© Oxford University Press 2002
Vol. 19, No. 4
Printed in Great Britain
Genital infection by Chlamydia trachomatis
in Lisbon: prevalence and risk markers
Armando Brito de Sá, João Paulo Gomesa, Sílvia Viegasa,
Maria Arminda Ferreiraa, Albertina Paulinoaand
Maria dos Anjos Catrya
Brito de Sá A, Gomes JP, Viegas S, Ferreira MA, Paulino A, Catry MA. Genital infection by Chlamydia
trachomatis in Lisbon: prevalence and risk markers. Family Practice 2002; 19: 362–364.
Background. There is little information about the prevalence and risk markers for Chlamydia
trachomatis infections in Portugal.
Objectives. Our aim was to assess the prevalence of C. trachomatis genital infection and to
study variables associated with this infection in a group of sexually active women aged ?30
years living in the Lisbon area and to estimate the prevalence of C. trachomatis infection among
partners of infected patients.
Methods. A systematic sample of women observed in general practice family planning and
teenager clinics was collected. A questionnaire was administered, followed by a pelvic exam-
ination. A first-catch urine sample was taken for polymerase chain reaction (PCR) Amplicor
assay. When a sample tested positive, the woman was invited to obtain a urine sample from her
partner. Socio-demograhic, behavioural and clinical variables were studied and their association
with the PCR Amplicor result was assessed.
Results. A total of 1108 women, aged between 14 and 30 years, were studied. Fifty-one women
(4.6% of total sample) tested positive for C. trachomatis. The prevalence of infection was slightly
higher in patients aged ?19 years (5.3%) than in age groups 20–25 (4.8%) and 26–30 years
(3.9%). African ethnicity was related to a higher percentage of infection than European ethnicity:
9.8% versus 3.8%, P = 0.0067. Use of condoms ‘sometimes/never’ was associated with a higher
prevalence of infection: 5.2% versus 2.3% in those responding ‘always/almost always’
(P = 0.0447). An altered cervix was associated with a higher prevalence of infection: 7.3% versus
3.7% with a normal cervix (P = 0.0106). Urine samples were obtained from 16 partners of infected
patients. Six partners (37.5%) tested positive for C. trachomatis.
Conclusions. A 4.6% prevalence of C. trachomatis genital infection was found. African ethnicity,
using condoms ‘sometimes/never’ and an altered cervix were associated with C. trachomatis
infection, but showed low positive predictive value for C. trachomatis infection. Younger age
may be associated with a slight increase in risk. Contact tracing for diagnosis and treatment
remains a difficult issue to approach effectively.
Keywords. Chlamydia trachomatis, contact tracing, epidemiology, polymerase chain reaction,
Genital Chlamydia trachomatisinfection is an established
cause of pelvic inflammatory disease (PID), ectopic preg-
nancy and infertility1among women. Epidemiological
studies show this infection to be one of the most wide-
spread sexually transmitted diseases in the world today.2
Attempts at a clinical diagnosis invariably have been
disappointing.3Laboratory testing of chlamydial genital
infection, on the other hand, continues to improve.4
There is a growing body of evidence in favour of genital
C. trachomatisscreening;5however, doubts remain about
whether this screening should be systematic, targeted to
specific female populations or opportunistic.6The issue
of contact tracing remains a relevant problem.7
There is little information about the prevalence and
risk markers for C. trachomatis infection in Portugal.8
Received 27 July 2001; Revised 11 December 2001; Accepted
11 March 2002.
Instituto de Medicina Preventiva, Faculdade de Medicina de
Lisboa, Av. Prof. Egas Moniz, 1649-028 Lisboa and aUnidade
de Clamídia, Instituto Nacional de Saúde, Av. Padre Cruz,
1649-016 Lisboa, Portugal. Correspondence to Armando Brito
de Sá; E-mail: email@example.com
Chlamydia trachomatis in Lisbon
This study had the following objectives:
(i)to describe a sample of sexually active women,
aged 30 or below, observed in family planning or
teenager primary care clinics in health centres in
the Lisbon area;
(ii) to assess the prevalence of C. trachomatis genital
infection in that sample;
