Skip Navigation

Repeated Chlamydia trachomatis Genital Infections in Adolescent Women

  1. Byron E. Batteiger1,3,
  2. Wanzhu Tu2,5,
  3. Susan Ofner2,
  4. Barbara Van Der Pol1,6,
  5. Diane R. Stothard1,a,
  6. Donald P. Orr4,
  7. Barry P. Katz2 and
  8. J. Dennis Fortenberry4
  1. Divisions of
  2. 1Infectious Diseases
  3. 2Biostatistics, Department of Medicine
  4. 3Department of Microbiology and Immunology
  5. 4Department of Pediatrics, Section of Adolescent Medicine, Indiana University School of Medicine
  6. 5Regenstrief Institute
  7. 6Marion County Health Department, Indianapolis, Indiana
  1. Reprints or correspondence: Dr. Byron E. Batteiger, Indiana University School of Medicine, Div. of Infectious Diseases, 545 Barnhill Dr., Emerson Hall, Rm. 435, Indianapolis, IN 46202-5112 (bbatteig{at}iupui.edu)

  • ↵a Present affiliation: Eli Lilly, Indianapolis, Indiana

 
Next Section

Abstract

BackgroundRepeated Chlamydia trachomatis infections are common among young sexually active women. The relative frequency of reinfection and antibiotic treatment failure is undefined

MethodsAdolescent women enrolled in a longitudinal cohort had behavioral and sexually transmitted infection assessments performed every 3 months, including amplification tests for C. trachomatis, ompA genotyping, and interviews and diary entries to document sex partner-specific coitus and event-specific condom use. Repeated infections were classified as reinfection or treatment failure by use of an algorithm. All infections for which treatment outcomes were known were used to estimate the effectiveness of antibiotic use

ResultsWe observed 478 episodes of infection among 210 study participants; 176 women remained uninfected. The incidence rate was 34 episodes/100 woman-years. Of the women who were infected, 121 experienced ⩾1 repeated infections, forming 268 episode pairs; 183 pairs had complete data available and were classified using the algorithm. Of the repeated infections, 84.2% were definite, probable, or possible reinfections; 13.7% were probable or possible treatment failures; and 2.2% persisted without documented treatment. For 318 evaluable infections, we estimated 92.2% effectiveness of antibiotic use

ConclusionsMost repeated chlamydial infections in this high-incidence cohort were reinfections, but repeated infections resulting from treatment failures occurred as well. Our results have implications for male screening and partner notification programs and suggest the need for improved antibiotic therapies

Chlamydia trachomatis is the most common cause of bacterial sexually transmitted infection (STI) and is associated with an increased risk of pelvic inflammatory disease, ectopic pregnancy, tubal infertility, and increased susceptibility to human immunodeficiency virus infection [1, 2]. Repeated chlamydial genital infections are common [3–6] and account for a substantial proportion of incident infections [7]. Repeated infections result from failure of antibiotic therapy or from reinfection due to unprotected sexual contact with either an untreated existing partner or a new infected partner. Distinguishing among these possibilities is important to focus treatment recommendations and disease control activities. For example, if many repeated infections are due to antibiotic treatment failure, then better antibiotic treatment regimens are needed. If most are reinfections, then strategies to either expedite partner treatment [8] or screen and treat men in high-risk networks [9] are necessary. The relative frequency of treatment failure and reinfection is not well defined

Studies identifying risk factors for repeated C. trachomatis genital infections [5, 10–14] have not used the biomarkers necessary to distinguish the different types of repeated infection and thus represent a composite of reinfections, treatment failures, and failure to receive treatment. The standard biomarker is serotype or genotype based on the chlamydial major outer membrane protein or the gene ompA [10, 15–17]. When different strains are detected for the initial and repeated episodes, the second episode is considered to be a reinfection. When the strains detected for each of the 2 episodes are identical, the repeated episode could be due to reinfection from an untreated partner, antibiotic treatment failure, or reinfection by a strain from a different partner that is indistinguishable from the original strain. Therefore, resolution of same-strain repeated infections into reinfection or treatment failure requires detailed treatment and behavioral information. For example, a same-strain repeated infection in the absence of coitus is more likely the result of treatment failure, whereas reinfection is more likely if coitus with the same untreated partner occurs between the initial and repeated episodes of infection

We report our experience with a cohort of women enrolled in a study during middle adolescence, for whom frequent biological and behavioral sampling occurred over a median follow-up of 3.1 years. Our study design included (1) longitudinal follow-up with regularly scheduled visits; (2) frequent, repeated determination of infection status; (3) documentation of treatment; (4) frequent, repeated determination of sexual behaviors, including sex partner-specific coitus and condom use; and (5) genotyping. We developed an algorithm to assess whether repeated infections represent reinfections, treatment failures, or failure to receive treatment, and we estimated the effectiveness of antibiotic therapy in our cohort

