Article Text
PDF
Abstract
Objectives Syphilis remains a public health challenge, particularly among people with HIV (PWH). This study aimed to examine the trends of syphilis and associated factors among PWH in Taiwan, 2016–2023, before the implementation of doxycycline postexposure prophylaxis (DoxyPEP).
Methods PWH aged 18 years or older who sought HIV care at a university hospital and had at least two serological tests for syphilis during the study period were included. Annual incidence rates of syphilis were calculated as the number of new syphilis cases per 100 person-years of follow-up (PYFU), while the prevalence was defined as the proportion of PWH who had a positive rapid plasma reagin (RPR) titre. Reinfection was defined as a ≥4 fold increase in RPR titre following a prior syphilis diagnosis. Multivariable logistic regression was used to identify factors associated with syphilis acquisition.
Results Among 3828 PWH, a total of 3201 incident syphilis cases were recorded during a total of 23 385.1 PYFU. The incidence rate decreased significantly from 16.78 per 100 PYFU in 2016 to 11.14 per 100 PYFU in 2023, a 33.6% reduction. The prevalence peaked at 45.0% in 2019 before declining to 41.6% in 2023. Reinfections constituted 66.3–85.0% of incident cases annually. Factors associated with acquiring syphilis included younger age (adjusted OR (AOR), per 10-year increase, 0.71; 95% CI, 0.67 to 0.75), men who have sex with men (AOR, 1.75; 95% CI, 1.32 to 2.32), a previous syphilis history (AOR, 7.26; 95% CI, 6.48 to 8.14) and no follow-up RPR data in the preceding year(s) (AOR, 3.02; 95% CI, 2.08 to 4.38).
Conclusions While the declines in incidence and prevalence of syphilis among PWH before the implementation of DoxyPEP were likely driven by an ageing population in Taiwan, regular serological testing for syphilis remains imperative for early diagnosis and treatment of syphilis to prevent further transmission.
- Post-Exposure Prophylaxis
- SYPHILIS
- Prevalence
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. Decoded data will be available upon reasonable requests.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Syphilis remains a significant public health concern among people with HIV (PWH), particularly among men who have sex with men (MSM). Doxycycline postexposure prophylaxis (DoxyPEP) has been shown to prevent syphilis and chlamydia among MSM.
WHAT THIS STUDY ADDS
This study shows a 33.6% reduction in syphilis incidence among PWH in Taiwan from 2016 to 2023, largely driven by the ageing of the PWH population. Incident syphilis was associated with younger MSM, prior syphilis and reduced adherence to recommendations of syphilis testing
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The findings suggest enhancing routine syphilis screening, linkage to effective treatment and targeted prevention for high-risk groups, while further research is needed to assess the impact of DoxyPEP on the trends of syphilis among MSM in Taiwan.
Introduction
Advances in antiretroviral therapy (ART) for HIV prevention—including pre-exposure prophylaxis (PrEP), postexposure prophylaxis (PEP) and treatment as prevention for HIV—have reduced new HIV infections globally. However, risk compensation, where preventive measures for HIV may encourage riskier sexual behaviours, emerged, particularly among men who have sex with men (MSM). Since the late 1990s, global sexually transmitted infection (STI) incidence has risen.1 2 A study analysing STI trends from 1990 to 2019 revealed that adolescents had the highest incidence rates, with varying trends across different geographical regions.3
Syphilis has dramatically resurged in the past two decades, reaching 7.1 million cases globally in 2020 and disproportionately affecting people with HIV (PWH) and MSM. The increased incidence of syphilis is associated with having multiple partners, chemsex, online social networking and condomless sex, especially within the context of biomedical HIV treatment and prevention.2 4 5 In 2019, England’s surveillance data showed that 30.5% of all syphilis diagnoses among MSM occurred in PWH.6 Globally, the pooled syphilis prevalence among MSM from 2000 to 2020 was 7.5%, with a slight decline in 2010–2020, possibly due to improved access to testing and treatment.4 A meta-analysis by Zheng et al indicated a significantly higher syphilis incidence (150.1 vs 79.5 per 1000 person-years of follow-up (PYFU)) and prevalence (22.2% vs 8.8%) among PWH who are MSM compared with their HIV-negative counterparts.7
In Taiwan, syphilis predominantly affects men, with over 9000 cases annually.8 A previous cohort study reported a decreasing trend in syphilis incidence among 4907 PWH from 2000 to 2010, with an incidence rate of 128 per 1000 PYFU.9 Syphilis and HIV share similar modes of transmission, and co-infection can enhance the acquisition and transmission of both infections. Among PWH, syphilis is associated with increased plasma HIV viral load, even among those on ART, and can present with unusual manifestations and accelerated progression.10 11 Therefore, prevention, early diagnosis and prompt treatment of syphilis are crucial in comprehensive HIV care.
