- Email: [email protected]
Post-therapeutic evolution of serum chlamydial antibody titers in women with acute salpingitis and tubal infertility*
r I
Vol. 62, No.2, August 1994
FERTILITY AND STERILITY
Printed on acid-free paper in U. s. A.
Copyright" 1994 The American Fertility Society
.
Post-therapeutic evolution of serum chlamydial antibody titers In women with acute salpingitis and tubal infertility*
Jeanine Henry-Suchet, M.D.t Myriam Askienazy-Elbhar, B.D.:j: Monique Thibon, M.D.§
Claire Revol, B.D. II Bernard A. Akue, M.D.1f
Hc5pital Jean Rostand, Sevres, and Laboratoire des Chlamydioses, Institut Pasteur, Paris, France
Objective: To determine if serologic data and, more particularly, antichlamydial immunoglobulin (Ig) M can be used for diagnosis of current chlamydial intrapelvic gynecologic infection. Design: Forty-two women with acute salpingitis (group A), 131 women with tubal factor infertility (group B), and 98 pregnant women (control group C) were studied. Setting: Hopital Jean Rostand, Sevres (patients), Laboratories Magenta and Eylau, Paris (serology), Institut Pasteur, Paris (cultures). Interventions: Study groups: endocervical/urethral swabs, pelvic samples; serologic study before and after treatment. Control group: Serologic study. Main Outcome Measures: Serum samples were collected from each patient initially and 6 to 9 weeks later; additionally, two to five sequential sera were obtained from 22 (group A) and 25 (group B) patients with positive cultures, evolutive serology, or positive antichlamydial IgM. Sera were tested for antichlamydial IgG by a microimmunofluorescence assay using Chlamydia trachomatis elementary bodies and for IgA and IgM by whole inclusion-fluorescent assay. Results: Before treatment, there was a correlation between the presence of antichlamydial IgM and positive cervical and/or intrapelvic chlamydia cultures. After treatment, antichlamydial IgM, when initially positive, rapidly disappeared in most subjects; its persistence after 4 months was significantly associated with tubal sequelae in group A patients and persistence of positive intrapeivic chlamydial cultures in group B women. Conclusion: Serologic analysis of women with acute salpingitis or tubal infertility, including antichlamydial IgM, may aid both in the before treatment diagnosis of chlamydial infection and in the follow-up evaluation. Fertil Steril 1994;62:296-304 Key Words: Chlamydia trachomatis, acute salpingitis, tubal factor infertility, intrapelvic cultures, antichlamydial antibodies
Chlamydia trachomatis gynecologic infections can be acute or asymptomatic (1, 2). Detection of
Received June 10, 1993; revised and accepted March 16, 1994.
* Presented at the VIIth International Symposium on Human Chlamydial Infections, Harrison Hot Springs, British Columbia, Canada, June 24 to 29, 1990. t Reprint requests: Jeanine Henry-Suchet, M.D., C.H.R. Jean Rostand, 141 Grand rue, 92311, Sevres Cedex, France (FAX: 42-56-06-50). Laboratoire Magenta. § Institut Pasteur, Paris. II Laboratoire d'Eylau, Paris. 1f Hopital Jean Rostand.
*
296
Henry-Suchet et aI.
Chlamydia serology in gynecology
intrapelvic C. trachomatis infection often requires either laparoscopy or laparotomy; cervical and urethral C. trachomatis cultures have poor predictive value (2). The value of serology in detecting upper genital tract infections is still debatable. A high antichlamydial immunoglobulin (Ig) G titer, although often detected in upper genital tract infections (13), cannot be considered as proof of current infection. Antichlamydial IgA has been shown to be significantly associated with acute salpingitis, ectopic pregnancy, and tubal factor infertility (4-6); antichlamydial IgM is considered as a marker of recent or current infection (7-9); however, only a few Fertility and Sterility
authors have detected positive antichlamydial IgM in chlamydial intrapelvic gynecologic infections. To further assess the predictive value of serum antichlamydial antibody titers for the diagnosis and therapeutic response of women with C. trachomatis infections, we evaluated these antibodies in sequential specimens before treatment and during the follow-up of women treated for acute chlamydial salpingitis and for tubal factor infertility of chlamydial origin. Antibody titers were compared with results of C. trachomatis cultures.
MATERIALS AND METHODS Study Population
Three groups of women admitted to the gynecologic unit of Hopital Jean Rostand, Sevres between January 1, 1987 and January 1, 1989 were included in this study. Group A comprised 42 patients with acute salpingitis, selected by laparoscopy (35 cases) or three of the following clinical and/or biologic signs (7 cases): pelvic pain, fever, vaginal discharge, abnormal leukocyte count, and/or increased erythrocyte sedimentation rate. Group B comprised 131 healthy patients, requiring treatment for infertility, without any ofthe precited signs, selected as having evidence of tubal lesions at hysterosalpingography and/or laparoscopy (hydrosalpinx, tubal stenosis, and/or peritubal adhesions); these two patient groups were included prospectively. Group C comprised 98 healthy pregnant women, without past history of acute salpingitis or infertility treatment, retrospectively taken as a control group for serology. Diagnosis of Current Chlamydial Infection Serum Specimens
Patients had one serum specimen taken at the identification of clinical disease (group A), at their first consultation for tubal factor infertility (group B), and all patients but one in group A had a second serum collected after 6 to 9 weeks. For control group C, two frozen sera obtained during the first trimester of pregnancy and 2 months later were used. Sera were tested for antichlamydial IgG by microimmunofluorescence using serovar L2 elementary bodies (10) and revealed by fluorescein-conjugated goat anti-human-IgG (Sanofi Pasteur, Marnes la Coquette, France). Antichlamydial IgM Vol. 62, No.2, August 1994
and IgA were detected by inclusion fluorescent antibody, using McCoy cells infected by C. trachomatis, serovar L2, as antigen (Electronucleonics; distributed by Eurobio, Les VIis, France) and revealed by fluorescein-conjugated goat anti-human IgM and IgA (Sanofi Pasteur) (7,8). Before testing for antichlamydial IgM and IgA, IgG and rheumatoid factor were removed from sera by rheumatoid factor absorbent (Behring, Rueil-Malmaison, France). Serial dilutions of sera were tested from 1:20 in the same assay for each patient; a titer of 1:32 was considered positive for antichlamydial IgG, a titer of 1:20 was considered positive for antichlamydial IgM and IgA; a fourfold change in antibody level was considered as evolutive serology. Twenty-eight patients with positive antichlamydial IgM were tested for anti C. pneumoniae IgG and IgM with species-specific elementary bodies and revealed by the same anti IgG and IgM conjugates. Genital Specimens
These were obtained only from subjects of groups A and B; 34 patients in group A and 112 in group B had endocervical and urethral swabs taken before treatment; the others, referred by another hospital, started antibiotic treatment before their first visit to our unit, which excluded endocervical and urethral sampling. Fourteen patients in group A and 89 in group B had intrapelvic cultures taken at laparoscopy before treatment; for each patient,. one to three of the following sites were sampled: cul-de-sac fluid, adhesions, and/or tubal biopsies. Twelve sexual partners of group A and 110 of group B patients had urethral cultures. Specimens were collected in Eagle Minimum Essential Medium supplemented with vancomycin, gentamicin, amphotericin B, and 20% fetal calf serum, using one container for both cervix and urethra and one for each laparoscopic specimen. These were stored at 4°C for <24 hours and inoculated in cycloheximide-treated L cells. Inclusions were identified with fluorescein-labeled monoclonal antibody (Syva Bio-Merieux, Marcy l'Etoile, France). No blind passages were done in this series. Diagnosis and Treatment of Current Infection
Patients with positive antichlamydial IgM and/ or evolutive IgA or IgG titers on two sera and/or positive endocervical or urethral or intrapelvic C. trachomatis cultures were considered as having a current chlamydial infection. All patients in groups A and B with positive antichlamydial IgG were conHenry-Suchet et aI.
