FALK, LARS MD*; LINDBERG, MARGRET MSc*; JURSTRAND, MARGARETHA MSc†; BÄCKMAN, ANDERS PhD†; OLCÉN, PER MD, PhD†; FREDLUND, HANS MD, PhD†
GENITAL Chlamydia trachomatis infection is a worldwide, often asymptomatic, treatable sexually transmitted disease (STD). 1 Untreated, it can cause tubal factor infertility and ectopic pregnancies. 2 Physicians in Sweden are therefore obliged to report C trachomatis infection by legislation passed in 1988, which also includes mandatory partner notification. 3 The rapid decline in incidence of C trachomatis infection in Sweden between 1988 and 1994 (284 and 155 cases per 100,000 inhabitants, respectively) 4 is believed to be a result mainly of the widely used screening, mandatory partner notification, and treatment. 5 However, there are doubts about the efficacy of partner notification, and since 1997 there has been an approximately 15% annual increase in reported cases. 6,7 Possible reasons for inadequate partner notification are insufficient cooperation from C trachomatis–infected patients, the limited catchment area for each clinic, delays in testing and treating partners, the insufficient diagnostic sensitivity for C trachomatis, and the existence of more than one genotype of C trachomatis within sexual networks. A network is defined here as an identified path of sexual contacts from one person to every other person.
Efforts against transmission can be more effective if the epidemiology of sexually transmitted diseases is more profoundly understood. Characterization of C trachomatis strains can provide valuable information about how the variants circulate in the community, as has been shown for Neisseria gonorrhoeae. 8,9 Serotyping with monoclonal antibodies recognizing antigenic determinants in the major outer membrane protein (MOMP) is the reference method for typing C trachomatis, 10–12 but new methods such as polymerase chain reaction (PCR), restriction fragment length polymorphism (RFLP), and sequencing of the omp1 gene, which encodes the MOMP, have recently been used. 13–18
The aims of the current study were to determine whether genotyping of the omp1 gene could alter the interpretation of C trachomatis transmission within sexual networks and to determine the efficacy of partner notification, including duration of the process of contact tracing.
Materials and Methods
In all, 2195 individuals attended the outpatient STD clinic at Örebro University Hospital during one year, March 1, 1999, to February 29, 2000. Of these, there were 1054 women between 13 and 59 years old (mean, 25.7 years) and 1141 men between 14 and 68 years old (mean, 28.5 years). Positive test results for C trachomatis were obtained and strains successfully sequenced for 231 patients: 96 women aged 15 to 54 years (mean, 22.7 years) and 135 men aged 16 to 49 years (mean, 24.8 years). Standard antibiotic treatment for C trachomatis infection was with lymecycline, doxycycline, or azithromycin. The physician noted compliance to antibiotic treatment and omissions of unprotected sexual contacts in patients’ casebooks at the follow-up visit.
Sampling and C trachomatis Detection
Specimens (urethral for men and endocervical for women) were obtained with polyacrylamide swabs. These were directly transported in 2-SP medium to the laboratory for tissue culture processing. First-void urine specimens (5–10 ml) were collected from both men and women, were transported to the laboratory, and after storage at 2 °C to 8 °C were processed for PCR. The first 779 urine samples were tested with Chlamydia trachomatis Amplicor PCR and the rest with the COBAS Amplicor Chlamydia trachomatis Test (Roche Diagnostic Systems, Branchburg, NJ) because the laboratory switched to a PCR with higher capacity but involving the same PCR and ELISA methods. All C trachomatis–positive patients were requested to attend a follow-up visit 4 weeks after antibiotic treatment commenced. Sampling and detection as described above were performed.
A total of 237 specimens from 231 patients were included in the study. The typing procedure was previously described in detail by Jurstrand et al. 17 The primary sample used for the omp1 PCR was the urogenital specimen (n = 189), or if this was negative or not obtained, the urine sample (n = 48). One C trachomatis–negative patient sample per every C trachomatis–positive sample was randomly selected each day, and these samples were used as negative controls in the study.
DNA was isolated directly from the clinical samples. Omp1 PCR was carried out with use of the primers P 1 (5′-ATG AAA AAA CTC TTG AAA TCG G-3′) and OMP 2 (5′-ACT GTA ACT GCG TAT TTG TCT G-3′) (Scandinavian Gene Synthesis AB, Köping, Sweden). The PCR products were purified and mixed with one of the primers, S1 (5′-TTG AGT TCT GCT TCC TCC T-3′) or OMP 2 (Scandinavian Gene Synthesis AB), in separate reaction mixtures and were sequenced with the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit (Applied Biosystems, Warrington, United Kingdom). After removal of dye terminators with a Dye Ex Spin kit (catalog no. 63106; Qiagen, Hilden, Germany), the sequence of the omp1 gene was determined with use of an ABI PRISM 310 Genetic Analyser (Applied Biosystems). Each PCR product was sequenced twice in each direction, which gave an overlap of about 200 base pairs amid the omp1 gene. The individual consensus sequence (about 1100 nucleotides) of the clinical isolates was compared to omp1 nucleotide sequences of known serovars of C trachomatis strains, with use of the BLAST search tool at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov).
