Abstract
Background
Over the past decade, the incidence of primary and secondary syphilis has increased dramatically in the United States and western Europe. Men living with HIV and those at risk of HIV infection experience disproportionately high rates of early syphilis [ES]. We compared the odds of ES among HIV-positive and HIV-negative men participating in a status-neutral comprehensive HIV prevention and treatment program [CHP].
Methods
We conducted a retrospective analysis of men aged 18–65 years with ≥ one CHP visit and ≥ two rapid plasma reagin [RPR] tests performed between January 1, 2018, and December 31, 2021. ES was defined as newly reactive RPR with a minimum titer of ≥ 1:4 or a ≥ four-fold increase in the RPR titer. Multiple logistic regression analyses were performed to determine predictors of ES.
Results
A total of 2490 men met the inclusion criteria, of whom 1426 (57·3%) were HIV-positive and 1064 (42·7%) were HIV-negative. Of the 393 men with ES, 284 (72·3%) were HIV-positive and 109 (27·7%) were HIV-negative. HIV-positive men had higher adjusted odds of ES (aOR 2·86, 95% CI 2·45, 3·27) than HIV-negative men did. Chlamydia or gonorrhea [CTGC] infection did not differ according to HIV status (aOR 0·93, 95% CI 0·82, 1·04).
Conclusions
In our status-neutral care setting, HIV-positive status was associated with significantly higher odds of ES, but not CTGC. Our findings emphasize the vulnerability of HIV-positive men to syphilis in an era of effective HIV biomedical prevention.
Keywords: Syphilis, HIV, sexually transmitted infections, epidemiology, men
Short Summary
Among men participating in a status-neutral comprehensive HIV program in New York City, HIV-positive status was associated with higher odds of early syphilis, but not of chlamydia or gonorrhea.
INTRODUCTION
Over the past decade, the incidence of primary and secondary syphilis has increased dramatically in the United States [US]1. Men accounted for 80.7% of the early syphilis [ES] cases reported in 2020. In both the US and Western Europe, men who have sex with men [MSM] experience disproportionately high rates of ES2. In the US, racial and ethnic disparities in ES incidence persist, with reported rates of 34·1, 13·9, and 6·9 per 100,000 in Blacks, Hispanics, and Whites, respectively1. Existing public health measures have failed to mitigate the re-emergence of this fully preventable sexually transmitted infection [STI] once considered a target for elimination3.
Although the rates of ES and other bacterial STIs are at historic highs, the incidence of HIV has declined during the same timeframe4. Over the past decade, antiretroviral treatment as prevention [TasP] and HIV preexposure prophylaxis [PrEP] have been adopted as key strategies for HIV prevention. Multiple randomized and observational TasP and PrEP studies have demonstrated significant reductions in HIV transmission utilizing these biomedical approaches to HIV prevention5–8. Within this pharmacopreventive paradigm, shifts in the utilization of non-pharmacological HIV prevention strategies, including decreased consistency of condom use and increased number of sexual partners, may partially explain the rising incidence of bacterial STIs despite declining HIV rates9,10.
Observational studies conducted prior to the advent of TasP and PrEP demonstrated higher rates of syphilis infection in HIV-positive compared to HIV-negative MSM. The higher risk of syphilis in HIV-positive MSM has been attributed to a complex interplay between behavioral and immunological factors11. While multiple studies have shown high rates of bacterial STIs among HIV-negative MSM participating in PrEP programs12–14, comparisons with risk-matched HIV-positive MSM are limited15. Few studies have directly compared the risk of syphilis in HIV-positive and HIV-negative MSM in the current era of HIV biomedical prevention. In this retrospective cohort study, we compared the odds of ES in HIV-positive and HIV-negative men participating in a status-neutral comprehensive HIV prevention and treatment program [CHP] at an urban academic health center. Our main goal was to determine whether the elevated risk of syphilis historically observed in HIV-positive MSM persists in our status-neutral care setting. In addition, we report on the frequency of screening for chlamydia and gonorrhea [CTGC] in the context of an ES diagnosis and compare the odds of CTGC infection by HIV-status.
MATERIALS AND METHODS
Setting
The New York Presbyterian Hospital [NYP] is a tertiary care facility located in Northern Manhattan, New York City. NYP’s CHP provides HIV treatment and prevention services to approximately 2,500 HIV-positive and 1,000 HIV-negative patients annually. Our HIV prevention program is grant-funded, with the ability to offer PrEP to all patients regardless of insurance status.
Data
STI testing data were obtained from NYP’s Sexual Health Quality Improvement and Assurance Database [QIA]. Since January 1, 2018, the QIA database has recorded the following STI test results: rapid plasma reagin [RPR] and reflex RPR titers, HIV serology, viral loads and genotypes, Chlamydia trachomatis [CT] and Neisseria gonorrhea [GC] nucleic acid amplification [NAAT] tests [CTGC], and hepatitis C virus serology and viral assays. Data from tests performed between January 1, 2018, and December 31, 2021, were included in this analysis.