Chlamydia serology in gynecology
297
sidered to have a possible current C. trachomatis infection and were treated with a synthetic tetracycline (200 mgjd doxycycline) for 3 to 6 weeks. The usual partners of acute salpingitis patients received the same treatment for 3 weeks. Patients of this group without a single partner had full instructions to avoid new sexual contacts or to use condoms during treatment and the 3 following months; they complied fully, as assessed by repeated inquiries. Partners of tubal infertility patients with positive cultures were treated for 3 weeks. Patients with persistent positive antichlamydial IgM received repeated treatments (doxycycline + erythromycin or ofloxacine) for 6 weeks. Treatment Follow-up Test of Cure Cultures
All patients and partners with initial positive endocervical or urethral cultures had test of cure cultures of the same sites 1 month after treatment. Twenty-five patients in group A and 59 in group B had intrapelvic cultures taken at follow-up laparoscopy, 1 to 3 months after the completion of treatment. Most patients of group B had only one laparoscopy: 89 before treatment and 59 after treatment. Fifteen women with positive serology to C. trachomatis in this group had two laparoscopies, one before and one after treatment. Serologic Follow-up
A serological follow-up was proposed to 31 patients of group A and 30 patients of group B who were considered to have current C. trachomatis infection. Twenty-two patients of group A and 25 patients of group B accepted this protocol: they had two to five additional sequential sera collected at 2 to 6 month intervals for a 1 year follow-up period. A statistical analysis was made using the x2 test with Yates correction for continuity.
RESULTS Detection of C. trachomatis by Serology Before Treatment
tility (75.6%), and 31 of 98 healthy pregnant controls (31.6%). The geometric mean titers were 1:450, 1:190, and 1:94, respectively. Both patient groups had a similar prevalence of antichlamydial IgG, which was significantly different from the control group (P < 0.001). IgA Titers
Antichlamydial IgA antibodies were detected in 28 of 42 women with acute salpingitis (66.7%), in 76 of131 women with tubal infertility (58%), and in 16 of the 98 controls (16.3%). The geometric mean titers were similar in the three groups, 1:49, 1:46, and 1:36, respectively. Both patient groups had a similar prevalence of antichlamydial IgA, which was significantly different from the control group (P < 0.0001). IgM Titers
Antichlamydial IgM antibodies were detected in 26 of 42 women with acute salpingitis (61.9%) and 28 of 131 women with tubal infertility (21.4%). There was no detectable antichlamydial IgM in the control group; the geometric mean titers were 1:40 and 1:22 in the patient groups. A significant difference in the occurrence of IgM antibody to C. trachomatis was noted between the two patient groups (P < 0.001), as well as between A and B patients versus the control group (P < 0.0001) The specificity of the IgM antibodies for C. trachomatis was evaluated by testing 28 of the sera for IgM antibody to C. pneumoniae, 11 sera from group A and 17 from group B. Of 11 patients with acute salpingitis who were antichlamydial IgM positive, one was also IgM positive for C. pneumoniae. Similarly, only 3 of 17 women with tubal infertility were IgM positive for both C. trachomatis and C. pneumoniae. Antibodies to C. trachomatis in Two Sequential Sera
In group A, one woman who was positive for antichlamydial IgG, IgA, and IgM did not have a second serum taken. Forty-one cases were studied in sequential sera (Table 1). IgG Titers
IgG Titers
Before treatment, IgG antibodies to C. trachomatis were detected in 32 of 42 women with acute salpingitis (76.2%), 99 of 131 women with tubal infer298
Henry-Suchet et al.
