Molecular epidemiology and treatment outcomes of vulvovaginal candidiasis in Namibian women

Abstract

Vulvovaginal candidiasis (VVC) is a common condition in women of childbearing age worldwide and usually presents as vaginal discharge syndrome (VDS). Other important causes of VDS are bacterial vaginosis (BV) and sexually transmitted infections (STIs). The most common STI causing VDS include Trichomonas vaginalis. Vaginal dischare syndrome (VDS) can also be caused by Neisseria gonorrhoeae, Chlamydia trachomatis and Mycoplasma genitalium. However, these are more commonly associated with endocervical infections, and may not present with VDS. Syndromic treatment is the standard of care for VDS, i.e. empirical antibiotics are provided without diagnostic testing. Performance of the syndromic approach depends on the aetiology of VDS and presence of antimicrobial resistance but there is little information from sub-Saharan Africa. The aim of this PhD study was to describe the aetiology of VDS and determine outcome of syndromic treatment of VDS in Namibian women. This PhD study had three components: first, a cross-sectional evaluation was completed to determine the aetiology of VDS. De-identified vaginal swabs (n=253) sent to the routine laboratory at the Namibia Institute of Pathology in Windhoek were tested for STIs using real-time PCR, BV by smear microscopy and VVC by culture and drug susceptibility testing. Second, a prospective cohort study of women (n=109) with VDS at Katutura Intermediate Hospital in Windhoek was conducted to determine incidence, risk factors and microbial aetiology of treatment failure. Last, comprehensive molecular microbiological analysis of phenotypic and genotypic resistance including multi-locus sequence typing was conducted through whole genome sequencing analysis of all Candida glabrata isolates (n=21) obtained in the two studies. Vulvovaginal candidiasis was the main aetiology (43%) of VDS in the cross-sectional study followed by BV (39%) and STIs (36%); multiple infections were present in 34% of women. Candida albicans was the most common fungal species (79%); most isolates (98%) were susceptible to fluconazole. In contrast, no non-albicans Candida species were susceptible to fluconazole. Vulvovaginal candidiasis aetiology at baseline was similarly high in the prospective cohort study with 41% with VVC, 43% of women diagnosed with BV and 40% with STI. At 30 days follow-up, treatment failure was reported by 31% of women, with 18% reporting recurrent and 13% persistent VDS after syndromic treatment. Incidence of treatment failure was 3.6 per 100 person-days at 7 days follow-up and 1.0 per 100 person-days at 30 days follow-up. Vulvovaginal candidiasis was the main risk factor for treatment failure (OR 4.3, 95% CI 1.7‒11, p=0.002). Microbial evaluation of treatment failure attributed most cases to untreated (31%) and azole-resistant (23%) VVC. Twenty-one fluconazole-resistant C. glabrata isolates were obtained from the two studies. Whole genome sequencing analysis showed non-synonymous single nucleotide polymorphisms in antifungal resistance genes in 95% of isolates. Single nucleotide polymorphisms were also detected in genes associated with polyene, echinocandin and multiple class antifungal resistance despite full phenotypic susceptibility to these drug classes. Multilocus sequence typing classified isolates in eight sequence types. The results show that VVC is an important cause of VDS in Namibian women, with C. albicans as the main species followed by C. glabrata. Untreated and fluconazole resistant VVC constitute an important risk factor for VDS treatment failure. Therefore, clinical consideration of VVC in women with VDS and access to fungal culture and susceptibility testing is warranted, especially among women with treatment failure.