Objective The contribution of sexual transmission to genital Candida infection remains unclear. This study sought to investigate whether sexual behaviours were associated with the presence of genital Candida species among a cohort of women who have sex with women (WSW) in addition to determining the genetic concordance of genital Candida spp. among WSW in sexual partnerships.
Methods WSW ≥18 years of age presenting to the Mississippi State Department of Health STD Clinic during 2009–2010 completed a sexual behaviour survey. Culture of vaginal fluid was performed for Candida spp. identification; associations with participant characteristics were determined using logistic regression analysis. Random amplified polymorphic DNA (RAPD) PCR was performed on DNA extracted from yeast cultures of WSW in sexual partnerships in which both partners had isolates of Candida spp. identified and among a set of age/sexual behaviour matched controls. RAPD genetic fingerprints were evaluated by hierarchical cluster analysis for concordance.
Results Genital Candida spp. were isolated in 105/196 (53.6%) of women: 13/105 (12.4%) had symptomatic vulvovaginal candidiasis while 92/105 (87.6%) had asymptomatic vaginal colonisation. Bisexual identity, sex with women and men during the past 12 months and numbers of male sexual partners during the past 12 months were the only significant predictors of genital Candida spp. in bivariate analysis. 13 pairs of WSW in sexual partnerships in which both partners had genital Candida spp. and 11 WSW with genital Candida spp. not in sexual partnerships were identified. Candida spp. RAPD banding patterns were discordant for all isolates among WSW within partnerships and in controls.
Conclusions This study found no evidence supporting sexual transmission of genital Candida spp. between women.
Statistics from Altmetric.com
Vulvovaginal candidiasis (VVC) is a common vaginal infection affecting up to 75% of women.1 It is second only to bacterial vaginosis (BV) in frequency of vaginal infections.2 ,3 Studies have shown that VVC may be more common among African-Americans.4 Candida albicans is the yeast species most likely to cause VVC, although it may be caused by C glabrata, C tropicalis and rarely other Candida spp.5 In addition to causing symptomatic disease, Candida spp. also colonise the vagina in approximately 15–20% of asymptomatic women.1 ,5
The contribution of sexual transmission to genital Candida spp. and VVC remains unclear.1 Epidemiological studies have shown that VVC is associated with increased frequency of vaginal sex,6 receptive orogenital sex4 ,6 and increased numbers of heterosexual partners.7 Molecular genetic evidence also suggests that colonisation with specific strains of C albicans predispose women to the development of VVC; strains with similar molecular signatures have been found in both men and women with active genital yeast infections.8 Correlation of candidal infection between sexual partners in heterosexual relationships has been observed,9 and studies using genotype comparison techniques have suggested that genital C albicans isolates may be sexually transmitted.8 Nevertheless, conflicting evidence exists regarding the treatment of male partners of women with recurrent VVC.10 ,11 Studies among heterosexual partnerships have not provided evidence beyond indirect correlation that yeast can be sexually transmitted.9
Recently, the plausibility of sexual transmission of genital Candida spp. among women who have sex with women (WSW) has been suggested in a study showing increasing odds of VVC (either symptomatic or asymptomatic) with greater numbers of female partners.12 Additional epidemiological data and partner studies are needed to clarify whether genital Candida spp. can be sexually transmitted between women. The objectives of this study were to investigate whether sexual behaviours were associated with the presence of genital Candida spp. among a cohort of WSW in addition to determining the genetic concordance of Candida spp. among WSW in sexual partnerships in which both partners had a positive yeast culture.