Previous SectionNext Section

Subjects and Methods

Study participantsA convenience sample of young women 14–17 years of age who were receiving care at 3 primary care clinics in Indianapolis was enrolled, as described elsewhere [18–21]. Written informed consent from each participant and parental permission were obtained at enrollment. This research was approved by the institutional review board of Indiana University-Purdue University Indianapolis/Clarian Health

Study design and proceduresData were collected as part of a longitudinal observational study of the risk and protective factors associated with STIs in women in middle adolescence. Enrollment began in April 1999 and ended in July 2005; active participants were followed through July 2009. Clinical examinations and face-to-face interviews to obtain behavioral data were conducted at enrollment and every 3 months thereafter. Clinical examinations included cervical and vaginal swab sampling for diagnosis of STIs. Results of quarterly tests were available in 48–72 h, to provide guidance for clinical management. We attempted to treat all participants with infections. Retention was excellent, with data from only 5% of possible quarterly follow-up visits missing [21]. Goal participation for each participant was 27 months in an initial 5-year project period; in 2004, the project was extended to 10 years to allow additional enrollment and to continue follow-up of enrolled participants to a maximum of 8.2 years

In alternating quarters during each year of participation, participants completed daily behavioral diaries and submitted weekly self-obtained vaginal swab samples [21]. Up to 12 weekly home visits were conducted by research personnel to collect the diaries and vaginal swab samples. In the next quarter, the collection periods were followed by a rest period during which no diaries or weekly vaginal samples were collected. Weekly vaginal swab samples were considered to be research samples and were not tested at the time that they were received in the laboratory. Rather, they were stored at −20°C and then run in batches immediately before the next quarterly visit. Results were promptly communicated to clinicians so that any incident infections identified from the weekly specimens were treated at the quarterly visit. If participants experienced symptoms suggestive of STI during the collection periods, they were instructed to seek evaluation

Partner enrollmentAt each visit, participants were asked about current male coital partners. Consenting sex partners provided urine specimens for STI diagnostic testing on a one-time basis and were treated if they or the index participant were infected. All chlamydial infections were reported to the Marion County Health Department for partner notification, but partner notification by disease intervention specialists was not a part of the study design

STI diagnostic testingWe used the Amplicor CT/GC polymerase chain reaction (PCR) (Roche Diagnostics) nucleic acid amplification test (NAAT) to analyze all study specimens for C. trachomatis and Neisseria gonorrhoeae. Trichomonas vaginalis was detected using a modification of the CT/NG PCR assay that included primers and probes specific for T. vaginalis [22]. Because of reports of false-positive NAAT results for N. gonorrhoeae [23], samples for which positive results for N. gonorrhoeae were determined by CT/GC PCR had positive results confirmed by GenProbe APTIMA Combo 2 (AC2) [20], which amplifies a different molecular target

C. trachomatis ompAgenotypingDNA sequencing of polymerase chain reaction (PCR)-amplified ompA from clinical samples was performed according to the methods of Stothard et al. [16], except that DNA eluted from vaginal swab samples into 1 mL of molecular-grade water served as the starting material. PCR products were subjected to a reverse dot blot procedure to identify the presence of multiple serovars (eg, D and E) in study samples [24], with the limitation that the procedure cannot distinguish between strains with nucleotide polymorphisms (eg, D and D2). Procedures for sequencing, nucleotide sequence alignment, and sequence comparisons were performed as described elsewhere [16]

Laboratory reference strains were considered to be prototype strains [16, 24]. We identified 15 sequence variants based on nucleotide polymorphisms among samples from this study (Table 1). Each variant was confirmed by repeating the process described above with use of the original sample. There is no standard nomenclature for ompA sequence variants; for convenience, we named variants on the basis of the closest match with a given prototype, and we added numbers sequentially as variants were identified

Definition ofC. trachomatisinfection and infectionepisodes We considered any quarterly test demonstrating positive results for C. trachomatis to denote infection, without regard for the results of weekly tests, mirroring clinical practice. Weekly positive test results temporally associated with a quarterly test result were considered to be part of an infection episode, as defined below, and are illustrated in Figure 1. To more accurately classify results of quarterly tests, if the result of a quarterly test ending a weekly collection period was negative or missing, we considered an infection to be present if ⩾3 weekly test results were positive during the preceding collection period (Figure 1A). This requirement minimized misclassification resulting from a false-negative quarterly test result or, among the weekly samples, from a false-positive test result or transient DNA carriage resulting from coitus with an infected person. We required that these positive weekly test results occur ⩾3 weeks after treatment of a previous infection, to avoid misinterpreting as a new infection transient shedding of DNA occurring after successful treatment [19, 21]

Figure 1.
View larger version:
Figure 1.

