Clinical and Personal Lubricants Impact the Growth of Vaginal Lactobacillus Species and Colonization of Vaginal Epithelial Cells: An in Vitro Study

Abstract

Background 

Vaginal lubricants are commonly used during gynecological examinations, during sexual activities, or to alleviate vaginal dryness. Many lubricants contain potentially bacteriostatic or bactericidal agents (parabens, chlorhexidine gluconate, nonoxynol-9). Our objective was to evaluate the impact of lubricants that vary in formulation on the growth and viability of vaginal Lactobacillus species and vaginal epithelial cell (VEC) colonization in an in vitro model.

Methods 

Growth curve, disk diffusion, and minimal inhibitory assays were used to determine the impact of lubricants or excipients on the growth of Lactobacillus crispatus, Lactobacillus gasseri, Lactobacillus jensenii, and Lactobacillus iners. L. crispatus strain was used in VEC colonization assays. Statistical differences were determined by analysis of variance.

Results 

Lubricants containing chlorhexidine gluconate or nonoxynol-9 (N-9; Conceptrol, K-Y Jelly, and Surgilube) significantly inhibited Lactobacillus species growth (P < 0.05). In contrast, other clinical lubricants (E-Z Lubricating Jelly, McKesson Lubricating) and personal lubricants (Astroglide Liquid, Good Clean Love Almost Naked, K-Y Warming Jelly) did not exhibit this effect. Chlorhexidine gluconate had a detrimental effect on Lactobacillus growth and exhibited stronger antimicrobial activity compared with methylparaben and propylparaben (P < 0.0001). There were lubricants that did not induce cytotoxicity in VEC (Good Clean Love Almost Naked, E-Z Lubricating Jelly, McKesson Lubricating Jelly), but these products did substantially decrease the attachment of L. crispatus to VEC, particularly when VEC were preexposed to lubricants before inoculation with bacteria (P < 0.0001).

Conclusions 

This in vitro model indicates that select vaginal lubricants, particularly those with chlorhexidine gluconate, have potentially adverse effects on women’s health by reducing growth and recolonization of vaginal Lactobacillus species.

In Brief

An in vitro study found that vaginal lubricants containing chlorhexidine gluconate can deleteriously affect protective Lactobacillus species, which may lead to increased risk for vaginitis or other adverse gynecologic outcomes.

Women commonly report use of personal lubricants in sexual practices and to help alleviate vaginal dryness and the genitourinary syndrome of menopause.¹ In addition, clinicians frequently use vaginal lubricants in the conduct of gynecologic examination and recommend them to patients with gynecologic cancer to help mitigate adverse effects of treatments, which cause vulvovaginal atrophy and an overall reduction in quality of life.² However, the effect of lubricants on the cervicovaginal microenvironment, including local microbiota, has not been comprehensively studied. In most healthy, reproductive-aged women, the vaginal microbiota is dominated by one or few Lactobacillus species, such as L. crispatus, L. gasseri, L. jensenii, or L. iners.³ However, in postmenopausal women, as a result of hormonal changes, the vaginal microbiota frequently lacks Lactobacillus dominance.³ Colonization of the vagina with lactic-acid producing Lactobacillus species has broadly been associated with vaginal health because the acidic microenvironment consequently protects the host from invading pathogens, including sexually transmitted infections (STIs).³

Studies investigating the impact of lubricants on the human mucous membranes are limited. Two reports using in vitro biomimetic models demonstrated that lubricants with high osmolality reduce epithelial barrier integrity, cause cellular damage, and alter inflammatory responses.^4,5 Relative to other body sites, hyperosmolar lubricants have been shown to cause rectal epithelial cell damage or denudation.⁶ Furthermore, consistent use of hyperosmolar lubricants during anal intercourse has been associated with higher prevalence of STIs among men who have sex with men.⁷ The World Health Organization guidelines published in 2012 state that lubricants’ osmolality should not exceed 1200 mOsm/kg and by comparison, but most lubricants on the market exceed this osmolality recommendation and the osmolality of vaginal secretions is 260 to 290 mOsm/kg.⁸