Clinical procedures and specimen collection
This study was approved by the Institutional Review Boards at the University of Mississippi Medical Center (UMMC), the Mississippi State Department of Health (MSDH) and the University of Alabama at Birmingham (UAB). African-American WSW aged ≥18 years presenting to the MSDH STD Clinic in Jackson, MS, between February 2009 and October 2010 and participating in a study of STI prevalence and risk behaviours among WSW13 provided specimens. A written survey on sociodemographics, sexual history, sexual behaviour characteristics and sexual partnership characteristics was completed by participants, as previously reported,13 ,14 followed by a standardised pelvic exam with collection of vaginal fluid for pH measurement, saline microscopy, potassium hydroxide evaluation and Gram stain (three of four Amsel's criteria15 were necessary for the clinical diagnosis of BV). During the exam, a polyester swab was used to collect secretions from the posterior vaginal fornix in order to perform yeast culture. Vaginal fluid was also collected for Trichomonas vaginalis InPouch culture, Chlamydia trachomatis and Neisseria gonorrhoeae testing using the Aptima Combo 2 assay, and Mycoplasma genitalium testing using a modified Aptima assay, as previously reported.13 ,14 Serum was collected for syphilis and HIV testing, as previously reported.13
The yeast collection swab was used to inoculate a liquid culture (brain heart infusion (BHI) broth supplemented with penicillin and streptomycin) within 24 h postcollection. Positive cultures were archived by cryopreservation in BHI with 10% glycerol. A cryopreserved archive of each positive liquid culture at UMMC was subsequently sent to UAB for formal speciation. A sample of the cryopreserved specimen was cultivated on Sabouraud Dextrose Agar (BD Diagnostics, Sparks, Maryland, USA) supplemented with gentamicin and CHROMagar Candida (supplemented with chloramphenicol; DRG International, Springfield, New Jersey, USA). Colony morphology, colour and germ tube growth characteristics were noted. Colonies were phenotypically categorised by gross observation and microscopic growth characteristics on corn meal agar (BD Diagnostics), and yeast species were confirmed by the API 20C AUX biochemical panel (Biomérieux, Durham, North Carolina, USA).
We identified WSW in active sexual partnerships with another woman in the study in which both women had a positive yeast culture; a representative of each matching yeast colony phenotype observed in culture from each partner was selected for DNA extraction. An additional group of WSW not in active sexual partnerships with another woman in the study was randomly matched to the women in sexual partnerships, selected via distribution frequencies by age and sexual contact with men. This control group was selected to provide a sampling of the potential genetic diversity of all yeast isolates within the entire cohort of women.
A 3–5-day-old colony was picked from positive yeast culture plates and suspended in sterile phosphate buffered saline, pH 7.5. The suspension was washed and centrifuged to remove the supernatant. The pellet was resuspended in a sterile cocktail of Tris/EDTA/sodium chloride (TEN) buffer (pH 7.5) and 50 mM dithiothreitol (DTT). Zymolase (G-Biosciences, St. Louis, Missouri, USA) was added to a final concentration of 0.15 U/µL and incubated at 37°C for 1 h. The spheroplasts were pelleted and digest supernatant was removed by centrifugation. The spheroplast pellet was resuspended in 100 mM DTT containing 0.075 mAU/µL of Proteinase K (Qiagen, Valencia, California, USA) and incubated at 70°C for 10 min. DNA from the treated spheroplasts was extracted using Generation Capture Column chromatography (Qiagen). The concentration and purity of the DNA was determined by NanoDrop (Thermo Scientific, Wilmington, Delaware, USA). All colony isolate DNAs were adjusted to a working concentration of 5 ng/µL.
A total of 25 ng of template DNA was used per random amplified polymorphic DNA (RAPD) reaction. The Illustra Ready-To-Go RAPD Analysis Beads kit (GE Healthcare Biosciences, Pittsburgh, Pennsylvania, USA) was used according to manufacturer's instructions. The kit contains six RAPD primers; these primers are specifically designed to be arbitrary and used alone in each reaction. Amplicons were electrophoresed in 1X Tris/Acetate/EDTA (pH 8.3) on 2% agarose gels prestained with ethidium bromide at 45 V for 16 h at 4°C. The resultant gels were imaged, and the nucleotide base-pair sizes of the banding patterns produced from the RAPD amplicons were estimated by comparison to a molecular weight DNA ladder (Hi-Lo DNA Marker, Minnesota Molecular, Minneapolis, Minnesota, USA) using Quantity One software V.4.6.9 (Bio-Rad, Hercules, California, USA). DNA band sizes were entered into a spreadsheet for cluster analysis with SPSS V.21 (IBM Corporation, Armonk, New York, USA).