Examples of infection episodes. Brackets above each example encompass an infection episode. A Incident infection defined by ⩾3 weekly tests in the absence of a positive quarterly test result and apparent spontaneous resolution but treated at a subsequent quarterly visit. B Infection detected at a quarterly visit and first weekly collection with treatment and clearance. C Infection detected at a quarterly visit with delayed treatment but then prompt clearance. D Asymptomatic incident infection emerging in a weekly collection period and treated at the subsequent quarterly clinic visit. E Infection detected at a quarterly visit, treatment within 2 weeks, and clearance; then subsequent unprotected coitus with an untreated partner infected with same genotype resulting in repeated incident infection; and then treatment of participant and partner with clearance. Q, quarterly visits; W, weekly home visits for sample collections; empty boxes, negative Chlamydia test result; hatched boxes, positive Chlamydia test result; letters in boxes, genotype; missing boxes, missing weekly samples; inverted triangles, azithromycin treatment; hearts, unprotected coitus

Examples in Figure 1 illustrate our approach to defining infection episodes by use of both quarterly and weekly test results. A commonly observed pattern of results was a positive result of a quarterly test at the beginning of a collection period, followed by positive results of weekly tests before treatment and for <3 weeks after treatment; these findings constituted a single episode of infection (Figure 1B and 1C). Another commonly observed pattern was several positive results of weekly tests before a quarterly clinic sample. These samples, marking the onset of an incident infection, plus the positive test result at the quarterly visit, constituted an episode of infection (Figure 1D) [18]. Because genotype was constant in different samples from the same infection episode (not shown), we selected another sample from that episode to define genotype if we were unable to amplify ompA from the primary quarterly sample

We identified the entire set of infection episodes and used these to determine the incidence of chlamydial infection and to identify participants with repeated infections

Documentation of treatmentFrom April 1999 through January 2005, study participants with chlamydial infections were given directly observed treatment with azithromycin. Subjects subsequently received prescriptions for azithromycin. Infections with T. vaginalis and N. gonorrhoeae were treated with single-dose therapy, largely by prescription according to published guidelines [31]. Antibiotics given and dosage and date of treatment were recorded. To identify effective treatments provided outside of study participation [21], all antibiotic orders and pharmacy transactions were extracted from the Regenstrief Medical Record System, an electronic medical record system that serves the clinics and associated healthcare system from which the participants were recruited [32]

Definition of episode pairsWith the high rate of return for the quarterly visit [21], and with weekly samples provided from 2 collection periods each year, we detected most incident chlamydial infections in the cohort. Episode pairs were defined as 2 successive infection episodes. Documented treatment was the primary data used to separate one episode from another, although in many instances multiple negative test results between infection episodes were documented as well. Sequences of positive weekly test results within a single collection period were defined as 2 infection episodes if treatment was documented during that period (Figure 1E)

We identified the entire set of episode pairs among participants with ⩾2 infection episodes. Episode pairs with ompA genotyping at both episodes were used to populate the classification algorithm. For example, if a participant had 4 total episodes making 3 episode pairs but genotyping was available for episodes 1, 3, and 4, only episode pair 3–4 was classified

Behavioral dataAt each quarterly visit, trained research personnel conducted a face-to-face interview to identify individual sex partners, occurrence of coitus with specific partners, and condom use with specific coital events occurring during the previous 3 months. These data were supplemented by daily diary entries obtained during the weekly sample collection periods, which identified specific sex partners, days on which coitus occurred, and condom use with each coital event. If coitus was documented in either source, we considered that coitus was documented by the available data. We limited evaluation of behavioral data to the relevant interval of time between each infection episode in a pair, with this interval defined as the 3 months before the second episode of infection

Classification algorithm and definitionsThe classification algorithm is depicted in Figure 2. Classifications were made by considering each episode pair individually; if the participant had >1 episode pair, each pair was evaluated separately. Because there is inherent uncertainty in the accuracy of diagnostic, treatment, and behavioral data, we defined repeated infections as either reinfections or treatment failures in a graded fashion, considering the strength of supporting data. “Definite reinfection” was defined as episode pairs with different genotypes, regardless of reported behaviors. “Probable reinfection” was defined as episode pairs with the same genotype, with unprotected coitus occurring with the same partner during the interim. “Possible reinfection” was defined as episode pairs with the same genotype, with unprotected coitus occurring with a different partner during the interim. “Persistence without treatment” was defined as episode pairs with the same genotype for which treatment could not be documented. “Probable treatment failure” was defined as episode pairs with the same genotype, with no coitus documented between episodes. “Possible treatment failure” was defined as episode pairs with the same genotype, with condom-protected coitus only

Figure 2.
View larger version:
Figure 2.