Chlamydia serology in gynecology
Of the 31 women initially positive for antichlamydial IgG and having sequential sera taken, all remained IgG positive, 28 of them (90.3%) with an equivalent titer and a significant evolution was Fertility and Sterility
Table 1
Evolution of Chlamydial Antibody Titers on Two Sequential Sera Taken at 0 and 3 Months First serum
Patient groups
No. of patients
No. of positive cases
Evolution on second serum No. of seroconversions
No. with significant increase
No. with significant decrease
No. without change
1 2 0
28 97 31
1 1 0
25 75 16
7 6 0
18 22 0
IgG A* B C
41 131 98
31 99 31
2 0 0
2 0 0 IgA
A* B
C
41 131 98
27 76 16
1 1 0
1 0 0 IgM
A* B C
41 131 98
25 28 0
1 0 0
0 0 0
* In group A, one patient with positive IgG, IgA, and IgM did not have a second serum taken.
observed for antichlamydial IgG in five cases: two seroconversions, two increases, and one decrease. In group B, 97 of99 (98%) women with tubal factor infertility who were antichlamydial IgG positive remained positive with an equivalent titer on second serum and two had a significant decrease in IgG level. There were no changes in the incidence or titer of antichlamydial IgG in the control group. IgA Titers
In group A, 25 of 27 (92.6%) patients who were IgA positive also remained positive with an equivalent titer on second serum. A significant evolution was observed for IgA in three cases: one seroconversion, one increase, and one decrease. In group B, 75 of 76 patients (98.7%) with antichlamydial IgA antibody remained positive with an equivalent titer in the second serum; there was one seroconversion and one significant decrease. There were no changes in the incidence or titer of antichlamydial IgA in the control group. IgM Titers
In contrast to IgG and IgA, of the 25 patients with acute salpingitis who initially were antichlamydial IgM positive, only 18 of 25 (72%) remained positive with an equivalent titer on second serum. The titers decreased in seven cases (28%) and became seronegative in five cases (20%) at 1 to 3 months; there was also one seroconversion but no Vol. 62, No.2, August 1994
increase in titer. Six of 28 women (21.4%) with tubal infertility who were initially antichlamydial IgM positive exhibited decreased titers and 5 (17.9%) became seronegative. During this time interval, there were no seroconversions and no increases in titer. In the control group, no antichlamydial IgM was detected in the second serum. Interpretation of Antibody Change in Sequ~ntial Sera
In group A, of 42 women with acute salpingitis, 32 were considered to be of C. trachomatis origin: 27 having positive antichlamydial IgM of which one was on the second serum only and 5 having significant evolution of antichlamydial IgG (2 cases) or both IgG and IgA (3 cases) in the absence ofIgM. Of the 10 other cases, 8 remained negative for chlamydial antibodies on two sera and had negative endocervical and urethral and intrapelvic cultures, two were positive for antichlamydial IgG at equivalent titers on both sera, had negative cultures, and were negative for antichlamydial IgA and IgM. These 10 patients were considered as having no current chlamydial infection. In this group, the sensitivity of antichlamydial IgM in the detection of current chlamydial acute salpingitis was 27 of 32 (84.4%), the sensitivity of antichlamydial IgA was 29 of 32 (90.6%), the sensitivity of antichlamydial IgG was 100% with 2 of 34 (5.9%) false positives. In group B, 30 women were considered as having a current C. trachomatis infection, 28 because of positive antiHenry-Suchet et al.
Chlamydia serology in gynecology
299
Table 2
Evolution of Chlamydia Antibody Titers on 12 months Follow-up of Groups A and B (n
o to 8 mo
o to 4 mo Antibody toCT
Total number
No. with significant decrease
IgG IgA IgM
47 44 42
1 (2)* 1 (2) 13 (31)t
No. with decrease to zero 0(0) 1 (2) 11 (26)t
* Values in parentheses are percentages.
t IgM
significantly different from IgG and IgA decrease, P < 0.001.
=
47)
o to 12 mo
No. with significant decrease
No. with decrease to zero
No. with significant decrease
No. with decrease to zero
9 (19) 4 (9) 26 (62)+
1 (2) 4 (9) 26 (62)+
15 (22) 13 (29) 32 (76)+
2 (4) 7 (16) 32 (76)+
+
IgM significantly different from IgG and IgA decrease to zero, P < 0.0001.
chlamydial IgM and two because of positive intrapelvic cultures and evolutive antichlamydial IgA and IgG or evolutive IgG only.
seronegative at 8 and 12 months, respectively. The difference is significant versus antichlamydial IgG and IgA seronegativation (P < 0.0001).
Antibodies to C. trachomatis in Long-term Follow-up Sera
Detection of C. trachomatis by Culture
Changes in antibody titers at 4, 8, and 12 months were not significantly different between the 22 women with acute salpingitis and the 25 women with tubal infertility. Therefore, the data for the two groups (47 patients in total) were combined for analysis (Table 2). IgG Decrease
The antichlamydial IgG antibody titer decreased in 1 of 47 (2%) patients at 4 months, 9 of 47 (19%) at 8 months, and 15 of 47 (31.9%) by 12 months. However, only 2 of 47 women (4%) reverted from IgG seropositive to seronegative over this 1-year interval. IgA Decrease
Similarly, antichlamydial IgA antibody titers decreased in 1 of 44 (2%) women at 4 months, 4 of 44 (9%) at 8 months, and 13 of 44 (29%) at 12 months. Only 7 of 44 women (16%) became antichlamydial IgA negative after 12 months. IgM Decrease
The titer of IgM antibodies to C. trachomatis, in contrast to the other antibody isotypes, was significantly decreased in 13 of 42 women (30.9%) at 4 months (P < 0.001), 26 of 42 women (61.9%) at 8 months, and 32 of 42 women (76%) at 12 months (P < 0.001). Eleven of 42 women (26%) converted to antichlamydial IgM seronegative at 4 months, 26 of 42 (61.9%) and 32 of 42 women (76%) became 300
Henry-Sucbet et al.
