This 1 hour lunch time poster session presented an overview of different studies presenting STI epidemiology amongst MSM & heterosexual populations.
- Florence Lot: STI co- infections at HIV diagnosis in France
Aim was to analyse the frequency of STIs in new HIV infections between 2012- 2015, using the mandatory HIV surveillance system which has, since 2012, collected data on bacterial STIs (CT, NG, LGV, syphilis). These had to be reported if detected at time of or in the 12 months prior to HIV diagnosis.
Analysis by transmission group and trends. Reported as ‘HIV/STI co-infection’.
Frequency of STI/HIV co-infections in adults by HIV transmission group & sex: 14.6% globally (26.4% MSM; 11.3% male & 5.8% females born in France; 5.1% male & 2.5% females born abroad; 7.2% male & 8.7% female IDU).
Frequency of STI/HIV co-infections in adults by HIV transmission group & year of HIV diagnosis: Significantly increased over time with 12.6% in 2012 to 18.3% in 2016. By transmission group, this increase was only significant for MSM from 22.1% in 2012 to 31.9% in 2016.
Frequency of STI/HIV co-infections in adults by HIV transmission group and bacterial STI: Syphilis on the rise especially in MSM, heterosexual men were more often co-infected with syphilis and NG than heterosexual women who were more frequently infected with CT. Rectal LGV dx only in MSM.
HIV & STI co- infection has increased over time and affect almost 1/3 of MSM newly dx with HIV. Highlight importance of testing, treating index + partner.
There was a question around testing frequency in France- was once per year and is now 3 monthly, including viral hepatitis screen (not sure if this had increased from yearly..). Another Q around high syphilis prevalence- confirmed that asymptomatic people are screened.
A comment from a clinician from the UK regarding HCV, they are seeing a significant increase in acute HCV infections not associated with IDU amongst HIV negative MSM population, same in France? A- don’t have the data.
- Flavia Kiweewa Matovu: STI acquisition among women using a variety of contraceptive options in Uganda
LARCs are being widely promoted, not much data on STIs in LARC use. High risk female population in in Uganda, established STI increases risk of HIV transmission.
Prospective cohort study- ASPIRE study.
Analysis population: 2264 women (50.2% from Sth Africa). 817 cases of STIs detected over 3,440 person years of follow up.
CT: 408 cases/ incidence of 11.86/100 person years
NG: 196 cases/ incidence 5.70/100 person years
T.vaginalis: 213 cases/ incidence 6.19/100 person years
(No mention of HIV, syphilis etc).
Incidence of CT & NG were not different across contraceptive methods.
Incidence of T.vaginalis was significantly lower for DMPA (medroxyprogesterone acetate injectable contraceptive), implant and NET-EN (norethisterone enanthate injectable contraceptive) users compared to IUD.
Significantly lower rates of T.vaginalis among users of progestin- based methods, likely due to hypoestrogenic states.
Limited by lack of randomisation to contraception method.
Question around extra- genital sites being tested- yes, serology and genital swabs.
Another question around relative risks being adjusted for baseline sexual risk- yes, adjusted for baseline age, sexual risk.
3) Jeffrey Parsons: Differences in biological and behavioural HIV risks before, during and after PrEP use among a national sample of GBM in the USA
PrEP & STIs- Does going on PrEP lead to increased condomless sex (CAS) and thus higher rates of STIs OR are the increases seen in STI rates due to the required quarterly testing…?
Limitations to these data: Predominately Caucasian, employed and well educated population. Half were in a relationship, half were single. The sample were also early PrEP adopters.
Cross sectional between group analysis: the 823 PrEP naïve men had significantly lower STI infection rate (4.2%) than those 77 currently (10.4%) or 17 formerly (11.8%) on PrEP (p < 0.02), with men on PrEP also reporting more acts of CAS (p <0.001).
Within- person longitudinal analyses of 181 men reporting PrEP use indicated a non- significant increase in the odds on an STI diagnosis while on PrEP and after discontinuing (OR= 1.25, p= 0.55; OR= 1.43, p= 0.53 respectively), compared to before starting PrEP. There were also no significant changes in CAS while on PrEP (OR = 1.09, p = 0.76) or after PrEP discontinuation (OR = 0.48, p = 0.10) compared to pre-uptake levels.
Using between- subjects comparisons of all participants, some evidence was found that GBM on PrEP have higher levels of both behavioural and biological risks, though findings were mixed when examining multiple time points.
Using within- subjects comparison over time among only those who had been on PrEP during at least one of the three visits, the rates of CAS increased while on PrEP but returned to pre- PrEP levels after discontinuation. They did not see a statistically significant increase in odds of STI infection.
Important not to lump all GBM into one category. Differences in behaviour, risk and motivation for accessing and using PrEP.