Most clinical and personal lubricants also contain excipients with antimicrobial properties, such as parabens and chlorhexidine gluconate (CHG). Parabens are commonly used preservatives with a broad-spectrum activity against fungi and bacteria.⁹ The CHG is a broad-spectrum antiseptic effective against gram-positive and gram-negative bacteria, as well as fungi.⁹ The mechanism of parabens’ and CHG’s action relates to the damage of cell membrane and wall integrity.⁹ Intriguingly, CHG has been banned by the US Food and Drug Administration (FDA) from use in over-the-counter (OTC) health care antiseptics and hand sanitizers in 2018 and 2020, respectively.^10,11 Ultimately, the effect of lubricants containing these excipients on the healthy constituents of the vaginal microbiota is unknown. In this study, we aimed to determine the effects of a broad range of personal and clinical lubricants on the growth of vaginal Lactobacillus species and investigated the impact of lubricants on Lactobacillus colonization of vaginal epithelial cells (VECs).

MATERIALS AND METHODS

Lubricants and Bacterial Strains

Personal lubricants were obtained from a local drugstore, and clinical lubricants were obtained from clinics located in Phoenix, AZ, and Baltimore, MD (Table 1). All lubricants were used before the expiration date. Bacterial strains were obtained from the American Type Culture Collection or the Biodefense and Emerging Infections Research Resources Repository and included 5 vaginal Lactobacillus strains: L. crispatus JV-V01, L. gasseri JV-V03, L. jensenii JV-V16, and L. iners AB107, as well as L. crispatus type strain VPI 3199 (the latter isolated from eye). L. crispatus, L. gasseri, and L. jensenii were grown on de Man, Rogosa, and Sharpe (MRS) agar or in MRS broth at 37°C under 5% CO₂. L. iners was grown on tryptic soy agar (TSA) supplemented with 5% defibrinated sheep blood (Quad Five, Ryegate, MT) or in tryptic soy broth (TSB) supplemented with 5% horse serum at 37°C under anaerobic conditions, generated with a GasPak EZ Anaerobe Container System. All bacterial culture media and supplements were purchased from Becton, Dickinson and Company (Franklin Lakes, NJ).

TABLE 1 - Formulations of Tested Clinical and Personal Lubricants

Lubricant	Osmolality, mOsm/kg	Ingredients	Manufacturer
Clinical lubricants
E-Z Lubricating Jelly	2243	Water, glycerin, carbomer, sodium hydroxide, PEG-150, methylparaben, propylparaben	Athena Medical Products
McKesson Lubricating Jelly	2125	Water, glycerin, sodium hydroxide, carbomer 140 g, polyethylene glycol, propylparaben, methylparaben	McKesson Medical-Surgical Inc.
Surgilube Surgical Lubricant	Not tested	Water, hydroxypropylmethylcellulose, propylene glycol, chlorhexidine gluconate	HR Pharmaceuticals Inc.
Personal lubricants
Astroglide Liquid	6100	Purified water, glycerin, propylene glycol, polyquaternium 15, methylparaben, propylparaben	Biofilm Inc.
Conceptrol	1257	Nonoxynol-9 (4%), lactic acid, methylparaben, povidone, propylene glycol, purified water, sodium carboxymethycellulose, sorbic acid, sorbitol solution	Caldwell Consumer Health LCC
Good Clean Love Almost Naked	270	Organic aloe barbadensis leaf juice, xanthan gum, agar, potassium sorbate, sodium benzoate, sodium lactate, lactic acid, natural flavors	Good Clean Love Inc.
K-Y Jelly	2500	Water, glycerin, hydroxyethylcellulose, chlorhexidine gluconate, gluconolactone, methylparaben, sodium hydroxide	Reckitt Benckiser Group plc
K-Y Warming Jelly	10300	Propylene glycol, PEG-8, hydroxypropylcellulose, tocopherol	Reckitt Benckiser Group plc

Excipients exhibiting potential antimicrobial properties against vaginal Lactobacillus species are indicated in bold.