Chlamydia serology in gynecology
Before treatment, C. trachomatis was detected in 16 of 34 endocervical or urethral cultures (47%) and 4 of 14 pelvic cultures (28.6%) in women with acute salpingitis (group A). Of the 16 patients with positive endocervical or urethral cultures, only 2 had pelvic cultures before treatment and they were positive. In total, 18 acute salpingitis patients had either endocervical, urethral, or pelvic positive cultures. Among the women with tubal infertility (group B), 4 of 112 endocervical and urethral (3.6%) and 9 of 89 pelvic (10.1 %) cultures were positive; there was a significant difference between the two groups in endocervical and urethral cultures (P < 0.01). In group A, of 12 usual partners who were sampled, 6 had positive cultures for C. trachomatis, 4 of them having no clinical signs. Of 110 usual partners in group B who were sampled, 4 had positive cultures, all of them being asymptomatic. One month after treatment, all endocervical and urethral cultures were negative with the exception of one acute salpingitis case whose cervix remained positive after a first course of treatment and became neg~tive after a second course. Her husband had positive pretreatment cultures, which remained positive despite three different courses of synthetic tetracycline treatment for 3 weeks. The isolate was kept and its sensitivity to antibiotics was tested in vitro, without showing any resistance to usual minimal inhibitory concentrations. Intrapelvic cultures performed during follow-up laparoscopy were still positive for C. trachoma tis in 6 of 25 women after acute salpingitis (24%) and 10 of 59 women with tubal infertility (16.9%), without significant difference between the two groups. Fertility and Sterility
Table 3
Chlamydia Cervical-Urethral and Intrapelvic Cultures Before Treatment Correlated to Serology Group A
Group B
Serology
Cervix -urethra
Intrapelvic
Cervix -urethra
IgG IgA IgM
N+/N*
N+/N* Cul-de-sac Adhesion Tube
N+/N*
n+/N*
o ofS o of3
10fS o of2 10fl
o of 20 o of2S o of47
o of 30 o of 19
+ + + +
+ + Total
+ +
1of3 o of 1t 15 of 19t 16 of 34
2of3 4 of 14
o of5 o of2
o of 1
10fl 1 of 1 2of9
o of1
* N + /N, number of cases with positive cultures/total number of cases sampled.
Relation Between Before Treatment Cultures and Serology Cervical- Urethral Cultures
In group A, of six women with acute salpingitis who were positive for antichlamydial IgG and/or IgA antibodies and negative for antichlamydial IgM, only one (16.7%) had a positive endocervical or urethral culture. In significant contrast, 15 of 20 women (75%) with antichlamydial IgG, IgA, and IgM (19 cases) or IgG and IgM (1 case) antibodies were endocervical or urethral culture positive (P < 0.001). In group B, among the women with tubal infertility, the small number of positive cervical cultures, 4 of 112, precluded statistical analysis. However, each of the culture-positive women had antichlamydial IgG, IgA, and IgM antibodies (Table 3). Intrapelvic Cultures
In group A, among the four women with acute salpingitis whose intrapelvic cultures were positive for C. trachomatis, one was negative on first serum and had a seroconversion to antichlamydial IgG antibodies in her second serum, one had antichlamydial IgG plus IgA with significant evolution on second serum and two were positive for anti C. trachomatis IgG, IgA, and IgM (Table 3). In Group B, positive intrapelvic cultures were obtained in nine women with tubal infertility, of which two had antichlamydial IgG and IgA without IgM; one had IgG and IgM without IgA (the IgA became positive on second serum); and six had IgG, IgA, and IgM antibodies. The incidence of positive intrapelvic cultures in this group was significantly higher in Vol. 62, No.2, August 1994
1of2
o of2
2 of 3 30f7
4of27 4 of 112
Intrapelvic
2of26 1 of It 6 of 13t 90fS9
Cul-de-sac Adhesion
o of 25 o of 14 o of 21 lofl 5 of 12 6of73
Tube
o of5 o of2
o of5 o ofS
2 of 15
1 of 12
2of5 4of27
lof2 2of27
t Cases with positive IgM versus cases without IgM, P < 0.001.
women who were antichlamydial IgM seropositive (7/14, 50%) versus women who were IgG or IgA seropositive and IgM seronegative (2/45, 4.4%; P < 0.001). Relation Between After Treatment Intrapelvic Cultures and Serology
In group A, among the six women who had acute salpingitis and still were chlamydia culture positive 3 months after treatment, five remained seropositive for IgG, IgA, and IgM antibodies to C. trachomatis; the sixth woman was only IgG positive. In group B, similarly, ofthe 10 women with tubal infertility who remained culture positive, 8 had IgG, IgA, and IgM antichlamydial antibodies and two had IgG and IgA without IgM antibody (Table 4). Women in group B who were positive for IgM antibody to C. trachomatis 3 months after treatment had a significantly higher incidence of positive intrapelvic cultures for this organism after treatment (8/15, 53.3%) than women who were chlamydia IgM negative (2/44, 4.5%; P < 0.01). In group B, there was no significant difference in the rate of positive cultures before and after treatment among women with positive antichlamydial IgM. Of the 15 tubal factor infertility cases who had intrapelvic cultures taken before and after treatment, 5 of 8 who initially vyere antichlamydial IgG, IgA, and IgM positive had positive cultures before treatment. In contrast, the other seven patients who were antichlamydial IgM negative and IgG or IgG and IgA positive had negative initial cultures. After treatment, antichlamydial IgM persisted in four cases; this was associated with positive cultures after treatment in all cases; of the 11 IgM-negative women, only 2 (18.2%) had positive cultures after treatment. Henry-Suchet et al. Chlamydia serology in gynecology
301
r Table 4
Chlamydia Intrapelvic Cultures After Treatment at Follow-up Laparoscopy Correlated to Serology Group A
Serology IgG
IgA
IgM
N+/N*
Cul-de-sac
Adhesion
Tube
N+/N*
Cul-de-sac
1 of 2 o of 7 3 of 14 4 of 23
o of2 o of 5
o of2
o of 2 o of 7
o of2 o of4
2of9 2 of 16
2 of 5 2 of 7
2 of 35 8 of 15t 10 of 59
1 of 30 5 of 12 6 of 48
o of 0 + + +
+ + Total
+
Group B
1 of 3 o of 7 5 of 15 6 of 25
Adhesion
Tube
o of 3 o of 12
o of 2
2of5 20f20
2 of 11 2 of 5 4 of 18
* N+/N, number of cases with positive cultures/total number of cases sampled.
t Cases with positive IgM versus cases without IgM, P < 0.001.
