Vaginal washing behavior and fecundability among Kenyan women in a prospective preconception cohort

Abstract

Background:

Intravaginal washing, practiced by a significant proportion of women globally, is associated with acquisition of human immunodeficiency virus (HIV), sexually transmitted infections, and bacterial vaginosis (BV). A single prior study among women in the United States found that vaginal washing was associated with lower fecundability.

Objective:

To examine the association between vaginal washing and fecundability among Kenyan women.

Methods:

HIV-negative Kenyan women who were trying to conceive and reported no history of infertility care-seeking were followed prospectively for incident pregnancy for up to six months. At monthly visits, participants reported the first day of last menstrual period, sexual behavior, vaginal washing behavior, underwent pregnancy testing, and provided vaginal swabs for detection of BV by Gram stain (Nugent score ≥7). Discrete time proportional probabilities models were used to estimate fecundability ratios (FR) and 95% confidence intervals (CI) comparing menstrual cycles when women reported vaginal washing to menstrual cycles when no vaginal washing was reported.

Results:

Four hundred fifty-eight women contributed 1,376 menstrual cycles and 255 pregnancies. At enrollment, a third (35.2%, 161 of 458) of participants reported vaginal washing with the majority using water only (73.9%, 119 of 161). After adjustment for age, frequency of unprotected intercourse, and study site, vaginal washing in the prior four weeks was associated with a 29% lower fecundability (adjusted FR (aFR) 0.71, 95% CI 0.53, 0.94), which did not change after further adjustment for BV at the visit prior to each pregnancy test (aFR 0.71, 95% CI 0.54, 0.95).

Conclusions:

Periconceptual vaginal washing may reduce fecundability. Potential mechanisms include vaginal washing-associated changes in the vaginal microbiota, inflammation, disruption of cervical mucus, and effects on sperm function. Vaginal washing has no known health benefits and cessation may improve women’s likelihood of conceiving.

SOCIAL MEDIA QUOTE

Vaginal washing was associated with a lower per-menstrual cycle probability of pregnancy among Kenyan women planning pregnancies. Vaginal washing has no known health benefits, and cessation may improve women’s likelihood of conceiving.

BACKGROUND

Intravaginal washing practices include cleansing with water or products such as vinegar, soaps, or douching agents.¹ Many women globally engage in vaginal washing to improve hygiene, prevent pregnancy, prevent or treat vaginal infections, or promote sexual pleasure.^1–7 Vaginal washing is associated with increased risk of bacterial vaginosis (BV), sexually transmitted infections (STI), human immunodeficiency virus (HIV), and pelvic inflammatory disease (PID).^8–10 Only one prior study has assessed the association between vaginal douching and fecundability, finding lower fecundability among those engaging in vaginal douching.¹¹ Our objective was to prospectively assess the association between vaginal washing and fecundability in Kenyan women planning pregnancy.

METHODS

Cohort selection and study procedures

Women in the Microbiota and Preterm Birth study, which enrolled women in Nairobi and Mombasa, Kenya between April 18, 2017 and March 18, 2020, were eligible for this fecundability analysis.¹² The parent study included HIV-seronegative women who were ≤45 years old, were non-contracepting (other than condoms for STI prevention), reported a menstrual period in the prior three months (or recently discontinued hormonal contraception), and were planning to become pregnant in the next six months. Women were excluded if they were not at risk for pregnancy including having a depot medroxyprogesterone acetate (DMPA) injection within three months, were at increased risk for preterm birth (ex: autoimmune diseases), used antibiotics in the last four weeks (common exclusion criterion for vaginal microbiota studies), or previously sought care for infertility. For this fecundability analysis, we excluded participants who did not contribute ≥1 menstrual cycles, those with history of PID treated in hospital, ectopic pregnancy, polycystic ovary syndrome, or endometriosis, and those trying to conceive for >3 menstrual cycles prior to enrollment (eFigure 1).^13,14