Growth Curve Assay

To determine the bacteriostatic and/or bactericidal effect of lubricants on vaginal Lactobacillus species, the growth of bacteria in liquid media with or without lubricants was analyzed. Bacteria were inoculated into MRS or supplemented TSB liquid media and cultured overnight at 37°C under 5% CO₂ or anaerobic conditions. The broths containing 10% (v/v) lubricants were inoculated with Lactobacillus species at the final optical density at 600 nm (OD₆₀₀) of 0.5 and cultured as described earlier. The broth without any lubricant was used as a positive control. The growth of bacteria with or without lubricants was determined after 4 and 24 hours by measuring the OD₆₀₀ of bacterial cultures and the standard plating assay. For the standard plating assay, bacterial cultures were serially diluted in phosphate-buffered saline (PBS) and spotted on MRS or supplemented TSA plates. The agar plates were incubated at 37°C for 48 to 72 hours for enumeration of colony-forming units (CFU). The concentration of viable bacterial cells in each culture was calculated and presented as CFU/mL.

Disk Diffusion Assay

The disk diffusion assay was used to determine antimicrobial properties of select excipients, such as parabens and chlorhexidine gluconate, on vaginal Lactobacillus species. MRS or supplemented TSA plates were inoculated with 1 × 10⁷ CFU of bacteria. Under aseptic conditions, 6-mm Whatman filter disks (GE Healthcare, Chicago, IL) were impregnated with 20 μL of 20% (w/v) solution of chlorhexidine gluconate in water (Tokyo Chemical Industry, Tokyo, Japan) or 20% (w/v) solutions of methylparaben or propylparaben in ethanol (Tokyo Chemical Industry). Bleach and ethanol were used as positive and negative controls, respectively. Disks were placed onto inoculated agar plates and incubated at 37°C under 5% CO₂ (for L. crispatus, L. gasseri, and L. jensenii) or anaerobic conditions (for L. iners). After a 24-hour incubation, zones of inhibitions were measured in millimeters.

Minimal Inhibitory Concentration Assay

Minimal inhibitory concentrations (MICs) of parabens and chlorhexidine gluconate were assessed using the broth microdilution method. Lactobacillus species were grown overnight on MRS or TSA plates and resuspended in sterile PBS. Bacterial suspensions were adjusted to the OD₆₀₀ of 1.0 and diluted in MRS or supplemented TSB liquid media. Two-fold serial dilutions of excipients (methylparaben, propylparaben, and chlorhexidine gluconate) in appropriate broths (50 μL) were aliquoted into respective wells in a sterile 96-well microtiter plate. Aliquots (50 μL) of bacterial inoculum containing 5 × 10⁵ CFU were added to wells with excipient dilutions. Broth without added excipients or bacterial inoculum were used as growth and sterility controls, respectively. The inoculated microtiter plate was incubated at 37°C under 5% CO₂ or anaerobic conditions. After a 24-hour incubation, the OD₆₀₀ was recorded using a Safire II Multi-Mode Microplate Reader (Tecan, Männedorf, Switzerland). The MIC was defined as the lowest concentration of the excipient that inhibits the visible growth of the tested bacteria.

Colonization Assays

Human vaginal epithelial (V19I) cells were cultured as monolayers in 1:1 (v/v) mixture of EpiLife and Keratinocyte Serum-Free Media (Life Technologies, Carlsbad, CA) at 37°C under 5% CO₂. For experimental manipulations, cells were quantified using trypan blue (0.25%; v/v) exclusion staining and seeded into 24-well tissue culture-treated plates at a density of 2 × 10⁵ cells/mL. L. crispatus strain JV-V01 was used for colonization assays. Bacteria were grown for 16 to 18 hours on MRS agar plates, resuspended in sterile Dulbecco PBS and used for in vitro colonization of VEC at a multiplicity of infection of 10. The impact of lubricants on L. crispatus colonization was tested using preexposure and postexposure approaches. For preexposure assays, VEC were preexposed to 10% (v/v) solution of select lubricants in cell culture media after immediate inoculation with bacteria. After adding bacterial inoculum to the cells, plates were centrifuged for 10 minutes at 900g to ensure bacterial inoculum reached the epithelial monolayers and incubated under standard conditions. For postexposure assays, VEC were precolonized with bacteria before exposure to lubricants. Vaginal epithelial cells were infected for 2 hours, washed 3 times with Dulbecco PBS to remove bacteria not attached to the cells and then treated with 10% (v/v) solution of lubricants. Dulbecco PBS or 10% (v/v) solution of glycerol was used as a negative treatment control. After a 4 or 24-hour incubation, VECs were washed 3 times with Dulbecco PBS, and trypsinized and resuspended in Dulbecco PBS. Suspensions of bacteria and VEC were vortexed for 5 minutes and serially diluted, plated on MRS agar, and incubated for 48 72 hours for bacterial quantification. All incubations were performed at 37°C under 5% CO₂.