Sequelae in Group A at Follow-up Laparoscopy
peated questions, it was presumed that cases with positive cultures after treatment were not due to recontamination. A serotyping of the strains should have given better proof of the absence of recontamination but was not done in this series. Before treatment, the detection of IgM antibodies to C. trachomatis, in contrast to IgG and IgA antibodies, was correlated with culture-positive chlamydial infections in endocervical and urethral samples of women with acute salpingitis and pelvic samples of women with tubal infertility. We may regret not having collected endocervical and urethral samples in the control series of pregnant women but, according to another series done in our unit during the same period (15), endocervical and urethral prevalence of C. trachoma tis in pregnant women was 2%. The absence of antichlamydial IgM in this control series is in total contrast to the two other groups; the 16.3% rate of antichlamydial IgA in pregnant women possibly is due to a previous silent infection, without tubal sequelae. The persistence of IgM antibody to C. trachomatis 3 months after the completion of treatment was correlated with continued positive intrapelvic cultures in tubal factor infertility cases and to tubal obstruction or adhesions in acute salpingitis cases. However, as only 3 of 11 women with sequelae had pretreatment laparoscopy, we cannot exclude a preexisting silent disease in the 8 other cases. Immunoglobulin G and IgA antibodies to C. trachomatis persisted in most seropositive women and were unrelated to antibiotic failure. Thus, in women with upper genital tract infections, IgM antibodies to C. trachomatis appear to indicate the current presence of this organism. However, the presence of IgA and/or IgG antibodies to C. trachomatis and the absence of IgM antibodies was observed in a minority of our culture-positive subjects and five acute salpingitis cases were considered of chlamydial origin for having significantly evolutive IgG and/or IgA
Among the 25 women of the acute salpingitis group who had follow-up laparoscopies, 3 had previous lesions that precluded the evaluation of sequelae. In the 22 other cases, the presence of tubal lesions (tubal obturation and/or peritubal adhesions) was correlated with the persistence of IgM antibody to C. trachomatis: 11 of 15 with persistent antichlamydial IgM had sequelae compared with 1 of 10 without IgM (P < 0.01). In contrast, no similar relationship was observed between tubal lesions and antichlamydial IgA antibodies in this series.
DISCUSSION
The present study documents the persistence of C. trachoma tis in the upper genital tract, but not in the cervix, several months after antibiotic therapy for chlamydia-associated acute salpingitis or tubal infertility. The pretreatment discrepancy between the rate of positive endocervical and urethral cultures in acute salpingitis versus tubal infertility cases is a possible consequence of a long-term contamination in the latter group. Other investigators also have reported evidence of intrapelvic infection in chronic cases (11) and after antibiotic treatment (12). We previously have observed the persistence of C. trachomatis-positive cultures in pelvic samples after treatment in 12% of women with acute salpingitis and 45% of women with tubal infertility who had, before treatment, positive intrapelvic cultures (13), and suggested that this was a cause for tuboplasty failure (14). Together, these reports strongly indicate that, in the absence of reinfection due to new sexual contact, C. trachomatis can persist as a chronic infection in the upper genital tract of some women despite standard treatment. On the basis of instructions given to patients, the fact that the partners were treated, and the answers to re302
Henry-Suchet et a1.
Chlamydia serology in gynecology
Fertility and Sterility
antibody titers in the absence of antichlamydial IgM. This lack of antichlamydial IgM detection in five cases possibly is due to a low response in these women. Previous efforts to detect IgM antibodies to C. trachomatis have met with variable success. Anestad et al. (9), using an assay similar to ours, reported a comparable incidence of antichlamydial IgM antibodies in their study population. Miettinen et al. (16), using enzyme immunoassay (EIA) and immunoblot analysis, observed antibodies to the major outer membrane protein of C. trachomatis elementary bodies in all patients who had C. trachomatis isolated. They concluded that acute salpingitis patients with and without upper genital tract C. trachomatis infection cannot be differentiated by reactivity of sera to specific chlamydial polypeptide antigens. The effective single test in their series was the determination of specific serum IgA antibody response by EIA. In a I-year followup of 69 antigen-positive patients, Clad et al. (17) found IgG and IgA antichlamydial antibodies that were correlated with positive cervical cultures and persisted in most cases despite treatment, suggesting the persistence of inflammation. Piura et al. (18) have emphasized the persistence of antichlamydial IgG and IgA antibodies without significant change after treatment in 63.6% of acute salpingitis cases and regretted not having correlated these findings to intrapelvic cultures. In our experience, antichlamydial IgM antibody determination by immunofluorescence using C. trachomatis-infected cells is easier to read and yields a higher incidence of positives than does immunofluorescence on purified elementary bodies. The use of infected cells allows detection of antibodies to a larger repertoire of chlamydial antigens than does analysis using elementary bodies, more particularly lipopolysaccharide and heat-shock proteins, which are genus antigens, related to deep-seated infection and delayed hypersensitivity. Inclusion of an absorption step to remove IgG and rheumatoid factor is effective in reducing the rate of false negatives due to competition between IgG and IgM antibodies for binding sites and false positives due to the binding of IgM rheumatoid factor to IgG that ha~ reacted with chlamydial antigens. Additional testing using C. pneumoniae reduces the incidence of cross-antigen reaction even further; in the present study, 4 of 28 women (14.3%) positive for the C. trachomatis IgM assay also were positive for C. pneumoniae IgM. Further studies are required to Vol. 62, No.2, August 1994
clarify the value of C. trachomatis antibody determinations in different populations. Our findings also reinforce previous observations that C. trachomatis is a major pathogen in both acute salpingitis and tubal infertility (1-3) and that analysis of cervical samples for C. trachomatis in tubal infertility cases does not reflect its possible occurrence in the upper genital tract (2). Because early detection and treatment of intrapelvic chlamydial infection is crucial to the preservation of fertility, the continued analysis and modification of serologic methods to detect asymptomatic C. trachomatis infections should receive the highest priority.
Acknowledgments. We thank Yvonne Perol, M.D., Ph.D. (Laboratoire de microbiologie, Hopital Saint Louis, Paris, France; Steven S. Witkin, Ph.D., Department of Obstetrics and Gynecology, Cornell University Medical College, New York, New York; and Michael P. Diamond, M.D., Department of Obstetrics, Gynecology, and Surgery, Vanderbilt University, Nashville, Tennessee for scientific advice.
