Risk of HIV or second syphilis infection in Danish men with newly acquired syphilis in the period 2000–2010
- Kirsten Salado-Rasmussen1,
- Terese Lea Katzenstein1,
- Jan Gerstoft1,
- Susan Alice Cowan2,
- Troels Bygum Knudsen3,
- Lars Mathiesen3,
- Steen Hoffmann4,
- Niels Obel1
- 1Department of Infectious Diseases, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
- 2Department of Epidemiology, Statens Serum Institut, Copenhagen, Denmark
- 3Department of Infectious Diseases, Hvidovre University Hospital, Copenhagen, Denmark
- 4Microbiology and Infection Control, Statens Serum Institut, Copenhagen, Denmark
- Correspondence to Kirsten Salado-Rasmussen, Department of Infectious Diseases, M5132, Copenhagen University Hospital, Rigshospitalet, Blegdamsvej 9, Copenhagen Ø 2100, Denmark;
- Received 16 March 2012
- Revised 17 November 2012
- Accepted 26 November 2012
- Published Online First 27 December 2012
Objectives Risk of subsequent diagnosis of HIV in persons diagnosed with newly acquired syphilis, and syphilis in HIV-infected persons, are of interest as these infections are markers of unsafe sex.
Methods From a nationwide register, all Danish men aged >16 years diagnosed with syphilis in the period 2000–2010 (n=1217) were identified, and subsequently data on HIV status was extracted from the Danish HIV Cohort Study. We used Kaplan–Meier analysis to estimate the 5-year risk of HIV and second syphilis infection, and Cox regression to determine incidence rate ratios (IRR).
Results The 5-year risk of HIV diagnosis was 9.8% (95% CI 7.0% to 12.6%). Those with a second diagnosis of syphilis had a higher risk of being diagnosed with HIV (IRR=3.1, 95% CI 1.2 to 8.0). The 5-year risk for a second diagnosis of syphilis was 14.8% (95% CI 12.1% to 17.4%) and HIV-infected persons had a higher risk of a second syphilis diagnosis (IRR=4.0, 95% CI 2.8 to 5.9). Sixty-five percent of the persons were men having sex with men (MSM). Thirty-four percent of the HIV-infected persons had viral load >1000 copies/ml at time of syphilis diagnosis.
Conclusions The substantial risks of syphilis and HIV infection in men diagnosed with one of these sexually transmitted diseases indicate a high frequency of unsafe sex in the Danish MSM population. As one-third of the HIV-infected persons diagnosed with syphilis had high viral loads, our data support initiation of antiretroviral therapy in all HIV-infected MSM to reduce HIV transmission.
Since 1999, the number of newly reported cases of syphilis in Denmark has increased dramatically. In Denmark, newly acquired syphilis, that is, primary and secondary syphilis, is notified to Statens Serum Institut (the National Institute for Health Data and Diseases Control).1 Furthermore, the serological diagnosis of syphilis is centralised at Statens Serum Institut, and all persons notified and/or diagnosed with syphilis are registered in a nationwide database. In 2011, a total of 434 cases were diagnosed with newly acquired syphilis. The vast majority of the persons were men who had been infected in Denmark.1 By contrast, only 22 cases were diagnosed in 1999, of whom the majority were infected in Eastern Europe, Africa and Asia.2 In 2003, an outbreak of the disease among men who have sex with men (MSM) constituted 78% of the notified cases, whereas only 33% of the cases notified from 1994 to 2002 were MSM.3 The proportion of MSM coinfected with HIV and syphilis has varied over the years, peaking in 2008 with 58% of the MSM being diagnosed with syphilis also being infected with HIV.4 By contrast, 62% of the MSM diagnosed with syphilis in 2011 were HIV-negative—that is, susceptible to HIV—at the time of the diagnosis of syphilis.1 As a consequence of the high rates of syphilis among HIV-infected persons, syphilis screening has been implemented at the specialised medical centres where these patients are seen at intended intervals of 12 weeks.
