Tubal damage in infertile women: prediction using chlamydia serology

Valentine A. Akande1,4, Linda P. Hunt2, David J. Cahill1, E.Owen Caul3, W.Christopher L. Ford1 and Julian M. Jenkins1

1 University Division of Obstetrics and Gynaecology, St Michael’s Hospital, Bristol BS2 8EG, 2 Division of Child Health, University of Bristol and 3 Public Health Laboratory, Myrtle Road, Bristol, UK

4 To whom correspondence should be addressed. e-mail: valentine.akande{at}bristol.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The study explores the relationship between serum chlamydia antibody titres (CATs) and detection of tubal damage in infertile women. METHODS: The tubal status and pelvic findings in 1006 women undergoing laparoscopy for infertility were related to CAT, which was measured using the whole-cell inclusion immunofluorescence test. RESULTS: A negative correlation between CAT and age was noted. A linear trend between serum CAT and the likelihood of tubal damage, including severe damage, was observed (P < 0.001). Titres in women with tubal damage (median 1:1024; range <1:64–1:4096) were significantly (P < 0.001) higher than in women with endometriosis alone (median <1:64; range <1:64–1:512) or those with a normal pelvis (median <1:64; range <1:64–1:1024). Women with positive titres were more likely to have pelvic adhesions than tubal occlusion unless titres were very high, when tubal damage was likely to be more severe. CONCLUSIONS: CATs are of predictive value in the detection of tubal damage and are quantitatively related to the severity of damage. For practical clinical purposes, Chlamydia serology is useful mainly as a screening test for the likelihood of tubal damage in infertile women and may facilitate decisions on which women should proceed with further investigations without delay.

Key words: chlamydia/infertility/pelvis/serology/tubal damage


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Couples may present to their family doctor complaining of infertility after failing to conceive for months or years. Those with a short duration of infertility will mainly be normal and have simply been unlucky so far (Hull, 1992Go). In primary or secondary care of infertile couples, the principal aim is to screen for a likely cause deserving early referral for specialist diagnosis and treatment. Other infertile couples can reasonably be managed by simple advice and encouragement before specialist help can be justified. The clinical challenge is deciding which couples deserve early referral to a specialist for further investigation or treatment.

Tubal damage is a common cause of infertility, and laparoscopy or hysterosalpingography (HSG) are accepted methods for diagnosing this condition. They are, however, both costly and invasive, and therefore unsuitable for screening on a large scale. Genital Chlamydia trachomatis infection has a worldwide distribution (Stamm, 1999Go) and is now recognized as the single most common cause of tubal peritoneal damage (WHO task force on the prevention and management of infertility, 1995Go; ESHRE Capri Workshop, 1996Go). Infection with C.trachomatis results in the formation of antibodies detectable in serum. In contrast to laparoscopy or HSG, detecting evidence of past chlamydial infection using serology is non-invasive, simple and quick to perform (Akande, 2002Go). As such, chlamydia serology may be used as a screening test for tubal damage in infertile women. This study explores the relationship between chlamydia antibody titres (CATs) and tubal damage in infertile women undergoing investigative laparoscopy.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Local research ethics committee approval was obtained for the study. The study group was drawn from infertile women whose initial attendance at the Reproductive Medicine Clinic at St Michael’s Hospital in Bristol was between 1985 and 1995. The clinic is run as a subspecialty service in a university teaching hospital, dealing with infertile couples referred by GPs in the Bristol area and providing tertiary care to couples referred by specialists in Bristol, the South West of England and parts of South Wales. In this cross-sectional study, all women included had a diagnostic laparoscopy for assessment of tubal patency, fibrosis, distortion, or the presence of endometriosis or pelvic adhesions. Women who had a distinct cause of infertility such as ovulatory dysfunction with no index of pelvic disease would not have had a routine laparoscopy and some others conceived before laparoscopy was necessary or arranged. Although different clinicians carried out the laparoscopies over this period of time, each clinician employed the same technique because they were supervised initially by one of two consultants (Professor M.G.R.Hull or Mr P.G.Wardle) prior to being allowed to assess the pelvis independently. All the clinicians were accredited specialists or senior trainees. Findings were recorded in a standardized way. Once laparoscoped, all women were studied, regardless of infertility diagnosis. A diagnosis of pelvic endometriosis was made if there were any typical superficial lesions or higher grade disease.