Statistical Analysis

All experiments were performed at least in triplicate. The statistical differences were determined using an analysis of variance (ANOVA) with Dunnett or Bonferroni adjustments for multiple comparisons. P values <0.05 were considered significant.

RESULTS

Herein, we tested 3 clinical lubricants and 5 commercially available personal lubricants, which differed in osmolality and formulation. Six of 8 lubricants contained an antimicrobial, such as methylparaben, propylparaben, polyquaternium 15, or CHG (Table 1). The impact of each lubricant on the growth of the 4 most predominant vaginal Lactobacillus species (L. crispatus, L. gasseri, L. jensenii, and L. iners) was evaluated. Initially, we assessed the bacterial growth by measuring optical densities of bacterial cultures containing lubricants. The presence of Conceptrol, K-Y Jelly, or Surgilube in liquid media significantly inhibited the growth of L. crispatus, L. gasseri, L. jensenii, and L. iners after 24-hour exposure when compared with controls (P ranging from <0.05 to <0.0001; Fig. 1). Then, we quantified viable bacteria after 4- and 24-hour exposure to each lubricant using standard plating assay. In the presence of K-Y Jelly or Surgilube, concentrations of viable Lactobacillus species were significantly lower compared with cultures without lubricant (P ranging from <0.05 to <0.001; Fig. 2). However, these numbers did not significantly decline relative to the numbers of viable bacteria at the time of inoculation, suggesting a bacteriostatic effect of lubricants on the Lactobacillus growth. In contrast, Good Clean Love (GCL) Almost Naked, McKesson Lubricating Jelly, and Astroglide Liquid had minor or no effect on the viability of tested vaginal Lactobacillus species (Fig. 2). Notably, the lubricants that significantly inhibited the growth of Lactobacillus species, that is, Conceptrol, K-Y Jelly, and Surgilube, were not distinguished by the highest osmolality; however, all contained antimicrobial agents, such as parabens or CHG.

Figure 1: Select lubricants (K-Y Jelly, Surgilube, and Conceptrol) inhibit the growth of vaginal Lactobacillus species. L. crispatus strain VPI 3199, L. crispatus strain JV-V01, L. gasseri strain JV-V03, L. jensenii strain JV-V16, and L. iners strain AB107 were grown in MRS or supplemented TSB broth with 10% (v/v) lubricants at 37°C under 5% CO₂ or anaerobic conditions. Bacterial growth was assessed by measuring OD₆₀₀ at 4 and 24 hours after inoculation. The broth without any lubricant was used as a positive control. OD₆₀₀ measurements of each culture with lubricants were compared with cultures without lubricants at respective time points. Data are shown as means ± SE from at least 3 independent experiments. P values were calculated using one-way ANOVA with Dunnett posttest (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

Figure 2: Select lubricants (K-Y Jelly, Surgilube, and Conceptrol) exhibit bacteriostatic effect on vaginal Lactobacillus species. L. crispatus strain VPI 3199 (A), L. crispatus strain JV-V01 (B), L. gasseri strain JV-V03 (C), L. jensenii strain JV-V16 (D), and L. iners strain AB107 (E) were grown in MRS or supplemented TSB broth with 10% (v/v) lubricants at 37°C under 5% CO₂ or anaerobic conditions. Bacterial growth kinetics were assessed by measuring the number of CFUs representing viable bacteria in the cultures in appropriate liquid media containing 10% solutions of lubricants at 4 and 24 hours after the exposure to 10% (v/v) lubricants using standard plating assay. The broth without any lubricant was used as a positive control. Concentrations of viable bacteria in each culture were calculated as CFU/mL and compared with culture without lubricants at respective time points. Data are shown as means ± SE from 3 independent experiments. P values were calculated using one-way ANOVA with Dunnett posttest (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).