REFERENCES 1. Mlhdh PA, Ripa T, Svenson L, Westrom L. Chlamydia trachomatis infection in patients with acute salpingitis. N Engl J Med 1977;296:1377-9. 2. Henry-Suchet J, Catalan F, Loffredo V, Serfaty D, Siboulet A, Perol Y, et al. Microbiology of specimens obtained by laparoscopy from controls and from patients with pelvic inflammatory disease or infertility with tubal obstruction. Am J Obstet Gynecol 1980;138:1022-5. 3. Henry-Suchet J, Utzmann C, De Brux J, Ardoin P, Catalan F. Microbiologic study of chronic inflammation associated with tubal factor infertility: role of Chlamydia trachomatis. Fertil Steril1987;47:274-7. 4. Sarov I, Sarov B, Lunenfeld E, Zion H, Chaim W, Piura B. The significance of Chlamydia specific serum IgA antibodies in Chlamydia trachomatis infections. In: Esculapio, editor. Proceedings of the European Society for Chlamydia Research Bologna: Italy, 1988:234-8. 5. Sarov I, Lunenfeld E, Sarov B, Hanuka N, Rosenzweig R, Potashnik G, et al. Chlamydia specific IgG and IgA antibodies in women with obstructive infertility as determined by immunoblotting and immunoperoxidase assays. Eur J Epidemiol 1998;4:216-23. . 6. Sellors JW, Mahony JB, Chernesky MA, Rath DJ. Tubal factor infertility: an association with prior chlamydial infection and asymptomatic salpingitis. Fertil Steril 1988; 49:451-7. 7. Scieux C, Colimon R, Perol Y. Detection des anticorps antichlamydiens par immunofluorescence indirecte sur inclusions de Chlamydia trachomatis. Med Mal Infect 1982; 8:466-9. 8. Scieux C, Colimon R, Bianchi A, Felten A, Perol Y. Interet diagnostique de la recherche des anticorps anti-chlamydiens au cours des salpingites. Presse MEld 1987;16:715-8.
Henry-Suchet et al.
Chlamydia serology in gynecology
303
9. Anestad G, Lunde 0, Moen M, Dalaker K. Infertility and chlamydial infection. Fertil Steril1987;48:787-90. 10. Wang SP, Grayston JT, Alexander ER, Holmes KK. Simplified micro-immunofluorescence test with trachoma lymphogranuloma venerum (Chlamydia trachomatis) antigens for use as screening test for antibody. J Clin Microbiol 1975;1:250-6. 11. Keilani A, Boulieu D, Raudrant D, Carraz M, Quenin P. Role de Chlamydia trachomatis dans les pathologies tubaires (salpingites aigues et srerilite tubaire). J Gynecol Obstet BioI Reprod (Paris) 1989;18:167-72. 12. Shepard MK, Jones RB. Recovery of Chlamydia trachomatis from endometrial and fallopian tube biopsies in women with infertility of tubal origin. Fertil Steril 1989;52:232-8. 13. Henry-Suchet J, Douieb P. Treatment of culture documented intrapelvic Chlamydia trachomatis infection in women selected for tuboplasties: preliminary studies. In: Oriel D, Ridgway G, Schachter J, Taylor-Robinson D, Ward M, editors. Chlamydial infections. New York: Cambridge University Press, 1986:201-4.
304
Henry-Suchet et al.
Chlamydia serology in gynecology
14. Henry-Suchet J, Veluyre M, Pia P. Etude statistique des facteurs influen~ant Ie pronostic des plasties tubaires. J Gynecol Obstet BioI Reprod (Paris) 1989;18:571-80. 15. Shafer MA, Pessione F, Scieux C, Bianchi A, Casin M, Athea N, et a1. Chlamydia trachomatis: facteurs de risques chez les femmes de la region parisienne. J Gynecol Obstet BioI Reprod (Paris) 1993;22:163-8. 16. Miettinen A, Heinonen PK, Teisala K, Punnonen R, Paavonen J. Antigen specific serum antibody response to Chlamydia trachomatis in patients with acute pelvic inflammatory disease. J Clin Pathol 1990;43:758-61. 17. Clad A, Baumgarten B, Petersen E. Long term persistence of disappearance of Chlamydial IgA and IgG in gynecologic patients. In: Mardh P A, La Placa M, Ward M, editors. Proceedings of the European Society for Chlamydia Research. Uppsala, Sweden: Uppsala University Center for STD Research, 1992:119. 18. Piura B, Sarov B, Sarov I. Persistence of antichlamydial antibodies after treatment of acute salpingitis with doxycycline. Eur J Obstet Gynecol Reprod BioI 1993;48:117-21.
Fertility and Sterility
Vol. 62, No.2, August 1994
FERTILITY AND STERILITY
Printed on acid-free paper in U. s. A.
Copyright" 1994 The American Fertility Society
.
Post-therapeutic evolution of serum chlamydial antibody titers In women with acute salpingitis and tubal infertility*
Jeanine Henry-Suchet, M.D.t Myriam Askienazy-Elbhar, B.D.:j: Monique Thibon, M.D.§
Claire Revol, B.D. II Bernard A. Akue, M.D.1f
Hc5pital Jean Rostand, Sevres, and Laboratoire des Chlamydioses, Institut Pasteur, Paris, France
Objective: To determine if serologic data and, more particularly, antichlamydial immunoglobulin (Ig) M can be used for diagnosis of current chlamydial intrapelvic gynecologic infection. Design: Forty-two women with acute salpingitis (group A), 131 women with tubal factor infertility (group B), and 98 pregnant women (control group C) were studied. Setting: Hopital Jean Rostand, Sevres (patients), Laboratories Magenta and Eylau, Paris (serology), Institut Pasteur, Paris (cultures). Interventions: Study groups: endocervical/urethral swabs, pelvic samples; serologic study before and after treatment. Control group: Serologic study. Main Outcome Measures: Serum samples were collected from each patient initially and 6 to 9 weeks later; additionally, two to five sequential sera were obtained from 22 (group A) and 25 (group B) patients with positive cultures, evolutive serology, or positive antichlamydial IgM. Sera were tested for antichlamydial IgG by a microimmunofluorescence assay using Chlamydia trachomatis elementary bodies and for IgA and IgM by whole inclusion-fluorescent assay. Results: Before treatment, there was a correlation between the presence of antichlamydial IgM and positive cervical and/or intrapelvic chlamydia cultures. After treatment, antichlamydial IgM, when initially positive, rapidly disappeared in most subjects; its persistence after 4 months was significantly associated with tubal sequelae in group A patients and persistence of positive intrapeivic chlamydial cultures in group B women. Conclusion: Serologic analysis of women with acute salpingitis or tubal infertility, including antichlamydial IgM, may aid both in the before treatment diagnosis of chlamydial infection and in the follow-up evaluation. Fertil Steril 1994;62:296-304 Key Words: Chlamydia trachomatis, acute salpingitis, tubal factor infertility, intrapelvic cultures, antichlamydial antibodies
Chlamydia trachomatis gynecologic infections can be acute or asymptomatic (1, 2). Detection of
Received June 10, 1993; revised and accepted March 16, 1994.