Infertility was defined as secondary if a previous conception had occurred, regardless of the outcome. A history of pelvic inflammatory disease (PID) was defined as a previous acute episode of pelvic infection. Smoking implied that the women were smokers at the time of consultation. If there was more than one apparent cause of infertility, priority was given to the condition that was more likely to be the immediate cause. Women with tubal damage (or pelvic adhesions not due to endometriosis) served as the ‘cases’ to be identified by the test (chlamydia serology) and women without damage served as ‘controls’ regardless of their other infertility diagnoses. Women were also categorized according to three main findings at laparoscopy: (i) tubal damage; (ii) endometriosis; or (iii) normal pelvis. Tubal damage was diagnosed by the finding of tubal occlusion, and/or distortion of the fimbriae, and/or restrictive tubal–ovarian adhesions, in the absence of visible endometriosis. Severe tubal damage was classified using the ‘Hull and Rutherford’ classification for tubal disease (Akande, 2002Go; Rutherford and Jenkins, 2002Go). This classification referred to women with bilateral tubal damage with extensive tubal fibrosis, and/or tubal distension >1.5 cm, and/or an abnormal tubal mucosal appearance and/or bipolar occlusion, and/or extensive dense pelvic adhesions.

Laboratory procedures
A clotted blood sample was obtained from the patient prior to the laparoscopy and sent to the Public Health Laboratory Service in Bristol for assay. Both clinical and laboratory personnel were blind as to the pelvic status of the woman at that time. Serum samples were assayed for chlamydia IgG antibody employing the single-antigen inclusion test using indirect immunofluorescence, as previously described by Richmond and Caul (1975Go), otherwise known as the whole-cell inclusion immunofluorescence (WIF) assay. This was applied in practice (Conway et al., 1984Go) and modified using C.trachomatis L2 serotype (Saikku and Paavonen, 1978Go) as antigen to infect McCoy cell monolayers and anti-human IgM–IgA–IgG–fluorescein conjugate. Dilutions of sera were expressed as antibody titres from 1:64 to >=1:4096, or negative (<1:64) (Chernesky et al., 1998Go).

Statistical analysis
Statistical analysis included non-parametric Spearman’s rank correlation coefficients (rs) and Mann–Whitney U-tests (to compare titres between pairs of groups), {chi}2-tests of trend and standard measures of diagnostic power.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 1119 infertile women (of all diagnostic groups) who underwent laparoscopic investigation for infertility were identified. Complete data including chlamydia serology were available for 1006, and subsequent analysis is based on these. The women’s ages ranged from 18 to 48 years (median 31 years). There was a significant negative correlation [rs = –0.175, 95% confidence interval (CI) –0.234 to –0.114; P < 0.0001] between CAT and age. The duration of infertility at the time of laparoscopy ranged from 1 to 20 years (median 3.8 years). It was not significantly correlated with CAT and there was no significant difference in the duration of infertility between women with or without tubal damage.

CATs were significantly higher in women who had conceived previously compared with those with primary infertility. However, there was no significant difference between titres according to the couples’ past history of pregnancy. Smokers were found to have significantly higher titres than non-smokers (Table I). Although <7% of the women revealed a past history of PID, CATs in these women were significantly higher than in those with no previous history (Table I).


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of the infertile women undergoing laparoscopy studied in relation to median CATs
 
The antibody titres in women with tubal damage were significantly higher than in women without tubal damage (P < 0.001; Table I) as also evident in the frequency distribution of women with and without damage shown in Figure 1. Table II shows the diagnostic groups of the women in the study. Three hundred and ninety women had tubal damage as their main cause of infertility, and 44 women had tubal damage as an additional cause (secondary diagnosis), so that a total of 434 of the 1006 women (43%) had tubal damage or pelvic adhesions not related to endometriosis on the basis of laparoscopic findings. Of the 434 women with tubal damage, 417 (96%) had some form of pelvic adhesions and 296 (68%) had at least one tube occluded. Women with tubal damage but no tubal occlusion had significantly lower median antibody levels than those with at least one tube occluded (1:512 versus 1:1024; P < 0.001).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 1. Frequency (n) distribution of chlamydia antibody titres in women with and without tubal damage undergoing diagnostic laparoscopy for infertility.

