Chlamydial genital infection is the most common reported sexually transmitted bacterial disease (STD). The estimated number of new cases was 92 million in 1999.1 Most infected individuals are asymptomatic,2 therefore the infection often remains undetected. This contributes to the spread of a silent Chlamydia epidemic and underreporting of Chlamydia trachomatis infection.

C trachomatis is the most common cause of non-gonococcal urethritis and accessory gland infection in men. Women are at high risk of complications including endometritis, salpingitis, and pelvic inflammatory disease, with associated long-term sequelae such as pelvic pain, ectopic pregnancy, preterm birth, and infertility.3 C trachomatis infection is also associated with an increased risk of cervical cancer.4 5

In Nordic countries, C trachomatis rates are increasing. In Denmark, Finland, Norway, and Sweden, the incidence rates were 245, 444, 441, and 376 cases per 100 000 in 2005 compared with 206, 262, 297, and 188 in 1999.1 In Finland C trachomatis infections have increased by 60% in 10 years.6 Approximately 3.5% of adult Finnish women are asymptomatic carriers of C trachomatis.7

We studied the epidemiology of C trachomatis in Finland. Seroprevalence data collected in pregnant women since 1983 were used to represent C trachomatis seroprevalence rates on maps over time.

METHODS

Finnish Maternity Cohort

Following informed consent, over 98% of pregnant Finnish women (altogether 750 000) have participated in the serological screening of congenital infections (syphilis, HIV, and hepatitis B) during the first trimester at municipal health centres. Since 1983, the screening samples have been collected and stored in the Finnish Maternity Cohort (FMC) serum bank of the National Public Health Institute. The FMC serum bank comprises approximately 1.5 million serum samples stored at −25°C. After the first pregnancy, approximately 50% of women become pregnant again within five years, donating another serum sample to the serum bank.

Altogether 275 505 women (<29 years of age) with a minimum of two serum samples withdrawn at two consecutive pregnancies before 2005 were identified and divided into 28 strata according to three-year periods (1983–1985, 1986–1988, 1989–1991, 1992–1994, 1995–1997, 1998–2000, 2001–2003) and age (<20, 20–22, 23–25, and 26–28 years) at the midpoint of the two samplings. A stratified random cohort of 8400 women was selected. A total of 400 women were excluded as a result of missing data, and the final number of women with paired sera tested for C trachomatis IgG antibodies was thus 8000. In this study, we used the result of the first serum sample for each woman, and the data were re-organised according to age and calendar year. The mean age of the women at the time of the first serum sampling was 22 years (range 14–28 years).

Serology

Serum IgG antibodies to C trachomatis were analysed (serum dilution in 1:10) by a commercial enzyme immunoassay technique based on a synthetic peptide containing an immunodominant B-cell epitope in the C trachomatis major outer membrane protein (AniLabsystems, Helsinki, Finland) according to the manufacturer’s instructions.5 A peptide-based enzyme immunoassay test is considered to be species specific with minimal cross-reactivity with Chlamydia pneumonia antibodies.8 The reproducibility of the test as reported by the manufacturer is high (SD 0.061, coefficient of variation 4.9%) and correlation of the result between the AniLabsystems and another peptide-based method is greater than 0.9 indicating the high specificity and sensitivity of the test.8 The results were expressed as the mean absorbance (optical density 450 mm) of the sample minus the mean absorbance of the reagent blank divided by the cut-off value. An absorbance value of greater than 1.4 was considered positive.

Statistical analysis

C trachomatis seropositivity was studied at the time of the first pregnancy sampling. Statistical analyses were performed separately for two age groups: women aged 14–22 and 23–28 years. The proportions of seropositive individuals (seroprevalence rates) with the 95% CI were estimated for three time periods: 1983–1989, 1990–1996, and 1997–2003. The statistical significance of the trends over time was tested using the chi-square test for a linear trend.

Municipality-specific seroprevalence rates for the same age and year groups were calculated. Random sampling of the 8000 subjects resulted in coverage of almost all of the 446 communities in Finland. The relative proportions of the subjects of all pregnant women within a community varied from 0 (less than 30 communities) to 10% per community. Statistical analyses were performed using SPSS 13.0 (SPSS Inc., Chicago, Illinois, USA) and R 2.1.1 (R Development Core Team, Vienna, Austria).

