PREVALENCE OF INFECTIOUS VAGINITIS: ANALYSIS OF DATA FROM VAGINAL SWABS EXAMINED IN THE CENTRAL LABORATORY OF KORLE BU TEACHING HOSPITAL, GHANA

* Ali Samba 1 , Kareem Mumuni 1 Opuni A 2 , Dankwah T 2 , Sowah A 2 and Mashud M 2 . 1. Department of OBGYN, Univ. of Ghana School of Medicine and Dentistry. Box 4236, Korle-Bu. Accra. Ghana. 2. Department of Microbiology, Korle-Bu Teaching Hospital. Box 77. Korle-Bu. Accra. Ghana. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History

159 Some degree of vaginal discharge is normal and healthy for women in their childbearing years [FSRH, 2012]. It derives from the physiological secretion of cervical and Bartholin's glands and desquamation of vaginal epithelial cells resulting from bacterial action in the vagina . The quantity and type of physiological vaginal discharge vary during the menstrual cycle as a result of hormonal fluctuations [FSRH, 2012]. However, vaginal discharge is considered abnormal or pathological when it is more abundant, has an unpleasant odor and/or, abnormal colour, consistency and is accompanied by vulval or vaginal itching, dysuria, and/or dyspareunia .
In women aged 15 to 49, abnormal vaginal discharge is the second leading cause of lost years of healthy life [da Fonseca, 2013] and imposes a significant negative impact on their health-related quality of life, especially on the physical domain [Sami, 2014;Valsangkar, 2014]. This gynecological morbidity was also shown to be a strong predictor for future HIV acquisition [Abbai, 2016]. Furthermore, during pregnancy, pathological vaginal discharge can cause serious harm to the mother and their children including prematurity, low birth weight, chorioamnionitis, post-partum endometritis, and post-cesarean wound infection [da Fonseca, 2013].
Vaginal infection, also known as vaginitis, including bacterial vaginosis, aerobic vaginitis, Candida albicans and trichomoniasis accounts for the majority of the cases of vaginal discharge [Donders, 2002;Mylonas and Bergauer, 2011;FSRH, 2012;Fan, 2013;Frobenius and Bogdan, 2015;Rice, 2016] and results from an increased colonization by different pathogenic microorganisms. Numerous studies/reviews attempted to determine the frequency of the different types of vaginitis and provided percentages that varied considerably between countries and world regions, and across various groups within countries. The frequency for bacterial vaginosis ranged between 17 to 19% in family-planning or student health clinics, 24 to 37% in sexually-transmitted disease clinics, and 10 to 29% among pregnant women [Sobel, 1997;Kenyon, 2014]. Vulvovaginal candidiasis is diagnosed in up to 40% of women with vaginal complaints [Ilkit and Guzel, 2011]. The prevalence of diagnosed aerobic vaginitis was estimated between 5 and 10.5% [Tansarli, 2013] while the annual incidence of trichomoniasis (per 1 000 persons) in the 15-49 year-oldwomen ranged between 11.7 in East Asia and Pacific regions to 119.4 in Sub-Saharan Africa [Kenyon, 2014].
The identification of the type of vaginitis and of the causative organism(s) involved in these infections is therefore recommended before initiating any treatment. It generally consists of assessing the clinical and sexual history of the woman, and performing a vulvo-vaginal examination with measurement of the vaginal pH, and simple laboratory investigations, including microscopy and microbiological culture of the vaginal discharge [FSRH, 2012]. This classic approach has been promoted by professional organizations [FSRH, 2012] and an abundant clinical literature that emphasizes the importance of making a precise diagnosis before treating the patient [Anderson and Karasz, 2005].
However, in routine clinical practice, vaginal discharge is not investigated microbiologically in all settings, especially in poor-resource ones, because it is costly and due to the limited availability of well-equipped facilities for fully diagnosing these infections [Rekha and Jyothi, 2010;Kapoor, 2016]. Most of the time, a presumptive diagnosis is made based on the nature of the discharge, without any laboratory investigations and treatment is initiated based on clinical findings alone [Rekha and Jyothi, 2010]. This approach is consistent with the FSRH guidance which indicates that women experiencing vaginal discharge who are at low risk of sexually-transmitted infection can be treated by syndromic or empirical management [FSRH, 2012].
Consistently, in Ghana, a careful risk assessment of women with vaginal discharge is generally used to exclude other potential diagnosis (e.g. sexually-transmitted infections), and to identify the most likely etiology of the discharge as well as the appropriate treatment regimens. Microbiological culture is usually reserved to cases of recurrent vaginal discharge and/or treatment failure [Ghanaian Ministry of Health, Standard Treatment Guidelines, 2010].
In this context, it may be helpful for those physicians who have no laboratory facility to recognize the different presentations of vaginal discharge but also to know the epidemiological data in their geographical area so that their diagnosis can be as accurate as possible and the symptoms treated according to its corresponding etiology. Although numerous studies have been carried out in developed countries to document both the "normal" vaginal flora and the flora associated with vaginal discharge, there is insufficient knowledge of the etiology of infective vaginal discharge in many developing countries. The aim of the present study was to determine the microbiological profile of 160 (symptomatic) vaginal discharge in the urban area of Accra (Ghana) in order to ensure a rationale choice for both empirical and definitive antimicrobial treatment of vaginal discharge.

Materials and Methods:-
Study design:-This retrospective study was conducted at the Microbiology Department of the Central Laboratory of the Korle Bu Teaching Hospital, which is a referral hospital located in the capital of Ghana, Accra. The approval and permission for the study was granted by the departments of Microbiology and Reproductive Health of the Korle-Bu Teaching hospital. High vaginal swabs routinely delivered at this department for microbiological examination from January to December 2014 were further used for this research work. The data was extracted from record books (specimen log books ) on HVS from 1 st January -31 st December 2014. Variables collected included age of patients and culture results with types of isolates. The records did not have other sociodemographic information such as educational level, occupation etc. No data on the requesting department or ward was available for 61.4% (n=892) of the HSVs. Out of the HSVs with information on the requesting department (n=561, 38.6%), most were outpatient cases from the Korle Bu Teaching Hospital, only three requests came from other medical facilities. Samples were obtained from women of different age groups (mean age: 31, range: 1-90). However, more than 87% of samples was collected from patients within the childbearing age range (15-45 years).   The distribution of the identified microorganisms among the different age group is shown in Figure 1. The highest frequency of Candida or bacteria positive samples was obtained in the 20-29 year age group followed by the 30-39 and 40-49 year age groups.

Discussion:-
This study found 86.45 % prevalence of positive cultures , most of which were commensals, this though not unexpected, it is difficult to determine whether they could be associated with the abnormal vaginal discharge since an overgrowth of commensals would cause abnormal discharge. Other studies such as that of Masand D et al (2015) found similar or slightly higher ( 99%) rates of positive cultures.
There was a 13.6% of no growth or negative cultures which is slightly higher than the 11% found by Masand D et al (2015). No growth can be attributed to several factors including non-infectious causes, use of inappropriate culture media for fastidious and anaerobic organisms. It may also be due to inappropriate collection and transport of specimen leading for instance to drying of swabs.
In this study Candida species were isolated from 33.0% ( The prevalence of bacteria associated vaginal discharge was found to be 15.7%, which is higher than the 4.5% found by Di Bartolomeo et al(2000), but lower than the 23% found by Klufio et al (1995) and even much lower than the 53% recorded by Masand et al (2015).This wide variation could be due in part to the differences in specimen collection and culture media available and techniques used as well as age distribution of the study populations. For instance the highest frequency of Candida or bacteria positive samples was obtained in the 20-29 year age group. Similar peak age incidence in case of infectious vaginitis was noted in other studies done by Rekha et al (2010), and in Masand, Patel and Gupta (2015). And so it could be inferred that the higher this age group as part of the study population the more likely the prevalence of bacteria isolation will be high.

162
The most common bacteria isolated was E.coli at 19% which is lower than the 25.8% was recorded by Deepthy et al (2016) , but lower than the 13.7% by Shamim et al (2008). The least isolated was N. gonorrhoeae at 1.1% which is consistent with the 1% by Wulu et al (2015). This is good news and indeed should be even much lower, since it is a sexually transmitted organism. However Wiesenfeld recorded a prevalence of 20%, reasons for this wide variation is not apparent and will remain speculative with the inherent limitations of retrospective studies. The limitations of this study include the non-availability of certain special culture media thus limiting isolation of certain organisms.

Conclusion:-
This study demonstrates the etiological predominance of Candida infection in infectious vaginitis, but also underlines the noticeable prevalence of bacterial vaginitis. Given this etiological heterogeneity of infectious vaginitis, a broad-spectrum combination treatments may be appropriate as first-line empirical therapy for infectious vaginitis.