 

View this table:
[in this window]
[in a new window]
 
Table II. Number of women in relation to main diagnosis causing infertility and frequency of tubal damage and pelvic adhesions found at laparoscopy
 
The distribution of antibody titres for all women and according to pelvic findings at laparoscopy is shown in Figure 2. The antibody titres for all women with endometriosis (n = 252, i.e. n = 222 as primary infertility diagnosis and n =30 as secondary diagnosis) were examined separately to determine the association between the adhesions found in women with endometriosis and evidence of previous chlamydial infection. This figure illustrates that titres were significantly higher in women with tubal damage compared with those with a normal pelvis or endometriosis (P < 0.001, Kruskal–Wallis test). However, if severe tubal damage was found (n = 225) then their titres were highest, and significantly higher than those with less severe damage (P < 0.001, data not shown).



View larger version (27K):
[in this window]
[in a new window]
 
Figure 2. Chlamydia antibody titres for all infertile women studied and according to pelvic findings at laparoscopy (horizontal lines = medians; shaded boxes = quartiles; whiskers = ranges; ovals = outliers; * = extremes).

 
A linear relationship between serum CAT and the likelihood of tubal damage was observed (Figure 3). The trends of increasing chlamydia antibody levels in relation to frequency of any tubal damage or severe tubal damage were both highly significant (P < 0.001; {chi}2 for trend). Figure 3 clearly illustrates that 17% (95% CI 13–20%) and 6.1% (95% CI 4.1–8.7) of women with negative titres (<1:64) had tubal damage or severe tubal damage, respectively. In women with the highest titres of >=1:4096, 100% (95% CI 87–100%) had tubal damage while 73.1% (95% CI 52.2–88.4%) had severe tubal damage. This diagram also illustrates that at higher titres, a greater proportion of women are likely to have severe tubal damage than at lower titres. The diagnostic ability of the immunofluorescence test used in detecting women with any degree of tubal damage and severe tubal damage is shown in Table III. Increasing the cut-off level improved specificity and the positive likelihood ratio (LR+) at the expense of sensitivity and the negative likelihood ratio (LR–).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 3. Frequency (%) of any tubal damage or severe tubal damage according to chlamydia antibody titres.

 

View this table:
[in this window]
[in a new window]
 
Table III. Diagnostic test analysis for all women with any tubal damage and pelvic adhesions and for women with severe tubal damage and pelvic adhesions in relation to serum CAT (immunoflouresence test) amongst infertile women undergoing laparoscopy for infertility (n = 1006)
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This is the first report of its kind employing a large population of infertile women for comparative analysis. Our study using laparoscopy on all patients confirms that past infection with C.trachomatis is associated with a significantly increased risk of women suffering tubal infertility, as shown by others (Conway et al., 1984Go; Westrom, 1987Go; Forsey et al., 1990Go; Mol et al., 1997Go; Thomas et al., 2000Go; Veenemans and van der Linden, 2002Go). Negative chlamydia serology (<1:64) does not, however, preclude the diagnosis of tubal damage. Conversely, high titres do not necessarily indicate the presence of tubal damage, as shown by the high titres observed in some women with a normal pelvis in the present study.

Laparoscopy is the accepted gold standard for the diagnosis of tubal damage (RCOG, 1998Go). The high prevalence of tubal damage observed may reflect the prolonged duration of infertility (3.8 years) of the women studied. The present study, however, was on a select group of women undergoing detailed infertility investigations in a tertiary setting. The studies’ use of controls from non-tubal cases theoretically meets the requirement for controls. Even so, there is a possibility that excluded cases who were not laparoscoped had conditions which precluded the use of laparoscopy or indeed that many minor cases of tubal damage may not have been picked, as they became pregnant spontaneously prior to laparoscopy. Nonetheless, a significant proportion of patients who underwent laparoscopy had no pelvic damage and were infertile due to other causes such as sperm or ovulatory dysfunction and unexplained infertility (Figure 2 and Table II).