Participants
We identified men aged between 18 and 65 years who had ≥ two RPR assays recorded in the database and who were evaluated at CHP at least once during the study period. We limited the analytic group to CHP participants to allow for better comparability of sexual and care-seeking behaviors between HIV-positive and HIV-negative men. The history of CHP participation was determined using clinical codes recorded during each laboratory visit. The baseline visit and end of follow-up were defined as the first and last dates on which STI test results were reported, respectively.
Primary outcome
The primary outcome was the incidence of ES. ES was defined as a newly reactive RPR result with a minimum titer of ≥1:4 or a ≥ four-fold increase in the RPR titer compared to the preceding assay. The primary outcome was operationalized as a binary outcome: ‘ES’ versus ‘no ES’. The ‘no ES’ group included those patients whose RPR serology was never reactive as well as those whose serology was reactive but did not meet criteria for ES.
Although participants could have had more than one ES episode during the observation period, our analysis was limited to the first ES episode captured by our diagnostic algorithm. The absence of treatment data limited our ability to distinguish between untreated and recurrent ES in patients with multiple occurrences of a ≥ four-fold increase in RPR titer.
Independent variables
Demographic information, including sex, date of birth, race, and ethnicity, was recorded as part of the routine clinical care. Age was defined as the baseline visit. A composite race/ethnicity variable was created using the following categories: ‘Asian/Pacific Islander’, ‘Black’, ‘Hispanic/Latino’, ‘White’, and ‘Unknown’.
HIV status was determined through a review of HIV serology and quantitative HIV-1 viral load [qVL] assays. High-level viremia was defined as qVL >1000 copies/ml, whereas virological suppression was defined as qVL ≤200 copies/ml. To assess for loss of virological control in HIV-positive participants, we created a binary outcome, ‘ever high-level viremia’ (yes/no), capturing patients with ≥ one episode of high-level viremia during the observation period. We determined the temporal relationship between ES and high-level viremia in the 3- and 6-month windows surrounding the ES diagnosis.
Composite test results for GC and CT were available in the QIA database as binary variables that combined all assays performed at any anatomical site during the study period. We created separate variables for CT or GC based on the anatomical site. STI co-testing was defined as any CTGC assay performed in the 4-week window surrounding the ES diagnosis (i.e., 2 weeks preceding or postdating the ES diagnosis). Co-detection was defined as any positive CTGC result within the testing window.
Statistical analysis
We conducted univariate and bivariate descriptive analyses based on ES status. For categorical data, Pearson chi-square tests or Fisher’s exact tests were performed when the expected cell frequency was less than five. For continuous variables, we used t-tests, ANOVA, or Wilcoxon rank-sum tests based on variable distributions. Unadjusted and adjusted multiple logistic regression analyses were performed to determine the odds of ES according to HIV status alone and after controlling for the potential confounding variables.
Owing to the uncertainty in the diagnostic classification of syphilis using retrospective data, we performed two sensitivity analyses of ES criteria. First, the minimum titer required for inclusion in the incident ES case definition was increased from 1:4 to 1:16. Second, we excluded patients with a baseline titer of ≥1:16 to avoid misclassification of preexisting ES as an incident infection.
Longitudinal PrEP adherence data was not available. However, we performed a sensitivity analysis comparing HIV-positive men to a subset of HIV-negative men who reported taking at least one dose of PrEP. We determined this subset by cross-referencing our HIV-negative cohort with an existing PrEP-engagement [PrEP-E] database. The PrEP-E database captures HIV-negative CHP participants who were documented by a program coordinator to have taken at least one dose of PrEP following program enrolment. Finally, we repeated our primary analysis after excluding patients with documented seroconversion. Documented seroconversion was defined as a discordant HIV serology with evidence of high-level viremia during the observation period.
A complete case analysis was performed. All statistical analyses were conducted using R (version 4·1·2 (2021-11-01) The R Foundation for Statistical Computing) and R Studio (version 2021·09·1; packages used: dplyr, tidyverse, janitor, arsenal, and broom).
Privacy and ethics
The institutional review board of the Columbia University Irving Medical Center approved the study protocol (protocol number: IRB-AAAS1368). The requirement for informed consent was waived. All data were identified prior to the planned analysis.
RESULTS
Cohort description
Between January 1, 2018, and December 31, 2021, 360,375 STI test results performed in 191,125 unique patients were recorded in the QIA database (Figure 1). Of these, 107,903 were never tested for syphilis, and 50,311 had a single RPR test result, leaving 32,911 patients with ≥ two RPR assays. Of those with ≥ 2 RPR assays, 5,972 were men, 25,893 were women, and 1,046 had missing sex information. Of the 5,972 men with ≥ 2 RPRs performed, 5,284 (88.5%) were between the ages of 18 and 65 years. Of these, 2,490 men were evaluated at the CHP at least once and were included in the analysis.
