During 1999 to 2008, cases of early syphilis (including primary and secondary [P and S] and early latent [EL] syphilis) in San Diego County increased over 1000%, from 47 to 524. During that time, cases among men who have sex with men (MSM) increased over 2800%, from 14 to 411, accounting for 76.6% of cases overall in the County (County of San Diego, unpublished data).
Sexually transmitted disease (STD) prevention and control efforts, including those for syphilis, have traditionally focused on core groups of transmitters.1 In other jurisdictions, MSM diagnosed with syphilis have been shown to be at high risk for repeat cases of syphilis. In San Francisco, 6.7% of MSM who were diagnosed with early syphilis during 2001–2002 had another syphilis diagnosis within 1 year, with HIV-infected MSM more likely to have repeat syphilis (adjusted odds ratio[OR]: 5.2; 95% confidence interval [CI]: 2.0–13.7).2 In Philadelphia, 3.4% of persons over 14 years old who were diagnosed with any stage of syphilis during 2002–2008 had another diagnosis of syphilis during that period. In that group, among men, repeat syphilis was significantly more common among MSM and those younger than 30 years of age.3 During 2000–2008, in Florida, 2.5% of persons with syphilis had had a prior syphilis diagnosis, which was more common among MSM and persons who were HIV-infected, white, 25 to 29 years old, or resided in Miami-Dade or Broward Counties. In that study,4 as well as a study in Seattle,5 HIV-infected MSM were more likely to have EL syphilis when diagnosed with repeat syphilis than they were when diagnosed initially. Finally, among persons diagnosed with syphilis at 2 Baltimore City STD clinics in 1988, 23% had a prior syphilis diagnosis. In that study, MSM and men who used injection drugs were significantly more likely than other men to have had a prior diagnosis.6
To help focus syphilis prevention and control efforts in San Diego County, we analyzed the 2-year period prevalence of repeat syphilis among MSM in the County during January 2004 to June 2007. The proportion of EL cases in repeat syphilis were compared with initial infections; the proportion of cases of repeat syphilis that were EL among HIV-infected were compared with HIV-uninfected MSM; investigated factors potentially associated with repeat syphilis; and constructed a prediction model to identify MSM at highest risk of repeat syphilis who could be targeted for more intense syphilis prevention and control efforts.
We conducted a retrospective cohort study. We defined MSM as men who reported having sex with another man in the year before diagnosis. We obtained demographic, clinical, and behavioral data on characteristics of MSM diagnosed with syphilis from records of syphilis interviews, which are routinely done by Disease Investigation Specialists for County residents diagnosed with early syphilis. We analyzed only characteristics obtained during interviews for initial episodes of syphilis. We defined HIV infection by self report or chart review. We defined repeat syphilis as a case of syphilis occurring within 2 years in a previously appropriately treated man who had a 4-fold titer decrease and then experienced either signs of syphilis or a 4-fold titer increase. We defined appropriate treatment and syphilis stages according to CDC guidelines relevant at the time of diagnosis.7,8 Because the objective of the study was to identify characteristics of MSM at high risk of repeat syphilis during a 2-year period, who could be targeted for enhanced interventions, we focused only on the initial diagnosis of syphilis and the first episode of repeat syphilis during the subsequent 2 years. Therefore, MSM diagnosed with >1 case of syphilis during 2004–2007 were only counted once, for their initial and first (if any) repeat syphilis diagnoses during that period.
We used the Cochran-Mantel-Haenszel test to compare the proportion of EL cases in initial infections compared with repeat infections among those with repeat syphilis infections. We used a prevalence ratio to compare the proportion of repeat syphilis cases that were EL among HIV-infected compared with HIV-uninfected MSM. We used univariate logistic regression to identify patient characteristics that significantly predicted repeat syphilis. We further entered characteristics associated with a repeat syphilis with P < 0.1 into a multivariate model, retaining those with P < 0.05 using a backward stepwise procedure, and calculated the area under the curve (AUC) for the final model. An AUC >0.5 indicates that a model predicts an outcome better than chance alone, with an AUC of 1 indicating perfect predictive ability.9 Statistical analyses were performed using Stata 9 (Stata Corp, College Station, TX) and SAS version 9.2 (SAS Institute Inc, Cary, NC). This study was determined to be a nonresearch public health surveillance project by the County of San Diego.