To test the antimicrobial properties of excipients listed as key components of lubricants (Table 1) against vaginal Lactobacillus species, we performed disk diffusion assays. We did not test polyquaternium 15 because this compound is not commercially available. In addition, lack of bacterial growth inhibition after exposure to Astroglide Liquid, containing polyquaternium 15, indicates that this compound does not impact the growth of Lactobacillus species (at least at the concentration present in the tested lubricant). The growth of L. crispatus, L. gasseri, and L. jensenii on agar plates had significant zones of inhibition from all excipients tested when compared with the negative control (P ranging from 0.0007 to <0.0001; Fig. 3A). For L. iners, CHG caused significant (P < 0.001) growth inhibition, whereas zones of inhibition from parabens did not reach statistical significance. Furthermore, CHG inhibited Lactobacillus species at a significantly higher magnitude (19–24 mm) compared with parabens (8–12 mm; P < 0.0001; Fig. 3A).

Figure 3: Chlorhexidine gluconate exhibit stronger antimicrobial properties than do parabens across all tested vaginal Lactobacillus species. A, The impact of each excipient on the growth of Lactobacillus species was tested using disk diffusion assay. Bacteria were grown on appropriate agar plates with 6-mm disks impregnated with 20% solutions (w/w) of methylparaben, propylparaben, and CHG. Ethanol (solvent for parabens) and bleach were used as negative and positive controls, respectively. Zones of inhibition were recorded 24 hours after inoculation and compared with a negative control. Data are shown as means ± SE from 3 independent experiments. P values were calculated using one-way ANOVA with Bonferroni posttest (**P < 0.01; ***P < 0.001; ****P < 0.0001). B, MICs of methylparaben, propylparaben and CHG were determined using the broth microdilution method. The MIC was defined as the lowest concentration of the excipient that inhibits the visible growth of the tested Lactobacillus species.

To better delineate the differences in antimicrobial potentials of parabens and CHG against vaginal Lactobacillus species, we determined the MICs for these compounds. Both parabens inhibited bacterial growth at 8000 mg/L for all tested Lactobacillus species except L. iners, which was inhibited at 2000 and 4000 mg/L of methylparaben and ethylparaben, respectively (Fig. 3B). In contrast, CHG inhibited the growth of Lactobacillus species in a species-specific manner at a concentration ranging from 1.25 to 10 mg/L, which are 200 to 6400 times (2.3–3.8 log) lower than parabens.

To assess the effect of lubricants on the colonization of the vaginal epithelium with Lactobacillus associated with optimal vaginal health, we exposed human VECs, grown as monolayers, to lubricants and infected the in vitro VEC cultures with L. crispatus. We were able to test only 3 lubricants: GCL Almost Naked, E-Z Lubricating Jelly, and McKesson Lubricating Jelly. The other lubricants used in this study induce substantial cytotoxicity, including condensation of chromatin and a loss of intercellular connections in the in vitro VEC model,⁵ which precluded testing the colonization. Glycerol was used as an additional control to mimic the viscosity of lubricants in the event that this property impacted colonization alone. We also used 2 different approaches to determine the impact of lubricants on bacterial colonization. First, we tested the effect of lubricants on VEC that were precolonized with L. crispatus before exposure lubricants. Four-hour exposure to lubricants did not significantly impact the colonization levels; however, 24-hour exposure to E-Z Lubricating Jelly or McKesson Lubricating Jelly significantly decreased L. crispatus colonization levels by 0.9 log (P < 0.05; Fig. 4A). In contrast, GCL Almost Naked did not impact the colonization at any tested time points compared with untreated or glycerol controls (Fig. 4A). Second, we tested the effect of lubricants on L. crispatus colonization levels when VECs were preexposed to lubricants after infection with L. crispatus. Preexposure to all tested lubricants for 4 or 24 hours significantly impacted the VEC colonization by L. crispatus compared with negative controls (VEC cultures without lubricants or with glycerol; P < 0.0001 and <0.001, respectively). The colonization levels were reduced by 1.7–2.3 logs at 4 hours and 1.3–1.4 logs at 24 hours when compared with untreated controls (Fig. 4B).

