Against a backdrop of increasing diagnoses of acute STIs and HIV prevalence
in the United Kingdom (UK), diagnoses of infectious syphilis have risen
dramatically since 1998 . This resurgence has been facilitated by
a number of outbreaks throughout the UK. The outbreaks have occurred
mainly in men who have sex with men (MSM), and, more recently, in heterosexual
men and women .
Comprehensive, existing routine STI surveillance systems from genitourinary
medicine (GUM) clinics are unable to provide timely data in the context
of a rapidly evolving epidemic; data often taking a minimum of 6 months
to collate. Therefore, enhanced syphilis surveillance systems were
developed and implemented in order to improve clinical case reporting.
Similar enhanced surveillance systems now operate throughout the UK
The aim of this article is to describe the epidemiology of infectious
syphilis in the UK in 2003. We also explore key features of the London
infectious syphilis outbreak.
Routine surveillance of syphilis
Routine surveillance data on STIs in the UK are derived from diagnoses
made in GUM clinics reported on the KC60 form (ISD(D)5 form in Scotland).
GUM clinics have had a statutory obligation to record data since 1917
. Reliable trend data on primary, secondary and early latent syphilis
diagnosed in GUM clinics extend back to 1931.
GUM clinics in England, Wales and Northern Ireland return quarterly
data to the Health Protection Agency (HPA) on total episodes by condition,
sex and for selected conditions, by sexual orientation and/or age group.
In contrast, Scottish data are episode based and returned to the Information
and Statistics Division (now Information Services) in Scotland. Reported
data includes primary and secondary syphilis, early latent syphilis,
other acquired syphilis (e.g. cardiovascular and neurosyphilis), congenital
syphilis, and epidemiological treatment of suspected syphilis .
Routine GUM data returns are often delayed; for example, complete
KC60 data for 2003 is only available in June 2004. Similarly, difficulties
extracting Scottish ISD(D)5 data have resulted in incomplete or no
data being available currently for 2001, 2002 and 2003.
Data on syphilis and other STIs diagnosed at GUM clinics are made
publicly available in a series of annual reports, on the HPA website
(UK data), and both the Information and Statistics Division (ISD) and
Scottish Centre for Infection and Environmental Health (now Health
Protection Scotland, HPS) websites (Scottish data only) [(3-5].
Enhanced surveillance initiatives
In response to the resurgence in syphilis since the late 1990s, a
number of enhanced surveillance initiatives were implemented. These
initiatives were designed to provide prompt demographic, behavioural
and clinical data in order to inform health planning and intervention
strategies. The first enhanced surveillance programme commenced in
Manchester in 1999 and was extended to cover the North West region
in 2003. The London Enhanced Syphilis Surveillance programme was established
in 2001. This was subsequently extended to the rest of England and
Wales in 2003 . A similar system, based on the London programme,
was established in Scotland in late 2002; data were collected retrospectively
to 2001 . Table 1* shows the data collected
for the London Enhanced Syphilis Surveillance Programme. Other initiatives
Epidemiological features of the London outbreak were analysed with STATA
v8.0, and chi squared tests were used to ascertain P-values for differences
Diagnoses in infectious (primary, secondary and early latent) syphilis
declined rapidly during the 1980s with the advent of HIV/AIDS and the
subsequent introduction of HIV prevention strategies aimed at sexual
behaviour modification. A relatively low level of diagnoses was maintained
through most of the 1990s; between 1995 and 1998 an average of only
300 diagnoses were seen annually throughout the UK. The first outbreak
of infectious syphilis occurred in Bristol in 1997 . This was followed
by outbreaks in the cities of Manchester , Brighton, Peterborough,
London, Newcastle upon Tyne, Glasgow, Edinburgh, Walsall and the regions
of south Wales and Northern Ireland [FIGURE 1].
Routine surveillance of syphilis in the UK
GUM diagnoses of infectious syphilis are now at their highest levels
in the UK since 1984. A total of 2233 diagnoses were made in GUM clinics
during 2003; 1028 in men who have sex with men (MSM), 860 in heterosexual
men, and 338 in women. Since 1998, there has been a 25-fold increase
in MSM (from 43 to 1028 diagnoses). Rises of a lower magnitude of 6-fold
(138 to 860) and 3-fold (112 to 338) were seen in heterosexual men
and women respectively (FIGURE 2). There is a continuing divergence
in male heterosexual cases and female cases since 2000. In 2000, the
ratio of heterosexual male to female cases was 1.2:1, in 2003 it was
Increases in infectious syphilis were mirrored in other forms of syphilis
in England, Wales, and Northern Ireland (Scottish ISD(D)5 data not
available in 2003). Other acquired syphilis rose by 108% (76 to 158)
between 1998 and 2003 in MSM, by 55% (564 to 874) in heterosexual men
and 117% (376 to 817) in women. This was accompanied by small numbers
of congenital syphilis cases. There were also increases in the epidemiological
treatment of suspected syphilis from 0 cases in 1998 to 147 cases in
2003 in MSM, from 20 to 109 case in heterosexual men and 36 to 67 cases
The distribution of regional patterns of syphilis diagnoses largely
reflects the impact of outbreaks. In 2003, London accounted for 42%
of diagnoses in MSM, 42% in heterosexual men and 52% in women. The
North West (18%, 12% and 8% in MSM, heterosexual men and women respectively)
and South East regions of England (12%, 10% and 11% in MSM, heterosexual
men and women respectively) also accounted for a high proportion of
Age group data are currently only available for primary and secondary
syphilis in England Wales and Northern Ireland. Unlike other STIs,
relatively few diagnoses were made in younger age groups. The highest
rates of syphilis were seen in men aged 25 to 34 years (13.5 per 100,000)
and 35 to 44 years (11.7 per 100,000). In women the highest rates were
seen in those aged 20 to 24 years (2.5 per 100,000) and 25 to 34 years
(1.9 per 100,000) (8).