The relatively high sero-prevalence of positive CAT and the relatively low proportion of women who give a history of previous PID attest to chlamydial infection being mainly asymptomatic (Bjercke and Purvis, 1993Go; Stamm, 1999Go). Smokers were also more likely to have a higher CAT, which may be due to socio-economic and other risk factor associations (Shafer et al., 1993Go). Ectopic pregnancies are a reflection of tubal damage, and smoking is recognized to be an independent risk factor (Saraiya et al., 1998Go; Bouyer et al., 2003Go), as is previous chlamydial infection (Brunham et al., 1986Go; Sheffield et al., 1993Go). It is therefore plausible that smoking predisposes women to tubal damage and thus ectopic pregnancies via increased susceptibility to chlamydia infection; this however remains to be tested.

The sensitivity of chlamydia serology in detecting tubo-peritoneal damage has been demonstrated by others (Dabekausen et al., 1994Go; Meikle et al., 1994Go; Land et al., 1998Go), including a meta-analysis (Mol et al., 1997Go). However, these studies included women with endometriosis considered as positive cases. The present study found that the majority of women with endometriosis had negative titres (median CAT <1:64), implying that adhesions found at laparoscopy in these women are unlikely to be of infective origin and therefore are unlikely to be detected using chlamydia serology. Like Anestad et al. (1987Go), we found that pelvic adhesions (not due to endometriosis) were the most frequent sequelae associated with a high CAT. The findings of their study (Anestad et al., 1987Go) suggested that adhesions were the most likely consequence of chlamydial infection, with occlusion being a manifestation of more severe infection associated with higher titres, consistent with our findings. chlamydia and gonorrhoea are both common causes of PID and often co-exist (Rice and Schachter, 1991Go), It is therefore plausible that in the women who had a history of acute PID and had negative chlamydia serology, this was caused by gonorrhoea or other organisms (Clausen et al., 2001Go).

A meta-analysis (Mol et al., 1997Go) showed that the performance of chlamydia antibody testing depended on the assay used, and found the WIF test with the enzyme-linked immunosorbent assay (ELISA) and micro-immunofluorescence (MIF) test to be superior to the immunoperoxidase assay. However, the studies examined were not strictly comparable because some were based on tubal damage diagnosed by HSG alone, and non-uniform cut-off levels were used. The immunofluorescence test employed in the present study is highly sensitive, as shown by a blinded comparative study of other serological tests for C.trachomatis antibody carried out in three international centres including one of ours (Chernesky et al., 1998Go; Jones et al., 2003Go). The test has the ability to detect both group lipopolysaccharide and species-specific antibodies, therefore high titres are indicative of highly specific C.trachomatis species antibody (Black, 1997Go). Consequently, women with positive serology but with a normal pelvis may have had non-genital chlamydia infection. In these cases, cross-reactive responses to past infection with other species of chlamydia such as Chlamydia pneumoniae (Gijsen et al., 2001Go) or Chlamydia psittaci (Bergstrom et al., 1996Go; Jones et al., 2003Go) is a possibility, but difficult to account for. Time-related antibody titre decline is a possible reason for false negatives (i.e. negative serology but positive laparoscopy), but this issue may be controversial. Previous studies have suggested a chronological decline in titres (Puolakkainen et al., 1986Go; Henry-Suchet et al., 1994Go); however, a more recent study revealed no significant decline (Gijsen et al., 2002Go). Another explanation for false negatives may be that the immune-mediated reaction responsible for adhesions or tubal occlusion may not have occurred in these women for unknown reasons (Witkin et al., 2000Go). Nonetheless, it is recognized that the C.trachomatis organism can be found in apparently healthy normal-looking Fallopian tubes (Menchaca et al., 1988Go; Shepard and Jones, 1989Go; Marana et al., 1990Go; Stacey et al., 1990Go). This suggests that its presence in the genital tract does not always result in damage and is consistent with the findings of the present study.