Figure 1.
Study flow chart
Baseline characteristics of the CHP cohort are described in Table 1.
Table 1.
Baseline characteristics, CHP cohort (n=2490).
| HIV-positive (n=1426) | HIV-negative (n=1064) | Total (n=2490) | |
|---|---|---|---|
| Age, mean, years (range) | 42 (18 – 65) | 32 (18 – 65) | 37 (18 – 65) |
| ≥18 - ≤25 | 76 (5·3%) | 211 (19·8%) | 287 (11·5%) |
| >25 - ≤29 | 169 (11·9%) | 280 (26·3%) | 449 (18%) |
| >29 - ≤34 | 259 (18·2%) | 266 (25%) | 525 (21·1%) |
| >34 - ≤45 | 329 (23·1%) | 216 (20·3%) | 545 (21·9%) |
| >45 | 593 (41·6%) | 91 (8·6%) | 684 (27·5%) |
| Race/Ethnicity, n (%) | |||
| Hispanic/Latino | 570 (40·0%) | 408 (38·3%) | 978 (39·3%) |
| Black | 486 (34·1%) | 139 (13·1%) | 625 (25·1%) |
| White | 130 (9·1%) | 231 (21·7%) | 361 (14·5%) |
| Asian/Pacific Islander | 3 (0·2%) | 33 (3·1%) | 36 (1·4%) |
| Unknown | 237 (16·6%) | 253 (23·8%) | 490 (19·7%) |
| Years of follow up (person years) | 7002 | 6640 | |
| >1 year follow up, n (%) | 1124 (85·8%) | 727 (68·3%) | 1951 (78·4%) |
| Syphilis serology | |||
| RPRs performed, mean (range) | 5·4 (2 – 26) | 6 (2 – 24) | 5·7 (2 – 26) |
| RPRs per year follow-up, mean | |||
| ≤1 | 115 (8·1%) | 19 (1·8%) | 135 (5·4%) |
| 1 - ≤2 | 532 (37·3%) | 125 (11·7%) | 657 (26·4%) |
| 2 ≤ 4 | 611 (42·8%) | 540 (50·8%) | 1151 (46·2%) |
| > 4 | 168 (11·7%) | 380 (35·7%) | 548 (22%) |
| RPR reactive at baseline, n (%) | 461 (32·3%) | 116 (10·9%) | 577 (23·2%) |
| Chlamydia/gonorrhea testing | |||
| NAATs performed, mean (range) | 5·96 (0 – 35) | 6·73 (0 – 27) | 6·29 (0 – 35) |
| Ever positive | 523/1183 (44·2%) | 529/1061 (49·9%) | 1052/2244(46·8%) |
| Never tested | 243 (17%) | 3 (0·003%) | 246 (9·9%) |
| HIV control | |||
| Ever high-level viremia, n (%) | 622 (43·6%) | ·· | ·· |
| ≥ 2 high-level viremia*, n (%) | 57/1402 (0·04%) | ·· | ·· |
| Always ≤200 copies/ml*, n (%) | 715/1402 (51%) | ·· | ·· |
CHP, comprehensive HIV prevention and treatment program; RPR, rapid plasma reagin; NAAT, nucleic acid amplification test;
In participants with ≥2 HIV VL assays during the observation period.
The mean duration of follow-up was 2·24 (±1·19) years, and the total follow-up time was 5576 person-years. The mean age of the patients was 37 years (median, 34 years; range, 18–65 years). Race/ethnicity was distributed as follows: Hispanic/Latino 39·3%, Black 25·1%, White 14·5%, Asian/Pacific Islander 1·4%, and Unknown 19·7%. Overall, 1,426 (57·3%) participants were HIV-positive and 1,064 (42·7%) were HIV-negative. The mean number of RPR tests performed per patient was 5·7 (median, 5; range, 2–26). RPR results were non-reactive in 1,673 (67·2%) and reactive in 817 (32·8%) patients. Among those with reactive RPR results, 393/817 (48·2%) met the criteria for ES. Fluorescent treponemal antibody [FTA-ABS] test results were available for 392 cases, all of which were positive. The 393 men with ES identified at the CHP accounted for 81·8% of all males with ES at our institution.
HIV-positive men were older (mean age 42 vs. 32 years), more likely to be non-Hispanic Black (34·1% versus 13·1%) and have a follow-up of >1 year (85·8% versus 68·3%) than HIV-negative men. HIV-positive men had a lower mean number of RPR tests performed (5·4 versus 6), were less likely to have >4 RPRs per year of follow-up (11·8% versus 35·7%) and were significantly more likely to have reactive RPR at baseline (32·3% versus 10·9%). Nearly half of all HIV-positive men had high-level viremia at least once during the observational period. Of the 1185 HIV-positive men with ≥ two qVL assays performed, 51% were always virologically suppressed.