At enrollment, participants completed an interview including socio-behavioral characteristics and medical history including first day of last menstrual period (LMP) and how long they had been non-contracepting and attempting pregnancy, and underwent a pelvic examination with vaginal swab collection.¹² At monthly preconception visits, participants underwent urine pregnancy testing, self-collected vaginal swabs, and completed an interview reporting sexual and vaginal washing behavior and first day of LMP. Participants were asked “Do you use anything to clean inside your vagina?” (no/yes). If “yes”, they were asked “In the four weeks prior to today, which of the following did you use to clean inside of your vagina?” including water only, soap and water, antiseptic, detergent, and other (specify). Most participants were eligible for up to six monthly preconception visits; those discontinuing DMPA <6 months before enrollment were eligible for up to 9 months.¹⁵

Participants with genital symptoms were assessed and treated per Kenyan syndromic management guidelines.¹⁶ Enrollment vaginal samples were tested for Neisseria gonorrhoeae, Chlamydia trachomatis, and Trichomonas vaginalis (Aptima Combo-2 CT/NG Detection System, Aptima Trichomonas vaginalis assay; Hologic Incorporated). Directed treatment was provided for STIs detected at enrollment. Vaginal samples from all visits were assessed for BV using Nugent criteria.¹⁷

Exposure

Two time-varying measures of vaginal washing were considered: 1) any vaginal washing in the last four weeks (no/yes), and 2) type of vaginal washing in the last four weeks (none, water only, soap and water). Two cycles were characterized by antiseptic use and were included with soap and water. Three cycles with an “other” type of vaginal washing were included with either water only or soap and water based on review of the substances specified.

Outcome

For each participant, number of menstrual cycles to first incident pregnancy was calculated using reported first days of LMPs and monthly urine pregnancy test results.¹³ In July 2018, a question was added to the monthly interview for report of additional first days of LMPs (“interim menstrual cycles”) for women who missed preconception visits (n=82 cycles reported). One-hundred thirty participants had visits prior to July 2018 and in these cases “interim menstrual cycles” were derived (n=41 cycles), as previously described.¹³ Participants were censored during follow-up for biomedical infertility treatment (n=3), not conceiving during follow-up (n=73), withdrawal or loss to follow-up (n=49), and at the onset of the COVID-19 pandemic (n=35).

Statistical Analysis

We used discrete time proportional probabilities models to generate fecundability ratios (FR) and 95% confidence intervals (95% CI) estimating the association between vaginal washing and fecundability.^14,18 For participants with prior conception attempt time, we utilized delayed entry to enter participants into the analysis at their current menstrual cycle of trying to reduce left truncation bias. Based on existing literature, knowledge of the study population, and consideration of causal relationships, maternal age (years: <25, 25–29, 30–34, 35–39, 40–45), frequency of unprotected intercourse in the prior four weeks (none, 1–4, 5–8, ≥9), and study site were included a priori in multivariable models.^1,19 In a second adjusted model, we included BV (Nugent score ≥7) at the visit prior, as women may have engaged in vaginal washing because of BV.¹³

Missing data:

For 123 derived or reported “interim menstrual cycles” (8.9% of cycles) and for rare data missingness for attended visits (0.03% of cycles), missing vaginal washing and unprotected intercourse data were imputed using data from the last visit carried forward.

Sensitivity analyses:

First, we excluded those with potential sub-fecundity (N. gonorrhoeae, C. trachomatis, T. vaginalis or PID at enrollment, any history of PID diagnosis or treatment for N. gonorrhoeae, C. trachomatis, T. vaginalis, or syphilis; self-report of fibroids or unknown uterine abnormality; DMPA use within six months of enrollment, HIV-seropositive partner). Second, we excluded derived and reported menstrual cycles. Third, we included women reporting ≤6 menstrual cycles of pre-enrollment conception attempt time.

Ethics Approval

The Kenyatta National Hospital-University of Nairobi and University of Washington ethics committees approved this study. Participants provided written informed consent.