* Presented at the VIIth International Symposium on Human Chlamydial Infections, Harrison Hot Springs, British Columbia, Canada, June 24 to 29, 1990. t Reprint requests: Jeanine Henry-Suchet, M.D., C.H.R. Jean Rostand, 141 Grand rue, 92311, Sevres Cedex, France (FAX: 42-56-06-50). Laboratoire Magenta. § Institut Pasteur, Paris. II Laboratoire d'Eylau, Paris. 1f Hopital Jean Rostand.
*
296
Henry-Suchet et aI.
Chlamydia serology in gynecology
intrapelvic C. trachomatis infection often requires either laparoscopy or laparotomy; cervical and urethral C. trachomatis cultures have poor predictive value (2). The value of serology in detecting upper genital tract infections is still debatable. A high antichlamydial immunoglobulin (Ig) G titer, although often detected in upper genital tract infections (13), cannot be considered as proof of current infection. Antichlamydial IgA has been shown to be significantly associated with acute salpingitis, ectopic pregnancy, and tubal factor infertility (4-6); antichlamydial IgM is considered as a marker of recent or current infection (7-9); however, only a few Fertility and Sterility
authors have detected positive antichlamydial IgM in chlamydial intrapelvic gynecologic infections. To further assess the predictive value of serum antichlamydial antibody titers for the diagnosis and therapeutic response of women with C. trachomatis infections, we evaluated these antibodies in sequential specimens before treatment and during the follow-up of women treated for acute chlamydial salpingitis and for tubal factor infertility of chlamydial origin. Antibody titers were compared with results of C. trachomatis cultures.
MATERIALS AND METHODS Study Population
Three groups of women admitted to the gynecologic unit of Hopital Jean Rostand, Sevres between January 1, 1987 and January 1, 1989 were included in this study. Group A comprised 42 patients with acute salpingitis, selected by laparoscopy (35 cases) or three of the following clinical and/or biologic signs (7 cases): pelvic pain, fever, vaginal discharge, abnormal leukocyte count, and/or increased erythrocyte sedimentation rate. Group B comprised 131 healthy patients, requiring treatment for infertility, without any ofthe precited signs, selected as having evidence of tubal lesions at hysterosalpingography and/or laparoscopy (hydrosalpinx, tubal stenosis, and/or peritubal adhesions); these two patient groups were included prospectively. Group C comprised 98 healthy pregnant women, without past history of acute salpingitis or infertility treatment, retrospectively taken as a control group for serology. Diagnosis of Current Chlamydial Infection Serum Specimens
Patients had one serum specimen taken at the identification of clinical disease (group A), at their first consultation for tubal factor infertility (group B), and all patients but one in group A had a second serum collected after 6 to 9 weeks. For control group C, two frozen sera obtained during the first trimester of pregnancy and 2 months later were used. Sera were tested for antichlamydial IgG by microimmunofluorescence using serovar L2 elementary bodies (10) and revealed by fluorescein-conjugated goat anti-human-IgG (Sanofi Pasteur, Marnes la Coquette, France). Antichlamydial IgM Vol. 62, No.2, August 1994
and IgA were detected by inclusion fluorescent antibody, using McCoy cells infected by C. trachomatis, serovar L2, as antigen (Electronucleonics; distributed by Eurobio, Les VIis, France) and revealed by fluorescein-conjugated goat anti-human IgM and IgA (Sanofi Pasteur) (7,8). Before testing for antichlamydial IgM and IgA, IgG and rheumatoid factor were removed from sera by rheumatoid factor absorbent (Behring, Rueil-Malmaison, France). Serial dilutions of sera were tested from 1:20 in the same assay for each patient; a titer of 1:32 was considered positive for antichlamydial IgG, a titer of 1:20 was considered positive for antichlamydial IgM and IgA; a fourfold change in antibody level was considered as evolutive serology. Twenty-eight patients with positive antichlamydial IgM were tested for anti C. pneumoniae IgG and IgM with species-specific elementary bodies and revealed by the same anti IgG and IgM conjugates. Genital Specimens
These were obtained only from subjects of groups A and B; 34 patients in group A and 112 in group B had endocervical and urethral swabs taken before treatment; the others, referred by another hospital, started antibiotic treatment before their first visit to our unit, which excluded endocervical and urethral sampling. Fourteen patients in group A and 89 in group B had intrapelvic cultures taken at laparoscopy before treatment; for each patient,. one to three of the following sites were sampled: cul-de-sac fluid, adhesions, and/or tubal biopsies. Twelve sexual partners of group A and 110 of group B patients had urethral cultures. Specimens were collected in Eagle Minimum Essential Medium supplemented with vancomycin, gentamicin, amphotericin B, and 20% fetal calf serum, using one container for both cervix and urethra and one for each laparoscopic specimen. These were stored at 4°C for <24 hours and inoculated in cycloheximide-treated L cells. Inclusions were identified with fluorescein-labeled monoclonal antibody (Syva Bio-Merieux, Marcy l'Etoile, France). No blind passages were done in this series. Diagnosis and Treatment of Current Infection
Patients with positive antichlamydial IgM and/ or evolutive IgA or IgG titers on two sera and/or positive endocervical or urethral or intrapelvic C. trachomatis cultures were considered as having a current chlamydial infection. All patients in groups A and B with positive antichlamydial IgG were conHenry-Suchet et aI.
