The value of Chlamydia trachomatis antibody testing in predicting tubal factor infertility

L.M.W. Veenemans1 and P.J.Q. van der Linden1,2

1 Department of Obstetrics and Gynaecology, Deventer Ziekenhuis


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The objective of the present study was to compare the likelihood of abnormal Chlamydia trachomatis antibody test results with that of abnormal hysterosalpingography (HSG) test results in patients with tubal factor infertility. METHODS: Anti-C. trachomatis immunoglobulin G antibodies were determined prospectively in 295 infertility patients by means of an indirect fluorescent antibody technique. In 48 of the 295 patients both HSG and laparoscopy with chromotubation were performed. The results of C. trachomatis antibody testing were compared with the results of HSG with respect to their predictive value of tubal factor infertility. Likelihood ratios for abnormal C. trachomatis antibody and HSG test results were determined in infertility patients, as assessed by laparoscopy. RESULTS: The positive likelihood ratio for C. trachomatis antibody testing was 1.8. This was comparable with the HSG, which had a positive likelihood ratio of 1.7. CONCLUSIONS: The predictive value of C. trachomatis antibody testing was equal to that of HSG, but ratios of 1.7 and 1.8 indicate a poor test, so both C. trachomatis antibody testing and HSG have a poor predictive value. C. trachomatis antibody testing causes minimal inconvenience to the patient, in contrast to HSG, and therefore should be maintained in infertility examinations.

Key words: Chlamydia trachomatis testing/likelihood ratios/receiver operating curve/tubal infertility


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
About 25% of couples in The Netherlands consult their physician for infertility. This study focuses on infertility due to tubal pathology. A total of 14% of female infertility is associated with tubal factor infertility. Inflammatory disease is the most important cause of tubal pathology (Dabekausen et al., 1994Go). The incidence of Chlamydia trachomatis increases worldwide. It can cause intraluminal adhesions and fibrosis, phimosis, hydrosalpinx and adhesions. Due to the serious consequences C. trachomatis infection can have on a woman's fertility, C. trachomatis antibody testing is part of the infertility work-up suggested by the Dutch Society for Obstetrics and Gynaecology. The aim of this study was to investigate the predictive value of the C. trachomatis antibody test in screening for tubal factor infertility, compared with the predictive value of hysterosalpingography, when used as a standard procedure during fertility work-up. The C. trachomatis antibody test and hysterosalpingography (HSG) were compared by calculating likelihood ratios and constructing a receiver operating characteristic (ROC) curve.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 295 female infertility patients who presented consecutively in our clinic participated in the present study. The basic data on these 295 patients (age, duration of infertility, primary or secondary infertility) are provided in Table IGo. Blood was drawn for determination of the Chlamydia antibody titre. An immunofluorescence test was used, the C. trachomatis-spot-IF-test (BioMérieux's Hertogenbosch, The Netherlands). For the C. trachomatis antibody test (CAT), a threshold point of 1:32 was used, i.e. a titre >=1:32 was considered positive. In patients with a positive C. trachomatis titre, a laparoscopy with chromotubation was performed for tubal patency testing. In patients with a negative C. trachomatis titre (<1:32), an HSG was performed. In patients with an abnormal HSG, an additional laparoscopy was performed. In patients with a normal HSG, no further steps were taken. If, after 6 months, those patients had not become pregnant, a laparoscopy with chromotubation was performed. From the 295 patients, 18 were excluded from further analysis. These were patients with tuboperitoneal abnormalities not caused by C. trachomatis or patients with abnormal HSG results caused by having only one tube. Hence, 277 patients remained available for analysis. In 48 of the 277 patients both HSG and laparoscopy with chromotubation were performed. In 97 of the 277 patients only a laparoscopy with chromotubation was performed. Therefore, the results of a CAT could be compared with laparoscopy in 145 patients, and HSG results could be compared with laparoscopy in 48 patients. HSG was performed after menstruation, during the follicular phase and before ovulation, using Lipiodol (Laboratoire Guerbet, Aulnay-Sous-Bois, France). An HSG was considered abnormal if one or both tubes did not allow passage of contrast medium. A laparoscopy with tubal patency testing was performed using Methylene Blue dye. Patients were classified as having tuboperitoneal abnormalities if evidence of adhesions, obstruction of one or both tubes or hydrosalpinx were present. A differentiation was made between an abnormal laparoscopy result with and without adhesions. The diagnostic value of the CAT was compared with the value of HSG in tubal pathology, using likelihood ratios (LRs). LRs, in contrast to positive and negative predictive values, are not affected by the prevalence of disease in the population studied, and therefore can be used to compare the outcome of the same test in different populations. In addition, comparison of different tests of the same disease entity in the same population is possible. The LR of a positive test result (LR+) indicates the likelihood of a positive test in a patient with the disease over the likelihood of a positive test in a patient without the disease. The LR– indicates the likelihood of a negative test in a patient with the disease over the likelihood of a negative test in a patient without the disease. The LR+ is calculated as [sensitivity/(1–specificity)]. The LR– is calculated as [(1–sensitivity)/specificity]. Calculation of LRs yields a score that allows categorization of test results: an LR+ of 2–5 indicates a fair clinical test, 5–10 is good, and >10 is excellent. A LR– of 0.5–0.2 indicates a fair clinical test, 0.2–0.1 is good, and <0.1 is excellent. A (ROC) curve is a graph that correlates true- and false-positive rates (sensitivity and 1–specificity respectively) for a series of threshold points for any test (Griner et al., 1981Go). The graph can be used to decide the optimum threshold point according to the purpose of the test. The LR+ and LR– were calculated for the CAT and HSG, and a ROC curve was constructed for the CAT. The LRs for the CAT were compared with the LRs for HSG, and the optimum threshold point for the CAT was decided.