Main results
Compared to those without ES, participants with ES were significantly younger (mean age 34 versus 38 years, p<0·001) and more likely to be black (OR 1·62, 95% CI 1·31, 1·92) or Hispanic/Latino (OR 1·36, 95% CI 1·11, 1·6) than White (Table 2).
Table 2.
Predictors of early syphilis, univariate and multivariable analyses.
| Early syphilis (n=393) | No early syphilis (n=2097) | OR (95%CI) Unadjusted | aOR (95% CI) Adjusted | |
|---|---|---|---|---|
| Age, years, mean (range) | 34 (18 – 64) | 38 (18 – 65) | * | |
| >18 - ≤25 | 40 (10·2%) | 247 (11·8%) | Ref | Ref |
| >25 - ≤29 | 72 (18·3%) | 377 (18%) | 1·18 (0·93, 1·43) | 1·16 (0·89, 3·27) |
| >29 - ≤34 | 114 (29%) | 411 (19·6%) | 1·71 (1·37, 2·06) | 1·75 (1·36, 2·14) |
| >34 - ≤45 | 118 (30%) | 427 (20·4%) | 1·71 (1·37, 2·05) | 1·8 (1·4, 2·21) |
| >45 | 49 (12·5%) | 635 (30·3%) | 0·48 (0·37, 0·58) | 0·81 (0·6, 1·03) |
| Race/Ethnicity, n (%) | ||||
| Hispanic/Latino | 165 (42·1%) | 813 (38·8%) | 1·36 (1·11, 1·6) | 1·12 (0·9, 1·35) |
| Black | 122 (31%) | 403 (24%) | 1·62 (1·31, 1·92) | 1·29 (1·01, 1·57) |
| White | 47 (12%) | 314 (15%) | Ref | Ref |
| Asian/Pacific Islander | 4 (1%) | 32 (1·5%) | 1·2 (0·4, 4·87)^^ | - |
| Unknown | 54 (14%) | 435 (20·7% | 0·84 (0·67, 1·02) | - |
| Syphilis serology | ||||
| RPRs performed, mean (range) | 8·2 (2 – 24) | 5·2 (2 – 26) | *** | - |
| Per year follow up, mean (%) | ||||
| ≤1 | 6 (1·5%) | 128 (6·1%) | Ref | Ref |
| >1 - ≤2 | 77 (19·6%) | 580 (27·7%) | 2·83 (1·6, 4·06) | 3·0 (1·52, 4·49) |
| >2 - ≤4 | 251 (63·9%) | 900 (42·9%) | 5·94 (3·42, 8·46) | 5·04 (2·6, 7·49) |
| >4 | 59 (15·0%) | 489 (23·3%) | 2·57 (1·44, 3·7) | 6·19 (2·97, 9·4) |
| Duration of follow up (years) | ||||
| 0 - ≤1 | 26 (6·6%) | 513 (24·5%) | Ref | Ref |
| >1 - ≤2 | 47 (12%) | 452 (21·6%) | 2·05 (1·53, 2·57) | 2·44 (1·72, 3·15) |
| >2 - ≤3 | 81 (20·6%) | 468 (22·3%) | 3·41 (2·61, 4·21) | 3·77 (2·66, 4·88) |
| >3 - ≤4 | 239 (60·8) | 664 (31·7%) | 7·1 (5·58, 8·63) | 6·29 (4·53, 8·03) |
| HIV status | ||||
| Negative | 109 (27·7%) | 955 (45·5%) | Ref | Ref |
| Positive | 284 (72·3%) | 1142 (54·5%) | 2·18 (1·92, 2·44) | 2·86 (2·45, 3·27) |
| Ever high-level viremia | 145 (51·1%) | 477 (41·8%) | 1·45 (1·26, 1·65) | 1·62 (1·36, 1·88)* |
| CTGC testing ^ | ||||
| Tests performed, mean (range) | 10 (0 – 35) | 5·6 (0 – 28) | *** | - |
| Ever positive | 303/384 (78·9%) | 749/1856 (40·3%) | 5·56 (4·81, 6·3) | 3·99 (3·38, 4·6) |
| Missing n (%) | 9 (2·3%) | 235 (12·7%) | * | - |
CHP, comprehensive HIV prevention and treatment program; ES, early syphilis; RPR, rapid plasma reagin.
CTGC, Chlamydia trachomatis/Neisseria gonorrhea; OR, odds ratio; CI, confidence interval; Ref, reference group.
p<0·001 by Pearson chi-squared,
p = 0·004 by ANOVA;
p<0·00001 by Wilcoxon rank sum test;
Fisher’s exact test; NS, not significant.
at any time during the observation period.