To address the syphilis resurgence among PWH, interventions like behavioural changes, improved healthcare access, partner notification, STI screening and health literacy campaigns have been implemented. The latest intervention, doxycycline PEP (DoxyPEP), has shown over 80% efficacy in reducing syphilis risk.12 13 In Taiwan, DoxyPEP has been gradually implemented in designated hospitals for HIV care since late 2023. Given the limited data on syphilis trends among PWH in Taiwan and the need for baseline data before the widespread adoption of DoxyPEP, this study aimed to assess the prevalence and incidence of syphilis and its associated factors among PWH at a university hospital in 2016–2023.
Methods
Study design and setting
This retrospective cohort study was conducted at the National Taiwan University Hospital (NTUH). PWH aged 18 years or older who received HIV care at NTUH with at least two rapid plasma reagin (RPR) tests at different time points from 1 January 2016 to 31 December 2023 were included. The follow-up duration was the interval between the first and last available serological tests. PWH without an RPR titre in a specific year were excluded from that year’s analysis and did not contribute follow-up time to the denominator for incidence calculations. Demographic and clinical data were retrospectively retrieved from the electronic medical records.
Definition of incident and prevalent syphilis
Syphilis was diagnosed based on a positive RPR titre (BD Macro-Vue RPR Card Test) in conjunction with a reactive Treponema pallidum particle agglutination (TPPA) assay (FTI-SERODIA-TPPA; Fujirebio Taiwan, Taoyuan, Taiwan).14 Incident syphilis was defined when (1) PWH without prior syphilis tested positive for both RPR and TPPA; or (2) PWH with previously treated syphilis experienced at least a fourfold increase in RPR titres. According to Taiwan’s HIV treatment guidelines, PWH on stable ART undergo laboratory tests every 3–6 months with syphilis screening at least annually or every 3–6 months for those at increased risk.15 Syphilis reinfection was defined as a fourfold RPR rise ≥12 months after prior treatment and was counted as one case per year per individual. Annual syphilis prevalence was defined as the number of PWH with one or more positive RPR titres, divided by the total number of PWH followed in the same year, excluding cases without TPPA confirmation or a documented history of syphilis. Annual incidence was determined by dividing the number of incident cases (including reinfections) by the total PYFU. Incidence rates were calculated annually, by age-specific subpopulations (PWH aged ≤30, 31–40, 41–50 and >50 years), by the follow-up status of PWH and by syphilis status during 2016–2017. The proportion of annual syphilis reinfection was also assessed.
Statistical analysis
The demographic and clinical characteristics of the included PWH were summarised. Non-categorical variables were compared using the Wilcoxon rank-sum test, while categorical variables were compared using the χ² test between PWH with and those without incident syphilis. Syphilis incidence rates and prevalence were calculated with 95% CIs. Differences in the incidence rates between 2016 and 2023 were assessed using the mid-p adjustment for exact p values.16 To assess the impact of PWH who dropped out, we compared syphilis incidence rates during 2016–2018 between those remaining in follow-up after 2019 and those lost to follow-up. Incidence rate ratios (IRR) and 95% CIs were calculated using Poisson regression models with person-time as an offset variable in Stata. Age-adjusted Poisson regression was performed to assess trends over time. Logistic regression models examined the associations between variables and the acquisition of syphilis as well as syphilis reinfections, with each syphilis episode as the unit of observation. Stepwise regression with a removal threshold of p=0.05 was used to select covariates for inclusion in the multivariable model. Results from the logistic regression models were presented as ORs. All analyses were performed using Stata/SE V.17.0 (StataCorp, College Station, Texas, USA) with two-tailed tests and p<0.05 considered statistically significant.