Repeated infection classification algorithm. *Three paired episodes were not classified, because each involved mixed infections that could not be resolved into same/different: F→F/Ja, D/Ja→D2, and D→D/F. **Three episode pairs with all coitus reported as condom protected were separated by multiple negative chlamydia test results, making treatment failure less likely: 1 quarterly and 9 weekly tests over 6.7 months, 5 quarterly and 19 weekly tests over 17.9 months, and 4 quarterly and 32 weekly tests over 13.2 months

Estimation of effectiveness of antibiotic useWhether treatment of the first episode of a paired episode is considered to be a cure or a failure is conditional on the outcome of the second episode in the pair. For example, if the second episode is a definite, probable, or possible reinfection, then this defines treatment of the first episode as resulting in a cure. Estimating effectiveness of use requires that we consider all episodes for which an outcome can be determined, including those not followed by another episode. Thus, in participants with only 1 episode, or for the last episode of participants with ⩾2 episodes, treatment was considered to be successful if it was followed by ⩾2 negative quarterly test results. Effectiveness of use was calculated as the percentage of successful treatments among all evaluable episodes of infection

Data analysisThe cohort was described using summary statistics, including means, medians, ranges, frequencies, and proportions. All paired episodes of infection with complete genotyping data were classified using the algorithm; repeated infections meeting the definitions mentioned above were counted and reported. A χ2 test was used to test the difference in the proportion of participants with incident infections. A multiple regression model was used to assess the effects of potential correlates of incident infection. To compare paired episodes due to same and different genotypes, we used bootstrap techniques to construct 95% confidence intervals for the mean difference of percentages of classified episodes [33]. Confidence intervals for use effectiveness were obtained from logistic regression analysis using generalized estimating equations to accommodate multiple infection episodes contributed by the same subject [34]

Previous SectionNext Section

Results

Incident and prevalent chlamydial infectionsThe demographic, behavioral, and clinical characteristics of study participants are reported in Table 2. Of the 386 participants enrolled in the study, 365 had ⩾1 quarterly follow-up visit and were included in calculation of the rate of incident infection. We identified 478 episodes of chlamydial infection in 210 participants; 42 were prevalent infections identified at study entry. The incidence rate was 34 cases/100 woman-years. Incident infections occurred more commonly in participants with infections at baseline than in those without infections at baseline (78.1% vs 51.7%; P<.001). Incident infection was associated with having >1 sex partner in the 3 previous months (odds ratio [OR], 2.15 [95% confidence interval {CI}, 1.61–2.85]; P<.001), concurrent infection with N. gonorrhoeae (OR, 3.75 [95% CI, 2.56–5.49], P<.001), and a history of STIs, as determined from yearly questionnaires (OR, 1.52 [95% CI, 1.16–1.99]; P<.002). Table 3 shows cumulative prevalence and quarterly visit point prevalences, which remained high throughout the study. The distribution of episodes of infection among participants is shown in Table 4; 176 participants never acquired a chlamydial infection during 477 woman-years of follow-up

Table 1
View larger version:
Table 1

ompA Genotypes Identified for 359 Infection Episodes Occurring among Adolescent Women

Table 2
View larger version:
Table 2

Demographic, Behavioral, and Clinical Characteristics of Participants

Table 3
View larger version:
Table 3

Point Prevalences of Chlamydia trachomatis at Quarterly Visits

Table 4
View larger version:
Table 4

Distribution of Chlamydia trachomatis Infections

We enrolled 313 unique individuals as sex partners; 66.7% of women had at least one sex partner enrolled. Infected participants named 1387 sex partners in the interviews and diaries. We estimated that 313 (22.6%) of 1387 possible sex partners were enrolled and tested; 82 (26.2%) of these 313 men were infected. We could not verify infection or treatment status in sex partners who were not enrolled

ompAgenotyping of episodesWe determined genotype from 692 discrete Chlamydia-positive samples, representing 359 (75.1%) of 478 episodes of infection (Table 1). All 15 identified variants were observed in multiple discrete samples obtained from single participants and often in both epidemiologically linked and unlinked samples from other participants or sex partners. We found no difference in the distribution of genotypes among participants with a single infection versus those with ⩾2 infections (not shown). Of the 268 episode pairs among 121 participants with ⩾2 episodes of infection, we attempted genotyping for 245 pairs (91.4%) and were successful in identifying genotypes at both episodes in 186 (74.9%) of the 245 pairs. Three episode pairs contained mixed serovars, leaving 183 to classify in the algorithm (Figure 2)

