( ) 8 (800) 100-28-84

Vulvovaginal candidiasis: histologic lesions are primarily polymicrobial and invasive and do not contain biofilms

:Alexander Swidsinski (MD), Alexander Guschin (PhD), Qionglan Tang (MD), Yvonne Dörffel (MD), Hans Verstraelen (MD, PhD), Alexander Tertychnyy (MD), Guzel Khayrullina (PhD), Xin Luo ( MD), Jack D. Sobel (MD), Xuefeng Jiang (MD).


The recent demonstration of a vaginal biofilm in bacterial vaginosis and its postulated importance in the pathogenesis of recurrent bacterial vaginosis, including relative resistance to therapy, has led to the hypothesis that biofilms are crucial for the development of vulvovaginal candidiasis. The histopathology and microbial architecture of vulvovaginal candidiasis have not been previously defined; neither has Candida, containing biofilm been reported in situ. The present study aimed at clarifying the histopathology of vulvovaginal candidiasis including the presence or absence of vaginal biofilm.

Study Design

In a cross-sectional study, vaginal tissue biopsies were obtained from 35 women with clinically, microscopically, and culture-proven vulvovaginal candidiasis and compared with specimens obtained from 25 healthy women and 30 women with active bacterial vaginosis. Vaginal Candida infection was visualized using fluorescent in situ hybridization with ribosomal genebased probes.


Candida microorganisms were confirmed in 26 of 35 biopsies obtained from women with vulvovaginal candidiasis; however, Candida containing biofilm were not detected in any of the cases. Histopathological lesions were exclusively invasive and accompanied by co-invasion with Gardnerella or Lactobacillus species organisms.


Histopathological lesions of vulvovaginal candidiasis are primarily invasive in nature and polymicrobial and do not resemble biofilms. The clinical significance of Candida tissue invasion is unknown.

Key words

Atopobium ;biofilm ;Candida ;Gardnerella ;Lactobacillus crispatus and iners ;polymicrobial invasion ;vulvovaginal candidiasis

Vulvovaginal candidiasis (VVC), specifically its recurrent form, is a highly problematic and a common clinical therapeutic challenge.1In clinical practice the diagnosis is mainly based on clinical signs and symptoms and a typical picture apparent on wet mount saline and 10% KOH microscopy.

AJOG at a Glance

Why was this study conducted?

Biofilms are hypothesized as crucial for the development of vulvovaginal candidiasis (VVC). We investigated vaginal biopsies from 35 women with VVC using fluorescent in situ hybridization and compared with specimens from healthy women and women with bacterial vaginosis.

Key findings

Contiguous Candida adherence was not detected in any of the cases or in controls. Histopathological lesions in 26 of 35 biopsies from VVC were exclusively invasive and accompanied by bacterial co-invasion. Lactobacilli (including L. iners and L. crispatus), Gardnerella, and Atopobium were most frequently co-invading.

What does this add to what is known?

Polymicrobial mucosal invasion is an unrecognized feature of Candida vaginitis. Our results do not support lactobacilli being beneficial or protective. Different from bacterial vaginosis, we found no biofilm elements in vaginal biopsies obtained from women with VVC.