Genotype E was the most common strain (n = 112; 47.3%), followed by F (n = 41; 17.3%) and K (n = 21; 8.9%). The prevalences of Ba, D, D/B-120, D/B-185, G, H, Ia, and J genotypes were 0.4% to 5.9%. One woman had a mixed E + F infection. 17
In this report the term network is used for designation of the anamnestically revealed part of the total sexual network including all individuals and their known sexual links in the society. Network is synonymous with the term component proposed by Wylie and Jolly and consists of two or more persons and applies where an identified path from one person to every other person exists. 18 Thus, not only attendees at the STD clinic but also partners tested at other clinics, partners identified but not tested, and partners identified but not possible to trace were included in the networks.
In the beginning of the study period the genotyping was delayed, and the network was scheduled only from the anamnesis of index patients and our knowledge of who was C trachomatis–positive and who was not and the estimated time for transmission; hence the data were provisional. Later in the study the genotyping was performed shortly after the routine testing, and hence it was possible to schedule the network with known genotypes.
We also use the term chain to describe the transmission of one C trachomatis strain between individuals within the networks.
Partner notification was performed according to Swedish standard methods. 5 The C trachomatis–positive patient, the index patient, was obliged to facilitate the notification. In most cases the index patient contacted all recent partners, but a letter sent to the partners by the contact tracer was an alternative way. The identity of the index patient was not revealed to the partner in the latter situation. The contact tracer asserted by letter or personal contact that notified partners had been tested. However, if the partner attended another clinic, revealing the result of C trachomatis testing to the tracer was prohibited by Swedish law. The goal of partner notification is to find every C trachomatis–positive partner.
The index patients in this study were advised to ask partners to attend the Örebro STD clinic, but they were free to choose any clinic. A social worker (M. L.) at the STD clinic performed the contact tracing. She always met the patient at the follow-up visit and customarily at the first visit if a C trachomatis infection was suspected. A network with a time axis was outlined. The duration of C trachomatis infection was estimated for each patient. The genotypes were established, in most networks, after partner notification was completed.
To determine contact-tracing efficacy, the following definitions were used. (1) Success means that all individuals within a network were traced and tested and that the transmission of infection was understood. (2) Possible success means that probably all individuals were tested but the transmission of infection was not absolutely determined. (3) No success means that there were missing links and/or it was uncertain whether all individuals were tested.
A total of 161 networks comprising 688 individuals could be identified, of which 328 individuals were shown to be C trachomatis–positive (Table 1). Samples from 231 of those were omp1 gene–sequenced. Sequencing of C trachomatis strains was not performed in samples from 83 persons because they attended and tested C trachomatis–positive at other clinics and their specimens therefore were not available at our laboratory for sequencing. A further 14 C trachomatis–positive samples from patients attending the Örebro STD clinic and belonging to networks were not sequenced because the patients attended before or in some cases after the inclusion period for sequencing.
The genetic congruence in the omp1 sequence was high. Mutations were seen in a single position in 21 of 237 clinical strains (genotypes D, D/B-120, E, F, J, K, and G). All six clinical specimens with genotype H had mutations in two positions. Four identical mutations were seen in five of six clinical specimens with genotype G. 17
Specimens were sequenced from at least two patients in 47 of 161 networks. In seven of these 47 networks (15%) there were discrepant genotypes (Table 2), and in two of these seven networks three different genotypes were found.
Only one case of mutation was seen within a chain, which comprised one man and one woman, both asymptomatic. The man harbored a C trachomatis genotype strain E with a mutation at position 997 (G→A). Both were tested within the same week.
Examples of Networks
In the following examples the term index patient is used to describe the first C trachomatis–positive patient attending the STD clinic and from whom the partner notification starts.
