Objectives

Nylon-flocked and Dacron swab anal cytology collection procedures were evaluated for detecting high-grade anal intraepithelial neoplasia.

Materials and methods

Cross-sectional data for 42 HIV-infected and 16 uninfected men who have sex with men have been used. Sequentially collected anal cytology specimens, high-resolution anoscopy, and medical biopsy evaluated the sensitivity and specificity of cytology for predicting high-grade anal intraepithelial neoplasia. Men showing atypical squamous cells (ASC) or more severe findings by cytology were compared with those showing negative for intraepithelial lesions.

Results

The prevalence of high-grade anal intraepithelial neoplasia was 35% (21/58), and findings were approximately 1.5 times higher among HIV-infected compared with uninfected men. Unsatisfactory cytology was twice as common among Dacron compared with nylon-flocked swab protocol specimens (14% [8/58] vs 7% [4/58]). Sensitivity and specificity for the nylon-flocked protocol cytology showing ASC or more severe findings were 81% (58%-95%) and 73% (50%-89%), respectively. Dacron protocol specimens showed 52% (30%-74%) and 58% (34%-80%) sensitivity and specificity, respectively. Men showing ASC or more severe findings using the nylon-flocked protocol cytology showed 3-fold higher odds for high-grade anal intraepithelial neoplasia compared with men with negative results (p < .05), but no statistically significantly higher odds of high-grade anal intraepithelial neoplasia for men showing ASC or more severe findings compared with those with negative results for Dacron protocol cytology (p > .05).

Conclusions

The nylon-flocked protocol better detects high-grade anal intraepithelial neoplasia than does the Dacron protocol, yields more interpretable results, and classifies men with high-grade anal intraepithelial neoplasia as cytologically abnormal 2.5 times more often, even in this small clinical trial.

Clinical trials registration number

NCT00955591.

Background

For women, the order of penile insertion, condom use, and ejaculation by orifice during sexual events affects the probability of HIV transmission and design of HIV prevention methods.

Methods

From October 2006 to June 2009, 431 women in Los Angeles and Baltimore in a rectal health study reported the sequence of penile insertion, condom use, and ejaculation by orifice location by computer-assisted self-interview. Multinomial logistic regression identified predictors of condom use by orifice among women who reported vaginal intercourse (VI) during their last anal intercourse (AI) event.

Results

Of the 192 reporting on a last AI event, 96.3% (180/187) reported VI. Of these, 83.1% had VI before AI. Including the 36% who ejaculated in both the rectum and vagina, 66% report any ejaculation in the vagina and 45% in the rectum. One-third used a condom for both VI and AI, <10% for VI only or AI only, and half used no condoms. After adjusting for race, partner type, and substance use, compared with women who used condoms for both VI and AI at last AI, being older (units = 5 years) [adjusted odds ratio (AOR) = 0.76; 95% confidence interval (CI): 0.60 to 0.96], with serodiscordant partners (AOR = 0.22; 95% CI: 0.08 to 0.61), and HIV-positive with seroconcordant partners (AOR = 0.15; 95% CI: 0.04 to 0.54) were associated with not using condoms.

Conclusions

For most of the women in our study VI accompanied AI, with AI usually occurring after VI. This evidence for use of multiple orifices during the same sexual encounter and low use of condoms across orifices supports the need for a multicompartment HIV prevention strategy.