Candidaby culture or polymerase chain reaction validates clinical findings but is not routinely obtained or justified. Approximately 1020% of asymptomatic healthy women harbor culturableCandidasp. and other yeasts in the vagina. ,
The pathogenesis of acute VVC is currently thought to reflect a microbiome imbalance or dysbiosis in the vagina as well as an abnormal host mucosal immune response to theCandidaorganism. The reasons for and factors involved in the development of dysbiosis are poorly understood. The explanation as to why yeast that normally asymptomatically colonize the vagina and can cause symptoms and inflammation is controversial. ,
A more recent hypothesis includes a switch from unstructured planktonic yeast growth to biofilm formation that facilitates transition from saprophytic to pathogenic yeast behavior. , , , , Potentially, vaginal biofilm could explain acute sporadic VVC or be more relevant in recurrent VVC (RVVC) as a vaginal reservoir for yeast organisms following antifungal therapy and explain vaginal recolonization. Indeed, planktonic and biofilm forms ofCandidaseem to be different entities.Candida-biofilms have been demonstrated in a variety of experimental conditions invitro as well as on prosthetic surfaces and endovascular and urethral catheters invivo.
As a result of growing interest in the topic, numerous investigations have been dedicated to unraveling factors responsible for the development ofCandidabiofilms. , , However, despite the widespread belief and broad acceptance of the possibility thatCandidabiofilms are a critical factor in VVC pathogenesis, we could find no publication that actually demonstrates a microscopic picture of theCandidabiofilm on the vaginal surface. The only publication claimingCandida albicansforms biofilms on the vaginal mucosa exclusively includes pictures of smears from vagina without vaginal epithelium. Confirmation of the presence ofCandidamucosal biofilm invivo is thus lacking.
Accordingly, we investigated the histopathology of VVC using fluorescent in situ hybridization (FISH). Biopsies from healthy women, women with bacterial vaginosis (BV), and women with VVC were comparatively investigated using FISH probes specific for fungi and bacteria.
Materials and Methods


The candidiasis group consisted of 35 randomly selected premenopausal women with confirmed vulvovaginal candidiasis (aged 1937 years, mean 27 years), 25 women from Guangzhou, China, and 10 women from the Friedrichhain Hospital in Berlin, Germany. Five women from Berlin and 8 from China had RVVC, and all others had sporadic VVC.
The diagnosis was based on the clinical appearance and microscopic evaluation of smears and culture. None of the women received antifungal treatment for 2 months prior to the investigation. In the candidiasis group, FISH-performing researches were not blinded to group diagnosis, but the investigators were not aware of individual data including results of culture, swab investigations, and clinical course.