In Figure 1 the network comprises nine individuals: seven C trachomatis–positive (five sequenced) and two negative. One woman had a genotype F strain in the urine sample and a genotype E strain in the endocervix sample. Figure 2 shows an example of an uncommon genotype and long duration of intrapersonal C trachomatis infection. Figure 3 shows the network comprising the largest number of individuals (n = 26) in this study, involving 14 C trachomatis–positive specimens. The duration of C trachomatis infection in this network was estimated to be 13 months or more. All C trachomatis–positive genotype E strains were identical and showed no mutation in comparison with the GenBank strain. The two D/B-120 strains were identical and had a mutation at position 1045 (T→C) in comparison with the GenBank reference strain.
Geographic Distribution of Genotypes
Twenty-two patients probably contracted their C trachomatis infection abroad, namely, one in Australia (genotype D), two in Thailand (Ba, E), two in the United States (Ia), two in Austria (D/B-120, D), two in Bulgaria (E), three in Greece (F , E), four in the Netherlands (F , D), two in Norway (F, D), two in Spain (Ia, K), one in Switzerland (D), and one in the United Kingdom (E).
Genotype Ia was detected in seven persons. Two contracted their infection in the United States, one in Spain, two in a small town in another Swedish county, one from a Zairian man living in Stockholm, and one from an unknown source. The distribution of the other genotypes in the current study showed no clear geographic pattern.
A follow-up visit at the STD clinic was requested of all C trachomatis–positive patients. Out of 133 men tested at the STD clinic with sequenced strains, 120 (90%) reattended for a control assessment (a median of 4 weeks [range, 3–41] after first visit and first treatment day). The corresponding figures for 93 women tested at the STD clinic were 84 (90%), at a median of 4 weeks (range, 3–24). Tests were C trachomatis–positive for three women and two men at the follow-up visit (one man was positive three times). Two harbored a new genotype and thus had contracted a new C trachomatis infection. 17
Partner Notification Efficacy
In 71 of 161 networks (44%), the duration of contact tracing could be calculated. The median duration was 9 weeks (mean, 12.5) and the range of duration (until the last traced partner was tested) was 1 to 69 weeks, starting from the index patient's first day of antibiotic treatment.
In 90 of 161 networks (56%) the duration of contact tracing could not be calculated; 85 networks comprised only 1 individual attending the STD clinic. In the other five networks, partners attended the STD clinic and tested C trachomatis–negative before the index patient attended.
In 10 of 30 completed networks comprising 3 to 6 individuals, the duration could be calculated, and the mean time until testing of each partner was 4 weeks (range, 1–14); the mean time until completion was 8 weeks (range, 1–28).
Most networks (n = 131; 81%), however, could not be completed because the index patient withheld or provided insufficient information to the contact tracer for partner identification (Table 3). In many cases the patient lacked adequate information for partner identification by the tracer but had enough information to contact the partner himself or herself. Thus, it could be difficult for the contact tracer to trace partners or to know if they had attended any clinic for C trachomatis testing.
Duration of C trachomatis Infection
In 137 of 161 networks (85%) the median duration of C trachomatis infection could be estimated as 15 weeks (mean, 23 weeks), with a plausible range from 1 to 160 weeks.
In 24 of 161 networks (15%) the duration could not be calculated, for a variety of reasons: discrepant genotypes (n = 4); all examined partners were C trachomatis–negative (n = 3); a long relationship with a partner may also comprise another partner unknown to contact tracer (n = 6); asymptomatic index patients and partners not possible to trace (n = 9); and index patient lost to follow-up and partner notification not completed (n = 2).
The results obtained in this study indicate a high rate of different chains within the scheduled networks, and in the seven networks with discrepant genotypes this knowledge would have altered and intensified the contact tracing. Not all specimens from C trachomatis–positive patients in this study were included for sequencing because some patients attended the STD clinic before or after the inclusion period and some patients belonging to the networks attended other clinics. In fact, only 47 of 161 networks involved more than one sequenced C trachomatis–positive specimen. Genotyping of all specimens from C trachomatis–infected persons in the sexual networks would probably have revealed more discrepant genotypes and new ways of interpretation.
The distribution of genotypes differs between different geographic regions. In Sweden it seems that genotype E is the most frequent C trachomatis strain at present, as shown in this study and others. 19 Other studies in Portugal and Senegal showed a similar pattern with a predominating genotype E, 20,21 whereas an American study showed a different and more even pattern, with genotypes F, D, and H as the most prevalent. 22 If a network comprises an uncommon genotype, it can be established that the partners belong to the same network, as shown in Figure 2, where all individuals with sequenced C trachomatis strains had the C trachomatis genotype K strain. The use of genotyping in contact tracing gives less information in the case of a common genotype, such as genotype E in this study, which is characterized by its high genetic stability. This has also been shown by other investigators, who found a high degree of identical strains of the genotypes E, F, and Ia (in comparison with GenBank reference strains) in specimens from women obtained over a 20-year period. 23
Genotyping might also be useful in legal issues, for instance, in cases of suspected sexual abuse involving identical or discrepant C trachomatis genotypes. Furthermore, it could be useful in follow-up of unrelenting chlamydial infection, for determining whether a new infection, reinfection, or treatment failure has occurred. 17,24 Discrepancy of genotypes in comparison with the first sample would implicate a new infection.