View this table:
[in this window]
[in a new window]
 
Table I. Patient characteristics (n = 295)
 

    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Blood samples were drawn from 295 patients to determine the C. trachomatis antibody titre. After this, 18 patients were excluded from further study. Four of these 18 patients were seronegative and had tuboperitoneal abnormalities due to previous appendicitis, peritonitis or endometriosis. These patients had abnormalities not caused by C. trachomatis, which could therefore not be detected with the CAT. Six patients had an abnormal HSG test result caused by having only one tube. This would have been disadvantageous for the final outcome of the HSG. Eight patients were excluded due to inadequate registration of their C. trachomatis antibody titre. Eventually, 277 patients remained available for analysis. In 84 of 277 patients studied (30.3%), the CAT was positive [anti-C. trachomatis immunoglobulin (Ig)G titre >=1:32, Table IIaGo] and in 193 patients the CAT was negative (anti-C. trachomatis IgG titre <=1:16, Table IIbGo). In 145 of 277 patients a laparoscopy was performed (Table IIIaGo). Of the 145 patients, 78 were seropositive and 67 were seronegative. Twenty-eight of the 78 seropositive (35.9%) had tuboperitoneal abnormalities. In 50 of the 78 seropositive (64.1%) no tuboperitoneal abnormalities were detected with laparoscopy, and therefore an expectant management was advised. Eighteen of these 50 patients became pregnant after the laparoscopy. Of the 67 seronegative patients, seven (10.4%) had tuboperitoneal abnormalities. In six of the total 84 seropositive patients a laparoscopy was not performed, because these six patients became pregnant before the date of the laparoscopy or did not continue work-up. In 48 of 277 patients, an HSG as well as a laparoscopy was performed (Table IIIbGo). In 31 of these 48 patients the results corresponded, but in 17 patients a discrepancy between HSG and laparoscopy was noted. In 114 of 277 patients a normal HSG was noted and hence an expectant management was carried out. Fifty-four of these 114 patients became pregnant after the HSG. Of the other 60 patients, some will have become pregnant without coming to our clinic again, and were therefore lost to follow-up. The LR+ of the CAT was 1.8, indicating a patient with tubal factor infertility to be 1.8 times more likely to have a positive test result (i.e. titre >=32) than a patient without tubal factor infertility. In comparison, the LR+ for HSG in the same group of patients was 1.7 (Table IVGo). The LR– for the CAT was 0.4, indicating a patient with tubal factor infertility to be 0.4 times as likely to have a negative test (i.e. titre <=32) as a patient without the disease. The LR– for HSG was 0.7. The optimum threshold point was decided at 1:32, using the ROC curve.


View this table:
[in this window]
[in a new window]
 
Table IIa. Patients with a positive C. trachomatis antibody titre (n = 84)
 

View this table:
[in this window]
[in a new window]
 
Table IIb. Patients with a negative C. trachomatis antibody titre (n = 193)
 

View this table:
[in this window]
[in a new window]
 
Table IIIa. Chlamydia antibody titre compared with laparoscopy with chromotubation
 

View this table:
[in this window]
[in a new window]
 
Table IIIb. Hysterosalpingography compared with laparoscopy with chromotubation
 

View this table:
[in this window]
[in a new window]
 
