|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
The bulk of infections remains undetected and untreated because most infected people are oligo- or asymptomatic and do not seek medical attendance. If untreated, chlamydiae may reach the upper genital tract of affected women and cause pelvic inflammatory disease (PID) with the risk of severe reproductive complications, such as tubal factor infertility and ectopic pregnancy.2 After one episode of PID, the ratio of infertility has been estimated at 11%, which increases to 23% and 54% after 2 and 3 episodes, respectively.3
Due to the high prevalence rates of C. trachomatis and the asymptomatic course of infection, screening programs have been established in some industrialized countries to reduce the rate of PID and prevent development of reproductive sequelae.2,4,5 Sensitivity and specificity of nucleic acid amplification tests (NAATs) for C. trachomatis on non-invasively collected specimens such as first-void urine (FVU) have been proven comparable to those obtained on samples collected directly from the cervix or urethra, making NAATs attractive diagnostic tools for screening asymptomatic individuals.6 Serological testing of uncomplicated C. trachomatis infections of the lower genital tract has not been recommended, but antibody titers are especially high in women with PID.7 In addition, C. trachomatis-specific antibodies have been associated with tubal damage and infertility.8–10 The microimmunofluorescence test (MIF) developed by Wang and Grayston is still considered the method of choice for serodiagnosis of chlamydial infections. However, due to the technically demanding and time-consuming assay format, the MIF is less useful for large-volume screening. New enzyme immunoassays based on synthetic peptides derived from the variable domain 4 (VD 4) of the major outer membrane protein of C. trachomatis demonstrated similar or even slightly better performance characteristics than the MIF for the detection of antibodies to C. trachomatis.11 In addition, they offer a more automated workflow and a more objective endpoint reading of titers, which make them valuable for epidemiologic studies.
In developing countries, data about the prevalence of genital C. trachomatis infections and their complications, such as infertility, is scarce because reliable test assays are too expensive and too complex for routine use in resource-limited settings. However, better knowledge about the epidemiologic situation could be the basis for implementation of treatment and screening strategies. Especially in poor countries in Africa, infertility is considered as an enormous social problem because siblings are still important to retain a families economic standard in great parts of the population.
The aim of this study was to determine the prevalence and seroprevalence of C. trachomatis among healthy women and women suffering from infertility to evaluate the association of previous C. trachomatis infections on infertility in a developing country.
| MATERIALS AND METHODS |
|
|
|---|
Laboratory methods. PCR testing from FVU as well as antibody testing for C. trachomatis were performed in the laboratory of the Institute of Medical Microbiology and Hygiene, University Hospital of Ulm, Germany. The laboratory has established a quality control program according to the standards of the International Organisation for Standardization (ISO 15189)12 that specifies requirements for quality and competence particular to medical laboratories. In addition, the laboratory participates continuously and successfully in an external proficiency program for C. trachomatis testing, approved by the German Society for Hygiene and Microbiology (Deutsche Gesellschaft für Hygiene und Mikrobiologie, DGHM).
Frozen samples were shipped in liquid nitrogen. After they had thawed, serum samples were tested for IgG and IgA antibodies specific to C. trachomatis by using a commercially available peptide-based enzyme-linked immunosorbent assay.11 The antigen used in this assay is a synthetic peptide based on species-specific epitopes of the major outer membrane protein of C. trachomatis. The absorbance was determined at 450 nm. Cut-off indices (COI) were calculated according to the instructions of the manufacturer (Savyon Diagnostics, Ashdod, Israel). Serum samples yielding a COI more than 1.1 were regarded positive. FVU samples were tested using the COBAS AMPLICOR PCR (Roche Diagnostics, Basel, Switzerland).13 The COBAS AMPLICOR PCR assay targets a specific nucleotide sequence of the 7.5-kbp cryptic plasmid of C. trachomatis and uses an internal inhibitory control. Specimens were processed and stored by the laboratory according the instructions of the manufacturer (Roche Diagnostics).
Statistical analysis.
As bivariate analysis, Pearson
2 test was performed to assess the association between infertility and various patients characteristics including the PCR and serological results. Continuous variables were converted to categorical variables based on the distribution of the data in the full study sample. The parameters and their categories are shown in Tables 1
and 2
. We then conducted a logistic regression model with fertility status as the dependent variable. In addition to IgG and IgA, further predictor variables were age and parity as known risk factors for infertility.14–16 SPSS Version 12.0 (SPSS, Inc., Chicago, IL) was used for all statistical procedures.
|
|
| RESULTS |
|
|
|---|
The IgG seroprevalence rate was 19.4%, and the IgA seroprevalence rate was 3.6% for healthy pregnant women. Seroprevalence rates among women suffering from infertility were 39.3% for IgG (P < 0.001; comparison with the control group) and 14.1 for IgA (P < 0.001; comparison with the control group), respectively. The seroprevalence rates increased to 47.3% for IgG and 14.5% for IgA among infertile women when only those were considered where a normal sperm count of the partner was available (N = 55; data not shown).
Bivariate analysis.
The results of the bivariate analysis are summarized in Tables 1
and 2
. Further parameters that were significantly associated with infertility were older age (P < 0.001), rural residential area (P = 0.002), sexual behavior only vaginal (P = 0.005), lower parity (P < 0.001), higher number of sex partners (P = 0.009), and younger age of first coitus (P = 0.006). Alcohol use and nicotine use did not show a significant association (data not shown).
Logistic regression analysis.
