PREVALENCE AND RISK FACTORS ASSOCIATED WITH VULVOVAGINAL CANDIDIASIS DURING PREGNANCY IN SANA'A, YEMEN

Background and objective: Women at pregnancy are further susceptible to vaginal colonization and yeast infection. The responsibility of Candida colonization in the incidence of preterm birth is correctly established. Knowing regional epidemiology and identifying risk factors of preterm birth are important for management and preventive strategies. The aim of the study was to reveal the prevalence of Candida species in vaginal swabs of pregnant women and determine odds ratio of risks for vulvovaginal candidiasis (VVC). Methods: Pregnant women attendance routine antenatal visits in Al-Olify –family Center in Sana'a city were registered into a cross-sectional study carried out from June 2018 to March 2019. The laboratory works were carried out in the National Center of Public Laboratories (NCPHL). Samples of vaginal swabs were taken from contributors after obtaining oral consent. The classification of the Candida species was done by culture on HiCrome Candida Differential Agar at 35 °C for 48 h for making of species-specific colors. Data of demographic, clinical and risk factors were collected in a pre-designed questionnaire. Results: A total of 190 pregnant women are included. The rate of VVC was 51.6%. Candida albicans accounted for 39.5% and non-Candida albicans accounted for 12.1% of isolates, mainly C. glabrata (4.7%), C. lipolytica (3.2%), and C. famata (2.1%). When VVC risk factors were considered, there were significant risk factors with the age group 20-24 years (61%, odds ratio (OR) = 1.8), first trimester of pregnancy (61.1%, OR = 1.7), Multipara of parity (61.8%, OR = 1.9), low socioeconomic level (60.1%, OR =2.4), and illiteracy (68%, OR = 2.2). When clinical symptoms were considered, only 86.2% of affected females had clinical signs of VVC. Conclusions: There is a high probable rate of VVC is found among pregnant women in Yemen, undiagnosed and unnoticed; and this highlights the need for health authorities to develop strategies for diagnosing VVC, including vaginal swabs for candidiasis as a routine procedure for all pregnant women. This study also revealed a steady increase in time with a non-C. albicans species prevalence rate. VVC syndrome management guidelines in Yemen should be revised to include a special protocol for pregnant women.


INTRODUCTION
VVC is produced by overgrowth of Candida yeasts, especially Candida albicans, which is an essential component of vaginal flora 1 . VVC symptoms comprise vaginal discharge, pain, itching and swelling of vulvar. In addition, vulgar erythematosus and edema with obstruction are common results. Typical vaginal secretions in VVC are described as cottage cheese-like in character 2 . It is possible that 75.0% of women may experience VVC during their lifetime 3 . Microbiological and epidemiological studies indicate that intrauterine infection accompanies up to 40% of preterm births 4 . The Candida colonization role was confirmed in the preterm birth occurrence 5 , as the Candida were isolated from amniotic fluid in spontaneous women preterm birth 5, 6 . Candida has been found to increase metalloproteinase 9 production by chorioamniotic membranes 6 . Metalloproteinase nine is a protein that remodels connective tissue and has an essential role in the origin of premature labor and early rupture of membranes 7 . Also, the mother's vulvovaginal candidiasis is a most important risk factor for neonatal Candida colonization 8 . There is evidence that eliminating Candida species during pregnancy may reduce the risk of spontaneous premature birth 9 . In epidemiology, VVC is the next most common cause of vaginosis worldwide, following a bacterial infection. In nonpregnant women, the risk of VVC is about 20%, but it increases by 30% during pregnancy 10 . Usually (90%) of infections due to Candida albicans are undisruptive and do not cause any symptoms. However, vaginal or vulvar infection may be linked with symptoms and signs for instance severe itching, pain, pruritis, irritation, bad odor, dyspareunia, and dysuria and burning in urination 10 . In the Arabian Peninsula, the epidemic of VVC infection varies in diverse countries, even from regions inside the same country 11-13 . Thus, local studies are essential to obtain related epidemiological data and features of Candida sensitivity to antifungal drugs with the purpose of help manage and treat patients with Candida infection. Regarding the causative species of VVC, some researchers report an increased rate of non-Candida albicans species, particularly C. glabrata, C. krusei and C. parapsilosis 14 . Some researchers have also found that the incidence of C. albicans infection decreases and other Candida types including C. glabrata. C tropicalis, C. kefyr, C. parapsilosis, C. africana, C. famata, C. dubliniensis, C. guilliermondii and C. lusitaniae associated mainly with vulvovaginitis from patients with immunodeficiency or in pregnant women 10 . In Yemen, vaginitis is one of the most common conditions for seeking medical care. In the city of Sana'a, vaginal infections have spread in Yemen among 37.6% of women of childbearing age, with VVC associated significantly with less than 25 years of age and use of intrauterine contraceptives 15 , also in Hadramaut three hundred and seventy two (39.2%) of the 950 pregnant women suffered from abnormal vaginal discharge and were positive for bacterial vaginosis 16 . Pregnant women are more likely to experience vaginal colonization and Candida infection. The responsibility of Candida colonization in the happening of preterm birth is well established. Knowing local epidemiology and identifying risk factors for preterm birth are important for prevention and management strategies, so the purpose of the current study was to determine the prevalence of Candida species in the vaginal swabs of pregnant women and to identify the risk factors associated with VVC.