Statistical analysis for epidemiological data was conducted using Stata/IC V.12.1 (College Station, Texas, USA). Percentage distributions were calculated for all categorical variables; continuous variables were categorised for ease of interpretation. Unadjusted associations between each demographic/behavioural factor and the presence of genital Candida spp. were examined using logistic regression analysis.
One hundred and ninety-six African-American WSW were enrolled; age ranged from 18 to 40 with the majority of women being <25 (mean=24.5, SD=5.0). Of these 196 women, 111 (58%) reported sex with women only during the previous 12 months, while 80 (42%) reported sex with both women and men (WSWM) during this time frame. Other demographic and behavioural characteristics of this sample have been presented previously.13 In total, 105/196 (53.6%) women were found to have genital Candida spp. isolated by culture: 13/105 (12.4%) were diagnosed with symptomatic VVC while 92/105 (87.6%) had asymptomatic vaginal colonisation. Characteristics of study participants, stratified by the presence of genital Candida spp., are presented in table 1.
Few variables were associated with the presence of genital Candida spp. Women claiming bisexual identity were more likely to have Candida spp. than lesbian-identified women (OR 2.3; 95% CI 1.2 to 4.5; p=.01). Likewise, women who reported both female and male partners over the past 12 months were significantly more likely to have Candida spp. (OR 3.5, 95% CI 1.9 to 6.4; p=.01) than women who reported female partners only. The number of male partners over the past 12 months was also associated with Candida spp., although there was no difference associated with having 1 vs >1 male partners. None of the other sexual partner or sexual behaviour variables were associated with the presence of genital Candida spp. nor was diagnosis with BV or any STI.
In total, 104/105 women had archived yeast cultures available. Table 2 demonstrates the prevalence of various Candida spp. among these yeast cultures. Also, 71/104 (68.3%) women had only one yeast phenotype; the remaining participants (n=33; 31.7%) had two (26.0%), three (3.8%) or four (1.9%) distinct yeast phenotypes isolated in culture. C albicans represented 83.7% of all yeast isolates, C tropicalis 17.3%, C parapsilosis 19.2%, C glabrata 11.5%, C kruseii 5.8% and C lusitaniae 1.9%.
Among the 104 WSW with positive genital yeast cultures available for further study, 25 women (in a total of 13 sexual partnerships) were identified in which both female partners had a yeast isolate grown in culture; 11 of these sexual partnerships did not share a common partner while 2 shared a common partner who was also enrolled in the study. Four partnerships consisted of exclusive WSW during the past 12 months, one partnership of WSW reporting sex with both women and men (WSWM) during this time frame and eight partnerships a mixture of both WSW and WSWM. RAPD patterns using three of the six available primers in the kit resolved discordance in 66% of the Candida spp. phenotypes between female partners (figure 1), that is, the yeast strains between partners were not genetically identical. The initial number of RAPD reactions was 75; three reactions with three unique primers for the 25 women in sexual partnerships. Because several women had multiple yeast isolates, cross-tab testing was performed on all identical yeast phenotypes shared between members of the partnerships; this approach incurred an additional 27 RAPD reactions, which resolved three of the six potential phenotype combinations using the original primer set. The remaining three combinations of yeast isolates required the use of the additional three primers in the RAPD kit not used in the original RAPD amplifications. Overall, RAPD banding patterns were discordant in all 13 WSW sexual partnerships.
A dendogram plot was produced using hierarchical clustering analysis of the amplicon patterns produced by the series of RAPD reactions (figure 2). Candida spp. fell into two distinct clusters. Cluster 1 contained C albicans and C tropicalis while cluster 2 contained C parapsilosis, C glabrata and C krusei. Of note, 7 of the 13 female sexual partnerships were a C albicans/C parapsilosis pairing. C albicans was shared between the remaining six female sexual partnerships, with some individuals having >1 C albicans strain, as defined by phenotype and genotype results.