The results of the regression model are shown in Table 3
. IgG and IgA antibodies were significantly associated with infertility (P = 0.004 and P = 0.030, respectively). Adjusted odds ratios (OR) for a patient with a positive result were 2.1 and 2.8, respectively.
|
| DISCUSSION |
|
|
|---|
Previous studies have shown that specific antigens or DNA of C. trachomatis can be detected in fallopian tube tissue in women with tubal infertility; however, invasive sampling procedures are required to obtain appropriate specimens.17 To overcome the problem of invasive specimen collection, serological testing has been suggested to predict tubal-factor pathology by C. trachomatis.18,19 We found significantly higher prevalences of IgG and IgA antibodies against C. trachomatis among women suffering from infertility. This is in agreement with data from den Hartog and others, who used laparoscopy as a reference and who found that C. trachomatis IgA and IgG antibodies against chlamydial heat-shock protein 60 were significantly more prevalent in women with tubal pathology.9 In contrast to data from Logan and others20 who showed that C. trachomatis IgG antibody testing by ELISA failed to predict tubal-factor infertility in subfertile women, C. trachomatis-specific IgG antibodies had a 2.1-fold increased OR for infertility in our population. Specific IgA antibodies have been associated with chronic inflammation and infection,21,22 and it has been suggested that serum IgA antibodies may be more reliable markers of persistent chlamydial infections.23 However, it is difficult to estimate whether the presence of specific IgG and IgA antibodies reflects an acute, chronic, or a past C. trachomatis infection because little is known how long specific antibodies may persist in individuals with resolved infections.
In our study population, PCR testing revealed a relatively low prevalence among infertile women (1.6%) as well as among healthy pregnant women (2.4%). This could indicate that most of the women with positive IgG and IgA antibodies might have become previously infected with C. trachomatis, for example, as adolescents or young adults. Nucleic acid amplification assays are rarely used in resource-limited settings mainly because of their high cost and technical complexity. In this study, FVU samples from a total of 439 Ghanaian women were investigated with a licensed PCR test that is commonly used in many routine clinical laboratories. Although we cannot completely rule out that transport conditions might have somewhat reduced the detection rates, our data are in good agreement with prevalence rates of 3.6% and 4.9%, respectively, reported from former studies from Ghana, using markedly less specific laboratory techniques.24,25 In addition, the PCR detection rates in our study were nearly identical with the prevalence of C. trachomatis infection in another West African country26 and surprisingly comparable with those of asymptomatic women and pregnant women in Europe, North America, and Australia.27,28
The low detection rates of C. trachomatis specific DNA in urine samples seem to stand in stark contrast to the high seroprevalence rates of C. trachomatis in our patients. However, we have to consider that, although PCR testing of FVU is excellent for diagnosis of C. trachomatis infections of the cervix,6 a negative urine sample does not rule out infection of the fallopian tubes. In addition, we have to consider that the average age in our study population was relatively high, and more than 90% of women in our population were married. Both factors may substantially reduce the risk for C. trachomatis infection of the lower genital tract. Especially in patients from rural regions, positive IgA and IgG antibodies against C. trachomatis may have resulted from trachoma and have increased the C. trachomatis seroprevalence in our study population. However, this remains a speculation.
One limitation of our study is the unknown prevalence of real existing tubal damage among our study population because women attending our hospital had not undergone systematic laparoscopy with tubal testing. Because other causes than tubal damage must be additionally considered for infertility, the diagnostic value of antibody testing for C. trachomatis-associated tubal infertility could be substantially improved if a more selected study population were investigated. This becomes especially evident when only those infertile women were considered where a normal sperm count of the partner was available.
Reproductive-organ disease with loss of reproductive function may be a large burden for many women in developing countries.29 In this study, we could demonstrate that C. trachomatis infection must be recognized as an important risk factor for infertility of Ghanaian women. Further studies are needed to obtain more epidemiologic data about the spread of C. trachomatis among adolescent and young African women and to guide effective prevention and intervention measures.
Received April 17, 2007. Accepted for publication November 23, 2007.
Acknowledgments: We thank the nurses from the Outpatients Department of Obstetrics and Gynaecology of the Komfo Anokye Teaching Hospital, Kumasi, Ghana, for great support of our study. We are thankful for the help from Thomas F. Kruppa, director of the Kumasi Centre for Collaborative Research, having solved our logistical problems. The expert technical support of Mrs. Sonja Rothen-berger is greatly acknowledged.
* Address correspondence to Jorn Siemer, University Hospital, Department of Obstetrics and Gynaecology, Universitätsstr. 21–23, 91054 Erlangen, Germany. E-mail: kaiserschnitt{at}gmx.de ![]()
Authors addresses: Jorn Siemer and Peter A. Fasching, University Hospital, Department of Obstetrics and Gynaecology, Universitätsstr. 21–23, 91054 Erlangen, Germany, Telephone: +49–9131–8533553, Fax: +49–9131–8533492, E-mail: kaiserschnitt{at}gmx.de. Oliver Theile and Rolf Kreienberg, University Hospital, Department of Obstetrics and Gynaecology, Ulm, Germany. Yaw Larbi and K. A. Danso, Komfo Anokye Teaching Hospital, Department of Obstetrics and Gynaecology, Kumasi, Ghana. Andreas Essig, University Hospital, Institute of Medical Microbiology and Hygiene, Ulm, Germany.
Reprint requests: Jorn Siemer, University Hospital, Department of Obstetrics and Gynaecology, Universitätsstr. 21–23, 91054 Erlangen, Germany, Telephone: +49 9131-8533553, Fax: +49 9131-8533492, E-mail: kaiserschnitt{at}gmx.de.
| REFERENCES |
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |