as normal flora.
There are saprophytic bacterial organisms that help in maintenance of a healthy vaginal environment and others whose presence is not beneficial and cause no obvious harm. Nature has provided for the preservation of a healthy vagina in the reproductive age by creating an acidic environment maintained by the presence of lactic acid among other factors. The lactic acid is supplied by some of the useful bacteria resident in the vagina of pre-menopausal women. Key in this liberation of lactic acid is the lactobacilli species which help not only in maintaining healthy vagina but in preventing genital infections [1]. Some groups of bacteria found in the vagina have a propensity to cause undesirable symptoms as soon as they find conducive environment to overgrow [2]. Although prevalence as low as 12% had been documented [5], it depends on the studied population. A recent systematic review and meta analysis estimated a high global prevalence of 23% -29% [6]. Bacterial vaginosis can be asymptomatic in significant proportion of infected population [7] [8], yet a cause of a range of complications in pregnant and nonpregnant women. It is associated with adverse conditions such as sexually transmitted infections and human immunodeficiency (HIV); infertility, preterm birth, premature rupture of membranes etc.
[9]- [14]. Because a significant number of women with bacterial vaginosis are asymptomatic, it may imply that the untoward effects of these organisms may get worse with further multiplication leading to the well documented adverse outcomes if undiagnosed over time. A recent study showed that asymptomatic bacterial vaginosis is associated with increased susceptibility to genital infections [17]. This study was undertaken, therefore, to determine the rate of occurrence of bacterial vaginosis in asymptomatic women, taking advantage of high specificity of Pap smear and its comparable accuracy with the Amsel's criteria and Nugent's scoring in the diagnosis of bacterial vaginosis. Also, the consideration of the numerous complications of infection by this group of organisms made it pertinent to study the rate at which they occur in asymptomatic women. Our expectation was that the findings from the study would improve the care to our gynaecological patients.

Aims
To determine the prevalence of bacterial vaginosis and its associated factors in asymptomatic women through cervical cytology smear.

Methods
The study was carried out in a well woman's health facility established and ma- Sample size was determined using the single proportion formula: n = z 2 pq/d 2 , where n = sample size; z = standard normal deviate, usually set at 1.96 which corresponds to 95% confidence level; p = the proportion in the target population estimated to have a particular characteristic i.e., expected prevalence (18.4%, from a previous study [16]); q = 1 − p; d = precision or degree of accuracy desired, set at 0.05 for 5% precision.
Thus, for our sample size, p = 18.4% = 0.184; q = 1 − 0.184 = 0.816 Substituting the above in the formula, we obtained a sample size of 230.7.
However, 285 consenting participants were recruited to enhance the power of the study. Data entry and analysis were done using statistical package for social sciences (SPSS) computer software version 21.0. Means and standard deviations were calculated for continuous variables while categorical variables were presented as frequency and percentages. Tests for significance were performed using chi-square for categorical and students t-test for continuous variables respectively. Level of significance was set at p value <0.05. Illustration of results was done using tables and charts as appropriate.

Results
A total of 285 eligible women were recruited during the first six months (Feb-July 2018) of the 2-year project. As shown in Table 1, among other sociodemographic parameters, the mean age of the subjects was 46.17 (SD12.41) years. 50.9% of the study subjects were above reproductive age (>45 years), 38.2% were in second half of reproductive age (31 -45), while 10.9% were in the first half of reproductive age (15 -30). Majority of them were married and had  Table 2. Study subjects above the reproductive age although constituted more than half of the study population had comparatively less occurence of bacterial vaginosis than those within reproductive age. Thus, there was a significant association of bacterial vaginosis with reproductive age (p value = 0.004). However, no significant association of bacterial vaginosis was observed with age at coitarche, parity or contraceptive as illustrated in Table 2. There was also no association between abnormal cervical cytology and asymptomatic bacterial vaginosis.

Discussion
Bacterial vaginosis' prevalence of 10.9% among asymptomatic women that presented for Pap smear in our study is lower than figures obtained in the reviewed studies that used pap smear for diagnosis of bacterial vaginosis. 18.4% [16] and For studies using Gram stain to diagnose bacterial vaginosis in nonpregnant patients, the prevalence from our study was still lower than 23% (using Amsel's criteria) and 27% (with Nugent's method) observed in female students at Calabar, Nigeria whose age ranged between 16 and 35 years [18]. [20], we found no association with age at coitarche, parity, contraceptive or abnormal cervical cytology. In contrast to our findings, some studies demonstrated significant association of bacterial vaginosis with contraceptive use. For instance, anatomic sites contraceptives were reported as having more occurrences of bac-Open Journal of Obstetrics and Gynecology terial vaginosis in some studies [19] [20] [21] while some systematic reviews and meta analysis showed that hormonal contraceptives were associated with significantly reduced prevalence, incidence and recurrence of bacterial vaginosis [22] [23]. With regards to parity, some authors, like in our study found no significant association with bacterial vaginosis [24] while some documented positive associations [13] [25]. Bacterial vaginosis is known to be more commonly seen in sexually active women. Hence, there seems to be insignificant relationship between bacterial vaginosis and age at coitarche >18 yrs [24] [26] compared to strong association which exists between it and the number of sex partners or being sexually active [3] [5] [27]. All the affected women in our study were sexually active and over 90% of them had their coitarche at age >15 years. We found no significant association between bacterial vaginosis and cervical intra epithelia neoplasms. This is at variance with previous studies including meta-analyses which reported existence of unconfirmed relationship between bacterial vaginosis and the risk of development of cervical precancerous conditions [28] [29] [30] [31].
Heterogeneity in study designs, participants, settings and lack of uniform designated gold standard method for diagnosis may account for these disparities observed among the reviewed studies as well as ours.

Conclusion
Prevalence of bacterial vaginosis is significant in our study population who were asymptomatic. Cervical smear should be taken advantage of as a tool for both cervical pre-cancer and bacterial vaginosis screenings since same sample can simultaneously be used to test for both conditions during cytology. Bacterial vaginosis may be unlikely involved in pathogenesis of cervical intraepithelial neoplasia.

Limitations
The fact that clients who presented for screening constituted the entire study subjects might not reflect the actual condition in the general population.
Some of the variables depended on the participants' ability to recall the required response which might not be accurately supplied.