Figure 4: Lubricants reduce the colonization of VECs with L. crispatus particularly when VEC are colonized with bacteria after the lubricant exposure. Three noncytotoxic lubricants (GCL Almost Naked, E-Z Lubricating Jelly, McKesson Lubricating Jelly) were tested to determine their impact on colonization of in vitro VEC model with L. crispatus strain JV-V01. A, VECs were precolonized with bacteria at the multiplicity of infection (MOI) 10 for 2 hours before exposure to 10% (v/v) lubricants. B, VECs were exposed to 10% (v/v) lubricants and immediately colonized with L. crispatus for 4 or 24 hours as described previously. Colonization levels were reported as number of viable bacteria (CFU) attached to VEC per well. Data are shown as means ± SE from at least 3 independent experiments. P values were calculated using two-way ANOVA with Bonferroni posttest (**P < 0.01; ****P < 0.0001).

DISCUSSION

The vaginal microbiota play a critical role in women’s health and disease.^3,12 Particularly, Lactobacillus-dominant communities contribute to homeostasis and protect the cervicovaginal microenvironment from invading pathogens.³ In contrast, the depletion of Lactobacillus species and the overgrowth of diverse anaerobes (characteristic of bacterial vaginosis [BV]) can lead to numerous gynecologic and obstetric sequalae, including increased risk of STIs, preterm birth, spontaneous miscarriages, pelvic inflammatory disease, and gynecologic cancer.^12,13 Multiple factors have been shown to affect the vaginal microbiome, including sexual practices and use of lubricants, sex toys, and feminine hygiene products.¹⁴ Indeed, lubricants, which frequently contain antimicrobial preservatives, may directly impact bacterial communities in the cervicovaginal microenvironment.

The available epidemiological data suggest that the use of lubricants is associated with increased risk of vaginal dysbiosis or BV.^15–17 However, mechanistic in vitro studies investigating vaginal lubricants or feminine hygiene products are still very limited. Previously, we demonstrated the detrimental effect of hyperosmolar lubricants on the vaginal epithelium using in vitro models.⁵ Herein, we sought to determine the impact of personal and clinical lubricants, which varied in osmolality and formulations, on health-associated vaginal Lactobacillus species. We examined lubricants routinely used in clinics for vaginal ultrasounds and pelvic examinations (E-Z Lubricating Jelly, McKesson Lubricating Jelly, K-Y Jelly, Surgilube), as well as OTC personal lubricants used for sexual practices, alleviating vaginal dryness (Astroglide Liquid, GCL Almost Naked, K-Y Jelly, K-Y Warming Jelly) or prevention of pregnancy (Conceptrol; Table 1).

In this study, we demonstrated that certain lubricants, such as K-Y Jelly, and Conceptrol, exhibited antimicrobial properties against vaginal Lactobacillus species in vitro, despite a lack of association with high osmolality, which relates to the concentration of glycols and glycerin (Table 1). These antimicrobial properties could potentially be due to their excipients, that is, CHG, parabens, polyquaternium-15, or a known spermicide, N-9. Intriguingly, lubricants containing CHG (K-Y Jelly, Surgilube) or N-9 (Conceptrol) inhibited the most bacterial growth, whereas lubricants containing parabens or polyquartenium-15 (Astroglide Liquid, E-Z Lubricating Jelly) did not have this effect (Figs. 1, 2) in our in vitro systems. These findings strongly suggest that CHG or N-9 in these lubricants is responsible for the bacterial growth inhibition. The disk diffusion and MIC assays performed in this study confirmed that CHG exhibits stronger antimicrobial properties against vaginal Lactobacillus species compared with parabens (Fig. 3).