Material from paraffin-embedded biopsies from 25 healthy women (aged 2035 years, mean 26.4 years) investigated for routine care and 30 women with bacterial vaginosis (aged 1840 years, mean 27.2 years) served as controls. All women in control groups were premenopausal. These materials had been preserved from previous studies onbacterial vaginosis (BV) and described. ,
Biopsies of about 35 mm diameter were taken from the middle side wall of the vagina with biopsy forceps (No. ER 058 R; Schubert, Aesculap, Tuttlingen, Germany) and fixated with modified nonaqueous Carnoy solution (6/1/2 volume ethanol/glacial acetic acid/chloroform). The fixated materials could be stored in Carnoy at room temperature for up to 6 months. Usually the time used was convenient for the investigated subject and laboratory staff. Carnoy-fixated material was processed and embedded into paraffin blocks using standard techniques.
Four-micrometer-thick sections were placed on Super Frost plus slides (R.Langenbrinck, Emmendingen, Germany). Sections of vaginal biopsies were hybridized with ribosomal RNAbased FISH probes specific for all bacteria,Gardnerellacluster,Atopobiumcluster, lactobacilli,Lactobacillus inersandLactobacillus crispatus, all yeasts, andCandida albicans(Table1)
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Table 1.jpg
All hybridizations were performed at 50C using a protocol described previously.
A Nikon e600 fluorescence microscope was used. The images were photodocumented with a Nikon DXM 1200F color camera and software (Nikon, Tokyo, Japan).
The study was reviewed and approved by the Institutional Review Board of Jinan University.
Hybridizations signals positive for yeasts were detected in biopsies from 26 of 35 women with VVC/RVVC. No yeasts were observed in the healthy group and the BV group, regardless of the protocol used (Figure1).
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 1.jpg
Within the candidiasis group, signals were positive forCandida albicansin 18 women, and 8 further samples were negative forCandida albicansbut were hybridized with the universal for most yeasts PF2 probe (Table2andFigure2,Figure3,Figure4,Figure5,Figure6). The histological patterns of candidiasis were identical in both sporadic as recurrent VVC.
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 2.jpg
a The fungal invasion was always accompanied by bacterial coinfiltration as detected with the Eub 338 FISH probe, which represents all bacteria. Such a bacterial coinfiltration is characteristic for VVC and never occurred in healthy women or in BV. Gardnerella, even when highly concentrated, accompanied Candida in VVC but did not build a characteristic biofilm as seen in BV.
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 2.2.jpg
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 3.jpg
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 4.jpg
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 5.jpg
Vulvovaginal candidiasi histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Figure 6.jpg
We found no yeast in the biopsies of 9 women from the VVC group (5 from China and 3 from Berlin). This could be due to the nonuniform fungal invasion over the epithelial surface, withsome regions remaining uninvolved as demonstrated inFigure3(a1a3).
All fungal cells detected by hybridization were mainly invasive, with variable hyphae penetrating more or less deeply into the epithelial surface of the biopsy (Figure2,Figure3,Figure4,Figure5,Figure6) and leaving some of the biopsy areas completely free (Figure3[a1a3]. The invasive character of the infection was especially well seen when fungal fluorescence hybridization was 4′,6′-diamino-2-phenylindole (DAPI) counterstained, revealing all eukaryotic cells (Figure2,Figure3,Figure4,Figure5,Figure6).
Occasionally single fungal cells or blastospores could be seen in slime-covering biopsies; however, a fungalbiofilm embedded in its own extracellular matrix was never observed.
Fungal infiltration was always accompanied by co-invasion with bacterial components. Bacteria were either evenly distributed over the depth of the fungal invasion (typical forGardnerellaand some lactobacilli, 18 cases total) or concentrated at the forefront of the fungal invasion (Lactobacillus inersco-invasion,Figure5,Figure6, 8 cases total).
Co-invading bacteria were polymicrobial, representing a broad spectrum of vaginal microbiota. Lactobacilli,Gardnerella, andAtopobiumwere most frequently seen.Gardnerellawas always associated either with considerable amounts of lactobacilli ranging from 108to 1011(80%) and/orAtopobium(60%). A high concentration of lactobacilli could accompany the fungal invasion in the absence ofGardnerellaorAtopobium. (Figures5and6).
Co-infiltrating lactobacilli were often but not exclusively represented byLactobacillus iners(Table1).
Many excellent reviews are dedicated to different aspects ofCandidaresearch. , , , , , , Our perception of VVC is based only on interpretation of indirect indices: epidemiology of colonization and disease, macroscopic appearance of lesions, symptoms, investigation of vaginal smears using microscopy, culture or molecular genetic identification of microorganisms, and simulation of infection/colonization invitro and in experimental animal models.