Increasing rates of syphilis have also been seen throughout Europe, North America and Australia, and the majority of the new diagnoses were found among MSM.5 A possible change in behaviour may be attributed to factors such as loss of fear of contracting HIV due to improved outcome after highly active antiretroviral therapy (HAART).6 Furthermore, the reduced transmission of HIV during successful HAART7 may have led to a more risky sexual behaviour.
Since syphilis infection facilitates acquisition and transmission of HIV,8 and because an increase in risky sexual behaviour has been reported,9 concerns about a potential increase in HIV transmission have been raised—especially among MSM. We used two nationwide registers to describe the risk of HIV and syphilis among Danish men from 1 January 2000 to 1 January 2011. These infections are of interest as they are markers of sexual risk behaviour. We estimated the risk of HIV acquisition in persons who had been diagnosed with syphilis. Furthermore we estimated the risk of reinfection with syphilis.
We used the unique 10-digit Central Person Registration number (CPR number) assigned to all individuals in Denmark at birth or upon immigration, to link data from the following registers: the Danish National Syphilis Registration System and the Danish HIV Cohort Study.
Denmark has a population of 5.6 million people with an estimated HIV prevalence in the adult population of 0.07%.10 ,11 Approximately 44% of the HIV-infected population reported MSM contact as route of infection, and 74% reported Denmark as the country of origin.11 The number of persons with newly diagnosed HIV has been stable over the last 10 years as well as the number of persons infected by homosexual transmission.12 Five hundred undiagnosed HIV-infected MSM are estimated to reside in Denmark, an estimate that has been stable over the last 10 years.13 Denmark's tax-funded healthcare system provides antiretroviral treatment free of charge to all residents. In Denmark, treatment of HIV infection is restricted to eight specialised medical centres, where patients are seen on an outpatient basis at intended intervals of 12 weeks. Syphilis screening at least once a year has been implemented as part of the routine testing in connection with these outpatient visits. During the study period, the national criteria for HAART initiation were any of the following: acute HIV infection, presence of a HIV-related disease, pregnancy, CD4 cell count <300 cells/μl until May 2008 and <350 cells/μl thereafter, and plasma HIV-RNA>100 000 copies/ml (until 2001). The National Board of Health recommends that patients tested for syphilis be offered HIV testing.
All persons in Denmark diagnosed with newly acquired syphilis, that is, primary and secondary syphilis are registered in the Danish National Syphilis Registration System which was established in 1993. According to Danish law, newly acquired syphilis is a notifiable disease and the notification to the register is done by the treating physician. Data collected include route of infection, country of infection and result of prior HIV testing. Patients may be registered with more than one episode of syphilis. In addition to the cases notified to the Danish National Syphilis Registration System, the register also includes persons indentified with syphilis by positive serology or PCR. To assure anonymity, only the last six numbers of the 10-digit CPR number are registered in the Danish National Syphilis Registration System. To obtain the full CPR number, the reporting forms were matched with CPR numbers from persons with positive serology. In case of multiple options, the persons were matched according to ethnicity and geographical data. The procedure was approved by the Danish Data Protection Agency.
The Danish HIV Cohort Study, which has been described in detail elsewhere,14 is a nationwide, prospective, population-based study of all Danish HIV-infected persons treated at Danish hospitals since 1 January 1995. The data is updated yearly and includes demographics, route of infection, CD4 cell counts, viral loads and antiretroviral treatment. The unique 10-digit CPR number is used to avoid multiple registrations and to track individuals in the Danish HIV Cohort Study. The study was approved by the Danish Data Protection Agency.
All male patients aged >16 years registered in the Danish National Syphilis Registration System with a diagnosis of newly acquired syphilis after 1 January 2000 were included. Date of study inclusion was the date of the first syphilis diagnosis after 1 January 2000. The study period was from 1 January 2000 to 1 January 2011.