Because there are justified constraints to the indiscriminate use of laparoscopy and HSG, there is a need to minimize the number of patients subjected to these diagnostic investigations who do not have disease (false positives). Diagnostic tests have three main uses: diagnosis of disease; screening; and patient management (Campbell and Machin, 1993Go). Screening is defined as a procedure that helps identify a specified disease or condition (in this context, tubal damage). As most screened individuals will be unaffected, the test must be safe and acceptable. As a screening test, chlamydia serology would also be allowed to possess higher margins of error and may be less accurate than diagnostic tests (Peters et al., 1996Go). If laparoscopy is readily available and the primary aim of screening is to avoid delay in referral for IVF or tubal surgery in those with significant tubal damage, false negatives have to be minimized. As such, a low cut-off may be the preferred option in view of its higher sensitivity. To achieve the objective of identifying a subgroup of infertile women for further investigation, a cut-off level is required. However, a universal single cut-off which splits women into two groups is likely to be controversial.

Approaches to dealing with the problem of a cut-off titres have to take into account the quantitative nature of serology in relation to frequency of disease. For each antibody titre, the sensitivity, specificity, and positive and negative likelihood ratios should be calculated (Land et al., 1998Go; Akande et al., 2002Go). These parameters can be used to dichotomize patients into those that are high risk and low risk of being found with tubal damage. However, in defining a cut-off titre, three main issues need to be considered (Coggon et al., 1993Go). First, a cut-off titre based on statistical analysis which is acceptable as a simple guide to the limits of what is common. Secondly, a cut-off titre based on the perceived or actual clinical importance may define the level of the antibody titre above which the finding of tubal damage becomes more frequent. Finally, a cut-off based on prognostic factors where high chlamydia antibody levels may be associated with a greater severity of disease and a lower chance of conception. Each of these approaches is suitable for different purposes and needs to be defined when being used (Coggon et al., 1993Go). However, it is not always necessary to state a value as the need to do so would depend on whether the aim is to make a diagnosis or a prognosis (Altman, 1991Go). Again, this is a clinical judgement and not a statistical issue. For example, in screening for lethal disease, high sensitivity is desirable though the trade off is usually lower specificity. However, when dealing with a non-life-threatening condition such as tubal damage, a high cut-off (lower sensitivity) may be chosen, but this may miss some cases but lead to fewer women who do not have the disease being subjected to invasive and costly laparoscopy (high specificity). The use of prevalence-independent likelihood ratios can also help overcome this problem; a suitable cut-off titre would have a positive likelihood ratio >5 and a negative likelihood ratio <0.2 (Land et al., 1998Go). Although the choice of cut-off level is discretionary, our findings (based on the likelihood ratios) suggest that with a titre of >=1:256 using the WIF test, the risk of a woman having tubal damage is increased sufficiently to consider early referral for further investigation. Titres <1:256 may justify early referral only if there were other risk factors such as previous abdominal surgery or history of pelvic infection and the couples had not conceived within a reasonable amount of time.

It is tempting to suggest that early detection of a disease is an end in itself. However, the spectrum of disease varies according to the severity and extent of lesions (Akande, 2002Go). The present study is clear in demonstrating that severe damage is more likely in women with higher titres. This implies that increasing antibody titres are quantitatively related to both the presence of tubal damage and the severity of tubal damage. Thus identification of trivial disease such as minor filmy adhesions or indeed untreatable conditions such as bilateral distended hydrosalpinges are important in terms of prognosis for fertility for different reasons (Akande, 2002Go). Consequently, identifying women who are at sufficiently high risk of having severe tubal damage impairing fertility may be more important than identifying women with minimal tubal damage. In these circumstances, a higher antibody titre cut-off would be required (Table III).