(iii) to identify possible risk markers associated with
the infection; and
(iv) to assess the prevalence of infection in the sexual
partners of the women who tested positive for
C. trachomatis infection.
A systematic sample of women observed in general
practice family planning and teenager clinics was studied
between April 1999 and February 2000. Inclusion
criteria were: (i) age ?30 years; (ii) sexually active;
(iii) a pelvic examination being already scheduled; and
(iv) informed consent given by the patient. Socio-
demographic variables studied were age, educational
level, professional status, marital status and ethnic
group. Behavioural variables included age of onset
of sexual activity; number of sexual partners to date, in
the 3 months previous to the study and presently; use
of a condom (always, almost always, seldom, never);
and use of antibiotics in the previous 30 days. Clinical
variables studied were reason for encounter of the
current appointment; gynaecological history including
number of pregnancies, spontaneous abortions and
voluntary terminations; history of PID or infertility;
presence and characteristics of vaginal discharge; and
presence and characteristics of cervical lesions. After
informed consent was obtained, a questionnaire was
administered, followed by a pelvic examination. A first-
catch urine sample was taken for polymerase chain
reaction (PCR) Amplicor assay (RMS™). The samples
were delivered by courier to a single laboratory where
all tests were performed and were processed in the
same day, or stored at 2–8°C for a maximum of 7 days.
The questionnaires were delivered directly to one of the
authors (ABS) in order to ensure confidentiality.
Chi-square, Student’s t-test and logistic regression
were used to check for associations between variables
and PCR positivity.
A total of 1108 women, aged between 14 and 30 years,
were studied. Mean age was 23.4 ± 4.1 years. Most
women (40.3%) had finished secondary school, 25.7%
had at least a basic education and 22.1% had a university
degree. Sixty per cent had a job and 23.0% were students.
Single and married/living together status were nearly
equally distributed (50.8 and 47.2% of the total sample).
Most women were of European ethnicity (85.2%), and
13.0% were of African ethnicity.
Mean age for first sexual intercourse was 17.8 ± 2.4
years. A single sexual partner ever was reported by
52.2% of the respondents, 25.3% reported two partners
and 12.8% reported three. Ninety-five per cent reported
a single sexual partner in the previous 3 months.
Condoms reportedly were used ‘always/almost always’
by 20.0% of women, and ‘sometimes/never’ by 80.0%.
Antibiotics had been taken by 12.2% of women in the
previous 30 days. Main reasons for encounter were clin-
ical check-ups and/or preventive procedures (55.3%).
Fifty-one women (4.6% of total sample) tested positive
for C. trachomatis. Table 1 shows data concerning the
association of some selected variables with C. trachomatis
infection. Age, educational level, professional situation,
marital status, age of onset of sexual activity, present
number of partners and gynaecological history did not
show an association with chlamydial infection. Mean
total number of sexual partners was higher in the PCR-
positive group, but the difference was not statistically
significant. The prevalence of infection was just slightly
higher in patients aged ?19 years (12/226; 5.3%) than
in age groups 20–25 (24/499; 4.8%) and 26–30 years
African ethnicity was related to a higher percentage
of infections than European ethnicity: 14/143 (9.8%)
versus 36/940 (3.8%), P = 0.0067. Use of condoms
Association of selected variables with PCR results
P = 0.0067
Age of onset of
18.0 ± 2.5 17.8 ± 2.4NS
Total no. of sexual
2.3 ± 2.41.9 ± 1.6 NS
Use of condoma
P = 0.0447
P = 0.0106
bStudent’s t-test; mean ± SD.
Family Practice—an international journal
‘sometimes/never’ was associated with a higher preva-
lence of infection—46/879 (5.2%) versus 5/219 (2.3%) in
those responding ‘always/almost always’ (P = 0.0447).
An altered cervix was associated with a higher preva-
lence of infection—25/345 (7.3%) versus 26/712 (3.7%)
with a normal cervix (P = 0.0106). However, logistic
regression only identified ethnicity as an independent
risk factor for PCR positivity.
Urine samples were obtained from 16 partners of
infected patients (31.4% of total positive cases), and
six (37.5%) tested positive for C. trachomatis.