The genotyping from routine C trachomatis–positive samples was feasible because all specimens except one could be genotyped and there were no problems in interpretation, whether invasive or urine samples were used. 17 Bandea et al 22 have recently published a report of a study in which they also showed that C trachomatis genotyping from urine samples was convenient and reliable. However feasible, the procedure is still laborious and hence expensive and not cost-effective for routine laboratory purposes. For epidemiologic studies and in special circumstances, e.g., sexual abuse, genotyping still could play an important role. The genotyping procedure could in the future probably be automated with a higher capacity and thus be more cost-effective.
The transmission dynamics, the reproductive rate of a sexually transmitted infection, depend on the duration of the infection, the infection's transmission probability, and the number of partner shifts per time unit, as shown in the transmission model by Anderson et al. 25 Hence, it is urgent to find the so-called core patients with many partner shifts. However, there is no well-defined threshold or boundary that separates the core group from other individuals, and the definition must be arbitrary. 26 An example of a radial network, 18 which emanates from persons with many partner shifts, in this study comprised 26 individuals, including 1 woman who had 9 partners during 8 months. Although most partners were very cooperative, exploration of the network took 50 weeks, and still some partners were not traceable.
The efficacy of partner notification among C trachomatis–infected patients in Sweden has previously been studied, and investigators have shown that social workers are more successful than physicians. 5 In this study a social worker (M. L.) performed the contact tracing. Partner notification is, despite legislation and index patients’ willingness to cooperate, not optimal at present because of the limited catchment area of each clinic. In this study only 30 of 161 networks (19%) could be completed. Although the failure of partner notification in many cases was due to the inadequacy of information given to the contact tracer, the limited catchment area for each clinic reduces the possibility of completing partner notification and hence ceasing the spread of infection. Partners can attend any clinic, and the result of C trachomatis testing is concealed from the tracer. A centralized STD control program would probably be more efficient, as proposed by Wylie and Jolly. 18 Contact tracing is also time-consuming, 27 both in contact with patients and in the compiling of the networks, which is a necessity for understanding the transmission of infection. Nevertheless, we believe that contact tracing, also implicating the unique opportunity for counseling and secondary STD prevention in the direct contact with the patient, is a most important tool in reducing the incidence of C trachomatis infection and other sexually transmitted infections in society.
The range of duration of C trachomatis infection within the networks and among individuals was very wide in this study. It could be established that one man infected with genotype K had been harboring the infection for at least 15 months. In one network, a couple might have been infected for 160 weeks. However, there might have been another source of the infection during this period. Current data do not allow one to reliably estimate the duration of genital infections with C trachomatis. 28
In conclusion, we have by means of a feasible and reliable method shown that genotyping of C trachomatis could be useful in a partner-notification context. At present, the method is expensive and not cost-effective. If there were wide distribution of different genotypes in society, genotyping would be more helpful in identifying circulating strains. However, the dominance of the C trachomatis genotype E strain and its high genetic stability reduced the effectiveness in this study.
1. Cates WJ Jr, Wasserheit JN. Genital chlamydial infections: epidemiology and reproductive sequelae. Am J Obstet Gynecol 1991; 164( 6Pt2): 1771–1781.
2. Kamwendo F, Forslin L, Bodin L, Danielsson D. Programmes to reduce pelvic inflammatory disease: the Swedish experience. Lancet 1998; 351( suppl 3): 25–28.
3. The Communicable Diseases Act. Stockholm: Swedish Legislature, 1988: 1472.
4. Annual Report. Swedish Institute for Infectious Disease Control. Stockholm: SMI, 1994.
5. Eitrem R, Erenius M, Meeuwisse A. Contact tracing for genital Chlamydia trachomatis in a Swedish county. Sex Transm Dis 1998; 25: 433–436.
6. Annual Report. Swedish Institute for Infectious Disease Control. Stockholm: SMI, 2001.
7. Götz H, Lindbäck J, Ripa T, Arneborn M, Ramstedt K, Ekdahl K. Is the increase in notifications of Chlamydia trachomatis infections in Sweden the result of changes in prevalence, sampling frequency or diagnostic methods? Scand J Infect Dis 2002; 34: 28–34.
