Article Text


Prevalence and determinants of Chlamydia trachomatis infections in women from Bogota, Colombia
  1. M Molano1,2,
  2. E Weiderpass4,
  3. H Posso2,
  4. S A Morré3,
  5. M Ronderos2,
  6. S Franceschi4,
  7. A Arslan4,
  8. C J L M Meijer1,
  9. N Muñoz4,
  10. A J C van den Brule1,
  11. and the HPV Study Group*
  1. 1Department of Pathology, Vrije Universiteit Medical Center, Amsterdam, Netherlands
  2. 2Division de Investigacion, Instituto Nacional de Cancerologia, Bogota, Colombia
  3. 3Laboratory of Immunogenetics, Faculty of Medicine, VU University Medical Center, Amsterdam, The Netherlands
  4. 4International Agency for Research on Cancer, Lyon, France
  1. Correspondence to:
 Elisabete Weiderpass
 Unit of Field and Intervention Studies, International Agency for Research on Cancer, 150 cours Albert Thomas, F-69372 Lyon Cedex 08, France;


Objectives:Chlamydia trachomatis infection in the cervix and uterus has been hypothesised to be a co-factor for cervical cancer. We performed a cross sectional study in Bogota, Colombia, where cervical cancer rates are high, to determine the prevalence and determinants of C trachomatis infection, and in particular its association with human papillomavirus (HPV).

Methods: 1829 low income sexually active women were interviewed and tested for C trachomatis, using an endogenous plasmid PCR-EIA, and for 37 HPV types, using a general primer GP5+/6+ mediated PCR-EIA.

Results: The overall prevalence of C trachomatis was 5.0%, and it did not differ substantially between women with normal (5.0%) and those with abnormal (5.2%) cervical cytology. Women infected with any HPV type (15.1%) had a slightly increased risk of being simultaneously infected with C trachomatis (adjusted OR 1.3, 95% CI: 0.8 to 2.4). This association was stronger when multiple HPV infections (adjusted OR 2.5, 95% CI: 1.1 to 5.9) were present. No other lifestyle or reproductive characteristics were clearly associated with risk of C trachomatis infection.

Conclusions: HPV infected women, particularly women with multiple HPV infections, are at increased risk of being infected with C trachomatis.

  • Chlamydia trachomatis
  • human papillomavirus
  • Colombia
  • epidemiology
  • cervix uterus

Statistics from

Infections with Chlamydia trachomatis, a highly prevalent sexually transmitted agent worldwide, are mostly asymptomatic (70%–80%) and often remain undetected. Besides causing cervicitis and urethritis, these infections may result in serious secondary complications, such as pelvic inflammatory disease and pelvic pain (18%–24%), tubal infertility (6%–21%), and ectopic pregnancy (in 7%–9% of those who become pregnant).1–5 In addition, C trachomatis has been suggested to be a cofactor in the development of cervical cancer.6

In Colombia, the prevalence and determinants of C trachomatis infection, in particular its association with HPV infections, have not yet been described.

We report here on the prevalence and determinants on C trachomatis infection in a large sample from a low income population in Bogota, Colombia.


Study population

A prospective cohort study in Bogota, Colombia, was initiated in the early 1990s by the Colombian National Institute of Cancer and the International Agency for Research on Cancer (IARC). The aim of the study was to investigate the natural history of human papillomavirus (HPV) infections and cervical intraepithelial neoplasias (CIN) lesions among low income women.

From November 1993 to November 1995, the Colombian National Cervical Cancer Institute conducted a census in four health districts in Bogota, which had no cervical screening programme previously implemented. Women aged 18–44 years were initially invited to participate. Subsequently, the upper age limit was expanded to screen elderly women also.

For the study presented here, the first 2000 women registered in the above census were invited to participate. They were interviewed face to face by specially trained interviewers and answered a structured questionnaire on sociodemographic characteristics, lifelong sexual behaviour, reproductive history, smoking, and dietary habits. After interview they were offered a cervical examination, when a Pap smear was collected, and asked to donate a blood sample. Only women who had ever been sexually active were included for this study.

Informed consent was obtained from all participants included in the study. The local ethics committee and the ethics committee at IARC cleared the study protocol.

