Anal intercourse (AI) is a common sexual behavior among men who have sex with men (MSM) and is also practiced by many women.1,2 Recent reviews have reiterated that unprotected receptive anal intercourse (RAI) remains the highest risk method of HIV transmission.3–5 Many men and women report use of lubricant products during AI: 59% of respondents reported always using lubricant in a large internet survey of 6124 men and women reporting AI6 and 89% of MSM in a San Francisco survey reported always using lubricant.7 Although there are few international studies of lubricant use, in Peru, 48% of MSM reporting RAI in the past 3 months reported lubricant use at last sex,8 suggesting use is also high throughout the world among MSM. However, use may be low among heterosexuals, as men in heterosexual couples in a Zambian study reported never using a lubricant product for vaginal sex.9 Many individuals also report use of saliva as a lubricant during AI10 in addition to or instead of commercial products or oils or lotions. Lubricants are used to reduce friction during AI, an effect that not only increases sexual pleasure but also facilitates penile penetration.
Concerns about the effects of lubricants on the epithelium are not new. The COL-1492 trial provided evidence that vaginal application of nonoxynol-9 (N-9) use was associated with increased risk of HIV infection, and that rectal administration of N-9 was associated with sloughing of rectal epithelium.11–13 Although increased rectal transmission of HIV secondary to N-9 use has never been demonstrated, these findings raised concerns about the potential for other rectal products used during RAI to facilitate HIV transmission. Studies using explant biopsy or surgical samples from humans and animals14 showed some commercial lubricants increased the infection (using laboratory strains of HIV-1) of those tissues when infected in the laboratory and have toxic effects on rectal epithelium.15–17 In an important clinical study, gel products similar to those that are commercially available caused short-term denudation of rectal epithelium,18 thought to be induced by the lubricant's osmotic effect on the rectal mucosa. Such injury of the rectal epithelia has been hypothesized to enhance the probability of transmission of pathogens such as HIV and merits additional study.
There is a dearth of data on the frequency as well as types (aqueous, oil, silicone-based, or numbing) or specific brands of lubricants used in populations practicing AI. Safety data are limited because lubricant products are classified in the United States as “medical devices” and in Canada as “cosmetics,” thereby avoiding the safety testing that accompanies drug licensure. There is no existing empirical examination of an effect of rectal lubricant use on rectal health in the context of AI, nor on the probability of rectal infection by a sexually transmitted infection (STI). Given the widespread use of these products in the community and the current focus on rectal-specific formulations of potential new methods of HIV prevention such as microbicides (e.g., gels that may be similar to existing rectal lubricants), the examination of the association between lubricant use and STI is essential.
Between October 2006 and June 2009, a rectal health and behaviors study designed to compare the effect of RAI and rectal behaviors on rectal health by gender was conducted as part of the University of California, Los Angeles (UCLA) Microbicide Development Program in 2 community sites in Los Angeles: the AIDS Research Alliance and UCLA CARE Clinic and in Baltimore at The Johns Hopkins University (JHU). Eligible, interested individuals recruited from newspaper, internet, and clinic posted advertisements, and research registries were given further details on the study and provided written informed consent in a private room. The general eligibility criterion was men and women who were at least 18 years of age; willing to be tested for STIs including HIV; willing to undergo an anal exam; and mentally competent to understand study procedures and give informed consent. Criteria by RAI status was defined as no RAI in the past year for the non-RAI men and women. For the practicing RAI group, it was reported RAI in the past 30 days for men and reported RAI in the past 12 months for women. Men and women were excluded if they were <18 years of age; unwilling to be tested for STIs (and HIV); unwilling to undergo an anal exam; unwilling to complete study questionnaire; not mentally competent to understand study procedures and give informed consent; or if they were a male who had no RAI in the past month but did have RAI in the past year.
The study procedures were reviewed and approved by institutional review boards at UCLA, AIDS Research Alliance, and The Johns Hopkins University. All procedures were also reviewed by the Division of AIDS at National Institutes of Health. Following written informed consent, participants completed computer-administered self interviews about rectal sexual and hygiene behavior and anorectal symptoms, underwent perianal and anorectal examinations including high resolution anoscopy to detect anal and distal rectal clinical signs, and were tested for STIs. Rectal swabs were collected and tested for Neisseria gonorrhoeae (GC) and Chlamydia trachomatis (CT) with the Aptima Combo 2 assay. Other specimens were collected including blood for syphilis (tested for RPR and TPHA with titer). The study population recruited was intentionally half individuals who were HIV positive with HIV-1 status confirmed in clinic by rapid tests and confirmed by Western Blot. The study population also intentionally was evenly divided by gender and by RAI status (n = 431 reporting recent RAI).