This is the first study that directly visualizesCandidamicroorganisms in vaginal biopsies, and the results substantially contradict widespread assumptions. We found no biofilm elements in vaginal biopsies obtained from women with VVC. Results were identical in women with acute sporadic or recurrent VVC.
This is in contrast to biofilms seen in bacterial vaginosis. Moreover,Candidacells on the surface were single, nonconfluent, and what is most important nonadherent to the vaginal epithelium. This cannot be explained by processing sample biases.
Carnoy solution, which we used for fixation of the biopsy samples, demonstrated its high efficiency in preserving biofilms and even bacteria-free slime on different mucosal surfaces. However, in appropriate and representative specimens obtained in VVC, we were unable to detect any contiguousCandidalayer on the mucosal or epithelial surface. It is possible that the cementing properties of the extracellular matrix, which is produced byCandida, is very low, unable to maintainCandidaattachment to the vaginal surface and not different from those produced by any other microorganisms growing in a colony on a culture plate.
The fixation or adherence ofCandidamicroorganisms to the vaginal surfaceispossible or likely through hyphal mucosal invasion. Visually apparent white membranes, which cover the vaginal epithelium in vulvovaginal candidiasis and appear to be biofilms, are actually deep inflammatory infiltrates.
In the past, few studies have included vaginal biopsies in women with VVC. In the absence of such critical tissue status information, vaginal candidiasis has been considered as entirely superficial mucosal infection and the lack of even superficial invasion emphasized. In fact, some clinicians refer toCandidavaginitis as infection of vaginal secretions only. Moreover, the histological appearance of nonvaginalCandidalesions presented in some publications are all definitively invasive (Gow and Hube, page 407, Figure1, A, B, and C, and Figure2; Jabra-Rizk etal, page 2729, Figure9, B and C; Gao etal, page 737, Figure9; Allison etal, page 8, Figure6) and similar to our findings.
A most prominent feature in our study is a deep infiltration of vaginal tissue. The infiltration is not homogeneous with some regions affected more heavily than others and some remaining completely free. The removal of such lesions is impossible without stripping the integrity of the epithelial layer. In spite of the depth of infiltration, it is not full thickness, and disseminated infection is not reported clinically.
Candidacolonization and infection occur in polymicrobial environments. Our findings suggest possible bacteria-yeast interactions in tissue invasion. With regard to VVC,Gardnerellaspp. was claimed to promote and lactobacilli to supressCandidabiofilms. , While our data support a contributory promoting role forGardnerella, little was forthcoming to support lactobacilli being beneficial or protective.
On the contrary, lactobacilli were visualized accompanyingCandidainfection even more often thanGardnerella. The concentrations or density of lactobacilli within the vaginal epithelium was high, both forL.inersand otherLactobacillusspecies includingL.crispatus. Because the lactobacilli were homogeneously distributed within the lesion and in some cases even enriched at the forefront ofCandidainfiltration, a pathogenic fungal-bacterial symbiosis seems more likely than a secondary saprophytic relationship.
Obviously pathogenic consortia include a variety of different types of bacteria as demonstrated in polymicrobial BV biofilms. The interaction between fungi and bacteria may contribute to the switch from saprophytic to invasive forms of fungal growth. Quorum-sensing research investigates stimuli that synchronize a correlated response of the microorganism to population density. Presently it is mainly focused on cross talking within single groups of microorganisms, , , this but should be widened to polymicrobial populations.
In our series, 26% of the lesions did not hybridize with theCandida albicans-specific Caal FISH probe but were positive with the universal yeast probe. At present, we cannot designate the origin of these fungi. Either the Caal FISH probe hybridizes less stringently to other representatives ofCandida, or the infiltration is indeed caused by non-Candidayeasts in some cases, which were clinically diagnosed as VVC. However, the extent of the infiltration byCandida albicansand other yeasts was identical, even in the absence of hyphae (Figure6), indicating similar pathogenic potential.
Our data point to vaginal epithelial surface or mucosal invasion as an unrecognized feature of VVC. Although vaginal biopsies were not obtained in asymptomatic women without VVC but colonized withCandida, it is likely that such common saprophytic colonization occurs in the absence of any tissue invasion. Most importantly, no evidence of in situ, invivoCandidabiofilm existence emerged.