Biological specimens

Cervical scrapes were collected from each woman using two Ayre spatulas and two endocervical brushes. The first spatula and brush were used for routine Pap smear, which were classified according to the Bethesda system. The second spatula and brush, and the remaining cells of the first spatula and brush, were placed in a tube containing 5 ml of phosphate buffered saline (PBS 1X) +0.05% thiomersal. Cells were detached from the spatula and endocervical brush by vortexing, and centrifuged at 3000 g for 10 minutes. The cell pellet was resuspended in 1 ml buffer 10 mM TRIS-HCI pH 8.3 and stored at −70°C until use. For analysis, a 100 µl aliquot was boiled for 10 minutes at 100°C, cooled on ice, and centrifuged for 1 minute at 3000 g. 10 µl of this pretreated crude cell suspension was used for polymerase chain reaction (PCR) analysis.7,8

In order to assess the quality of the target DNA, a 209 base pair amplifying β globin PCR was performed using the primer combination BGPCO3 and BGPCO5 as previously described.9

C trachomatis detection by PCR

The detection of C trachomatis was performed as described previously.10 Plasmid endogenous specific primers Bio PL6.1 (Bio 5′-AGAGTACATCGGTCAACGA-3′) and PL6.2 (3-TCACAGCGGTTGCTCGAAGCA-5′) were used for PCR amplification. Briefly, the reaction was carried out in 50 µl of PCR solution containing 10 mM TRIS-HCl pH 8.3, 50 mM KCl, 200 µM of each deoxynucleotide, 1.5 mM of MgCl2, 1 U of DNA polymerase (AmpliTaq; Perkin-Elmer, USA), 25 pmol of each primer (Eurogentec, Belgium), and 5 µl of sample. The PCR amplification consisted of DNA denaturation at 95°C for 4 minutes followed by 40 cycles of amplification using a PE 9600 thermocycler (Perkin-Elmer, USA). Each cycle included a denaturation step at 95°C for 1 minute, one annealing step at 55°C for 1 minute, and a chain elongation step at 72°C for 1.5 minutes. The final elongation step was extended for another 4 minutes.

The biotinylated PL6.1/PL6.2 PCR products were detected using an enzyme immunoassay as described previously.11,12 Briefly, in this assay, 5 µl of the biotinylated PCR products were captured for 1 hour incubation in streptavidin coated wells of a microtitre plate (Roche, Mannheim, Germany). Subsequently, the wells were washed three times with 1X SSC; the captured DNA was denaturated by alkaline treatment with 0.1 M NaOH, and hybridised for 1 hour using the digoxigenin labelled type specific probe to PL6.1/PL6.2 amplified products. The unbound probe was removed by washing three times with 1X SSC and the hybrids were detected using an anti-dig Fab framents labelled with alkaline phosphatase (Roche, Mannheim, Germany) and paranitrophenyl phosphate (Sigma, USA) was used as substrate. Finally, the optical density (OD) was measured at 405 nm using a Labsystem Multiscan reader. In our assay a cut-off point was defined using three times mean OD of the negative controls. As a positive control, a 10-fold dilution series of C trachomatis L2 DNA was used as previously described,13 resulting in a detection sensitivity corresponding to 0.01–0.10 inclusion forming units (IFU).

HPV detection by PCR

HPV-DNA detection in these samples had been performed previously by a standard GP5+/GP6+ PCR-EIA based assay,9 and HPV results for women with cytomorphologically normal scrapes are presented elsewhere.14 In the present report, HPV results of abnormal cytology are additionally presented.

Briefly, HPV positives samples were subjected to EIA-HPV group specific analysis using cocktail probes for high risk (HR) and low risk (LR) HPVs.11 The HR HPV cocktail probe consisted of oligoprobes for HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68. The LR HPV consisted of oligoprobes for HPV 6, 11, 40, 42, 43, 44, HPV82 (MM4), HPV 83 (MM7), HPV 84 (MM8), Iso39, HPV 71 (CP8061), CP6108, HPV 81 (CP8304), HPV 26, 34, 53, 54, 55, 57, 61, 70, 72, and 73. However, HPV types with unknown oncogenic potential—namely, 26, 34, 53, 73, and Iso39—were considered in the results as HR types. This was based on both the alignment analysis of the E6 gene (modified from Myers et al15) and the risk estimates obtained for the various HPV types within a multicentre case-control study of cervical cancer conducted by the IARC.16–23 Additionally, HPV positivity was assessed by Southern blot hybridisation of GP5+/GP6+ PCR products with the general probe of specific (α−32 P)dCTP labelled DNA fragments from cloned DNA of HPV 6, 11, 16, 18, 31 and 33.7,9 Samples that were positive by Southern blot analyses and negative by HR/LR EIA were considered as HPV X or undetermined type.