Ref: HIV-positive without high-level viremia.
All multivariate analyses (except for ‘any CTGC infection’) were adjusted for the following variables: age, race/ethnicity, categorical number of RPRs performed per year, categorical follow-up duration, and any positive CTGC. Any CTGC infection was adjusted for age, race/ethnicity, categorical number of RPRs performed per year, categorical follow-up duration, and categorical number of GCCT tests performed.
The mean number of RPRs performed was higher in men with ES (8·2 versus 5·2, p<0·00001). The odds of ES increased with the duration of follow-up and number of RPRs performed per year. Participants with >3 years of follow-up had 7·1 (95% CI 5·58, 8·63) times the odds of ES than those with <1 year of follow-up. The odds of ES in men who had >2–≤4 RPR tests per year were 5·94 (95% CI 3·42, 8·46) times the odds of those who had <1 RPR per year. A history of any positive CT or GC result was a strong predictor of ES (OR 5·56, 95% CI 4·81, 6·3).
Of the 393 men with ES during the observation period, 284 (72·3%) were HIV-positive and 109 (27·7%) were HIV-negative. HIV-positive men had 2·18 (95% CI 1·92, 2·44) times odds of ES than HIV-negative men (Table 2). After adjusting for age, race/ethnicity, number of RPR tests per year, duration in care and any positive CTGC result, the odds of ES in HIV-positive men were 2·86 (95% CI 2·45, 3·27) times the odds of ES in HIV-negative men. HIV-positive men with ES had 1·62 (95% CI 1·36, 1·88) times the adjusted odds of ever high-level viremia than HIV-positive men without ES.
The mean number of CTGC tests performed per patient was 5·96 in HIV-positive and 6·73 in HIV-negative men (p<0.0001). Adjusting for age, race/ethnicity, duration in care, number of CTGC tests performed, and ES diagnosis, odds of any positive CTGC result in HIV-positive men were 0·93 (95% CI 0·82, 1·04) times the odds in HIV-negative men.
Subgroup analyses
Table 3 provides a description of CTGC and qVL cotesting in men with ES. CTGC test results were available for 323 (84·5%) men with ES. Among those tested, 56/323 (17·3%) had a positive GC assay, 59/323 (18·3%) had a positive CT assay, and 25/323 (7·7%) had CTGC coinfection. The most common site of detection is the rectum. We found no significant differences in the positivity rate or site of CTGC infection according to the HIV status. HIV-positive men with ES were significantly less likely to be tested for CTGC (21·5% versus 0% missing results, p<0·0001).
Table 3.
Select demographics, chlamydia and gonorrhea infection, and HIV-1 RNA quantification in men with ES (n=393).
| HIV-positive (n=284) | HIV-negative (n=109) | OR# (95%CI) | aOR@ (95%CI) | |
|---|---|---|---|---|
| Age, mean, years (range) | 35 (18–61) | 32 (19–64) | ** | ** |
| Race/Ethnicity | ||||
| White | 31 (10·9%) | 16 (14·7%) | Ref | Ref |
| Hispanic/Latino | 109 (38·4%) | 56 (51·4%) | 1·01 (0·71, 1·42) | 1·19 (0·76, 1·62) |
| Black | 106 (37·3%) | 16 (14·7%) | 3·42 (2·27, 5·14) | 4·11 (2·39, 5·84) |
| Gonorrhea codetection * | ||||
| Rectal | 26/193 (13·4) | 17/100 (17%) | 0·76 (0·5, 1·02) | ·· |
| Pharyngeal | 10/201 (5%) | 6/107 (5·6%) | 0·88 (0·28, 3·04) | ·· |
| Urethral | 4/207 (1·9%) | 2/108 (1·8%) | 1·04 (0·15, 11·7) | ·· |
| Any site positive | 34/223 (15·2%) | 22/109 (20·2%) | 0·71 (0·5, 0·93) | 0·71 (0·48, 0·94) |
| No testing at any site | 61/284 (21·5%) | 0 | ·· | ·· |
| Chlamydia codetection * | ||||
| Rectal | 35/193 (18·1%) | 16/100 (16%) | 1·16 (0·79, 1·49) | ·· |
| Pharyngeal | 6/201 (3%) | 1/107 (0·9%) | 3·25 (0·39, 150) | ·· |
| Urethral | 9/207 (4·3%) | 3/108 (2·8%) | 1·59 (0·39, 9·32) | ·· |
| Any site | 41/223 (18·3%) | 18 (16·5%) | 1·14 (0·79, 1·49) | 1·2 (0·81, 1·58) |
| No testing at any site | 61/284 (21·5%) | 0 | ·· | ·· |
| Any CTGC positive * | 59/223 (26·4%) | 31 (28·4%) | 0·91 (0·67, 1·14) | 0·93 (0·51, 1·34) |
| CTGC coinfection * | 16/223 (7·2%) | 9 (8·3%) | 0·86 (0·49, 1·23) | 0·92 (0·5, 1·34) |
| Coinciding high-level HIV viremia | ||||
| Within 3 months of ES | 61/252 (24·2%) | ·· | ·· | ·· |
| Within 6 months of ES | 71/268 (26·5%) | ·· | ·· | ·· |
ES, early syphilis; Ref, reference group; CTGC, Chlamydia trachomatis/Neisseria gonorrhea;
HIV negative is the reference group unless otherwise stated.