Classification of paired episodesDifferent genotype paired episodes were significantly more likely to be associated with a participant having a different partner at the time of the second episode (Table 5). Figure 2 shows the classification of all episode pairs. In summary, 25 (13.7%) of 183 repeated episodes were probable/possible treatment failures; 154 (84.2%) of 183 were definite (n=100), probable (n=32), or possible (n=22) reinfections; and 4 (2.2%) of the 183 episodes were not associated with documented treatment. Intermediate (serovars F and G), B (B, D, and E), and C (H, I, Ia, J, Ja, and K) serogroups were not associated with either reinfection or treatment failure. Of the same-genotype episodes for which treatment was documented, 25 (31.6%) of 79 were classified as possible/probable treatment failures, whereas 54 (68.4%) of 79 were classified as possible/probable reinfections

Table 5
View larger version:
Table 5

Comparison of Infection Episode Pairs of the Same and Different Genotypes

Partner genotype was available at the second episode for 24 paired episodes; in 20 of 24 instances, a partner genotype matched the genotype of the participant and served to corroborate our clinical classifications (9 of 12 definite reinfections, 8 of 8 probable reinfections, 1 of 2 possible reinfections, and 2 of 2 possible treatment failures)

Estimation of effectiveness of antibiotic useVirtually all women in the study were treated with single-dose azithromycin. We estimated the effectiveness of treatment use by using the set of 318 episodes for which treatment was documented and an outcome determined. Successful treatment was defined as (1) the subset of single episodes and last episodes (for those individuals with ⩾2 episodes) for which there were 2 negative quarterly test results (n=139) and (2) all first episodes where the second episode was classified in the algorithm as definite, probable, or possible reinfection (n=154) (Figure 2). Treatment was thus successful for 293 episodes. We considered as unsuccessful treatment all episodes classified as probable (n=10) or possible (n=15) treatment failures (Figure 2). When the data were combined, 25 (7.9%) of 318 evaluable episodes of infection were considered to be probable/possible treatment failures, providing an estimate of antibiotic use effectiveness of 92.1% (95% CI, 89.9%–96.0%)

Previous SectionNext Section

Discussion

Most repeated chlamydial infections were definite or probable/possible reinfections in this cohort, on the basis of our classification scheme. The consistently high point prevalences at clinic visits occurring every 3 months, the nearly 98% rate of documented treatment, and the high partner prevalence (26.2%) are also consistent with frequent reinfection. Our analysis suggests that frequent testing and treatment of women alone will not suffice to reduce prevalence in high-risk populations, highlighting the need for methods to expedite partner treatment and screening and treatment of networks of young men at high risk. Our results also indicate that little protective immunity is evident in this setting characterized by frequent testing and prompt treatment

Despite the preponderance of reinfections, probable/possible treatment failures accounted for 13.7% of paired repeated infections; by considering all episodes with an outcome, we estimate a 92.1% effectiveness of antibiotic treatment. This estimate is lower than the 95% effectiveness at 1 month reported in sexually transmitted disease clinics [35] and the 97% microbiologic cure rates reported in controlled trials [36]. These studies were limited by short durations (2–5 weeks) of follow-up after treatment and by use of culture rather than NAAT in some studies. Our analysis is unique because follow-up is of long duration, with repeated and systematic ascertainment of coitus and infection status. A study of expedited partner treatment using NAATs and 3- to 19-week follow-up also reported treatment failure rates of 8% [8]. These data suggest that, despite the accepted effectiveness of single-dose azithromycin, improved treatment regimens should be sought

The rate of incident chlamydial infection in the cohort was 34 episodes/100 person-years. Similar high rates have been found among African-American adolescent women in Denver (29.5 episodes/100 person-years) [13] and Baltimore (33.6 episodes/100 person-years) [3]. These studies, based on larger cohorts, relied on returns to care venues prompted by symptoms or being named as a contact of an person with an STİ, rather than by scheduled follow-up. Nevertheless, our results can likely be generalized to similar populations of urban adolescents. We found, as have others [13, 37], that a chlamydial infection at baseline was associated with a higher frequency of incident infection during follow-up

Our genotyping results are consistent with the few studies in which repeated infections have been characterized: same-serovar/same-genotype infections are common, especially early after initial infection [10, 15, 30, 38]. Theoretically, multilocus strain typing [39–42] might classify as definite reinfections some repeated infections due to the same ompA genotype, if sufficient variation is found in future studies among epidemiologically independent isolates of common serovars. Genotypes were stable within infection episodes and were distributed in a manner similar to that noted in other cohorts [16, 17, 43–47]. Serogroups were not associated with reinfection, treatment failure, or likelihood of repeated same-genotype infection