Statistical analysis

The association between C trachomatis infections and different risk factors was evaluated in all women regardless of their cytological diagnosis. Odds ratios (OR) and 95% confidence intervals (CI) were calculated using unconditional logistic regression models, considering C trachomatis infections as a dependent variable, and several known or hypothesised risk factors for cervical cancer as independent variables. The software used for data management and analysis was STATA (Stata Press, College Station, TX, USA). We performed both age adjusted (with age grouped as <25, 25–29, 30–34, 35–44, 45 or more years) and multivariate analyses. The following variables were included in the multivariate models: age (as categorised above), educational level (low, education up to primary level; intermediate, at least half of secondary school or complete secondary school; and high, technical school or university), number of lifetime regular (1, 2, 3 or more) and casual (yes or no) sexual partners, parity (nulliparous, 1–2 children, 3 or more), age first sexual intercourse (<17, 17–19, 20 or more years), oral contraceptive use (ever or never), condom use, and smoking patterns (ever smoked at least 100 cigarettes). No sample size calculation was done a priori for this study.


From the 2000 invited women, 53 refused to participate, seven were considered ineligible (mental illness, hysterectomy, history of cervical cancer), 29 did not provide cell specimens for HPV detection, and 82 had poor DNA quality as indicated by negative β globin PCR and were therefore considered inadequate for HPV-PCR and C trachomatis PCR testing. Thus, a total of 1829 women had PCR and valid HPV and Chlamydia results. However, for the analysis of characteristics of the studied population we had to further exclude 16 women who did not answer the questionnaire completely.


All Pap smears were initially classified both using the CIN classification and the Bethesda System terminology. In all, 1687 women (92.2%) had normal cytological findings (table 1). Of the 115 women with abnormal cytology, 63 (3.4%) were diagnosed with low grade squamous intraepithelial lesion (LSIL), 28 (1.5%) had atypical squamous cell of undetermined significance (ASCUS), 14 (0.8%) had atypical glandular cells of undetermined significance (AGUS), eight (0.4%) had high grade squamous intraepithelial lesions (HSIL), and two (0.1%) had invasive cervical cancer.

Table 1

HPV and C trachomatis infections and cytological diagnosis in women from Bogota, Colombia

Characteristics of the studied women

The median age of women participating in the study was 33 years (range 18–85 years). Most women had low (38.0%) or intermediate (36.9%) educational level, had their first sexual intercourse or first regular sexual partner before the age of 20 (62.2%), reported only one lifelong sexual partner (74.0%), and had at least two full term pregnancies (median 2). IUD was the most common contraceptive method ever used (59.1%), followed by oral contraceptives (49.9%) and condoms (31.6%). Less than one third (28.3%) had ever smoked regularly (table 2).

Table 2

Determinants of C trachomatis infection in women from Bogota, Colombia

C trachomatis prevalence and risk factors

The overall prevalence of C trachomatis infection was 5.0%, and did not differ substantially between women with normal or abnormal cytological results (table 1).

The highest prevalence of C trachomatis was observed in women aged 30–34 years (7.9%) and the lowest in women older than 45 years (1.5%) (table 2; fig 1).

Figure 1

Prevalence and 95% CI of C trachomatis and HPV DNA in women from Bogota, Colombia.

There was no clear association between educational level, sexual behaviour (number of regular or casual sexual partners), reproductive history (parity, ever use or duration of use of oral contraceptives, use of condoms or IUD), or smoking patterns and risk of C trachomatis infection (table 2).

HPV prevalence

The overall HPV DNA prevalence was 15.1%. Women presenting abnormal cytology had a substantially higher prevalence of HPV infections (39.1%) than women with normal cytology (13.6%) (table 1).

There were more women infected with HR HPV types (11.9%) than with LR types (2.9%). Infections with uncharacterised types (HPV X, 0.5%) were few (table 3).