RESULTS

The 458 participants were a median of 29 years old (IQR 25–34) and most reported no prior conception attempt time (80.1%, n=367) (Table 1). There were 255 pregnancies across 1,376 menstrual cycles. Twenty-six percent (359 of 1376) of cycles were exposed to vaginal washing (Table 2), with water accounting for 20.9% (287 of 1376) of cycles and soap and water 5.2% (42 of 1376) of cycles. After adjustment for age, frequency of unprotected intercourse, and study site, vaginal washing in the prior four weeks was associated with a 29% lower fecundability (adjusted FR (aFR) 0.71, 95% CI 0.53,0.94), which was similar after further adjustment for BV (aFR 0.71, 95% CI 0.54, 0.95) (Table 2). Compared to no vaginal washing, vaginal washing with water only was associated with a 17% lower fecundability (aFR 0.83, 95% CI 0.62, 1.10) and vaginal washing with soap and water was associated with a 78% lower fecundability (aFR 0.22, 95% CI 0.07, 0.71). Results were similar for sensitivity analyses excluding women with potential sub-fecundity, excluding derived and reported cycles, and including participants with up to 6 cycles of pre-enrollment conception attempt time (eTable 1).

Table 1.

Enrollment characteristics for 458 Kenyan women trying to conceive

Characteristic	Total N=458
Demographics
Age
<25	87 (19.0)
25–29	147 (32.1)
30–34	121 (26.4)
35–39	84 (18.3)
40–45	19 (4.2)
Education Level
<8 years	29 (6.3)
8–11 years	137 (29.9)
12–15 years	207 (45.2)
≥16 years	85 (18.6)
Monthly Household Income (KSh)a
<2,500	13 (2.9)
2,500–10,000	136 (29.9)
10,000–30,000	197 (43.3)
30,000–75,000	70 (15.4)
>75,000	39 (8.6)
Married or living with partner	439 (95.9)
Partner’s HIV-serostatusb
HIV-seronegative	345 (75.7)
HIV-seropositive	22 (4.8)
Unknown	89 (19.5)
Reproductive History
Most recent contraceptive methodc
None	102 (22.3)
Condoms	24 (5.2)
OCP	10 (2.2)
DMPA Injectable	18 (3.9)
Copper IUD	127 (27.7)
Implant	175 (38.2)
Other	2 (0.4)
Ever pregnant	430 (93.9)
Number of menstrual cycles of prior conception attempt timed
0	367 (80.1)
1	59 (13.0)
2	19 (4.2)
3	13 (2.8)
Abnormal uterus (fibroids or other/unknown pathology)e	8 (1.3)
Sexual behavior and vaginal washing in last 4 weeks
Any vaginal washing	161 (35.2)
Type of vaginal washing
None	297 (64.9)
Water only	119 (26.0)
Soap + Water	42 (9.2)
Antiseptics	0 (0.0)
Frequency of unprotected intercoursef
No unprotected intercourse	43 (9.4)
1–4	152 (33.3)
5–8	115 (25.2)
≥9	147 (32.2)
STIg and BV
History of PID with outpatient treatment	1 (0.2)
History of STIh	6 (1.3)
N. gonorrhoeae	3 (0.7)
C. trachomatis	34 (7.5)
T. vaginalis	4 (0.9)
BV (Nugent ≥7)	164 (35.8)

Abbreviations: BV–bacterial vaginosis, DMPA–depo medroxyprogesterone acetate, IUD–intrauterine device, KSh–Kenyan shillings, OCP–oral contraceptive pills, PID–pelvic inflammatory disease, STI–sexually transmitted infection

^aN=455

^bN=456

^cParticipants reporting OCP, contraceptive implant, or copper IUD discontinuation >2 months prior to enrollment or a last DMPA injection >6 months prior to enrollment were classified as ‘none’ for this analysis.

^dWomen reporting >3 cycles of conception attempt time prior to enrollment were excluded for this analysis.

^eSelf-reported

^fN=457

^gSTI results were missing for N=4 for N. gonorrhoeae, N=4 for C. trachomatis, and N=5 for T. vaginalis

^hSelf-reported history of syphilis, chlamydia, gonorrhea, and/or trichomoniasis

Table 2.