Demographics adjusted for age (ordinal categorical) and race/ethnicity (nominal categorical). CTGC testing was adjusted for age, race/ethnicity, and the number of CTGC tests performed (ordinal categorical).
Positive test in the 4-week window surrounding the date of ES diagnosis.
p<0·001 by ANOVA.
Among 284 HIV-positive men with ES, 268 (94·3%) underwent qVL testing three months before and after ES diagnosis. Within this 6-month window, 71/268 (26·5%) patients had high-level viremia. Of the 71 patients with high-level viremia in the 6 months window surrounding ES diagnosis, 59 (83.1%) had at least one qVL <200 copies/ml, 11 were viremic (≥200 copies/ml) on all available qVL assays, and one patient had a single qVL with high-level viremia during the observation period. Narrowing the testing window to three months did not substantially change the results: 252 (88·7%) underwent qVL testing and 61/252 (24·2%) had evidence of high-level viremia.
Sensitivity analyses
Increasing the minimum RPR titer required to meet the ES case definition from ≥1:4 to ≥1:16 did not alter the number of diagnoses identified. Excluding participants with a baseline RPR titer ≥1:16 (n=142) resulted in higher odds of ES for HIV-positive men (adjusted OR 3·11, 95% CI 2·56, 3·67).
After cross-referencing the 1,064 HIV-negative men in our cohort with the PrEP-E database, we identified 793 (74.9%) men documented to have taken at least one dose of PrEP following program enrolment, leaving 271 prevention program participants with no adherence documentation. Restricting our analysis to the 793 HIV-negative men with documented adherence, the adjusted odds of ES in HIV-positive men were 3·10 (95% CI 2·65, 3·62) times the odds in HIV-negative men.
Twenty-three HIV-positive men met the seroconversion criteria. After excluding seroconverters from the analysis, the adjusted odds of ES in HIV-positive men were 2·82 (95% CI 2·4, 3·32) times the odds in HIV-negative men. Among the 23 men who underwent seroconversion, 11 (47·8%) were diagnosed with ES during the observational period.
DISCUSSION
Among CHP participants, HIV-positive men had significantly higher odds of ES than HIV-negative men. The rates of CTGC infection did not differ according to HIV status. Given these observations, we believe that differences in individual sexual behaviors do not fully explain the higher incidence of ES observed in HIV-positive men. Possible explanations for our findings include assortative mixing with a higher background prevalence of syphilis, increased probability of syphilis transmission, and detection bias due to longer follow-up in HIV-positive men16.
Immunological mechanisms may lead to the increased transmission and acquisition of syphilis in HIV-positive men. A recent retrospective analysis from the Netherlands demonstrated higher rates of recurrent syphilis in HIV-positive MSM than in HIV-negative MSM with similar sexual behaviors15. HIV infection may lead to slow or blunted serological responses following syphilis treatment. However, the role of acquired immunity in protection against recurrent syphilis remains unclear17. In our study, we noted higher age-adjusted rates of baseline RPR reactivity in HIV-positive men than those in HIV-negative men. This may indicate higher rates of syphilis infection and reinfection, or persistence of serofast serology in adequately treated HIV-infected patients18,19.
HIV-positive men had a longer follow-up duration but fewer STI tests than HIV-negative men did. At CHP, prevention program participants undergo quarterly bacterial STI screening following a structured protocol, while testing in HIV-positive MSM is left to the clinician’s discretion20. We found higher rates of ES in HIV-positive men despite the lower number of RPR assays performed. All HIV-negative men with ES underwent CTGC testing in the 4-week window surrounding ES diagnosis, while 21·5% of HIV-positive men did not. High rates of CTGC detection have been observed in men with ES. Most CTGC infections were detected in rectal specimens, underscoring the need for extragenital STI screening in MSM at risk for HIV infection21,22.