Lubricant use was assessed as reported frequency (always, sometimes, never) in the past month during RAI and 620 of 896 male and female participants (69%) provided a response to this question (an additional 244 replied “does not apply” and 15 skipped the question); 240 of 620 (39%) were excluded because they responded they never used lubricant for RAI in the past month and reported no RAI in the past month. The final analysis was conducted on the 380 who reported having RAI in the reference period, provided a response to the lubricant use questions, and for whom there were STI test results. Final analyses used a dichotomous variable for lubricant use: those who reported always using lubricant in the past month were coded as “consistent users” versus those who either sometimes or never used lubricant as “inconsistent users.” “Commercial lubricants” were specified as those that participants “can buy in a store or online such as KY-jelly.” The definition also specified that using saliva or the lubricant that comes with condoms were not considered “commercial lubricant use.” Participants were then asked to report which lubricants they had used in the past month with choices of “silicone-based (like Eros brand); water-based (like KY and Wet); oil-based (like Crisco); numbing (lubricant that reduces feeling in your butt, vagina, or penis).” Participants who reported RAI in the past month were also asked whether the last time they had RAI they used a commercial lubricant, oil, spit, lotion, a desensitizing lubricant, nothing, or other. Condom use was reported for last 2 RAI events. STI (n = 20) was defined as a positive result on a rectal compartment-specific test for a GC (n = 6) or CT infection (n = 13) or syphilis (positive RPR and TPHA with titer 1:8 or greater and no history of previous diagnosis of syphilis) (n = 4).
Bivariate associations between STI and lubricant use in the past month, demographics, HIV status, number of acts of AI in the past month, numbers of rectal sex partners in the past month and other behaviors were analyzed using univariate logistic regression, χ2 tests, Fisher exact tests and t tests. Logistic regression was used for univariate and multivariable analyses. All analyses were performed in Stata version 8.0 (StataCorp, College Station, TX).
As described above, the sample was intentionally half male, from each city, and HIV positive (Table 1). Distribution of demographic characteristics shows an ethnically diverse sample with half black (53%), a quarter white (24%), 17% Hispanic/Latino, and 6% other race/ethnicity. About one-quarter reported being homeless in the past year, and 22% were currently unemployed. Half the participants reported a main partner in the past month and <5% reported a one-time, unknown, or trade partner in the past month. Test results for STI were available for 380 of those reporting RAI lubricant use frequency (227 males and 153 females). Among those reporting RAI in the past month, males reported a slightly higher frequency of acts of RAI in the past month, but overall these were not significantly different (mean 5.38, median 2 vs. females reported mean of 4.20, median 2.0, P = 0.20).
Lubricant Use With RAI
Thirty-six percent (137/380) of participants reported consistent lubricant use when they engaged in RAI in the past month and 64% (224/351) reported use of “commercial lubricant” at last RAI. There were significant differences by gender; fewer women reported consistent use of lubricant in the past month compared with men (24% of women vs. 45% of males [P < 0.000]). There were differences by race/ethnicity (χ = 9.6, df = 3, P = 0.02); 47% of whites, 32% of blacks, 40% of Hispanics, and 19% of “Other” reported consistent use of lubricants (Table 1). There was no difference in reported frequency of lubricant use among participants who were HIV-positive compared with those who were HIV-negative, or related to study location (Baltimore vs. Los Angeles). By partner type consistent use of lubricants in the past month was most often reported by those with acquaintance partners and one-time partners and less by those with main and one-time partners, but these differences were not significant (Table 1). Significantly, fewer of those who were homeless and unemployed reported consistent use than inconsistent use of lubricants. The number of RAI acts in the past month as well as numbers of partners in the past month did not differ among those reporting consistent use of lubricants and those using inconsistently. Those who reported consistent use of commercial lubricant during RAI in the past month were significantly older than those who did not (mean age of 40 vs. 38, t test P = 0.009).
Rectal STI Prevalence
Overall, the prevalence of STI was 5.3% (n = 20/380) of those reporting on rectal lubricant use in the past month (6.2% of males and 3.9% of females, difference not significant). Those positive for STI were significantly younger (35.4 years vs. 39.7, P = 0.05); this was true for males (36.7 vs. 41.2; P = 0.05) but borderline for females (32.3 vs. 37.7; P = 0.09). There was no significant difference by race/ethnicity, HIV status, study city, number of RAI acts in past month, having a main partner in the past month, or number of RAI partners in the past month (Table 2). A third of those with STI also tested positive for GC/CT in their urine-based NAAT test (3/17); 30% of those with STI had urethral GC/CT versus 3.7% that did not (P < 0.007); however, significantly more women were infected in both compartments (44% of those with STI also had cervical infection), whereas few men with rectal STI also had urethral infection (6.7%); none of the men reporting about rectal lubricant use in the past month were infected in both compartments.
Association of Rectal STI and Reported Lubricant Use
There were significantly more STIs detected among those who reported consistent use of lubricant for RAI in the past month than in those reporting inconsistent use (4.1% of those never reporting lubricant, 2.4% of those sometimes using lubricant, and 9.5% of those always using lubricant (P = 0.019 Fisher exact test). Consistent lubricant users in the past month also had a higher prevalence of STI than inconsistent users (9.5% vs. 2.9%; P = 0.006) (Figure 1).
Most participants who reported lubricant use in the past month reported using just 1 lubricant type at last RAI (64% [239/374]); however, 16.6% reported using at least 2 types of lubricants (Table 3). The number of different types of lubricant used in the past month was significantly associated with prevalent rectal GC or CT infection. Those with STI reported greater numbers of lubricants used (mean of 1.23) than those who did not have rectal GC or CT (mean of 0.97 types of lubricants used in the past month; t test P = 0.04). Males reported using significantly more lubricants than females in the past month (mean, 1.11 vs. 0.75, respectively; P = 0.000).
Of the participants using lubricants, most reported using a water-based lubricant (61%), whereas 20% used silicon-based products, 15% oil-based lubricants, and 7% said they had used numbing lubricants (Table 3). Efforts to correlate a specific type of reported lubricant use with STI were limited because of small sample sizes. Nevertheless, detected STI was higher in those lubricant users who reported exclusively using water-based lubricants in the past month compared with those used other types of lubricants (6.1% vs. 2.7%; P = 0.05) and those exclusively using silicone lubricants versus those using other lubricant products (9.2% vs. 3.3%, P = 0.02). There was no difference in those reporting “exclusively” using oil-based lubricants and those who used other lubricants. There were not enough cases of STI by type of lubricant used to assess these differences in multivariate analyses.
There was a difference between those reporting lubricant use at last RAI by condom use at last RAI: fewer condom users reported not using lubricant than noncondom users (23.8% vs. 43.9% respectively; P = 0.000). Among those using condoms at last RAI, more reported always lubricant use in the past month. There was no significant difference in STIs between those reporting condom use (4.6% of those reporting condom at last RAI had STI vs. 5.1% of those reporting not using a condom at last RAI).
In a multivariable logistic regression model, testing positive for STI was associated with consistent use of lubricant during RAI in the past month (adjusted odds ratio: 2.98; 95% confidence interval: 1.09–8.15) after controlling for age, gender, study location, HIV status, and numbers of RAI partners in the past month. A second model controlled for number of RAI acts in the past month with similar findings; a significant association with lubricant use (adjusted odds ratio: 3.41, 95% confidence interval: 1.22–9.51) (Table 4). When condom use at last RAI was included in the model, the findings remained consistent; however, it was not included in the final multivariable models because the time frame was different (not in the last month). Models with and without condom use are presented for reference in Table 4.
The use of rectal lubricants for RAI has not been previously assessed in a large observational study as a risk factor for rectal STIs. We report the first epidemiologic study to find an association between prevalent STIs and reported use of lubricants for recent RAI; nevertheless, this paper reports an association, not causation. While it is not possible to determine the exact act, the exact behaviors practiced, nor the exact route by which an STI was acquired, we used conservative definitions of STIs by restricting our outcome to nonviral STIs most likely acquired through rectal exposure. Our findings also suggest an association between the use of more types of commercial lubricants and prevalent STI among men and women in these 2 US cities.
Our findings are limited by a lack of a definitive temporal relationship between the reports of lubricant use and the timing of STI acquisition. Only a randomized controlled clinical trial or an observational longitudinal study could better determine such relationships. Although clinical trials conducted for rectal microbicide development may shed more light on this, their subjects are randomized to use either an active microbicide product or a gel placebo that is known to be minimally harmful to the epithelium (e.g., the universal HEC placebo) and participants are counseled to not use other rectal lubricant products. A longitudinal study would be better able to determine STIs that were incident; however, reports would remain based on recall given that assessments of STI would likely be at monthly, 6 monthly, or yearly intervals. In this study, detection of prevalent infections (during clinic visit) ensures that these bacterial infections were likely relatively recently acquired. It should be noted that most of the women with rectal GC or CT also had a positive urine test for GC or CT and it possible that the rectal specimens were contaminated. However, because all women reporting rectal infection also reported having vaginal intercourse, they were likely penetrated by the same infected person in more than one site in the same sexual event or during the same period of time.
Because about 17% of study participants reported using more than one type of lubricant in the past month, we could only assess lubricant type by STI among those who reported using just 1 type of lubricant. Our sample size was too small to allow analysis by those who used different combinations of lubricant types (i.e., used both water-based and silicone-based in the past month). Future studies may have to resort to other epidemiologic designs, more detailed behavioral data, and more specific measurement of lubricants types to clarify whether specific lubricant types or brands increase risk of STIs more than others.
AI has been clearly demonstrated as a behavior widely practiced by men and women and is an important factor in facilitating the HIV and STI epidemics. Clearly, there is a need for rectal microbicide prevention products. Modelers have demonstrated the potential for such interventions for HIV prevention among high-risk groups19 and among those who do not use condoms.20 The study findings reported here, while specifically defining associations, not causation, contribute to this science by identifying additional factors that may facilitate transmission of STIs, and provide information important to the promotion of better rectal safety and rectal health.
1.Chandra A, Mosher WD, Copen C, et al. Sexual behavior, sexual attraction, and sexual identity in the United States: Data from the 2006–2008 National Survey of Family Growth. National Health Statistics Reports 2011; 36.
2.Halperin D. Heterosexual anal intercourse: Prevalence, cultural factors, and HIV infection and other health risks, Part I. AIDS Patient Care STDS 1999; 13:717–730.
3.Fox J, Fidler S. Sexual transmission of HIV-1. Antiviral Res 2010; 85:276–285.
4.Dosekun O, Fox J. An overview of the relative risks of different sexual behaviours on HIV transmission. Curr Opin HIV AIDS 2010; 5:291–297.
5.Varghese B, Maher JE, Peterman TA, et al. Reducing the risk of sexual HIV transmission: quantifying the per-act risk for HIV on the basis of choice of partner, sex act, and condom use. Sex Transm Dis 2002; 29:38–43.
6.Javanbakht M, Murphy R, Gorbach P, et al. Preference and practices relating to lubricant use during anal intercourse: implications for rectal microbicides. Sex Health 2010; 7:193–198.
7.Carballo-Diéguez A, O'Sullivan LF, Lin P, et al. Awareness and attitudes regarding microbicides and Nonoxynol-9 use in a probability sample of gay men. AIDS Behav 2007; 11:271–276.
8.Kinsler JJ, Galea JT, Peinado J, et al. Lubricant use among men who have sex with men reporting receptive anal intercourse in Peru: implications for rectal microbicides as an HIV prevention strategy. Int J STD AIDS 2010; 21:567–572.
9.Jones DL, Weiss SM, Chitalu N, et al. Acceptability and use of sexual barrier products and lubricants among HIV-seropositive Zambian men. AIDS Patient Care STDS 2008; 22:1015–1020.
10.Butler LM, Osmond DH, Jones AG, et al. Use of saliva as a lubricant in anal sexual practices among homosexual men. J Acquir Immun Defic Syndr 2009; 50:162–167.
11.Van Damme L, Ramjee G, Alary M, et al. Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: A randomised controlled trial. Lancet 2002; 360:971–977.
12.Phillips DM, Sudol KM, Taylor CL, et al. Lubricants containing N-9 may enhance rectal transmission of HIV and other STIs. Contraception 2004; 70:107–110.
13.Tabet SR, Surawicz C, Horton S, et al. Safety and toxicity of nonoxynol-9 gel as a rectal microbicide. Sex Transm Dis 1999; 26:564–571.
14.Adriaens E, Remon JP. Mucosal irritation potential of personal lubricants relates to product osmolality as detected by the slug mucosal irritation assay. Sex Transm Dis 2008; 35:512–516.
15.Sudol KM, Phillips DM. Relative safety of sexual lubricants for rectal intercourse. Sex Transm Dis 2004; 31:346–349.
16.Russo J, Rohan LC, Moncla B, et al. Safety and Anti-HIV Activity of Over-the-Counter Lubricant Gels. Paper presented at: 2010 International Microbicides Conference, Pittsburgh, PA. Abstract 347.
17.Begay O, Jean-Pierre N, Abraham CJ, et al. Identification of personal lubricants that can cause rectal epithelial cell damage and enhance HIV type 1 replication in vitro. AIDS Res Hum Retroviruses 2011;27:1019–1024.
18.Fuchs EJ, Lee LA, Torbenson MS, et al. Hyperosmolar sexual lubricant causes epithelial damage in the distal colon: potential implication for HIV transmission. J Infect Dis 2007; 195:703–710.
19.Breban R, McGowan I, Topaz C, et al. Modeling the potential impact of rectal microbicides to reduce HIV transmission in bathhouses. Math Biosci Eng 2006; 3:459–66.
20.Boily M-C. Population-level impact of microbicides to prevent HIV: the efficacy that matters? In: Symposium of Microbicides 2010 Conference. Pittsburgh, PA: 176; 2010.