Striking a balance, in a target population between, on the one hand, the severity of the disorder affecting fertility and the prevalence of disease, and, on the other, the availability, costs, hazards and acceptability of invasive diagnosis is a practical necessity. This study shows that using chlamydia serology for screening provides a useful guide to the risk of tubal damage causing infertility but also exposes certain limitations of this method of screening. However, the choice of cut-off level used for screening would depend on the prevalence of the disease in the target population to which it is applied and whether one wants to identify most cases of women with tubal damage or mainly those with severe damage.


    Acknowledgements
 
We would like to acknowledge the contribution of the late Professor M.G.R.Hull in the initial supervision of this study. The study was carried out using research funds provided by the University of Bristol Centre for Reproductive Medicine.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Akande, V. (2002) Tubal pelvic damage: prediction and prognosis. Hum. Fertil., 5, S15–S20.

Akande, V.A., Mathur, R.S., Emovon, E.U., Hull, M.G.R. and Jenkins, J.M. (2000) Risk factors for tubal pelvic infective damage (TPID) and prediction using Chlamydia serology. Hum. Fertil., 3, 142.

Altman, D.G. (1991) Practical Statistics for Medical Research. Chapman and Hall, London.

Anestad, G., Lunde, O., Moen, M. and Dalaker, K. (1987) Infertility and chlamydial infection. Fertil. Steril., 48, 787–790.[ISI][Medline]

Bergstrom, K., Domeika, M., Vaitkiene, D., Persson, K. and Rightrdh, P.A. (1996) Prevalence of Chlamydia trachomatis, Chlamydia psittaci and Chlamydia pneumoniae antibodies in blood donors and attendees of STD clinics. Clin. Microbiol. Infect., 1, 253–260.[Medline]

Bjercke, S. and Purvis, K. (1993) Characteristics of women under fertility investigation with IgA/IgG seropositivity for Chlamydia trachomatis. Eur. J. Obstet. Gynecol. Reprod. Biol., 51, 157–161.[ISI][Medline]

Black, C.M. (1997) Current methods of laboratory diagnosis of Chlamydia trachomatis infections. Clin. Microbiol. Rev., 10, 160–184.[Abstract]

Bouyer, J., Coste, J., Shojaei, T., Pouly, J.L., Fernandez, H., Gerbaud, L. and Job-Spira, N. (2003) Risk factors for ectopic pregnancy: a comprehensive analysis based on a large case–control, population-based study in France. Am. J. Epidemiol., 157, 185–194.[Abstract/Free Full Text]

Brunham, R.C., Binns, B., McDowell, J. and Paraskevas, M. (1986) Chlamydia trachomatis infection in women with ectopic pregnancy. Obstet. Gynecol., 67, 722–726.[Abstract]

Campbell, M.J. and Machin, D. (1993) Medical Statistics: A Commonsense Approach. John Wiley and Sons, Chichester, UK.

Cates, W., Jr and Wasserheit, J.N. (1991) Genital chlamydial infections: epidemiology and reproductive sequelae. Am. J. Obstet. Gynecol., 164, 1771–1781.[ISI][Medline]

Chernesky, M., Luinstra, K., Sellors, J., Schachter, J., Moncada, J., Caul, O., Paul, I., Mikaelian, L., Toye, B., Paavonen, J. et al. (1998) Can serology diagnose upper genital tract Chlamydia trachomatis infections? Studies on women with pelvic pain, with or without chlamydial plasmid DNA in endometrial biopsy tissue. Sex. Transm. Dis., 25, 14–19.[ISI][Medline]

Clausen, H.F., Fedder, J., Drasbek, M., Nielsen, P.K., Toft, B., Ingerslev, H.J., Birkelund, S. and Christiansen, G. (2001) Serological investigation of Mycoplasma genitalium in infertile women. Hum. Reprod., 16, 1866–1874.[Abstract/Free Full Text]

Coggon, D., Rose, G. and Barker, D. (1993) Epidemiology for the Uninitiated. BMJ Books, London.