The CHG is a broad-spectrum microbicide that can be found in a variety of products, including OTC antiseptic mouthwashes, creams, wipes, toothpastes, deodorants, sunscreens, eye drops, hair conditioners, and more.⁹ In the context of the oral cavity, previous clinical studies demonstrated that use of mouthwash containing CHG has been linked to major shifts in the microbiota composition.¹⁸ Studies on the impact of CHG on skin microbiota have shown conflicting findings. On one hand, a minimal effect of CHG on the skin microbiota has been demonstrated, suggesting high stability and resilience of bacterial communities,¹⁹ whereas other studies showed decreased bacterial density on CHG-treated skin.²⁰

Data presented in this in vitro study suggest that intravaginal use of products containing CHG may also have a detrimental effect on the vaginal microbiota by decreasing the overall bacterial load, including health-associated Lactobacillus species. This might allow BV-associated bacteria to colonize the vagina, leading to vaginal dysbiosis or BV. Our findings are in accordance with a previous report showing that CHG inhibits the growth of not only genital pathogens (such as Neisseria gonorrhoeae or Trichomonas vaginalis) but also vaginal Lactobacillus species.²¹ A 2010 study also demonstrated that use of CHG-based Surgilube during pelvic examination decreased the detection of group B Streptococcus, a common vaginal opportunistic pathogen.²² Intriguingly, despite the links to anaphylaxis and recent FDA ban of CHG in health care antiseptics and classification as “not generally recognized as safe and effective for use,”¹⁰ the American College of Obstetricians and Gynecologists approved the off-label use of CHG for surgical vaginal preparation, as there is no other FDA-approved alternative to povidone-iodine.²³ This approval comes from reports demonstrating that CHG is more effective than povidone-iodine at killing vaginal bacteria,²⁴ and thus far, CHG has not been linked to signs of vaginal irritation.²⁵ The discordance in recommendations from different bodies highlights the urgent need for better safety screening of female hygiene products, for example, using in vitro 3-dimensional biomimetic models and larger epidemiologic studies and trials.⁵

The other antimicrobial excipient, N-9, is a surfactant spermicide and an active ingredient in Conceptrol. It has been shown in clinical and in vitro studies to cause genital inflammation and barrier breach,²⁶ and consequently has been associated with increased HIV-1 transmission in high-risk women.²⁷ Furthermore, the adverse effect of N-9 on vaginal Lactobacillus species was well characterized by several studies in the 1990s.^28,29 Adverse findings on N-9 were also validated in a 2012 study using 16S rRNA sequencing, which demonstrated shifts in composition from Lactobacillus-dominant communities to communities dominated by anaerobes associated with BV, as well as Streptococcus, Enterococcus, and Escherichia, after twice-daily vaginal application of 4% N-9.³⁰ In accordance to previous reports, Conceptrol, containing 4% N-9, also reduced the growth of vaginal Lactobacillus species in our study.

Our in vitro evidence suggests that lubricants might impact the attachment of Lactobacillus to the vaginal epithelium (Fig. 4). A limited number of lubricants were tested for bacterial colonization because most products induce substantial damage to human epithelial cells due to high osmolality.⁵ The viscosity of the tested lubricants could have been a limitation in experiments. However, the viscosity of the lubricants does not explain the effect of tested lubricants on colonization of VEC with bacteria. This was confirmed by the use of glycerol as a viscosity control, which did not elicit any substantial decrease in Lactobacillus colonization of epithelial cells. Therefore, the reason for the reduction of L. crispatus colonization on VEC was potentially due to specific ingredients or other physical properties of lubricants than the viscosity.

Overall, this in vitro study suggests that personal and clinical lubricants containing N-9 or CHG might exhibit adverse effects on the growth of vaginal microbiota species and highlights the need for consumers and clinicians to use these lubricants with caution. Future clinical studies, particularly with longitudinal study designs, are needed to show whether the use of these vaginally applied products have a long-lasting effect on the microbiota in vivo and to confirm the clinical relevance of our in vitro findings. Notably, the World Health Organization warned consumers to avoid certain ingredients that are commonly found in OTC lubricants (e.g., N-9 and polyquaternium), as these may increase the risk of HIV infection.^6,8 However, currently, they have made no comment of CHG regarding consumer products. Our data suggest that feminine hygiene products containing CHG could impact vaginal Lactobacillus growth and potentially recolonization of the vagina. Ultimately, additional clinical studies and mechanistic in vitro studies are required to investigate the impact of vaginal lubricants and other feminine products, including moisturizers, washes, wipes, creams, sprays, powders, douches, and probiotics/pharmaceutical vehicles and other intravaginal practices, on vaginal Lactobacillus species and cervicovaginal epithelium.