Abstract
Regarding people living with HIV (PLHIV), little is known about the epidemiological characteristics and management decisions for transgender individuals. This retrospective study compared transgender and cisgender (homosexual and heterosexual) PLHIV at both the S. Maria della Misericordia of Perugia and Careggi of Firenze Teaching Hospitals from 2000 to 2018. Multivariate logistic regression was performed to analyse possible relationships between viral suppression (dependent variable) and age, sexually transmitted infections (STIs), and hepatitis diagnosis (independent variables). After analysing and comparing epidemiological and clinical data for 124 transgender, 180 homosexual cisgender and 188 heterosexual cisgender PLHIV, we found that transgender PLHIV, mostly Latin American sex workers, were more likely to have other STIs. Likewise, this subgroup, on average, was younger at the time of HIV diagnosis and more likely to be less adherent to care, consequently jeopardizing the achievement of viral suppression. Finally, the use of hormone therapy and gender confirmation surgery in transgender PLHIV contributed to specific management issues. To date, major attention has focused on studying the epidemiological characteristics of homosexual and heterosexual PLHIV. Our analysis found that transgender PLHIV were the least likely group to be adequately retained in the continuum of care and presented specific issues in part due to social and behavioural realities.
Background
The term “transgender” is used to indicate persons who do not recognise themselves in the sex of their birth: natal males with a female gender identity, or MtFs, and natal females with a masculine gender identity, or FtMs. 1 For a longtime, in clinical trials, transgender persons (TGPs) were considered to be men who had sex with men (MSMs), while transgender individuals were never investigated as a distinct subgroup. In fact, a distinction was not made between gender identity and sexual behaviour. Moreover, national census data for Western Countries do not seek to identify this subgroup of people. However, recent reports have estimated that 0.6% of adults in the United States identify as transgender. 2 As a result, until 2012, there had not been a clear picture of the burden of HIV infection among TGPs. In the few countries where epidemiological data for TGPs are available it seems that this subgroup has a very high risk of acquiring HIV infection. In fact, a systematic review and meta-analysis on the worldwide burden of HIV in TGPs published by Baral SD et al. in 2013, 3 suggested a pooled HIV prevalence of 19.1% and an odds ratio of 48.8 (95% CI 21.2–76.3) for acquiring HIV, compared with all adults of reproductive age, across 15 countries without differences among low, middle and high-income countries. To date in Italy, limited data are available regarding the number of TGPs living with HIV and their adherence to care rates.
Methods
The principal aim of this retrospective study was to record and then compare epidemiological characteristics and management decisions for TGPs living with HIV between 2000 and 2018 at the S. Maria della Misericordia of Perugia and Careggi of Firenze Teaching Hospitals with a sample of homosexual and heterosexual males living with HIV followed at the same hospitals.
We collected epidemiological and clinical information between 2000 and 2018 and excluded heterosexual females. Moreover, we excluded from the cisgender control group all patients with unknown risk factor or non-sexually-related HIV transmission. Data regarding nationality, age at the time of HIV diagnosis, alcohol/drug abuse, smoking habit, involvement in sex work, CD4+ cell nadir, HIV RNA zenith, viral and immunological measures at the most recent visit, and diagnosis of sexually transmitted infections (STIs), especially syphilis, were collected from medical records and transferred anonymously to an Excel database. All people living with HIV (PLHIV) provided consent for the use of their data at the moment of admission to the hospital. The responsible ethics committees approved the study protocol.
Continuous data were analysed using one-way ANOVA, whereas comparisons between the two groups were carried out with Bonferroni correction. Categorical data were analysed-utilising chi-square test. Multivariate logistic regression was performed to analyse possible relationship between viral suppression (dependent variable) and age, STIs, hepatitis diagnosis (independent variables). Fit for logistic regression model was previously evaluated with Hosmer-Lemeshow test.
Results
Overall, we reviewed the profiles of 492 PLHIV, of which 491 were males.
Considering the major factors contributing to HIV risk, 124 were TGPs (123 MtFs, 1 FtM), 180 MSM and 188 heterosexual males.
The median age at diagnosis was 31.3 years (range 13–62, SD 10.7) for TGPs, 39.5 (range 20–73, SD 11.9) for MSM, and 39.7 (range 18–76, SD 12.9) for heterosexual males (P < 0.001).
Of the transgender PLHIV, 83% (103/124) were non-Italians, mostly Latin Americans (99/124, 79.83%). However, 10% of MSM (18/180) and 36.7% (69/188) of heterosexual males were also non-Italians.
Analysing laboratory data, we found that the average CD4+ cell count was <200 cells/mm3 for 37.7% (40/106) of TGPs, 31% (57/180) of MSMs and 51% (96/188) of heterosexual males (P < 0.001). Moreover at diagnosis, 48.9% of transgender PLHIV (46/94) had HIV-RNA > 100,000 copies/mL, whereas this value was 63.8% and 64.3% for MSM and heterosexual males, respectively (P = 0.026).
A year after starting combination antiretroviral therapy (cART) , viral suppression (i.e., HIV RNA < 50 copies/mL) was reached in 66.6% (74/111) of TGPs, 87.7% (158/180) of MSM and 81.38% (153/188) of heterosexual males (P < 0.01).
We used logistic regression to analyse possible variables influencing negatively viral suppression and found a significant relationship with younger age (P = 0.035). No statistical significance emerged analysing STIs and hepatitis co-infections. This analysis was only possible for 111 TGPs because the 13 others were lost to follow up and HIV RNA data were lacking.