Table 3

C trachomatis infection and HPV infection in women from Bogota, Colombia

Association between C trachomatis and HPV infections

The prevalence of C trachomatis infections was non-significantly higher among women infected with HPV of any studied type than among HPV uninfected women (6.9% v 4.7%; adjusted OR 1.3, 95% CI: 0.8 to 2.4). C trachomatis prevalence was higher among women with multiple HPV infections (9.5%) than women with single HPV infections (5.9%; adjusted OR 2.5, 95% CI: 1.1 to 5.9) (table 3).


This is the first study describing the prevalence and determinants of C trachomatis infections in the population of Bogota, Colombia.

The overall C trachomatis prevalence in this study (5.0%) was similar to the prevalence reported in women from the general population in Amsterdam, Colorado, Washington, and Copenhagen, which ranged from 4.5% to 9%.12,24–26 Our results also confirm a somewhat higher C trachomatis infection prevalence in young age groups, as observed in other populations.12,27 Behavioural factors (such as early age in the initiation of sexual activity, multiple partners, irregular use of condoms, use of contraceptive methods) were, unexpectedly, not associated with C trachomatis infection risk. Thus, our study is in partial disagreement with some previous reports,12,25,28,29 which found associations with C trachomatis infection and oral contraceptive use, parity, lifetime number of sexual partners, and smoking habits. Discrepancies between our results and those from studies conducted elsewhere may be because of differences in the characteristics of the studied populations (such as age, patterns of sexual behaviour or openess in reporting sexual behaviour), and variations in the sensitivity and specificity of laboratory methods used.

Duration of oral contraceptive use had no apparent influence on the prevalence of C trachomatis infection—as reported also by Munk et al12 and Jacobson et al.30 The lack of association between parity and C trachomatis infection observed by us is in line with the study by Munk et al12 but it contrasts with the study by Bagshaw et al.31

Besides young age, the only risk factor independently associated with C trachomatis infection in our study was HPV infection, particularly multiple infections. This is in agreement with some previous studies using PCR and serological assays.32,33 The clear association between C trachomatis and multiple HPV infections might be indicative of multiplicity of sexual partners from women themselves or their partners. The male partner(s) sexual practices (such as multiplicity of sexual partners or high risk sexual behaviour) is a plausible explanation for C trachomatis infection among women. Such information is, however, not available in our study.

Some studies reported an association between antibodies to C trachomatis and SIL lesions.34–37 However, analysis of the DNA presence of C trachomatis and its association with HPV infections and SIL lesions is still limited. We did not observe any substantial difference in C trachomatis infection prevalence between women with normal or abnormal cervical cytology. This suggests that C trachomatis does not itself induce cervical abnormalities. The role of C trachomatis as a potential cofactor in cervical disease, however, cannot be ruled out.

Key messages

  • C trachomatis infection is relatively common (about 5% prevalence) among adult women in Bogota, Colombia.

  • HPV infected women, particularly women with multiple HPV infections, are at increased risk of being infected with C trachomatis.

  • The overall prevalence of C trachomatis does not seem to differ substantially between women with normal and those with abnormal cervical cytology.


M Molano obtained a fellowship from Colciencias of the Colombia Government. Financial support was received from the Department of Pathology, Vrije Universiteit Medical Center, Netherlands, the Department of Epidemiology, Instituto Nacional de Cancerologia, Colombia, the International Agency for Research on Cancer (IARC), France, and the UNPD/UNFPA/WHO/World Bank Special Program of Research, Development and Research Training in Human Reproduction, Department of Reproductive Health and Research, WHO, Geneva (Grant 94053A).

 MM was responsible for laboratory testing and participated in the interpretation of results; EW, SF, and AA were responsible for statistical analysis, interpretation of the results and drafting of the manuscript; EW wrote the final version of the paper; HP and MR coordinated the study design and inclusion of patients; NM coordinated the overall study design and participated in drafting of the paper; SM participated in laboratory testing; CJLMM and AJCB coordinated the laboratory testing. The HPV Study Group participated in the inclusion of patients and cytology laboratory analysis.


View Abstract


 Mauricio Gonzales, Joaquin Luna, Gilberto Martinez, Edmundo Mora, Gonzalo Perez, Jose Maria Fuentes. Constanza Gomez, Eva Klaus, Constanza Camargo, Cecilia Tobon, Teodolinda Palacio, Carolina Suarez, Claudia Molina.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.