Multiple strategies have been proposed to address the increasing incidence of ES in MSM. Screening frequency has been identified as a key intervention to reduce syphilis transmission. In a modelling study of MSM transmission networks in Switzerland, increasing the screening frequency from yearly to biannually was predicted to reduce the incidence of syphilis among HIV-positive MSM by 63%23. An Australian study concluded that quarterly screening is required to reduce the prevalence of syphilis among MSM24. Antimicrobial post-exposure prophylaxis [PEP] studies conducted in MSM and transgender women have shown a significant reduction in the incidence of syphilis and other bacterial STIs25,26. Several bacterial STI PEP trials are currently underway to determine the optimal targeting and potential drawbacks of this strategy27.
High rates of HIV viremia were also observed in our cohort. Forty-four percent of HIV-positive men had at least one episode of high-level viremia during the follow-up. The risk of high-level viremia was greater in men with ES, 25% of whom were viremic within three months of ES diagnosis. The high rates of coinciding viremia suggest that shared factors may lead to simultaneous loss of ART adherence, poor engagement in care, and syphilis acquisition. Among the few HIV seroconversions identified, 48% had ES during the observation period.
Our study is one of the largest retrospective cohorts of patients with early syphilis and is unique in its comparison of HIV-positive and HIV-negative men participating in a status-neutral CHP. Although this single-center study limits its generalizability, we believe that it allows for stronger internal validity and better comparability between HIV-positive and HIV-negative men. However, our study had several limitations. Our ES case definition relied solely on laboratory data. Given the large number of ES cases identified, we did not perform a clinical chart review to confirm our diagnosis. Nevertheless, we believe that the proposed diagnostic algorithm was robust. Using criteria very similar to ours, Menza et al. found very high specificity of serology-based algorithms for the diagnosis of ES in HIV-positive MSM, using clinical chart review as the gold standard28. Our findings remained significant even after sensitivity analyses of the ES criteria were performed.
We assumed that participation in the CHP program was a proxy for sexual risk in HIV-negative men. The high rates of STIs observed in HIV-negative men in our cohort support this hypothesis. However, data on sexual behavior, identity, or orientation were unavailable, precluding a comparison of differences in sexual practices by HIV status. We believe that most men in the cohort were MSM, based on clinical experience from our practice setting and the syphilis serology inclusion criteria of the study. While longitudinal data on PrEP adherence was not available, prior analyses of HIV-negative patients at CHP demonstrated PrEP uptake of 80% at the first prevention visit29. In the current analysis, the odds of ES remained higher in HIV-positive men compared to HIV-negative men in whom PrEP adherence was documented.
In conclusion, our findings demonstrate the ongoing vulnerability of HIV-positive men to early syphilis in an era of effective HIV biomedical prevention. Protocolized syphilis prevention strategies, including an increased frequency of STI screening in HIV-positive MSM, are essential to ensure timely detection and treatment of syphilis across the HIV risk spectrum.
Funding Statement:
Dr. Zucker was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (award number: K23AI150378).
Footnotes
Conflict of Interest Declaration: The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interests in the subject matter or materials discussed in this manuscript.
References
- 1.https://www.cdc.gov/std/statistics/2020/overview.htm#Syphilis. Accessed September 19, 2022.
- 2.Abara WE, Hess KL, Fanfair RN, et al. Syphilis Trends among Men Who Have Sex with Men in the United States and Western Europe: A Systematic Review of Trend Studies. 2004. PLoS ONE 11(7): e0159309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.The National Plan to Eliminate Syphilis from the United States - Executive Summary. https://www.cdc.gov/stopsyphilis/exec.htm. Accessed November 10, 2022.
- 4.HIV Surveillance Reports. https://www.cdc.gov/hiv/library/reports/hiv-surveillance/vol-32/index.html Accessed July 2, 2022.