Based on data from the Danish National Syphilis Registration System, the yearly number of men diagnosed with syphilis was calculated. From the Danish HIV Cohort Study, we obtained data on HIV status, and the yearly number of men diagnosed with syphilis was stratified according to HIV status. For persons infected with HIV at the time of syphilis diagnosis, information on date of HIV diagnosis, start of antiretroviral treatment and viral load was extracted. HIV-RNA at date of first syphilis episode after HIV diagnosis was defined as the HIV-RNA closest to the date of syphilis diagnosis. We calculated the fraction of HIV-infected persons diagnosed with syphilis who had a viral load <1000 copies/ml. Our objective was to estimate the fraction with suppressed viral load, therefore, persons with missing HIV-RNA were allocated to the group with >1000 copies/ml. For the population diagnosed with syphilis, but not with HIV at the time of syphilis diagnosis, the time to HIV diagnosis was determined. In these analyses, the time from first diagnosis of syphilis within the study period to date of HIV diagnosis or 1 January 2011, whichever came first, was calculated. Time-to-event curves from Kaplan–Meier tables were constructed and stratified by the diagnosis of syphilis before and after 1 January 2006, respectively. We used Cox regression analysis with the date of the second syphilis diagnosis introduced as the time-updated variable to determine the incidence rate ratio (IRR) for risk of HIV diagnosis before and after a second diagnosis of syphilis. Likewise, the time to a second syphilis diagnosis was determined for the total population and Kaplan–Meier tables were stratified by syphilis diagnosis before and after 1 January 2006. We used Cox regression analysis with the date of HIV infection introduced as the time-updated variable to determine IRR for risk of a second syphilis infection.
The study was approved by the Danish Data Protection Agency. SPSS V.15.0 (SPSS Inc, Chicago, Illinois, USA) was used for data analysis.
Syphilis testing methods
The serological diagnosis of syphilis is centralised at Statens Serum Institut.15 Sera samples from patients were examined for the presence of syphilis antibodies. Two non-treponemal tests were used; Wassermann's reaction was done with a complement fixation technique, and rapid plasma reagin was determined by agglutination. Furthermore, treponemal tests were used on seroreactive samples; Antiflagel IgG was determined by a capture ELISA and IgM by a direct ELISA. The fluorescent treponemal antibody absorption test was done by immunofluorescence microscopy. If an individual had tested seropositive for syphilis several times, the serology was carefully evaluated and compared with the reporting forms (mandatory for the treating physician). The results were regarded as being diagnostic of a new infection if a substantial increase in syphilis antibodies was evident, preceded by a decrease in antibody titre documenting cure of the prior episode. A few patients with genital chankres were diagnosed by PCR testing of material from the lesions.
During the 11-year study period from 1 January 2000 to 1 January 2011, a total of 1536 episodes of syphilis were diagnosed among 1361 persons, including 144 women, leaving 1217 men in the present study. In our study population, the median age was 37.8 years (table 1). The yearly number of men diagnosed with syphilis increased substantially in the study period irrespective of HIV infection, from 38 in 2000 to 344 in 2010 (figure 1). Sixty-five percent of the men with syphilis were notified as MSM on the syphilis reporting form, 14% were notified as heterosexual and finally 21% of the reporting forms were lacking information regarding the route of transmission.
Men diagnosed with HIV prior to diagnosis of syphilis
A total of 359 men diagnosed with syphilis were diagnosed with HIV prior to diagnosis of syphilis, the majority of whom reported MSM as route of HIV infection (table 1). Sixty-six percent of the HIV-infected population had a viral load <1000 HIV-RNA copies/ml at the time of diagnosis of syphilis, and thereby, 33.7% of the HIV-infected population diagnosed with syphilis had viral loads >1000 HIV-RNA copies/ml. Data on viral load at the time of syphilis diagnosis were missing from 13 patients, and in these cases, the viral load was defined as >1000 HIV-RNA copies/ml. Other characteristics are described in table 1.