This mixed methods study found no evidence (epidemiological or laboratory-based) supporting sexual transmission of genital Candida spp. between women. This is in contrast to findings from a previous study that found increasing odds of VVC (either asymptomatic or symptomatic) among WSW with greater numbers of female partners during the previous year.12 Bisexual identity, sex with women and men during the past 12 months and numbers of male partners during the past 12 months were the only significant predictors of genital Candida spp. in this study. We found no associations between numbers of female partners (both lifetime and past 12 months) nor sexual practices with female partners (ie, vaginal penetration by female partners, shared use of vaginally inserted sex toys with female partners, inconsistent condom use on sex toys with female partners, etc.) and the presence of genital Candida spp. In addition, RAPD banding patterns were discordant for all yeast isolates among WSW within sexual partnerships in which both partners had a positive yeast culture as well as in a group of controls.
It has been hypothesised that penile-vaginal intercourse might facilitate the movement of Candida spp. into the vagina and result in minor local trauma that creates conditions suitable for tissue invasion.16 ,17 An association between frequency of heterosexual intercourse and symptomatic attacks of VVC has formed the basis of this hypothesis.16 ,17 In addition, epidemiological data suggest that orogenital and anogenital contact may also transmit yeast.4 ,6 ,18 Prior studies using molecular techniques to genotype strains of Candida spp. among women with VVC and their male partners have shown genetic similarities, further supporting the hypothesis of sexual transmission.8 ,19–22 With regards to WSW, it is hypothesised that sexual activities such as orogenital sex and the sharing of sex toys could facilitate cross-infection of Candida spp., leading to vaginal colonisation and subsequent development of infection.12 Indeed, sexual exchange of infected vaginal fluid has been found to be a possible mechanism for the acquisition of BV among WSW.23 ,24 However, taking into account the results of our study and the fact that genital Candida spp. can gain access to the vaginal lumen from the adjacent perianal area,3 it may be that the pathogenesis of BV and vaginal yeast infections is different among WSW, that is, WSW may acquire genital Candida spp. endogenously while BV is acquired exogenously from infected female partner(s). Thus, it remains unknown whether Candida spp. are truly sexually transmitted or whether sexual activities only potentiate Candida spp. colonisation while additional factors (ie, use of antibiotics, etc.) influence development of infection.3
To our knowledge, this is the first study using RAPD to perform genotypic characterisation of genital Candida spp. among WSW in sexual partnerships in which both partners had a positive yeast culture. We found no evidence of genetic concordance of yeast isolates among WSW in sexual partnerships. In keeping with the hypothesis mentioned above and our results, it may be that female–female sexual practices do not sufficiently facilitate the movement of Candida spp. into the vagina nor induce enough local trauma to create conditions suitable for tissue invasion of yeast. Alternatively, it could also be that repeated sexual exposures to female partner(s), similar to those that predispose WSW to a higher risk of colonisation of Gardnerella vaginalis25 and other BV-associated bacteria, are necessary to facilitate the transfer of yeast between women. Unfortunately, we did not have data available with regards to the frequency of sexual activities WSW in sexual partnerships participated in with their female partners. It is also important to note that several women in the partnerships were also having sex with men at the time of enrolment; how this influenced their genital yeast isolates is unknown as we did not have their male partners available for testing. Therefore, further research is necessary to confirm the results of our study.