Table IV. Comparison of the C. trachomatis antibody titre and hysterosalpingography
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Laparoscopy with tubal patency testing remains the most accurate method of diagnosing tuboperitoneal pathology at the moment. HSG is used for screening purposes. Patients tend to experience this procedure as a painful and annoying test. Furthermore, HSG has an infection-risk of 1–3%. The CAT, on the other hand, is a simple blood test and causes little inconvenience for patients. We investigated the value of the C. trachomatis test in screening for tubal factor infertility. The predictive value of the CAT was compared with the predictive value of HSG. LRs were used for this comparison, for they are prevalence-independent. The LR+ of the CAT was 1.8 and the LR– was 0.4. The sensitivity was 80% and the specificity 55%. Dabekausen and co-workers showed a better performance of the CAT (Dabekausen et al., 1994Go). A sensitivity of 74% was reported and a specificity of 92%, an LR+ of 9.1 and an LR– of 0.3 at a threshold level of <1:8 for the CAT, indicating a good clinical test. According to Mol et al. the current method of CAT is the most accurate test, but it is possible there are still many cross-reactions with other Chlamydia species and the test may not be that specific (Mol et al., 1997Go). Retrospectively, over a longer period of time, antibodies correlate well with episodes of infection. However, after infection, antibodies are not always found. Assuming that specific antibodies are only present after a real infection, a positive CAT is proof for at least one episode of infection in the past, but a negative CAT cannot exclude a C. trachomatis infection. According to Ossewaarde the border values and interpretation `Clear indication for an infection in the past' depend mostly on the type of test used, antibody, conjugativum, fluorescence lamp and on the population (Ossewaarde, 1998Go). Therefore, comparison of titres from different laboratories and different tests is not possible. Every diagnostic laboratory should take this into consideration.

In our study, the LR+ of the HSG was 1.7 and the LR– was 0.7. The sensitivity for tubal pathology was 57% and the specificity 66%. The likelihood ratios the CAT and HSG are comparable, but both show poor performance. It should be noted that in our study in only 48 of the 277 patients available for analysis, both an HSG and a laparoscopy with tubal patency testing were performed. Therefore, HSG results could be compared with laparoscopy results in only 48 patients.

According to two meta-analyses (Swart et al., 1995Go; Mol et al., 1997Go), the CAT and HSG are equally proficient in diagnosing tubal pathology, hence the CAT is not a better screening test than HSG. What clinical or practical implications does this have? In using a threshold point of 1:32, in 64% of the patients a laparoscopy has been performed unnecessary (the false-positive group). Furthermore, there is a false negative group. These patients (10%) had negative HSG results, and waited 6 months before a laparoscopy was performed. In this group, tuboperitoneal abnormalities were eventually found, but the false negative CAT did cause some delay in their infertility work-up. A disadvantage of the CAT is that tubal abnormalities not caused by C. trachomatis cannot be detected. Of the seronegative patients with abnormal laparoscopy results, four had abnormalities caused by appendicitis, peritonitis and endometriosis. During history taking, attention should be given to these specific items, because a negative CAT is of little value for such a patient. The specificity, sensitivity, LR+ and LR– have been calculated for different threshold points. When the threshold level is raised, the sensitivity decreases and the specificity increases. This means that a higher threshold point produces more false positive and less false negative results. On raising the threshold point, the LR+ also improves. But a disadvantage of raising the threshold point is that it may miss patients who would have shown tubal pathology at laparoscopy, but are not seropositive for C. trachomatis. If the threshold point is set very high, for example at 1:256, the specificity is very high, but the sensitivity very low. Using the CAT as a screening test with a threshold point of 1:256, tubal factor infertility could be missed. According to the ROC curve that was constructed, the optimum threshold point would be 1:32 or 1:64. If the threshold point 1:32 is used, the sensitivity of C. trachomatis antibody testing is 80%, the specificity 55%, LR+ 1.8 and LR– 0.4. The threshold point 1:64 produces a lower sensitivity (66%), a higher specificity (68%), and the LR+ shows some improvement (2.0). A threshold point of 1:64 could be considered in the future, in order to obtain a higher specificity without a too big decrease in sensitivity.

In conclusion, this study focused on the predictive value of serum anti-C. trachomatis IgG antibody screening in women presenting with infertility. The predictive value of the CAT was equal to the predictive value of HSG in screening tuboperitoneal pathology. The CAT causes minimal inconvenience to the patient in contrast to an HSG, and should be maintained in infertility work-up.


    Notes
 
2 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Deventer Ziekenhuis, P.O. Box 5002, 7400 GC Deventer, The Netherlands. E-mail: lindevdp{at}dz.nl Back

Submitted on 8 December 2000; resubmitted on August 13, 2001


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Dabekausen, Y.A.J.M., Evers, J.L.H., 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]

Ossewaarde, J.M. (1998) Antistoffen tegen Chlamydia trachomatis als marker voor infectie in het verleden. Nederl. Tijdschr. Med. Microbiol., 6, 3–6.

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

Swart, P., Mol, B.W.J., van der Veen, F., van Beurden, M., Redekop, W.K. and Bossuyt, P.M.M. (1995) The accuracy of hysterosalpingography in the diagnosis of tubal pathology: a meta-analysis. Fertil. Steril., 64, 486–491.[ISI][Medline]

Griner, P.F., Mayewski, R.J., Mushlin, A.I. and Greenland, P. (1981) Selection and interpretation of diagnostic tests and procedures. Ann. Internal Med., 94, 555–600.

accepted on October 29, 2001.