Men not diagnosed with HIV prior to diagnosis of syphilis—risk of HIV acquisition
At the time of diagnosis of syphilis, 858 men were not diagnosed with HIV. From this cohort, 64 men subsequently acquired HIV within the study period. After 5 years, 9.8% (95% CI 7.0 to 12.6%) of the population had been diagnosed with HIV. Figure 2 illustrates time from first syphilis diagnosis to HIV diagnosis stratified on men being diagnosed with syphilis before and after 1 January 2006, respectively. In a Cox model with the date of second diagnosis of syphilis introduced as a time-updated variable, the risk of being diagnosed with HIV was substantially higher after the second diagnosis of syphilis (IRR=3.1, 95% CI 1.2 to 8.0), that is, men with minimum two episodes of syphilis had a higher risk of being diagnosed with HIV compared with men with only one episode of syphilis within the study period.
Risk of reinfection with syphilis
After 5 years, 14.8% (95% CI 12.1 to 17.4%) of the population had been diagnosed with a second episode of syphilis. Figure 3 illustrates the time from the first diagnosis of syphilis to the second episode within the study period before and after 1 January 2006, respectively. In a Cox model with the date of HIV infection introduced as a time-updated variable, HIV-infected persons had a higher risk of being reinfected with syphilis compared with syphilis patients not diagnosed with HIV within the study period (IRR=4.0, 95% CI 2.8 to 5.9).
In a nationwide cohort of Danish men diagnosed with newly acquired syphilis, we observed a very high risk of both prior and subsequent diagnoses of HIV and of reinfection with syphilis during an 11-year follow-up period. A major strength of the study is our ability to link two nationwide registers of persons diagnosed with syphilis and HIV.
During the last decade, Denmark has experienced a substantial increase in the number of persons diagnosed with syphilis, the majority of whom were men. From our data, the dominant profile of the epidemic was MSM. This is in line with the findings from a number of studies which indicate that the resurgence of syphilis is geographically widespread and outbreaks have been seen throughout Europe, North America and New Zealand. Overall, the affected population was MSM.16–19 The marked increase in syphilis among MSM has mainly been attributed to behavioural factors, that is, a more risky sexual behaviour in the wake of the impressive impacts of antiretroviral therapy on HIV progression rates.20 ,21 Supporting this assumption is the rise in other sexually transmitted diseases among MSM, for example, Chlamydia trachomatis lymphogranuloma venereum variant and gonorrhoea.22 Whether the increase is due to a general increase in sexual risk behaviour among MSM or due to the establishment of a core group with high-risk behaviour is not known.23 Our finding that HIV-infected persons had a four times greater risk of acquiring a second case of syphilis compared with persons diagnosed with syphilis without HIV could support the existence of such a core group with a high-risk behaviour. But unsafe sex among all MSM in Denmark seems to be increasingly common.9
An alternative explanation has also been put forward by Grassly et al24 who suggested that the increase in syphilis was mainly due to endogenous oscillation in disease incidence predicted by the natural dynamics of the infection. They found support for this theory in the different dynamics of syphilis and gonorrhoea incidences. These conclusions have, however, been questioned.25
Thirty percent of the persons diagnosed with syphilis were diagnosed with HIV prior to being diagnosed with syphilis. The majority of these persons were under HAART treatment but only two-thirds of the HIV-infected persons had HIV-RNA levels below 1000 copies/ml. This figure is lower than that observed in 2004 in the national database for all HIV-infected persons.26 As previously mentioned, the HIV-population had a higher risk of a second syphilis diagnosis, which may indicate a more risky sexual behaviour. As one-third of the HIV-infected persons diagnosed with syphilis had high viral loads, our data supports initiation of antiretroviral therapy in all HIV-infected MSM to reduce HIV transmission
The aim of the current study was to describe the pattern of syphilis among men in Denmark, and also to investigate whether being diagnosed with syphilis was predictive for subsequent HIV acquisition. Almost 10% of the population diagnosed with syphilis was subsequently diagnosed with HIV within 5 years. Interestingly, the risk of being diagnosed with HIV subsequent to receiving a syphilis diagnosis decreased during the latter part of the study period. Similar trends where outbreaks of syphilis have not had a substantial impact on HIV incidence have been observed in North America.27 A possible explanation of the stagnation of the HIV transmission despite the increase in syphilis diagnoses could be that antiretroviral treatment has become more widely used in recent years with viral suppression as a consequence. Even without taking into account the decreasing HIV transmission risk, the finding could be explained by the increase in the number of persons diagnosed with syphilis during the study period.