Although not a primary objective, it should be noted that this is the first study detailing the prevalence of genital Candida spp. among a cohort of African-American WSW. Our observed prevalence (53.6%) was much higher than that seen in prior studies of premenopausal women (15–20%)3 and Caucasian WSW (18.4%).12 Potential explanations could be that all women in this study were of African-American race (a genetic factor that may predispose to colonisation or vaginitis with yeast),4 constituted a high-risk STD clinic population frequently diagnosed with BV/STIs and may have recently taken antibiotics or had additional risk factors predisposing them to yeast colonisation/VVC.1 Unfortunately, we did not collect the latter information and can only speculate that this may have contributed. Interestingly, in a study of 338 predominately African-American women at an STD clinic in Birmingham, Alabama, USA, the prevalence of yeast detected by wet prep/culture was also high at 39.1%.26
We also found that multiple species and mixed subspecies of Candida can be isolated from the vaginal vault. The four species associated with VVC in the USA (C albicans, C tropicalis, C glabrata and C parapsilosis)1 were observed in this population. However, while three of the four species were found in similar rates to those described in the literature,27 C parapsilosis was present at an unusually higher rate (19.2%). The prevalence of C parapsilosis in the USA is estimated to be 1.7% nationally and 1.8% in the South.27 Potential reasons for this deviation from the national prevalence of C parapsilosis among women in this study are unknown. Of note, a higher prevalence of C parapsilosis (12%) has been observed in a study of Costa Rican sex workers with or without VVC.28 In this study, the use of miconazole ≥4 times a year, for treatment or prophylaxis, was significantly associated with drug resistance among the C parapsilosis isolates. Given the recent increase in non-C albicans-associated VVC,29 perhaps attributed to the overuse/abuse of over-the-counter (OTC) antifungals,30 one could hypothesise that the increased prevalence of C parapsilosis observed in our study could be related to behavioural factors such as overuse of this type of medication. However, we did not have detailed behavioural information available with regards to frequency of OTC vaginal antifungal use to see whether such a correlation existed.
This study has several limitations. First, the relatively small sample size of African-American WSW presenting to an urban STD clinic in Jackson, MS, limits the generalisability of our results. Second, as this was a pilot study, we were unable to collect data on receptive oral sex with female partners or data on other specific sexual practices with female partners, which could have an association with the presence of genital Candida spp. In addition, although RAPD is a powerful molecular technique that allows for the investigation of genetic information without prior knowledge of the target organism's genome, it has several weaknesses. Template DNA may be contaminated from other sources. In addition, if RAPD conditions are not optimised and consistent from test to test, results may not be reproducible. Resultant banding patterns may be difficult to interpret in number, intensity and position to a standard. Nevertheless, RAPD can be a valuable tool under controlled and well-characterised settings such as those used in our current and prior14 studies.
In conclusion, in contrast to previous studies of BV, this study found no evidence of sexual transmission of genital Candida spp. between women. More research is needed to better understand the contribution of sexual transmission to genital Candida spp. isolates and VVC among WSW.
The contribution of sexual transmission to genital Candida spp. and vulvovaginal candidiasis remains unclear.
In contrast to previous studies of bacterial vaginosis, this study found no evidence (epidemiological or laboratory-based) supporting sexual transmission of genital Candida spp. between women.
The overall prevalence of genital Candida spp. in this study (53.6%) was much higher than that noted in prior studies of premenopausal women (15–20%).
Handling editor Jackie A Cassell.
Acknowledgements The authors would like to thank Heather King, Tina Barnes and the Crossroads STD Clinic nursing staff for their assistance in recruiting and enrolling patients for this study. In addition, the authors would like to thank the laboratory personnel of John Cleary, PharmD, and Edwin Swiatlo, MD, PhD, for archiving the yeast isolates at the University of Mississippi Medical Center prior to transfer to the University of Alabama at Birmingham. The authors would also like to thank Edwin Swiatlo, MD, PhD, for helpful discussions regarding manuscript preparation. CAM is supported in part by a Developmental Award from the American Sexually Transmitted Diseases Association.
Contributors CAM, CAR, LAM and JRS were responsible for conception and design.CAM, CAR, CJP and LAM were responsible for acquisition of data.CAM, CAR, ELA and JRS were responsible for analysis or interpretation of data.CAM, CAR and ELA were responsible for drafting of the manuscript.CAM, CAR, CJP, ELA and JRS were responsible for revising the manuscript for intellectual content. All authors were responsible for the final version of the manuscript to be published.
Funding Data from this study were presented in part at the 20th Biennial Conference of the International Society of Sexually Transmitted Diseases Research in Vienna, Austria, on July 16, 2013; abstract# P1.036.
Competing interests None.
Ethics approval University of Mississippi Medical Center, Mississippi State Department of Health, University of Alabama at Birmingham.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.