This study was conducted in the primary care setting.
Patients studied were those seen in the family planning
clinics of health centres on a normal basis. Only patients
who had a pelvic examination scheduled for any reason
were invited to participate in the study, in order to avoid
a possible selection bias. We thus think that the sample
obtained, despite its relatively small size, accurately
reflects the population reality of primary care family
planning clinics in the Lisbon area, run by family doctors.
The processing of all urine samples in the same
laboratory, using a highly specific and sensitive diag-
nostic method, also contributed to the validity of the
We found a prevalence similar to those described in
other Western European cities; the marked increase
in prevalence associated with an African origin of the
patient has also been found by others.9This increased
prevalence in the African ethnic group was found to be
independent of education, professional status and
number of sexual partners.
Age at first intercourse has been found by others to be
associated with genital infection by C. trachomatis.10An
inverse trend relating age and prevalence of infection
was in fact found in this study, but no statistically sig-
nificant association could be found, perhaps due to the
small sample size.
Using condoms ‘sometimes/never’ and the presence
of an altered cervix, though positively associated with
the infection, fail to provide an acceptable positive pre-
dictive value for infection: 6.5 and 6.8% positive predict-
ive values, respectively. This indicates that, as mentioned
before, a clinical approach is not adequate for our popu-
lation, leaving the prevalence as the sole epidemiological
indicator for a decision for or against any kind of formal
Contact tracing and treatment remain a major
problem in our study; asking the women with a positive
result for C. trachomatis to obtain a urine sample from
their partners for testing elicited a response from
fewer than one-third of the women. A high prevalence
of infection (almost 40%) was found among partners,
reinforcing the notion that active efforts to identify
sexual partners of infected women are essential, to
reduce both reinfection rates and new infections.
The authors wish to thank the 45 doctors who par-
ticipated in the data collection, and Dr Fernando Moura
Pires who provided statistical support. This study was
supported by the Comissão de Fomento da Investigação
em Cuidados de Saúde do Ministério da Saúde, Project
20/98, and by the Instituto Nacional de Saúde.
1Oakeshott P, Hay P. General practice update: Chlamydia infection
in women. Br J Gen Pract 1995; 45: 615–620.
2Stamm WE. Chlamydia trachomatis infections: progress and
problems. J Infect Dis 1999; 179 (Suppl 2): S380–S383.
3Passey M, Mgone CS, Lupiwa S, Tiwara S, Lupiw T, Alpers MP.
Screening for sexually transmitted diseases in rural women in
Papua New Guinea: are WHO therapeutic algorithms appro-
priate for case detection? Bull WHO 1998; 76: 401–411.
4Puolakkainen M, Hiltunen-Back E, Reunala T et al. Comparison of
performances of two commercially available tests, a PCR assay
and a ligase chain reaction test, in detection of urogenital
Chlamydia trachomatis infection. J Clin Microbiol 1998; 36:
5Kamwendo F, Forslin L, Bodin L, Danielsson D. Programmes to
reduce pelvic inflammatory disease—the Swedish experience.
Lancet 1998; 351 (Suppl III): 25–28.
6Boag F, Kelly F. Screening for Chlamydia trachomatis. Br Med J
1998; 316: 1474.
7Andersen B, Østergaard L, Nygård B, Olesen F. Urogenital
Chlamydia trachomatisinfections in general practice: diagnosis,
treatment, follow-up and contact tracing. Fam Pract 1998; 15:
8Guerreiro D, Borrego MJ, Teles LC, Catry MA. [Cervical
Chlamydia trachomatis infection in women attending family
planning clinics—screening, is it necessary?]. Act Med Port
1996; 9: 151–156.
9Winter AJ, Sriskandabaian P, Wade AA, Cummins C, Barker P.
Sociodemography of genital Chlamydia trachomatis in
Coventry, UK, 1992–6. Sex Transm Inf 2000; 76: 103–109.
10Suss AL, Homel P, Hammerschlag M, Bromberg K. Risk factors
for pelvic inflammatory disease in inner-city adolescents. Sex
Transm Dis 2000; 27: 289–291.