Between January 2004 and June 2007, 634 MSM in San Diego County were diagnosed with early syphilis and were interviewed by Disease Investigation Specialists. Stage of infection, cross tabulated by initial and repeat infection, is shown in Table 1. Within 2 years, 74 (11.7%; 95% CI: 9.3%–14.4%) had repeat syphilis. Those 74 cases included 16 primary cases (21.6%), 28 secondary cases (37.8%), and 30 EL cases (40.5%). EL cases were staged as such on the basis of a 4-fold titer increase in the previous year (n = 20; 66.7%); a negative serologic test for syphilis in the previous year (n = 7; 23.3%); presence of primary and secondary symptoms in the previous year (n = 2; 6.7%); or exposure to an individual with confirmed early infection in the previous year (n = 1; 3.3%). Among those with repeat infections, the proportion of EL cases in repeat syphilis compared with initial infections did not differ significantly (prevalence ratio of EL cases in repeat compared with initial infections, 1.18; 95% CI: 0.94–1.48, P = 0.14). The proportion of repeat cases that were EL among HIV-infected compared with HIV-uninfected MSM did not differ significantly (prevalence ratio of repeat cases that were EL among HIV-infected compared with HIV-uninfected MSM 1.39; 95% CI: 0.72, 2.67; P = 0.30).
Characteristics of MSM with and without repeat syphilis, and the association between those characteristics and repeat syphilis, are shown in Table 2. HIV status and condom use for most recent receptive anal sex were entered into a multivariable model. In a model with HIV status, condom use for most recent receptive anal sex were not significantly associated with repeat syphilis (P = 0.25). The final model, therefore, included only HIV status, with HIV infection associated with a greater OR of repeat syphilis (OR: 1.9; 95% CI: 1.1, 3.4). The AUC was 0.57.
These findings show that nearly 12% of MSM in San Diego County diagnosed with early syphilis had another case of early syphilis within 2 years of the first infection. HIV-infected MSM had nearly twice the OR of repeat syphilis within 2 years compared with HIV-uninfected MSM. The link between repeat syphilis and HIV infection might result from serosorting, which is the practice of seeking sex partners of the same HIV status. Factors that might increase risk of repeat syphilis among HIV-infected MSM who are engaged in serosorting include lack of condom use10 and a higher prevalence of syphilis among HIV-infected MSM compared with HIV-uninfected MSM.11 The AUC, 0.57, indicates that the prediction model can predict, using HIV status, which MSM will have repeat syphilis with an accuracy modestly better than chance alone.
Our findings are similar to those from the San Francisco study2 in which HIV infection was the only factor significantly associated with increased risk of repeat syphilis. Because of the different time periods of our study and the San Francisco study, we cannot directly compare the percentage of MSM experiencing repeat syphilis in San Diego County with that in San Francisco.2 The Philadelphia, Florida, and Baltimore studies did not specify a specific time interval for repeat syphilis, and included all persons diagnosed with syphilis.3,4,6 Unlike the Philadelphia study,3 we did not find an association between age and repeat syphilis. Unlike the Seattle and Florida studies, we did not find an increased proportion of EL cases among repeat compared with initial infections, or an increased proportions of repeat cases that were EL among HIV-infected MSM, respectively.4,5
This study is subject to at least 4 limitations. First, the association between HIV infection and repeat syphilis, and especially EL syphilis, might result from ascertainment bias because HIV-infected MSM are more likely to get tested after being diagnosed with syphilis compared with HIV-uninfected MSM.12,13 Second, because of a relatively small sample size, we had limited power to detect differences in proportions of EL cases in various groups. Third, we did not have adequate information on serosorting practices or condom use besides the most recent episode of receptive anal sex that might help explain the higher prevalence of repeat syphilis among HIV-infected MSM in this study. Fourth, we did not have sufficient data to both derive and validate our prediction model.