Conway, D., Glazener, C.M., Caul, E.O., Hodgson, J., Hull, M.G., Clarke, S.K. and Stirrat, G.M. (1984) Chlamydial serology in fertile and infertile women. Lancet, 1, 191–193.[CrossRef][Medline]

Dabekausen, Y.A., Evers, J.L., Land, J.A. and Stals, F.S. (1994) Chlamydia trachomatis antibody testing is more accurate than hysterosalpingography in predicting tubal factor infertility. Fertil. Steril., 61, 833–837.[ISI][Medline]

ESHRE Capri Workshop (1996) Infertility revisited: the state of ART today and tomorrow. Hum. Reprod., 11, 1779–1807.[ISI][Medline]

Forsey, J.P., Caul, E.O., Paul, I.D. and Hull, M.G. (1990) Chlamydia trachomatis, tubal disease and the incidence of symptomatic and asymptomatic infection following hysterosalpingography. Hum. Reprod., 5, 444–447.[Abstract]

Gijsen, A.P., Land, J.A., Goossens, V.J., Leffers, P., Bruggeman, C.A. and Evers, J.L. (2001) Chlamydia pneumoniae and screening for tubal factor subfertility. Hum. Reprod., 16, 487–491.[Abstract/Free Full Text]

Gijsen, A.P., Land, J.A., Goossens, V.J., Slobbe, M.E. and Bruggeman, C.A. (2002) Chlamydia antibody testing in screening for tubal factor subfertility: the significance of IgG antibody decline over time. Hum. Reprod., 17, 699–703.[Abstract/Free Full Text]

Henry-Suchet, J., Askienazy-Elbhar, M., Thibon, M., Revol, C. and Akue, B.A. (1994) Post-therapeutic evolution of serum chlamydial antibody titers in women with acute salpingitis and tubal infertility. Fertil. Steril., 62, 296–304.[ISI][Medline]

Hull, M.G.R. (1992) Infertility treatment: relative effectiveness of conventional and assisted conception methods. Hum. Reprod., 7, 785–796.[Abstract]

Jones, C.S., Maple, P.A.C., Andrews, N.J., Paul, I.D. and Caul, E.O. (2003) Measurement of IgG antibodies to Chlamydia trachomatis by commercial enzyme immunoassays and immunofluorescence in sera from pregnant women and patients with infertility, pelvic inflammatory disease, ectopic pregnancy, and laboratory diagnosed Chlamydia psittaci infection. J. Clin. Pathol., 56, 225–230.[Abstract/Free Full Text]

Land, J.A., Evers, J.L. and Goossens, V.J. (1998) How to use Chlamydia antibody testing in subfertility patients. Hum. Reprod., 13, 1094–1098.[Abstract]

Marana, R., Lucisano, A., Leone, F., Sanna, A., Del’Acqua, S. and Mancuso, S. (1990) High prevalence of silent Chlamydia colonization of the tubal mucosa in infertile women. Fertil. Steril., 53, 354–356.[ISI][Medline]

Meikle, S.F., Zhang, X., Marine, W.M., Calonge, B.N., Hamman, R.F. and Betz, G. (1994) Chlamydia trachomatis antibody titers and hysterosalpingography in predicting tubal disease in infertility patients. Fertil. Steril., 62, 305–12.[ISI][Medline]

Menchaca, A., Suarez, M., Friberg, J. and Gleicher, N. (1988) Incidence of Chlamydia trachomatis in peripartum fallopian tubes. J. Reprod. Med., 33, 199–201.[ISI][Medline]

Mol, B.W., Dijkman, B., Wertheim, P., Lijmer, J., van der Veen, F. and Bossuyt, P.M. (1997) The accuracy of serum chlamydial antibodies in the diagnosis of tubal pathology: a meta-analysis. Fertil. Steril., 67, 1031–1037.[CrossRef][ISI][Medline]

Peters, T.J., Wildschut, H.I.J. and Weiner, C.P. (1996) Epidemiologic considerations in screening. In Wildschut,H.I.J., Weiner, C.P. and Peters, T.J. (eds), When to Screen in Obstetrics and Gynaecology. Saunders, London, pp. 1–12.