Pregnant women attending routine antenatal visits in
Al-Olify -family Center in Sana'a city were enrolled into a cross-sectional study carried out from June 2018 to March 2019. Inclusion criteria for subject selection were healthy individuals with no systemic disease. In addition, pregnant women who currently taking antifungal, steroids, antibiotics, or immunosuppressive drugs in the past 6 months were excluded. The sample included 190 pregnant mothers. All pregnant women was examined clinically by specialist and vaginal swabs were taken. The vaginal swabs were sent to the National Center of Public Laboratories (NCPHL) where the laboratory works were carried out. Vaginal swabs were sampled from the participants after oral approval. Smears were inoculated in chloramphenicol Sabouraud's glucose agar and incubated at 37°C for 24 to 48 hours under aerobic conditions for performing fungal cultivation 17 . Candida species were identified by culturing HiCrome differential Agar at 35°C for 48 hours to produce species specific colors. Data of demographic, clinical and risk factors were collected in a pre-designed questionnaire.

Data analysis
The data was statistically analyzed using EPI-Info version 6. The difference in the distribution of Candida types among groups was based on a comparison of repeat distributions by chi-square test. The odds ratio associated with VVC risk factors was performed by 2x2 tables to obtain an OR, 95% CI, Chi squared and p value by uncorrected static tests where the value of p <0.05 was considered significant.

Ethical approval
We obtained written consent in all cases. Approval was obtained from the participants prior to collection of samples. The study proposal was evaluated and approved by the Ethics Committee, Faculty of Medicine and Health Sciences, University of Sana'a.  (Table 2). Table 4 shows the prevalence of Candida species according to presence of symptoms. The Candida was isolated in most pregnant women who had symptoms of Candida VV (86.2%), while only 22.3% was isolated in pregnant women who had no symptoms of Candida VV. There was a higher rate of isolated albicans Candida in women with symptoms (73.6%) compared to 10.7% in women without symptoms. On the other hand, the percentage of non-Candida albicans in symptomatic women (11.7%) is almost similar to 12.6% in asymptomatic women.

DISCUSSION
Information on the prevalence of VVC in Yemen is not known. Unfortunately, VVC is not a reportable disease and diseases are regularly diagnosed by sign and symptoms without supporting laboratory diagnosis. The same as a result, the spectrum of yeasts involved in causing the disease is unknown in Yemen. The prevalence of VVC among pregnant women is also unknown in Yemen, but it is known worldwide that it is the second most common infection in the vaginal vulva part in women with symptoms as it represents about 17% to 42.% 18-20 . Even though the prevalence of infection in the current study among pregnant women ranging in age from 17-37 years, with mean±SD of 25.4±4.9 years was 51.6%, slightly higher than the reported range, 18-20 , it was higher than the prevalence rates reported by Ahmed and others in India among pregnant and non-pregnant women 19 and Olowe et al., in Nigaria among pregnant women 20 , however lower prevalence was reported by ERylander et al., among sexually active young women and association with orogenital sex 21 . Differences in rates can be explained by identifying differences in the sociodemographic characteristics and immunity status of patients 22 , treatment of patients with extensive antibiotics, immunosuppressive drugs 23 and hormonal effects 24 as several of the factors for differences in prevalence and/or recurrent VVC between studies. Age, level of education, pregnancy stages, parity and sociodemographic status were investigated as potential risk factors for vulvovaginitis in the present study. There was no strong evidence for the association between level of education, gestational trimester, and prevalence of VVC. Among the sociodemographic characteristics, age appears to be an important factor in the incidence of VVC in general as the rate of VVC was 61% with OR =1.8, 95% CI =1-3.9, P =0.04 in the 20-24 years age group ( Table 2). The current result was similar with a preceding study by Sobel et al., 25 , they showed that VVC is rare in puberty (the first appearance of menstruation), but its frequency increases with the end of the second decade of life (10-19 years) and reaches its peak in the third (20-29 years) and the fourth decade (30-39 years old). Although in the current study, the odds ratio detected for the 20-24 age group was significant; the association of older age groups and vaginal candidiasis was not statistically significant as far as the adjusted odds ratio was considered (Table 2); this may point toward that the association was affected by other variables. The incidence was more in pregnant women who are illiterate (68%) than in patients with university education (39.7%). The improvement in personal hygiene and/or the economic situation resulting from education may explain the difference in the incidence rate among the illiterate and the people who got better education. Current results were consistent with those of Rathod et al., in India 26 , but in disagreement to the conclusion of Vadav and Prakash 27 . Among the six Candida species isolated in this study, the isolated rate for C. albicans was 39.5% , 9 (4.7%) for C. globrata, 2 (1.1%) for C. dubliniensis, 1 (0.52%) for C. rugosa, 6 (3.2%) for C. lipolytica and for C. famata 4 (2.1%) ( Table 3). The results of the current study of C. albicans as the predominant species were reliable with similar previous studies 25, 28 . Although many studies on the prevalence of different types of Candida have led to the general concurrence that C. albicans are the most common isolated species in patients with vulvovaginal candidiasis, there has been an increasing trend to recover non-albicans Candida species. This is evidenced by the current study in which the isolation rate of non-albicans  34 , they were discovered that C. glabrata was the predominant yeast among nonalbicans Candida species.  35 C. albicans proved to be the most common isolated species followed by C. glabrata, C. tropicalis, and C. parapsilosis. The significance of this result can be explained with caution that C. globrata may replace C. albicans under selective pressure of fluconazole, leading to inflammatory resistance to current fluconazole-based treatment in Yemen.

CONCLUSIONS
There is a high probable rate of VVC among pregnant women in Yemen, and this highlights the need for health authorities to develop strategies to diagnose VVC, including vaginal swabs for candidiasis as a routine procedure for all pregnant women. This study also revealed a steady increase in time with a non-C. albicans species prevalence rate. The guidelines for managing VVC syndrome in Yemen should be revised to include a special protocol for pregnant women.