The clinical relevance of these findings is largely unknown, but the absence of finding a biofilm in VVC implies that antibiofilm therapy is not indicated because it may be required in BV. The finding of tissue invasion in VVC may have relevance to recolonization of vagina with candida organisms following antifungal therapy (ie, the vaginal mucosa may serve as organism reservoir to explain recolonization following treatment ).
The work has not been published previously and is not under consideration for publication elsewhere. The publication is approved by all authors, and tacitly or explicitly by the responsible authorities in which the work was carried out, if accepted, it will not be published elsewhere in the same form, in English or in any other language, including electronically without the written consent of the copyright holder. Each author named in the byline participated actively and sufficiently in the study reported. The author contributions included the following: Drs Swidsinski, Guschin, XY, and Tertychnyy designed the study; Drs Dörffel, Tang, Tertychnyy, Luo, and Jiang conducted the study; Drs Sobel and Verstraelen critically revised the manuscript; Drs Swidsinski, Guschin, and Dörffel performed the fluorescent in situ hybridization; and Drs Tertychnyy, Tang, Luo, and Jiang analyzed the data. All authors contributed to the conception of the work, revising of the data, shaping of the manuscript, and approved the final draft submitted.
  1. Denning D.W. Kneale M. Sobel J.D. Rautemaa-Richardson R. Global burden of recurrent vulvovaginal candidiasis: a systematic review. Lancet Infect Dis. 2018; 18 (PE339-E347)
  2. Muzny C.A. Schwebke J.R. Biofilms: an underappreciated mechanism of treatment failure and recurrence in vaginal infections. Clin Infect Dis. 2015; 61: 601-606
  3. Nobile C.J. Johnson A.D. Candida albicans biofilms and human disease. Annu Rev Microbiol. 2015; 69: 71-92
  4. Harriott M.M. Noverr M.C. Importance of Candida-bacterial polymicrobial biofilms in disease. Trends Microbiol. 2011; 19: 557-563
  5. Höfs S.Mogavero S.Hube B.Interaction of Candida albicans with host cells: virulence factors, host defense, escapestrategies, and the microbiota. J Microbiol. 2016; 54: 149-169
  6. Gao M. Wang H. Zhu L. Quercetin assists fluconazole to inhibit biofilm formations of fluconazole-resistant candida albicans in in vitro and in vivo antifungal managements of vulvovaginal candidiasis.Cell Physiol Biochem. 2016; 40: 727-742
  7. Allison D.L. Willems H.M. Jayatilake J.A. Bruno V.M. Peters B.M. Shirtliff M.E. Candida-bacteria interactions: their impact on human disease. Microbiol Spectr. 2016; 4: 1-26
  8. Hirota K. Yumoto H. Sapaar B. Matsuo T. Ichikawa T. Miyake Y. Pathogenic factors in Candida biofilm-related infectious diseases. J Appl Microbiol. 2017; 122: 321-330
  9. Lohse M.B. Gulati M. Johnson A.D. Nobile C.J. Development and regulation of single- and multi-species Candida albicans biofilms. Nat Rev Microbiol. 2018; 16: 19-31
  10. Blankenship J.R. Mitchell A.P. How to build a biofilm: a fungal perspective. Curr Opin Microbiol. 2006; 9: 588-594
  11. Ganguly S. Mitchell A.P. Mucosal biofilms of Candida albicans. Curr Opin Microbiol. 2011; 14: 380-385
  12. Soll D.R. Daniels K.J. Plasticity of Candida albicans biofilms. Microbiol Mol Biol Rev. 2016; 80: 565-595
  13. Harriott M.M. Lilly E.A. Rodriguez T.E. Fidel P.L. Noverr M.C. Candida albicans forms biofilms on the vaginal mucosa.Microbiology. 2010; 156: 3635-3644
  14. Swidsinski A. Mendling W. Loening-Baucke V. et al. Adherent biofilms in bacterial vaginosis. Obstet Gynecol. 2005; 106: 1013-1023
  15. Swidsinski A. Loening-Baucke V. Mendling W. et al. Infection through structured polymicrobial Gardnerella biofilms (StPM-GB). Histol Histopathol. 2014; 29: 567-587
  16. Alexander Swidsinski, Loening-Baucke V. Evaluation of polymicrobial involvement using fluorescence in situ hybridization (FISH) in clinical practiceapplication guide. Liehr T (ed). Berlin, Germany: Springer-Verlag; 2017. p. 531-43.
  17. Gow N.A. Hube B. Importance of the Candida albicans cell wall during commensalism and infection. Curr Opin Microbiol. 2012; 15: 406-412
  18. Jabra-Rizk M.A. Kong E.F. Tsui C. et al. Candida albicans pathogenesis: fitting within the host-microbe damage response framework. Infect Immun. 2016; 84: 2724-2739
  19. Matsubara V.H. Bandara H.M. Mayer M.P. Samaranayake L.P. Probiotics vas antifungals in mucosal candidiasis. Clin Infect Dis. 2016; 62: 1143-1153
  20. Sobel J.D. Recurrent vulvovaginal candidiasis. Am J Obstet Gynecol. 2016; 214: 15-21
Article Info
Publication History
Published online: October 25, 2018
Accepted: October 17, 2018
Received in revised form: September 28, 2018
Received: July 3, 2018
The study was supported by a Charité University research promotion grant (2016) and The German Federation of Industrial Research Associations ZIM Project ZF4143701AJ5 . Both funding sources were not involved in the study design, collection, analysis and interpretation of data, in writing a report, or the decision to submit the article for publication.
The authors report no conflict of interest.
Cite this article as: Swidsinski A, Guschin A, Tang Q, et al. Vulvovaginal candidiasis: histologic lesions are primarily polymicrobial and invasive and do not contain biofilms. Am J Obstet Gynecol 2019;220:91.e1-8.