- 5.Cohen MS, Chen YQ, McCauley M, et al. Antiretroviral Therapy for the Prevention of HIV-1 Transmission. N Engl J Med. 2016;375(9):830–839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rodger AJ, Cambiano V, Phillips AN, et al. Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. Lancet. 2019;393(10189):2428–2438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Grant RM, Lama JR, Anderson PL, et al. Preexposure Chemoprophylaxis for HIV Prevention in Men Who Have Sex with Men. N Engl J Med. 2010;27:2587–2599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Molina JM, Capitant C, Spire B, et al. On-Demand Preexposure Prophylaxis in Men at High Risk for HIV-1 Infection. N Engl J Med. 2015;373(23):2237–2246. [DOI] [PubMed] [Google Scholar]
- 9.Traeger MW, Schroeder SE, Wright EJ, et al. Clinical Infectious Diseases Effects of Pre-exposure Prophylaxis for the Prevention of Human Immunodeficiency Virus Infection on Sexual Risk Behavior in Men Who Have Sex With Men: A Systematic Review and Meta-analysis. Clin Infect Dis. 2018;67(5):676–86. [DOI] [PubMed] [Google Scholar]
- 10.Ramchandani MS, Golden MR. Confronting Rising STIs in the Era of PrEP and Treatment as Prevention. Curr HIV/AIDS Rep. Published online 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zetola NM, Klausner JD. Syphilis and HIV Infection: An Update. Clin Infect Dis. 2007;44:1222–1230. [DOI] [PubMed] [Google Scholar]
- 12.Traeger MW, Cornelisse VJ, Asselin J, et al. Association of HIV Preexposure Prophylaxis With Incidence of Sexually Transmitted Infections Among Individuals at High Risk of HIV Infection. JAMA. 2019;321(14):1380–1390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kojima N, Davey DJ, Klausner JD. Pre-exposure prophylaxis for HIV infection and new sexually transmitted infections among men who have sex with men. AIDS. 2016;30(14):2251–2252. [DOI] [PubMed] [Google Scholar]
- 14.Nguyen VK, Greenwald ZR, Trottier H, et al. Incidence of sexually transmitted infections before and after preexposure prophylaxis for HIV. AIDS. 2018;32:523–530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nieuwenburg SA, Sprenger RJ, Schim Van Der Loeff MF, et al. Clinical outcomes of syphilis in HIV-negative and HIV-positive MSM: Occurrence of repeat syphilis episodes and non-treponemal serology responses. Sex Transm Infect. 2022;98(2). [DOI] [PubMed] [Google Scholar]
- 16.Arando M, Fernandez-Naval C, Mota-Foix M, et al. Early syphilis: risk factors and clinical manifestations focusing on HIV-positive patients. BMC Infect Dis. (2019) 19:727 doi: 10.1186/s12879-019-4269-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Stoltey JE, Cohen SE. Syphilis transmission: a review of the current evidence. Sex Health. 2015. April; 12(2): 103–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ren M, Szadkowski L, Walmsley SL. Deciphering the serological response to syphilis treatment in men living with HIV. AIDS. 2020;34(14). [DOI] [PubMed] [Google Scholar]
- 19.Lin KY, Yang CJ, Sun HY, et al. Comparisons of Serologic Responses of Early Syphilis to Treatment with a Single-Dose Benzathine Penicillin G Between HIV-Positive and HIV-Negative Patients. Infect Dis Ther. 2021;10(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.CDC STI Screening Recommendations. Available at: https://www.cdc.gov/std/treatment-guidelines/screening-recommendations.htm. Accessed September 17, 2022.
- 21.Abara WE, Llata EL, Schumacher C, et al. Extragenital Gonorrhea and Chlamydia Positivity and the Potential for Missed Extragenital Gonorrhea with Concurrent Urethral Chlamydia among Men Who Have Sex with Men Attending Sexually Transmitted Disease Clinics-Sexually Transmitted Disease Surveillance Network, 2015–2019. Sex Transm Dis. 2020;47(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Johnson Jones ML, Chapin-Bardales C, Bizune D, et al. Extragenital Chlamydia and Gonorrhea Among Community Venue–Attending Men Who Have Sex with Men — Five Cities, United States, 2017. MMWR Morb Mortal Wkly Rep. 2019;68(14):321–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Balakrishna SI, Salazar-Vizcaya L, Schmidt AJ, et al. Assessing the drivers of syphilis among men who have sex with men in Switzerland reveals a key impact of screening frequency: A modelling study. PLoS Comput Biol. 2021. 17(10):e1009529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gray RT, Hoare A, Prestage GP, et al. Frequent testing of highly sexually active gay men is required to control syphilis. Sex Transm Dis. 2010;37(5). [DOI] [PubMed] [Google Scholar]
- 25.Molina JM, Charreau I, Chidiac C, et al. Post-exposure prophylaxis with doxycycline to prevent sexually transmitted infections in men who have sex with men: an open-label randomised substudy of the ANRS IPERGAY trial. Lancet Infect Dis. 2018;18(3):308–317. [DOI] [PubMed] [Google Scholar]
- 26.Luetkemeyer A, Dombrowski J, Cohen S, et al. Doxycycline post-exposure prophylaxis for STI prevention among MSM and transgender women on HIV PrEP or living with HIV: high efficacy to reduce incident STI’s in a randomized trial. Presented at: AIDS 2022, Montreal. [Google Scholar]
- 27.Klausner JD, Grant JS, Stafylis C, et al. Clinical Infectious Diseases Doxycycline Prophylaxis for Bacterial Sexually Transmitted Infections. Clin Infect Dis. 2020;70(6):1247–1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Menza TW, Levine K, Grasso C, et al. Evaluation of four algorithms to identify incident syphilis among HIV-positive men who have sex with men engaged in primary care. Sex Transm Dis. 2019. April; 46(4): e38–e41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Zucker J, Patterson B, Ellman T, et al. Missed Opportunities for Engagement in the Prevention Continuum in a Predominantly Black and Latino Community in New York City. AIDS Patient Care STDS. 2018;32(11). [DOI] [PMC free article] [PubMed] [Google Scholar]