Our study has several limitations. Although infection prevalence among enrolled sex partners was high, we lack complete data (infection, genotype, and treatment status) for each sex partner during periods relevant to reacquisition of infection by participants. Incorporating such data into our algorithm could provide more certainty in classifying repeated infections due to the same genotype. We may have failed to identify some incident infections occurring during rest periods, although we would miss only those infections that resolved before the next clinic visit. Our results may not be representative of populations with lower incidence rates; repeated infections in such populations may have different proportions of reinfection and treatment failure. Finally, classification of same-genotype repeated infections depended on behavioral data obtained by self-report from interviews and diary entries. Because we considered single unprotected coitus occurring during the 3 months before the repeated episode as indicating probable reinfection, failure to report such a contact would cause us to misclassify the repeated episode as a probable treatment failure and thus underestimate the effectiveness of antibiotic therapy

Our characterization of a relatively small but intensively followed cohort of urban adolescent women indicates that reinfection is the predominant mode of repeated infection. Without effective interventions among sexual partners or relevant sexual networks, testing and treating high-risk adolescent women even at 3-month intervals may not materially reduce the prevalence of infection in similar populations. In addition, the estimated effectiveness of antibiotic therapy in this setting is lower than that observed in formal treatment trials with shorter follow-up periods, suggesting that improved treatment regimens for chlamydial genital infection should be sought

Previous SectionNext Section

Acknowledgments

We thank Pat Brooks, the clinical research team, and the participants, for their outstanding dedication to the project, as well as Greg Toth, David Wormser, and Angela Berger, for expert assistance with ompA sequencing

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: B.V.D.P is a consultant for Roche Diagnostics. All other authors: no conflict

  • Financial support: Sexually Transmitted Infections Cooperative Research Centers, National Institute of Allergy and Infectious Diseases, National Institute of Health (grants U19AI43924 [to D.P.O.] and U19AI031494 [to Stanley Spinola]); National Institute of Child Health and Human Development (grant R01HD044387 [to J.D.F.])