Mapping method

The spatiotemporal variation of C trachomatis seroprevalence rates was visualised as a series of maps by smoothing the community-level input data with a 2 × 2 km raster layer. The rates for the biggest cities (with 58 or more study subjects) were shown as coloured circles (with circle size indicating the population of the city). The seroprevalence rates were visualised using 19 colours varying from blue and green for low rates to yellow and red for high rates. This method has been used for human papillomavirus (HPV) and cancer mapping studies.9 10 Areas with less than one inhabitant per square kilometre were masked with a screen to stress the uncertainty related to these rates.

RESULTS

C trachomatis seroprevalence trends

We found decreasing C trachomatis seroprevalence rates in both age groups from 1983–1989 to 1997–2003 (table 1). The highest C trachomatis seroprevalences (20.8% and 19.6%) were seen in women under 23 years of age in 1983–1989 and in 23–28-year-old women in 1990–1996.

During the last period, 1997–2003, C trachomatis seroprevalence rates decreased in both age groups, 10.6% and 12.5%, respectively.

Spatiotemporal variation of C trachomatis seroprevalence

C trachomatis seroprevalence was highest in southwestern Finland among women under 23 years of age in the 1980s and among 23–28-year-old women in the 1990s (fig 1). The high C trachomatis seroprevalence area in northern Finland is hatched because of the sparsely populated area and thus the uncertainty of the observations. C trachomatis clusters emerged around larger cities and close to the Russian border crossing areas (fig 1). In the 2000s, the two highest C trachomatis seroprevalences were southeastern Finland close to the Russian border and central Lapland. From 1990s and up to 2003, decreasing high to moderate C trachomatis seroprevalence was seen close to the largest cities.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Chlamydia trachomatis seroprevalence (%) in pregnant Finnish women. C trachomatis IgG antibodies were determined in a random sample of first trimester sera from 14–22 and 23–28-year-old women in three consecutive periods. Areas covered with hatched lines in the north indicate sparsely populated areas (less than one inhabitant/km2).

DISCUSSION

High seroprevalences were recorded in southwestern Finland in the 1980s, especially in women who were under 23 years of age. C trachomatis seroprevalence has decreased among the younger women since 1990. Seroprevalence decreased in the subcohorts of 23–28-year-old women in 1990–1996 and 1997–2003, suggesting a possible birth cohort effect.

C trachomatis prevalence varies widely, depending upon the setting, country, and context. In asymptomatic European women, the prevalence has ranged from 1.7% to 25%.11 12 Only a few population-based serological studies have, however, been reported. In the early 1990s in Brazil and the Philippines, C trachomatis seroprevalences were 20.2% and 23.0%, respectively.13 In Japan, C trachomatis seroprevalence in pregnant women decreased over 10 years during 1987–1997.14

In our study, C trachomatis seroprevalence clusters were seen around the largest cities. High seroprevalence areas in the 1990s and 2000s were seen in southeastern Finland close to the Russian border and in central Lapland. These peaks may be partly explained by the collapse of the Soviet Union in 1989 and the increasing migration and travel between the countries. Tourism in Lapland has increased massively from the 1980s to the 2000s.

The population-based FMC is the world’s largest serum bank, which covers virtually the entire fertile-aged female population in Finland since 1983. A random subsample of the FMC with a minimum of two pregnancies has previously been used for a trend study of HPV infections.15 HPV-16 seroprevalence has increased, probably as a result of increasing risk-taking behaviour. It also indicates that serology is useful for STD epidemiology research, particularly in trend analyses.

The decreasing C trachomatis seroprevalence rate is surprising because increasing numbers of Chlamydia infections have been reported in Finland.6 One explanation for this discrepancy may include the current intensive STD testing with improved (PCR) methods, resulting in more cases being detected. Early detection and treatment might lead to poor humoral immune response and low C trachomatis antibody levels.16 Poor antibody response might lead to the rapid loss of antibodies reflected in the FMC cohort. This might explain the decreasing seroprevalence trend over time.

Although the FMC bank lacks the serum samples of truly infertile women, it is regarded as representing the whole population, and C trachomatis seroprevalence trends found in this study reflect the primary and secondary prevention efforts in the population. A comparison of serological and PCR-based prevalence and incidence trends reveals whether prevention programmes work.