Results
Clinical characteristics of the cohort
During the 8-year study period, 3981 PWH had syphilis serological data; of these, 153 were excluded for having only one RPR titre, leaving 3828 for analysis, contributing a total of 23 385.1 PYFU and median follow-up duration of 7.7 years (IQR, 4.5–7.9 years). The baseline characteristics of the included PWH, categorised by incident syphilis status, are presented in online supplemental table 1. Overall, PWH who acquired syphilis during the follow-up were younger (36.5 vs 41.1 years, p<0.001) and more likely to be male (99.8% vs 95.4%, p<0.001) and MSM (97.5% vs 87.8%, p<0.001), and to have prior syphilis (75.1% vs 34.3%, p<0.001), shorter testing intervals (4.7 vs 5.3 months, p<0.001) and more frequent serological tests (2.5 vs 2.2 tests per year, p<0.001) compared with those without incident syphilis; however, they were less likely to have hepatitis B surface antigen (HBsAg) positivity (12.0% vs 13.5%, p=0.02). From 2016 to 2023, the median age increased from 38.2 to 43.1 years (p<0.001), while male sex, MSM status and HBsAg positivity remained stable. The total number of PWH included in each year is shown in table 1. The number of PWH retained in the cohort peaked in 2019 with 3165 PWH being regularly followed. Notably, the percentage of PWH aged over 50 years increased from 7.4% in 2016 to 14.4% in 2023, while that of PWH aged 30 years or younger decreased from 27.8% to 7.6% (figure 1A). The number of annual incident HIV cases in the cohort decreased over the study period (online supplemental table 2).17 Annual dropout rates ranged from 3.2% to 5.0% throughout the study period (table 1).
Supplemental material
(A) Changes in the proportions of different age groups by study year. (B) Proportions of annual incident cases of syphilis by age groups, 2016–2023.
Annual incidence of syphilis (2016–2023)
Changes of incidence and prevalence of syphilis
During the study period, 3201 incident syphilis cases occurred among 1702 unique PWH, with an overall incidence rate of 13.69 per 100 PYFU (95% CI, 13.22 to 14.17). The annual incidence rate significantly decreased from 16.78 per 100 PYFU (95% CI, 15.33 to 18.36) in 2016 to 11.14 per 100 PYFU (95% CI, 9.91 to 12.53) in 2023 (figure 2 and table 1), with an IRR of 0.66 (95% CI, 0.57 to 0.77, p<0.001). This 33.6% reduction in the incidence of syphilis by 2023 (95% CI, 22.8% to 42.9%) corresponded to a decrease of 5.63 per 100 PYFU (95% CI, 3.63 to 7.63) compared with 2016. However, no significant changes in incidence rates were found between 2016 and 2023 by age group. Specifically, among those aged 30 years or younger, the IRR was 1.49 (95% CI, 0.95 to 2.46, p=0.07); for those aged 31–40 years, it was 1.18 (95% CI, 0.94 to 1.50, p=0.14); for those aged 41–50 years, it was 1.25 (95% CI, 0.94 to 1.66, p=0.11); and for those over 50 years, it was 1.44 (95% CI, 0.89 to 2.32, p=0.12) (online supplemental figure 1). Despite the absence of significant changes within individual age groups between 2016 and 2023, age-specific analysis consistently showed a higher syphilis incidence among those aged 30 years or younger compared with those over 50 years in all study years (p<0.001). Individuals aged 31–40 years contributed more than 40% of total cases annually. From 2016 to 2023, the proportion of incident cases among those aged 30 years or younger decreased from 27.8% to 7.9%. Moreover, the overall trend in syphilis cases was influenced by the shifting age distribution over the study period, with older age groups experiencing an increase in case numbers while younger age groups seeing a decrease. Specifically, in those aged 30 years or younger, both the number and proportion of cases declined annually, whereas both increased in those aged over 50 years (figure 1A and B). To account for the impact of age on the syphilis trends, an age-adjusted Poisson regression analysis was conducted, which confirmed that syphilis incidence decreased by 3.8% per 1 year increase (IRR=0.96; 95% CI, 0.96 to 0.97, p<0.001), while syphilis incidence also declined over time, with a 5.6% reduction per year (IRR=0.94; 95% CI, 0.93 to 0.96, p<0.001).
Annual incidence rate and prevalence of syphilis among PWH, 2016–2023. PWH, people with HIV; PYFU, person-years of follow-up.