By contrast to the risk of being diagnosed with HIV, the risk of being diagnosed for the second time with syphilis did not decrease from the first to the second part of the study period. This is in line with the increase in syphilis incidence that was observed especially in the latter part of the study period.
We found a substantially higher risk of being diagnosed with HIV after a second diagnosis of syphilis indicating that syphilis is a marker for sexual risk behaviour. This is in line with a study among high-risk HIV-uninfected MSM in Peru where newly acquired syphilis (and herpes simplex virus type 2), as well as sex with a casual partner, were found to be associated with incident HIV infection.28
The present study combines data from nationwide registers and provides a well-founded estimate about the current syphilis epidemic among men in Denmark. However, some limitations to our study must be addressed. First, a possible bias is caused by selective syphilis screening of HIV-infected persons as part of the routine testing of this group, thus underestimating the syphilis prevalence in low-risk groups. Therefore, an unknown number of cases of syphilis have evidently been left out since low-risk groups have not been screened. Second, our study includes both MSM and heterosexual men who have different risk behaviours and exposures. Even though information on route of transmission is supposed to be stated on the syphilis reporting forms, this data was rather incomplete since information on this variable was missing in 21% of the forms. We expect that the actual fraction of MSM may be significantly higher than the 65% who reported MSM as route of infection on the syphilis reporting form. Third, our data analysis did not account for mortality and emigration since this data was not available for the HIV-uninfected population. This could potentially have an impact on our risk estimates because of the competing risk of fatality making the actual risk higher. Another limitation of the study lies in the definition of a ‘case of syphilis’ as it requires interpretation to distinguish between a reinfection and resurgence of an incompletely treated infection.
In conclusion, our study demonstrates that men diagnosed with syphilis have a high risk of a subsequent HIV diagnosis, and men coinfected with HIV and syphilis have an increased risk of a second syphilis diagnosis. Furthermore, the study underlines that a substantial fraction of the HIV-infected men with confirmed HIV viremia are diagnosed with syphilis, thereby indicating that unsafe sex is prevalent in the Danish HIV-infected MSM population with the concomitant risk of transmitting HIV. This finding supports the thesis that offering antiretroviral therapy to all HIV-infected MSM could potentially reduce HIV transmission.
Finally, the high rates of syphilis and HIV among men who have already been in contact with the Danish healthcare system reflect a major missed opportunity for risk reduction in terms of postcounselling.
Danish men diagnosed with syphilis have a high risk of subsequent HIV diagnosis.
Danish men coinfected with HIV and syphilis have a higher risk of being reinfected with syphilis.
These patients are frequently in contact with the Danish healthcare system. Efforts must be made to reduce risk behaviour.
Contributors KS, TK, JG and NO collaborated in the writing of the manuscript. NO designed the study and performed statistical analyses. SC chose the main directions and provided data. SC, TB, LM and SH revised the manuscript before submission, and complemented it. All authors take responsibility for the integrity of the data.
Competing interests TK has received payment for board membership, consultancy and lectures from Gilead, Viiv, Bristol–Myers Squibb, Merck Sharo and Dohme, Abbott, GlaxoSmithKline, Roche, Boehringer Ingelheim, Janssen and Sanofi-pasteur. JG has received research funding from Gilead, Viiv, Bristol–Myers Squibb, Merck Sharo and Dohme, Abbott, GlaxoSmithKline, Roche, Boehringer Ingelheim, Janssen and Sanofi-pasteur. LM has received payment for consultancy for Gilead, Bristol–Myers Squibb, GlaxoSmithKline and Janssen. NO has received research funding from Bristol–Myers Squibb, Merck Sharo and Dohme, Abbott, Roche, Boehringer Ingelheim, Janssen–Cilag and Swedish Orphan Drugs.
Ethics approval The Danish Data Protection Agency.
Provenance and peer review Not commissioned; externally peer reviewed.