  • Received May 20, 2009.
  • Accepted August 18, 2009.
Previous Section
 

References

  1. ↵
    1. Hillis SD,
    2. Owens LM,
    3. Marchbanks PA,
    4. Amsterdam LF,
    5. MacKenzie WR
    . Recurrent chlamydial infections increase the risks of hospitalization for ectopic pregnancy and pelvic inflammatory disease. Am J Obstet Gynecol176:103-7.
  2. ↵
    1. Holmes KK,
    2. Sparling PF,
    3. Stamm WE,
    4. et al.
    1. Paavonen J,
    2. Westrom L,
    3. Eschenbach D
    . Pelvic inflammatory disease. In: Holmes KK, Sparling PF, Stamm WE, et al., editors. Sexually transmitted diseases. p. 1017-50.
  3. ↵
    1. Burstein GR,
    2. Gaydos CA,
    3. Diener-West M,
    4. Howell MR,
    5. Zenilman JM
    . Incident Chlamydia trachomatis infections among inner-city adolescent females. JAMA280:521-6.
  4. ↵
    1. Burstein GR,
    2. Zenilman JM,
    3. Gaydos CA,
    4. et al
    . Predictors of repeat Chlamydia trachomatis infections diagnosed by DNA amplification testing among inner city females. Sex Transm Infect77:26-32.
  5. ↵
    Burden of recurrent Chlamydia trachomatis infections in young women. Arch Pediatr Adolesc Med161:246-51.
  6. ↵
    1. Whittington WLH,
    2. Kent C,
    3. Kissinger P,
    4. et al
    . Determinants of persistent and recurrent Chlamydia trachomatis infection in young women. Sex Transm Dis28:117-23.
  7. ↵
    1. Brunham RC,
    2. Pourbohloul B,
    3. Mak S,
    4. White R,
    5. Rekart ML
    . The unexpected impact of a Chlamydia trachomatis infection control program on susceptibility to reinfection. J Infect Dis192:1836-44.
  8. ↵
    1. Golden MR,
    2. Whittington WLH,
    3. Handsfield HH,
    4. et al
    . Effect of expedited treatment of sex partners on recurrent or persistent gonorrhea or chlamydial infection. N Engl J Med352:676-85.
  9. ↵
    1. Rietmeijer CA,
    2. Hopkins E,
    3. Geisler WM,
    4. Orr DP,
    5. Kent CK
    . Chlamydia trachomatis positivity rates among men tested in selected venues in the United States: a review of the recent literature. Sex Transm Dis35:S8-S18.
  10. ↵
    1. Blythe MJ,
    2. Katz BP,
    3. Batteiger BE,
    4. Ganser JA,
    5. Jones RB
    . Recurrent genitourinary chlamydial infections in sexually active female adolescents. J Pediatr121:487-93.
  11. ↵
    1. Fortenberry JD,
    2. Brizendine EJ,
    3. Katz BP,
    4. Wools KK,
    5. Blythe MJ
    . Subsequent sexually transmitted infections among adolescent women with genital infection due to Chlamydia trachomatis, Neisseria gonorrhoeae or Trichomonas vaginalis. Sex Transm Dis26:26-32.
  12. ↵
    1. Orr DP,
    2. Langefeld CD,
    3. Katz BP,
    4. et al
    . Factors associated with condom use among sexually active female adolescents. J Pediatr120:311-7.
  13. ↵
    1. Rietmeijer CA,
    2. Van Bemmelen R,
    3. Judson FN,
    4. Douglas JMJ
    . Incidence and repeat infection rates of Chlamydia trachomatis among male and female patients in an STD clinic: implications for screening and rescreening. Sex Transm Dis29:65-72.
  14. ↵
    1. Xu F,
    2. Schillinger JA,
    3. Markowitz LE,
    4. Sternberg MR,
    5. Aubin MR
    . Louis ME. Repeat Chlamydia trachomatis infection in women: analysis through a surveillance case registry in Washington State, 1993–1998. Am J Epidemiol152:1164-70.
  15. ↵
    1. Batteiger BE,
    2. Fraiz J,
    3. Newhall WJV,
    4. Katz BP,
    5. Jones RB
    . Association of recurrent chlamydial infections with gonorrhea. J Infect Dis159:661-9.
  16. ↵
    1. Stothard DR,
    2. Boguslawski G,
    3. Jones RB
    . Phylogenetic analysis of the Chlamydia trachomatis major outer membrane protein and examination of potential pathogenic determinants. Infect Immun66:3618-25.
  17. ↵
    1. Millman K,
    2. Black CM,
    3. Stamm WE,
    4. et al
    . Population-based genetic epidemiologic analysis of Chlamydia trachomatis serotypes and lack of association between ompA polymorphisms and clinical phenotypes. Microbes Infect8:604-11.
  18. ↵
    1. Van Der Pol B,
    2. Williams JA,
    3. Orr DP,
    4. Batteiger BE,
    5. Fortenberry JD
    . Prevalence, incidence, natural history and response to treatment of Trichomonas vaginalis infection among adolescent women. J Infect Dis192:2039-44.
  19. ↵
    1. Chernesky M,
    2. Caldwell H,
    3. Chistiansen G,
    4. et al.
    1. Batteiger BE,
    2. Tu W,
    3. Katz BP,
    4. et al
    . Application of Chlamydia trachomatis diagnostic PCR to test multiple serial vaginal samples in a longitudinal study of adolescent women. In: Chernesky M, Caldwell H, Chistiansen G, et al., editors. Chlamydial infections. p. 573-6.
  20. ↵
    1. Shew ML,
    2. Fortenberry JD,
    3. Tu W,
    4. et al
    . Association of condom use, sexual behaviors and sexually transmitted infections with the duration of genital human papillomavirus infection among adolescent women. Arch Pediatr Adolesc Med160:151-6.
  21. ↵
    1. Tu W,
    2. Batteiger BE,
    3. Wiehe S,
    4. et al
    .
  22. ↵
    1. Van Der Pol B,
    2. Kraft CS,
    3. Williams JA