STIs were significantly more frequent in TGPs (85/104, 81.73%) and in MSM (131/180, 72.8%), compared to heterosexual males (39/188, 20.7%): P = 0.001. Epidemiological and clinical data are described in Table 1.
Epidemiological and clinical data about the 3 groups.
Additional information regarding drug/alcohol abuse, smoking habit, sex worker status, use of hormonal therapies and/or surgical procedures (gender confirmation surgeries including transition) were available for a limited number of transgender PLHIV and are listed above in Table 2.
Drug/alcohol abuse, smoking habit, sex worker status, use of hormonal therapies and gender confirmation surgeries in TGPs.
aAlso irregular use.
bTransdermal oestradiol, ethinyloestradiol/cyproterone per os, hydroxyprogesterone IM for MtFs; testosterone esters for the FtM.
Discussion
Worldwide, HIV prevalence in TGPs is estimated to be higher than the general population. 3 In Italy, there is a paucity of recent data regarding the prevalence of HIV for this subgroup as well as limited data on the risk behaviours associated with adherence to care rates.4–6 This is certainly due to the traditional practice of grouping together TGPs and MSM. However, TGPs have unique demographic, psychological and behavioural characteristics that influence medical manifestations and clinical outcomes.
Our study results suggest that, at diagnosis, TGPs were significantly younger and had the highest CD4 cell nadir. The reason behind these findings may be that TGPs know that they are at a very high risk of acquiring HIV infection and this has been reported to induce many to test frequently and therein could explain why they receive a diagnosis at an early age and stage.7,8 Previous studies have asserted that TGPs tend to underestimate their level of risk for HIV and do not believe they were at risk for having or acquiring HIV. 9 However, in our study, only 66.6% of TGPs achieved viral suppression after one year of cART, a result in agreement with previous studies.10–15 The failure to reach this target has been reported primarily due to a lack of retention in care, which is more common in non-Italians 16 and young people. 17 In fact, we observed that most of the TGPs were non-Italians (83%), most often lacking family and community support. Furthermore, logistic regression showed that viral detectability correlated with TGPs' younger age. These characteristics are not shared by MSM, which are mostly Italians and can appeal to organisations guarding their rights. Moreover, in contrast to MSM, disclosing their gender identity can be a barrier to finding legal employment. TGPs often resort to sex work, which often means not having a regular income, and being an unauthorized migrant constitutes a further barrier to medical care even in a country with a free national health system.18–20
Additional factors favouring poorer retention in care for TGPs include mental illness as well as drug and alcohol abuse. 21 In fact, noncontinuous care adherence is known to increase the probability of worse long-term outcomes for PLHIV as well as an increased risk of onward transmission.
Apart from lower continuum of care outcomes, some authors 22 highlighted that TGPs sometimes interrupted cART because of concerns about interference with their hormone therapy.
Overall, 72% (34/47) of TGPs participating in our study used hormone treatments, increasing the probability of drug-drug interactions, most importantly with cART, and other prescribed therapies (i.e., cardiovascular treatments). Unfortunately, we could not use data concerning hormone therapy for logistic regression because it was reported by few TGPs (47/124, 38%) and only patients that visited in the last period. This constitutes a bias to understand if there is a relationship between viral detectability and hormone treatments because the information was available only for patients still retained in care. This evidence corroborates the observation that, especially in the past, physicians have dedicated little attention to issues specific to transgender PLHIV. Specifically, the type and route of hormone therapy administration are known to have adverse effects: ethinyloestradiol per os, especially with cyproterone, has been associated with insulin resistance and a greater risk of venous thromboembolism, ischaemic stroke and myocardial infarction23–27; transdermal oestradiol has not been associated with these events. 28 Consequently, TGPs who take hormone treatment are at a greater risk of cardiovascular diseases, even if they abstain from smoking and/or drug use. In fact, it has been repeatedly reported that MtFs have a higher rate of cardiovascular events than either postmenopausal women or FtMs.27,29
In agreement with previous studies we observed a high rate of STIs in transgender sex workers.30–32 Moreover, gender confirmation surgeries and hormone therapy in TGPs were associated with infection. In fact, it has been reported that vaginoplasty can lead to the development of a new microenvironment lacking the immunological characteristics of a healthy vaginal mucosa, thereby making the organ more susceptible to STIs.32–34 Furthermore, erectile dysfunction, associated with the use of feminizing hormones, can interfere with the use of condoms. 35
Conclusions
Our study results highlight the issues surrounding an adequate access and delivery of healthcare for transgender PLHIV. We observed that they, compared to other subgroups, had social and behavioural realities that contributed to a poorer retention in care. Greater attention to the issues often experienced by transgender individuals is needed to address disparities in care.
Ethics approval and consent to participate
Informed consent to collect data from clinical records was obtained from all individual subjects included in the study at their first access to the hospital centres.
All procedures performed in this study were in accordance with the ethical standards of the ethical committees of both the Umbrian and Tuscany Regions (Comitato Etico Aziende Sanitarie dell’Umbria: TransHIV-16565/19/ON, approved on 13th June 2019; Comitato Etico Regionale per la Sperimentazione Clinica della Regione Toscana, Sezione Area Vasta Centro: 10452 (ex-oss.16.280) approved 28th February 2017) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards
Footnotes
Acknowledgements
We thank Mr Thomas Charles Kilcline for his important editorial assistance.
Author Contributions
CP, FL, ES, GS, and DF contributed to the study conception and design. CP and FL performed the data collection and analysis. CP wrote the first draft of the manuscript; FL, ES, GS, and DF commented on previous versions of the manuscript. CP, FL, ES, GS, and DF read and approved the final manuscript.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