Unadjusted and adjusted association between vaginal washing and fecundability among 458 Kenyan women trying to conceive

Exposure	Menstrual Cycles Exposed (N=1376)	Pregnancies Exposed (N=255)	Unadjusted FR	Model 1: Adjusted FRa	Model 2: Adjusted FRb
	n(%)	n(%)	(95% CI)	(95% CI)	(95% CI)
Primary Analysis
Any vaginal washing in last 4 weeks
No	1017 (73.9)	204 (80.0)	1.00 (reference)	1.00 (reference)	1.00 (reference)
Yes	359 (26.1)	51 (20.0)	0.72 (0.54, 0.96)	0.71 (0.53, 0.94)	0.71 (0.54, 0.95)
Secondary Analysis
Type of vaginal washing in last 4 weeks
None	1017 (73.9)	204 (80.0)	1.00 (reference)	1.00 (reference)	1.00 (reference)
Water only	287 (20.9)	48 (18.8)	0.85 (0.63, 1.13)	0.83 (0.62, 1.10)	0.83 (0.62, 1.11)
Soap and waterc	72 (5.2)	3(1.2)	0.21 (0.07, 0.67)	0.22 (0.07, 0.71)	0.23 (0.07, 0.71)

^aMaternal age, study site, and frequency of unprotected intercourse were included in Model #1.

^bMaternal age, study site, frequency of unprotected intercourse, and BV at the visit prior (time-varying) were included in Model #2.

^cIncludes n=2 cycles with report of antiseptic use.

COMMENT

Principal Findings

Among Kenyan women, recent vaginal washing was associated with a 29% lower per-menstrual cycle probability of pregnancy. Compared to those reporting no vaginal washing, vaginal washing with soap and water was associated with a 78% lower fecundability.

Strengths of the study

Strengths include the prospective design, monthly ascertainment of vaginal washing, and inclusion of the first menstrual cycle at-risk for pregnancy for most participants.

Limitations of the data

First, data on frequency, timing in relation to intercourse, and reasons for vaginal washing were not collected, so we were unable to assess these characteristics. Second, sexual frequency was not collected for a biologically-confirmed fertile window, which may contribute to residual confounding. Lastly, we were unable to assess mediation of the association between vaginal washing and fecundability by the vaginal microbiota.

Interpretation

Our results are similar to the only other study of vaginal washing and fecundability we are aware of.¹¹ In a retrospective cohort study of vaginal douching among 840 women in King County, WA, vaginal douching >2 times/year prior to pregnancy was associated with a 31% lower fecundability compared to never/rare vaginal douching.

Vaginal washing may reduce the presence or concentration of optimal Lactobacillus species and promote ascension of BV-associated bacteria to the upper reproductive tract causing sub-clinical or clinical PID.^7,8,20–22 There could also be an inflammatory response to vaginal washing independent of microbiota, as well as effects on cervical mucus.

Whether associations between vaginal washing and adverse reproductive outcomes are causal versus due to a BV episode that precipitated the vaginal washing is debated.^9,23,24 We previously reported associations between recent and persistent BV and lower fecundability in this cohort.¹³ Our current analysis assessing vaginal washing had similar results in adjusted models with or without BV. Due to the timing of sample collection visits and monthly assessment of vaginal washing behavior, we are unable to employ marginal structural models to assess mediation of the association between vaginal washing and lower fecundability by BV. A study addressing this question would need to measure both vaginal washing and BV at more than one point during each menstrual cycle.

CONCLUSIONS

Among Kenyan women attempting pregnancy, there was an association between recent vaginal washing and lower fecundability that was strongest among those using soap and water. Vaginal washing cessation may improve women’s likelihood of conceiving.

SYNOPSIS.

Study question:

Is vaginal washing associated with fecundability among Kenyan women trying to conceive?

What’s already known:

Intravaginal washing, practiced by a significant proportion of women globally, is associated with acquisition of HIV, sexually transmitted infections, and bacterial vaginosis.

What this study adds:

Any vaginal washing in the four weeks before pregnancy testing was associated with a 29% lower per-menstrual cycle probability of pregnancy. Compared to those reporting no vaginal washing, vaginal washing with soap and water was associated with a 78% lower fecundability. Vaginal washing has no known health benefits and cessation may improve women’s likelihood of conceiving.