8. Berglund T, Fredlund H, Giesecke J. Epidemiology of the reemergence of gonorrhea in Sweden. Sex Transm Dis 2001; 28: 111–114.
9. Unemo M, Berglund T, Olcén P, Fredlund H. Pulsed-field gel electrophoresis as an epidemiologic tool for Neisseria gonorrhoeae: identification of clusters within serovars. Sex Transm Dis 2002; 29: 25–31.
10. Lampe MF, Suchland RJ, Stamm WE. Nucleotide sequence of the variable domains within the major outer membrane protein gene from serovariants of Chlamydia trachomatis. Infect Immun 1993; 61: 213–219.
11. Wang SP, Kuo CC, Barnes RC, Stephens RS, Grayston JT. Immunotyping of Chlamydia trachomatis with monoclonal antibodies. J Infect Dis 1985; 152: 791–800.
12. Yuan Y, Zhang YX, Watkins NG, Caldwell HD. Nucleotide and deduced amino acid sequences for the four variable domains of the major outer membrane proteins of the 15 Chlamydia trachomatis serovars. Infect Immun 1989; 57: 1040–1049.
13. Frost E, Deslandes S, Veilleux S, Bourgaux-Ramoisy D. Typing Chlamydia trachomatis by detection of restriction fragment length polymorphism in the gene encoding the major outer membrane protein. J Infect Dis 1991; 163: 1103–1107.
14. Kaltenboeck B, Kousoulas KG, Storz J. Two-step polymerase chain reactions and restriction endonuclease analyses detect and differentiate ompA DNA of Chlamydia spp. J Clin Microbiol 1992; 30: 1098–1104.
15. Lampe MF, Wong KG, Stamm WE. Sequence conservation in the outer major membrane protein among Chlamydia trachomatis strains isolated from the upper and lower urogenital tract. J Infect Dis 1995; 172: 589–592.
16. Ostergaard L. Diagnosis of urogenital Chlamydia trachomatis infection by use of DNA amplification. APMIS 1999; 107( suppl l89): 5–36.
17. Jurstrand M, Falk L, Fredlund H, et al. A. Characterization of Chlamydia trachomatis omp1 genotypes among sexually transmitted disease patients in Sweden. J Clin Microbiol 2001; 39: 3915–3919.
18. Wylie J, Jolly A. Patterns of chlamydia and gonorrhea infection in sexual networks in Manitoba, Canada. Sex Transm Dis 2001; 28: 14–24.
19. Sylvan SP, von Krogh G, Tiveljung A, et al. Screening and genotyping of genital Chlamydia trachomatis in urine specimens from male and female clients of youth-health centers in Stockholm county. Sex Transm Dis 2002; 29: 379–386.
20. Borrego MJ, Gomes JP, Lefebvre JF, Eb F, Orfila J, Catry MA. Genotyping of Portuguese Chlamydia trachomatis urogenital isolates. Genitourin Med 1997; 73: 561–563.
21. Sturm-Ramires K, Brumblay H, Diop K, et al. Molecular epidemiology of genital Chlamydia trachomatis infection in high-risk women in Senegal, West Africa. J Clin Microbiol 2000; 38: 138–145.
22. Bandea CI, Kubota K, Brown TM, et al. Typing of Chlamydia trachomatis strains from urine samples by amplification and sequencing the major outer membrane protein gene (omp1). Sex Transm Infect 2001; 77: 419–422.
23. Martin DH, Bandea C, Wang X, Black C. Temporal distribution of Chlamydia trachomatis omp 1 genotypes from 1980 through 1999 in New Orleans. Int J STD AIDS 2001; 12( suppl 2): 39.
24. Pedersen LN, Kjaer HO, Moller JK, Orntoft TF, Ostergaard L. High resolution genotyping of Chlamydia trachomatis from recurrent urogenital infections. J Clin Microbiol 2000; 38: 3068–3071.
25. Anderson RM, Medley GF, May RM, Johnson AM. A preliminary study of the transmission dynamics of the human immunodeficiency virus (HIV), the causative agent of AIDS. IMA J Math Appl Med Biol 1986; 3: 229–263.
26. Liljeros F, Edling CR, Nunes Amaral LA, Stanley HE, Åberg Y. The web of human sexual contacts. Nature 2001; 411: 907–908.
27. Macke BA, Hennessy MH, McFarlane M. Predictors of time spent on partner notification in four US sites. Sex Transm Infect 2000; 76: 371–374.
28. Golden MR, Schillinger JA, Markowitz St. Louis ME. Duration of untreated genital infections with Chlamydia trachomatis: a review of the literature. Sex Transm Dis 2000; 27: 329–337.