Early adopters- so more study required to determine the behavioural differences in early and late PrEP adopters.
Question from the audience regarding the sites tested for STIs- in NYC, only 50% of participants had completed a rectal swab which could affect the data.
88% completed urine and serology test.. pharyngeal swab around 60% (sorry missed that comment).
Highlights the importance of a complete STI screen to ensure both a public health and epidemiological perspective.
4) Marie Suzan- Monti: Partner notification (PN) of STIs among MSM on PrEP: s sub- study of the ANRS- IPERGAY trial
In France, there are no PN specific guidelines, and scarce PN information. Data on 275 HIV negative men from the ANRS- IPERGAY PrEP trial who reported an STI were used.
Out of 275, 250 reported at least one previous STI. Of the 250, 172 had informed their partner (138 their occasional partner and 83 their main partner).
No significant socio-demographic difference between this who did and did not notify their partner.
Less likely to notify their main partner when most recent sexual contact was through condomless sex with an occasional partner (aOR(95%CI) 0.31 (0.14; 0.68), p=0.03).
Older MSM less likely to inform occasional partners (aOR(95%CI) 0.44(0.21;0.94), p=0.03).
Those participating in chemsex at most recent sexual encounter were more likely to inform sexual partners (aOR(95%CI) 2.56(1.07;6.09), p=0.03).
Condomless sex with occasional partners was identified as a barrier to PN, and chemsex a motivator for PN.
Not measures if health care workers were notifying partners.
Hopefully these data support the need for systematic PN services, support and information in France- highlights how well Australia (speaking from a Victorian perspective) undertake PN form a top down approach.
5) Kristin Wall: Predictors of genital ulceration in HIV negative sero-discordant couples in Lusaka, Zambia.
Genital ulcers are a known risk factor for HIV transmission, and little is known about the risk factors for genital ulcers, limiting early detection and treatment.
Exposure data were taken from HIV serodiscordant heterosexual couples every 3 months at ART uptake or HIV transmission. Associations were evaluated between exposures measured during the visit prior to the presentation with an ulcer. 18 year longitudinal cohort study (1994- 2012).
1393 M+F- couples were followed for 2756 couple- years, and 1656 M-F+ couples were followed for 3216 couple- years.
Risk for genital ulcer in HIV- women was associated (p<0.05) with bilateral inguinal adenopathy (BIA) (aHR=1.9), genital inflammation (GI) (aHR=1.5-1.9), male partners non- STI GI (aHR=2.9) and increasing number of previous pregnancies (aHR=1.1).
Risk for genital ulcer in HIV+ women was as above (BIA- aHR=1.5; GI- aHR=1.5-2.0; male non- STI GI- aHR=2.0), as well as late HIV vs early HIV (aHR=1.5) and being pregnant (aHR=0.7).
Risk in HIV- men was associated with BIA (aHR=1.8), STI GI (aHR=2.9) and non- STI GI (aHR=1.4), female partners ulcer (aHR=1.7), and being uncircumcised (aHR=1.7). Being uncircumcised with foreskin smegma was independently predictive (aHR=3.2).
Risk in HIV+ men was associated with STI GI (aHR=2.8), HSV-2 positivity (aHR=2.5), late HIV vs early (aHR=1.7) and being uncircumcised with foreskin smegma was independently predictive (aHR=2.4).
Ulcers were also tested for syphilis, prevalence of chancre 2-3%
BIA & GI may be early indicators/ risk factors for genital ulceration. Uncircumcised men with foreskin smegma either HIV +/- were at increased risk of ulceration.
HSV-2 positivity not a predictor once controlled for genital ulcers and only a predictor in HIV+ men.
Suggest: Targeted screening amongst those with advanced HIV infection.
6) Cari van Schalkwyk: Are associations between HIV & HPV transmission due to behavioural confounding factors or biological effects?
This presentation was a mathematical modelling study to assess whether confounding for behavioural factors and network effects sufficiently explain associations between HPV & HIV infection.
MicroCOSM is a dynamic individual- based network model and was used to simulate epidemics of HIV & 13 oncogenic HPV types.
The mean unadjusted hazard ratio of HIV acquisition after detection of an oncogenic HPV type is 3.2 (95% CI 2.6, 3.8); and the mean unadjusted hazard ratio for the effect of HIV on newly detected HPV is 3.7 (95% CI 3.4, 4.1).
The study results are similar to observational study unadjusted results, suggesting that observed associations between HPV & HIV transmission could be attributed to confounding by behavioural factors and network- level effects. The author concluded that primary prevention with the HPV vaccine may therefore not be beneficial in HIV prevention.
There was also no further increased risk in the presence of cervical lesions.
The study group have a proposal for a clinical trial using the HPV vaccine to determine if this decreases HIV transmission although no funding as yet.