To assess the impact of PWH who dropped out, we compared the syphilis incidence during 2016–2018 by follow-up status since 2019. Those lost to follow-up after 2019 had a significantly higher syphilis incidence (25.09 per 100 PYFU (95% CI, 20.52 to 30.68)) than those who continued follow-up (16.98 per 100 PYFU (95% CI, 16.12 to 17.89)) (IRR, 1.48; 95% CI, 1.19 to 1.82, p<0.001). Moreover, PWH with confirmed syphilis during 2016–2017 had a significantly higher incidence rate during 2018–2023 (21.73 per 100 PYFU (95% CI, 20.37 to 23.19)) compared with those with positive RPR titres but without confirmed syphilis (17.71 per 100 PYFU (95% CI, 16.16 to 19.42)) (IRR, 1.23; 95% CI, 1.10 to 1.38, p<0.001) and those with negative RPR titres (4.53 per 100 PYFU (95% CI, 4.11 to 4.99)) (IRR, 4.80; 95% CI, 4.27 to 5.41, p<0.001) (online supplemental figure 2). Notably, 26.8% of PWH with confirmed syphilis during 2016–2017 accounted for more than 50% of incident cases during 2018–2023 (online supplemental figure 2).
A significant proportion of annual incident syphilis cases were reinfections, ranging from 66.3% to 85.0% (online supplemental figure 3). Additionally, 25.9% of newly diagnosed HIV infections were concurrently diagnosed with syphilis, varying from 11.1% to 33.3% annually.
Syphilis prevalence was 38.5% (95% CI, 36.7% to 40.3%) in 2016, which increased and peaked in 2019 at 45.0% (95% CI, 43.2% to 46.7%) before declining to 41.6% (95% CI, 39.8% to 43.4%) by 2023 (figure 2).
Factors associated with acquisition of syphilis and syphilis reinfections
Factors associated with acquiring syphilis are presented in table 2. In multivariable analysis, independent factors associated with syphilis included younger age (per 10-year increase, adjusted OR (AOR), 0.71; 95% CI, 0.67 to 0.75), MSM (AOR, 1.75; 95% CI, 1.32 to 2.32), a previous history of syphilis (AOR, 7.26; 95% CI, 6.48 to 8.14) and no follow-up RPR data in the preceding year(s) (AOR, 3.02; 95% CI, 2.08 to 4.38). Factors associated with syphilis reinfections, compared with PWH with only one episode of syphilis during the study period, are presented in online supplemental table 3. In the multivariable analysis, younger age (per 10-year increase, AOR, 0.75; 95% CI, 0.67 to 0.84) was identified as an independent factor associated with reinfections.
Logistic regression analysis of factors associated with acquisition of syphilis
Discussion
In this cohort study, we observed a significant reduction in syphilis incidence and prevalence among PWH in Taiwan from 2016 to 2023, before DoxyPEP implementation. The overall decrease in the syphilis incidence in our cohort contrasts with trends in high-income countries such as Korea, Western Europe and the Americas, where syphilis rates have generally been increasing.18–20 However, similarly declining trends have been reported in England and Japan. Data from the UK Health Security Agency showed a 10.2% reduction in syphilis diagnoses among MSM with HIV from 2017 to 2019.6 A retrospective study by Naito et al in Japan found that the resurgence of syphilis in Japan was not attributed to transmission among PWH, as the percentage of syphilis cases among PWH decreased during the study period.21 The possible explanation for the reduction in syphilis incidence in our cohort before DoxyPEP implementation seems associated with the ageing PWH population. From 2016 to 2023, the average age of the cohort significantly increased, with no significant changes in other baseline characteristics. This phenomenon reflects not only the ageing PWH population due to improved survival rates but also the decreasing number of new HIV diagnoses, particularly among younger individuals.22 The new HIV diagnoses in Taiwan have decreased by 62.4% from the peak incidence in 2017 to 2023, with a 70.1% reduction among those aged 15–34 years.17 Age was a significant factor associated with incident syphilis in our study, with a 37% reduction in risk per 10-year increase in age. Compared with those aged 30 years or younger, individuals aged over 50 years had a 76% lower risk of incident syphilis. The lack of statistically significant differences in the incidence rates by age group between 2016 and 2023 supports the conclusion that the overall reduction in syphilis incidence was primarily driven by the ageing cohort.
The decline in syphilis prevalence since 2019 may be partly attributed to increased awareness among physicians and more frequent screening by following Taiwan’s HIV treatment guidelines,15 leading to earlier detection of latent syphilis and a subsequent reduction in incident cases, as observed in a previous study.4 23 Additionally, incident HIV cases were frequently diagnosed concurrently with syphilis in our cohort and another study in southern Taiwan, where 30.1% of PWH were diagnosed with syphilis at the time of HIV diagnosis.24 The overall decrease in new HIV cases also likely contributed to the reduction in syphilis incidence.