The County of San Diego and a community-based partner organization have designed and implemented a campaign, called “We All Test,” that encourages MSM to sign up at the campaign website to receive text message and/or email reminders every 3 to 6 months to get tested for syphilis.14 The campaign is aimed at all MSM in the County, regardless of HIV status or syphilis history. However, on the basis of the findings of this study, and in light of limited resources, the campaign initially offered a monetary incentive ($30) to register to receive testing reminders only to HIV-infected MSM diagnosed with early syphilis. However, because of a recent decline in syphilis cases in San Diego County, the campaign now has sufficient resources to offer incentives to all MSM diagnosed with early syphilis, regardless of HIV status.
In summary, MSM in San Diego County who were diagnosed with syphilis are at high risk of repeat syphilis within 2 years, particularly if they are HIV infected. An intervention such as “We All Test,” which follows the model of STD prevention and control that focuses on core transmitters, represents a potentially valuable tool to combat the syphilis epidemic in San Diego County.
1. Thomas JC, Tucker MJ. The development and use of the concept of a sexually transmitted disease core. J Infect Dis 1996; 174:S134–S143.
2. Phipps W, Kent CK, Kohn R, et al. Risk factors for repeat syphilis in men who have sex with men, San Francisco. Sex Transm Dis 2009; 36:331–335.
3. Newbern EC, Anschuetz G, Salmon M, et al. Syphilis again and again—syphilis re-infection in Philadelphia, 2002–2009. Poster session presented at: 2010 National STD Prevention Conference, March 8–11, 2010, Atlanta, GA.
4. Brewer TH, Peterman TA, Newman DR, et al. Reinfections during the Florida syphilis epidemic, 2000–2008. Sex Transm Dis. In press.
5. Kerani R, Lukehart S, Stenger M, et al. Is early latent syphilis more likely in patients with a prior syphilis infection? Paper presented at: 18th Conference of the International Society for STD Research, June 28–July 1, 2009, London, United Kingdom.
6. Hutchinson CM, Rompalo AM, Reichart CA, et al. Characteristics of patients with syphilis attending Baltimore STD clinics. Multiple high-risk subgroups and interactions with human immunodeficiency virus infection. Arch Intern Med 1991; 151:511–516.
7. CDC. Sexually transmitted diseases treatment guidelines. MMWR 2002; 51(RR-6):1–80.
8. CDC. Sexually transmitted diseases treatment guidelines. MMWR 2006; 55(RR-11):1–94.
9. Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: Issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996; 15:361–387.
10. Osmond DH, Pollack LM, Paul JP, et al. Changes in prevalence of HIV infection and sexual risk behavior in men who have sex with men in San Francisco: 1997–2002. Am J Public Health 2007; 97:1677–1683.
11. Peterman TA, Furness BW. The resurgence of syphilis among men who have sex with men. Curr Opin Infect Dis 2007; 20:54–59.
12. Marcus JL, Katz KA, Bernstein KT, et al. Syphilis testing behavior following diagnosis with early syphilis among men who have sex with men—San Francisco, 2005–2008. Sex Transm Dis. In press.
13. Horberg MA, Ranatunga DK, Quesenberry CP, et al. Syphilis epidemiology and clinical outcomes in HIV-infected and HIV-uninfected patients in Kaiser Permanente Northern California. Sex Transm Dis 2010; 37:53–58.
14. WeAllTest.com [Internet]. San Diego: Family Health Centers of San Diego and County of San Diego. Available at: http://www.wealltest.com/
. Cited June 23, 2010.