Enhanced syphilis surveillance in London
The London outbreak is the largest reported in the UK to date with 1910
diagnoses of infectious syphilis reported between April 2001 and end
September 2004. The characteristics of the outbreak were similar to
those seen throughout the rest of the UK, other areas of western Europe
and the United States . Infections are geographically clustered,
and associated with high rates of partner change in core risk groups,
and concurrent HIV infection.
Two epidemics are evident in London: one among MSM (1276 cases) and
one among heterosexual men (383 cases) and women (237 cases) (FIGURE
3). As seen in the routine surveillance data there is a disparity between
heterosexual male and female diagnoses.
In both MSM and heterosexuals the majority of cases attended GUM clinics
with symptoms (61% and 53% respectively) or for routine asymptomatic
screening (27% and 20% respectively). There were significant differences
in the other characteristics of MSM and heterosexuals diagnosed with
infectious syphilis [TABLE 2]. MSM with infectious syphilis were older
than heterosexuals, more likely to be HIV positive and more likely
to present with secondary syphilis [TABLE 3]. Sixty-five per cent of
MSM were born in the UK and 89% were of white ethnicity. In contrast,
only 46% of heterosexuals were born in the UK and 44% were of white
Most infections were acquired in London: 85% in MSM and 76% in heterosexuals.
MSM were significantly less likely to acquire their infection abroad.
In general, MSM reported higher prevalence of sexual risk behaviour.
They reported a higher number of sex partners (a median of three in
the last three months compared with one in heterosexuals); and a higher
proportion used social venues/sexual networks for acquiring new partners
(36% versus 7% in heterosexuals). Links with the commercial sex industry
were evident in heterosexuals. Thirteen percent of those diagnosed
with infectious syphilis were either commercial sex workers (CSW) or
CSW clients. Only three percent of MSM had links with CSW.
Syphilis in heterosexuals
When comparing syphilis cases in heterosexual men and women, men were
significantly older than women and more likely to present with primary
syphilis [TABLE 3].Two thirds of heterosexual men attended due to symptoms,
compared with just 30% of women. A further 29% of women attended due
to other reasons (e.g. a positive antenatal screen). A higher proportion
of heterosexual men reported using venues/sexual networks for acquiring
new partners (10% versus 3% in women), and oral sex being the likely
mode of transmission (11% versus 4% in women). Heterosexual men also
reported higher numbers of partners: a median of two in the previous
three months versus a median of one in women.
A common feature of syphilis outbreaks in England is the high proportion
of concurrent HIV infection in MSM diagnosed with infectious syphilis
. In London, 53% of MSM were HIV positive; this has remained fairly
stable throughout the outbreak. HIV co-infection in MSM with syphilis
was strongly associated with age group, stage of syphilis infection,
reason for attending, and use of sexual networks [TABLE 4]. Forty-one
per cent of those with concurrent HIV infection frequented sexual venues
compared with 31% in those who were HIV negative. There was no discernable
difference between HIV positive and negative MSM in terms of where
the infection was acquired, country of birth, ethnicity, oral sex as
mode of acquisition, CSW links or numbers of sexual contacts.
Routine surveillance data confirm continuing increases in syphilis
diagnoses during 2003 in MSM and heterosexual men, and to a lesser
extent in women. The balance of the epidemic remains in MSM, despite
data up to end September 2004 which suggests that the London epidemic
in MSM is plateauing. However, this is not consistent throughout the
UK, and preliminary 2004 data from enhanced surveillance in Scotland
indicate continuing increases in MSM in Glasgow and Edinburgh.
Nationally, our surveillance data confirm a continued divergence between
diagnoses in heterosexual men and women; a trend also observed in London
during 2002 and 2003 [FIGURE 2]. The excess male cases may have resulted
from the association between heterosexual outbreaks and the commercial
sex industry. The divergence may also be due to differences in clinical
presentation and health seeking behaviour [TABLE 3]. This conflicts
with London enhanced surveillance [FIGURE 3], where some of the convergence
may be due to reporting bias.
Key features of syphilis epidemiology in the UK include the geographical
isolation of outbreaks, especially amongst MSM where there was little
imported infection; localisation amongst CSW and their clients with
a steady increase in heterosexual transmission, and the high proportion
of concurrent HIV infection in MSM.
The potential impact of syphilis infection on HIV transmission is
concerning, and further studies examining the impact on HIV incidence
are now needed. Also worrying is the increased risk of congenital syphilis
cases which may accompany the rise in heterosexual transmission. Whilst
there have been ad hoc reports of congenitally acquired syphilis associated
with heterosexual outbreaks, there is a conspicuous lack of surveillance
activity in this area which needs to be tackled urgently.
We would like to thank the staff of all the GUM clinics that contributed
to the enhanced surveillance initiative. We would also like to thank
Dr. Chalmers and staff at Information Services, NHS National Services
Scotland for providing ISD(D)5 data, Dr. Thomas at CDSC, National Public
Health Service Wales, Dr. Gorton at HPA North East, Mr. Ashton at HPA
North West, Dr. Fox at CDSC Northern Ireland and Dr Joseph at The Manor
Hospital from providing data on local outbreaks.
1 has been replaced by reference 1, with the authors' agreement; the Table
collected in the London enhanced syphilis surveillance programme' was
omitted and is now included, named Table 1; therefore all tables have been
editorial office, 17 January 2005]
Tables 2, 3 and 4 have been corrected by the
authors [Eurosurveillance editorial office, 24 January 2005]