In this study, we found that a significant proportion of annual incident syphilis cases were reinfections, ranging from 66.3% to 85.0%, a rate even higher than that reported in a 2018 US study, which revealed that approximately 50% of syphilis cases among male PWH were reinfections.25 Younger age remained a strong factor associated with reinfection. Furthermore, we demonstrated that recent syphilis at baseline significantly increased the likelihood of incident syphilis in the following 6 years. In previous studies conducted in Taiwan, similar factors associated with the acquisition and reinfection of syphilis were identified, along with others not assessed in our study, such as improved immune status, lower educational achievement, serosorting and recreational drug use.9 26 These findings have important implications for clinical practice and public health policy. In addition to the routine syphilis screening, more frequent testing should be emphasised for high-risk groups, particularly younger PWH, MSM, those with irregular follow-up and individuals with a history of syphilis—especially recent infections—as part of comprehensive HIV care. Our findings also highlight potential target groups for DoxyPEP in Taiwan, particularly younger PWH who are MSM with a history of syphilis. This aligns with the findings of Traeger et al, which suggest that DoxyPEP prescribing strategies based on STI history may be more effective than those based solely on HIV status or PrEP use.27
DoxyPEP has shown over 80% efficacy in reducing syphilis risks, particularly among MSM.12 13 However, a randomised controlled trial in cisgender women showed no significant benefit.28 International guidelines on DoxyPEP use vary, reflecting inconsistent recommendations.29 Through a questionnaire survey, we found high willingness (85.9%) to use DoxyPEP among the participants including PWH and PrEP users.30 A pilot DoxyPEP programme was implemented in November 2023 for at-risk populations, including PWH with a history of syphilis or STIs in the previous year and PrEP users, who underwent routine testing for syphilis and other bacterial STIs every 3–6 months. Preliminary data have revealed a declining incidence of C. trachomatis infection and syphilis, while the rates of gonorrhoea and Mycoplasma genitalium infection remain unchanged as of February 2025 (Hung CC, unpublished data). Continued monitoring of this pilot DoxyPEP programme is expected to provide valuable data to inform future national policies and guidelines on DoxyPEP in Taiwan.
Several limitations should be considered for this study. The retrospective design may have introduced selection bias by including only individuals with multiple syphilis tests. Additionally, the absence of a behavioural survey limited the identification of high-risk individuals who may benefit from DoxyPEP. Furthermore, conducting the study at a single centre may limit the generalisability of the findings. The drop-out of PWH may also have influenced the syphilis incidence observed. We found that PWH who were lost to follow-up after 2019 had a significantly higher incidence rate compared with those who continued follow-up during 2016–2018. These individuals might have contributed to a higher syphilis incidence had they remained in the cohort after 2019. Additionally, annual variability in syphilis prevalence could have been affected by both the dropout of PWH and the inclusion of newly diagnosed PWH during the study period. Finally, variations in serological testing frequency and intervals may have influenced the detection of incident syphilis, as testing tends to be more frequently conducted after a syphilis diagnosis or in individuals with symptoms or higher risk.
In conclusion, this study revealed that the decline in syphilis incidence among PWH in our cohort was largely driven by the ageing of the PWH population. Our findings highlight the need for targeted interventions and ongoing surveillance, particularly for younger PWH and those with prior syphilis and reduced adherence to HIV and STI care.
Abstract translation
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. Decoded data will be available upon reasonable requests.
Ethics statements
Patient consent for publication
Ethics approval
This study was approved by the Research Ethics Committee of NTUH (registration number: NTUH-202402083RINC).
References
Footnotes
Handling editor Jane S Hocking
Contributors Conceptualisation: K-HC and C-CH; Methodology: K-HC, K-YL and C-CH; Formal analysis and investigation: K-HC and L-HS; Writing—original draft preparation: K-HC; Writing—review and editing: all authors; Funding acquisition: C-CH; Resources: K-YL, Y-SH, S-HH, W-DL, Y-CC, AC, H-YS, C-CH; Supervision: C-CH; All authors contributed to the article and approved the submitted version. C-CH acted as the guarantor.
Funding This research received a grant from National Taiwan University Hospital Yunlin Branch, Yunlin, Taiwan (grant number: NTUHYL.112.HC001).
Competing interests K-HC, H-YS and C-CH have received research support from Gilead Sciences.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.