Puolakkainen, M., Vesterinen, E., Purola, E., Saikku, P. and Paavonen, J. (1986) Persistence of chlamydial antibodies after pelvic inflammatory disease. J. Clin. Microbiol., 23, 924–928.[ISI][Medline]

RCOG (1998) The Management of Infertility in Tertiary Care. Royal College of Obstetricians and Gynaecologists, London.

Rice, P.A. and Schachter, J. (1991) Pathogenesis of pelvic inflammatory disease. What are the questions? J. Am. Med. Assoc., 266, 2587–2593.[Abstract]

Richmond, S.J. and Caul, E.O. (1975) Fluorescent antibody studies in chlamydial infections. J. Clin. Microbiol., 1, 345–52.[ISI][Medline]

Rutherford, A.J. and Jenkins, J.M. (2002) Hull and Rutherford classification of infertility. Hum. Fertil., 5, S41–S45.

Saikku, P. and Paavonen, J. (1978) Single-antigen immunofluorescence test for chlamydial antibodies. J. Clin. Microbiol., 8, 119–22.[ISI][Medline]

Saraiya, M., Berg, C.J., Kendrick, J.S., Strauss, L.T., Atrash, H.K. and Ahn, Y.W. (1998) Cigarette smoking as a risk factor for ectopic pregnancy. Am. J. Obstet. Gynecol., 178, 493–498.[ISI][Medline]

Shafer, M.A., Pessione, F., Scieux, C., Bianchi, A., Casin, M., Athea, N., Vexiau, D., Kermanach, M., Spira, A., Perol, Y. et al. (1993) Chlamydia trachomatis: risk factors in women in the Parisian region. Importance of smoking and cervical ectropion. J. Gynecol. Obstet. Biol. Reprod., 22, 163–168.[Medline]

Sheffield, P.A., Moore, D.E., Voigt, L.F., Scholes, D., Wang, S.P., Grayston, J.T. and Daling, J.R. (1993) The association between Chlamydia trachomatis serology and pelvic damage in women with tubal ectopic gestations. Fertil. Steril., 60, 970–975[ISI][Medline]

Shepard, M.K. and Jones, R.B. (1989) Recovery of Chlamydia trachomatis from endometrial and fallopian tube biopsies in women with infertility of tubal origin. Fertil. Steril., 52, 232–238.[ISI][Medline]

Stacey, C., Munday, P., Thomas, B., Gilchrist, C., Taylor-Robinson, D. and Beard, R. (1990) Chlamydia trachomatis in the Fallopian tubes of women without laparoscopic evidence of salpingitis. Lancet, 336, 960–963.[CrossRef][ISI][Medline]

Stamm, W.E. (1999) Chlamydia trachomatis infections of the adult. In Holmes, K.K., Sparling, P.F., Mardh, P.E., Lemon, S.M., Stamm, W.E., Piot, P. and Wasserheit, J.N. (eds), Sexually Transmitted Diseases. McGraw-Hill, New York, pp. 407–432.

Thomas, K., Coughlin, L., Mannion, P.T. and Haddad, N.G. (2000) The value of Chlamydia trachomatis antibody testing as part of routine infertility investigations. Hum. Reprod., 15, 1079–1082.[Abstract/Free Full Text]

Veenemans, L.M. and van der Linden, P.J. (2002) The value of Chlamydia trachomatis antibody testing in predicting tubal factor infertility. Hum. Reprod., 17, 695–698.[Abstract/Free Full Text]

Westrom, L. (1987) Pelvic inflammatory disease: bacteriology and sequelae. Contraception, 36, 111–128.[CrossRef][ISI][Medline]

WHO task force on the prevention and management of infertility (1995) Tubal infertility: serologic relationship to past chlamydial and gonococcal infection. Sex. Transm. Dis., 22, 71–77.[ISI][Medline]

Witkin, S.S., Linhares, I., Giraldo, P., Jeremias, J. and Ledger, W.J. (2000) Individual immunity and susceptibility to female genital tract infection. Am. J. Obstet. Gynecol., 183, 252–256.[CrossRef][ISI][Medline]

Submitted on October 17, 2002; resubmitted on March 14, 2003; accepted on May 8, 2003.