    . Use of an adaptation of a commercially available PCR assay aimed at diagnosis of chlamydia and gonorrhea to detect Trichomonas vaginalis in urogenital specimens. J Clin Microbiol44:366-73.
  23. ↵
    1. Whiley DM,
    2. Tapsall JW,
    3. Sloots TP
    . Nucleic amplification testing for Neisseria gonorrhoeae: an ongoing challenge. J Mol Diagn8:3-15.
  24. ↵
    1. Stothard DR
    . Use of a reverse dot blot procedure to identify the presence of multiple serovars in Chlamydia trachomatis urogenital infection. J Clin Microbiol39:2655-9.
    1. Dean D,
    2. Millman K
    . Molecular and mutation trends analyses of omp1 alleles for serovar E of Chlamydia trachomatis. J Clin Invest99:475-83.
    1. Sayada C,
    2. Denamur CE,
    3. Elion J
    . Complete sequence of the major outer membrane protein-encoding gene of Chlamydia trachomatis serovar Da*. Gene120:129-30.
    1. Peterson EM,
    2. Markoff BA,
    3. de la Maza LM
    . The major outer membrane protein nucleotide sequence of Chlamydia trachomatis serovar E. Nucleic Acids Res18:3414.
    1. Zhang Y-X,
    2. Morrison SG,
    3. Caldwell HD
    . The nucleotide sequence of the major outer membrane protein gene of Chlamydia trachomatis serovar F. Nucleic Acids Res18:1061.
    1. Hamilton PT,
    2. Malinowski DP
    . Nucleotide sequence of the major outer membrane protein gene from Chlamydia trachomatis serovar H. Nucleic Acids Res17:8366.
  30. ↵
    1. Dean D,
    2. Suchland RJ,
    3. Stamm WE
    . Evidence for long-term cervical persistence of Chlamydia trachomatis by omp1 genotyping. J Infect Dis182:909-16.
  31. ↵
    1. Workowski KA,
    2. Berman SM
    . Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep55:1-94.
  32. ↵
    1. McDonald CJ,
    2. Overhage JM,
    3. Tierney WM,
    4. et al
    . The Regenstrief Medical Record System: a quarter century experience. Intl J Med Inform54:225-53.
  33. ↵
    1. Efron B,
    2. Tibshirani R
    . An introduction to the bootstrap. New York: Chapman & Hall/CRC.
  34. ↵
    1. Diggle PJ,
    2. Liang KY,
    3. Zeger SL
    . Analysis of longitudinal data. Oxford, United Kingdom: Clarendon Press.
  35. ↵
    1. Hillis SD,
    2. Coles FB,
    3. Litchfield B,
    4. et al
    . Doxycycline and azithromycin for prevention of chlamydial persistence or recurrence one month after treatment in women: a use-effectiveness study in public health settings. Sex Transm Dis25:5-11.
  36. ↵
    1. Lau C-Y,
    2. Qureshi AK
    . Azithromycin versus doxycycline for genital chlamydial infections. Sex Transm Dis29:497-502.
  37. ↵
    1. Richey CM,
    2. Macaluso M,
    3. Hook EW
    . Determinants of reinfection with Chlamydia trachomatis. Sex Transm Dis26:4-11.
  38. ↵
    1. Brunham RC,
    2. Kimani J,
    3. Bwayo J,
    4. et al
    . The epidemiology of Chlamydia trachomatis within a sexually transmitted diseases core group. J Infect Dis173:950-6.
  39. ↵
    1. Pannekoek Y,
    2. Morelli G,
    3. Kusecek B,
    4. et al
    . Multilocus sequence typing of Chlamydiales: clonal groupings within the obligate intracellular bacteria Chlamydia trachomatis. BMC Microbiol8:42.
  40. ↵
    1. Klint M,
    2. Fuxelius H-H,
    3. Goldkuhl RR,
    4. et al
    . High-resolution genotyping of Chlamydia trachomatis strains by multilocus sequence analysis. J Clin Microbiol45:1410-4.
  41. ↵
    1. Pedersen LN,
    2. Podenphant L,
    3. Moller JK
    . Highly discriminative genotyping of Chlamydia trachomatis using omp1 and a set of variable number tandem repeats. Clin Microbiol Infect14:644-52.
  42. ↵
    1. Gomes JP,
    2. Bruno WJ,
    3. Nunes A,
    4. et al
    . Evolution of Chlamydia trachomatis diversity occurs by widespread interstrain recombination involving hotspots. Genome Research17:50-60.
  43. ↵
    1. Millman K,
    2. Black CM,
    3. Johnson RE,
    4. et al
    . Population-based genetic and evolutionary analysis of Chlamydia trachomatis urogenital strain variation in the United States. J Bacteriol186:2457-65.
  44. ↵
    1. Lan J,
    2. Melgers I,
    3. Meijer CJLM,
    4. et al
    . Prevalence and serovar distribution of asymptomatic cervical Chlamydia trachomatis infections as determined by highly sensitive PCR. J Clin Microbiol33:3194-7.
  45. ↵
    1. Gao X,
    2. Chen X-S,
    3. Yin Y-P,
    4. et al
    . Distribution study of Chlamydia trachomatis serovars among high-risk women in China performed using PCR-restriction fragment length polymorphism genotyping. J Clin Microbiol45:1185-9.
  46. ↵
    1. Bandea CI,
    2. Debattista J,
    3. Joseph K,
    4. Igietseme J,
    5. Timms P
    . Chlamydia trachomatis serovars among strains isolated from members of rural indigenous communities and urban populations in Australia. J Clin Microbiol46:355-6.
  47. ↵
    1. Mossman D,
    2. Beagley KW,
    3. Landay AL,
    4. et al
    . Genotyping of urogenital Chlamydia trachomatis in regional New South Wales, Australia. Sex Transm Dis35:614-6.
| Table of Contents

This Article

  1. J Infect Dis. (2010) 201 (1): 42-51. doi: 10.1086/648734
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most