Chlamydia serology in gynecology
297
sidered to have a possible current C. trachomatis infection and were treated with a synthetic tetracycline (200 mgjd doxycycline) for 3 to 6 weeks. The usual partners of acute salpingitis patients received the same treatment for 3 weeks. Patients of this group without a single partner had full instructions to avoid new sexual contacts or to use condoms during treatment and the 3 following months; they complied fully, as assessed by repeated inquiries. Partners of tubal infertility patients with positive cultures were treated for 3 weeks. Patients with persistent positive antichlamydial IgM received repeated treatments (doxycycline + erythromycin or ofloxacine) for 6 weeks. Treatment Follow-up Test of Cure Cultures
All patients and partners with initial positive endocervical or urethral cultures had test of cure cultures of the same sites 1 month after treatment. Twenty-five patients in group A and 59 in group B had intrapelvic cultures taken at follow-up laparoscopy, 1 to 3 months after the completion of treatment. Most patients of group B had only one laparoscopy: 89 before treatment and 59 after treatment. Fifteen women with positive serology to C. trachomatis in this group had two laparoscopies, one before and one after treatment. Serologic Follow-up
A serological follow-up was proposed to 31 patients of group A and 30 patients of group B who were considered to have current C. trachomatis infection. Twenty-two patients of group A and 25 patients of group B accepted this protocol: they had two to five additional sequential sera collected at 2 to 6 month intervals for a 1 year follow-up period. A statistical analysis was made using the x2 test with Yates correction for continuity.
RESULTS Detection of C. trachomatis by Serology Before Treatment
tility (75.6%), and 31 of 98 healthy pregnant controls (31.6%). The geometric mean titers were 1:450, 1:190, and 1:94, respectively. Both patient groups had a similar prevalence of antichlamydial IgG, which was significantly different from the control group (P < 0.001). IgA Titers
Antichlamydial IgA antibodies were detected in 28 of 42 women with acute salpingitis (66.7%), in 76 of131 women with tubal infertility (58%), and in 16 of the 98 controls (16.3%). The geometric mean titers were similar in the three groups, 1:49, 1:46, and 1:36, respectively. Both patient groups had a similar prevalence of antichlamydial IgA, which was significantly different from the control group (P < 0.0001). IgM Titers
Antichlamydial IgM antibodies were detected in 26 of 42 women with acute salpingitis (61.9%) and 28 of 131 women with tubal infertility (21.4%). There was no detectable antichlamydial IgM in the control group; the geometric mean titers were 1:40 and 1:22 in the patient groups. A significant difference in the occurrence of IgM antibody to C. trachomatis was noted between the two patient groups (P < 0.001), as well as between A and B patients versus the control group (P < 0.0001) The specificity of the IgM antibodies for C. trachomatis was evaluated by testing 28 of the sera for IgM antibody to C. pneumoniae, 11 sera from group A and 17 from group B. Of 11 patients with acute salpingitis who were antichlamydial IgM positive, one was also IgM positive for C. pneumoniae. Similarly, only 3 of 17 women with tubal infertility were IgM positive for both C. trachomatis and C. pneumoniae. Antibodies to C. trachomatis in Two Sequential Sera
In group A, one woman who was positive for antichlamydial IgG, IgA, and IgM did not have a second serum taken. Forty-one cases were studied in sequential sera (Table 1). IgG Titers
IgG Titers
Before treatment, IgG antibodies to C. trachomatis were detected in 32 of 42 women with acute salpingitis (76.2%), 99 of 131 women with tubal infer298
Henry-Suchet et al.
Chlamydia serology in gynecology
Of the 31 women initially positive for antichlamydial IgG and having sequential sera taken, all remained IgG positive, 28 of them (90.3%) with an equivalent titer and a significant evolution was Fertility and Sterility
Table 1
Evolution of Chlamydial Antibody Titers on Two Sequential Sera Taken at 0 and 3 Months First serum
Patient groups
No. of patients
No. of positive cases
Evolution on second serum No. of seroconversions
No. with significant increase
No. with significant decrease
No. without change
1 2 0
28 97 31
1 1 0
25 75 16
7 6 0
18 22 0
IgG A* B C
41 131 98
31 99 31
2 0 0
2 0 0 IgA
A* B
C
41 131 98
27 76 16
1 1 0
1 0 0 IgM
A* B C
41 131 98
25 28 0
1 0 0
0 0 0
* In group A, one patient with positive IgG, IgA, and IgM did not have a second serum taken.
observed for antichlamydial IgG in five cases: two seroconversions, two increases, and one decrease. In group B, 97 of99 (98%) women with tubal factor infertility who were antichlamydial IgG positive remained positive with an equivalent titer on second serum and two had a significant decrease in IgG level. There were no changes in the incidence or titer of antichlamydial IgG in the control group. IgA Titers
In group A, 25 of 27 (92.6%) patients who were IgA positive also remained positive with an equivalent titer on second serum. A significant evolution was observed for IgA in three cases: one seroconversion, one increase, and one decrease. In group B, 75 of 76 patients (98.7%) with antichlamydial IgA antibody remained positive with an equivalent titer in the second serum; there was one seroconversion and one significant decrease. There were no changes in the incidence or titer of antichlamydial IgA in the control group. IgM Titers
In contrast to IgG and IgA, of the 25 patients with acute salpingitis who initially were antichlamydial IgM positive, only 18 of 25 (72%) remained positive with an equivalent titer on second serum. The titers decreased in seven cases (28%) and became seronegative in five cases (20%) at 1 to 3 months; there was also one seroconversion but no Vol. 62, No.2, August 1994
increase in titer. Six of 28 women (21.4%) with tubal infertility who were initially antichlamydial IgM positive exhibited decreased titers and 5 (17.9%) became seronegative. During this time interval, there were no seroconversions and no increases in titer. In the control group, no antichlamydial IgM was detected in the second serum. Interpretation of Antibody Change in Sequ~ntial Sera
In group A, of 42 women with acute salpingitis, 32 were considered to be of C. trachomatis origin: 27 having positive antichlamydial IgM of which one was on the second serum only and 5 having significant evolution of antichlamydial IgG (2 cases) or both IgG and IgA (3 cases) in the absence ofIgM. Of the 10 other cases, 8 remained negative for chlamydial antibodies on two sera and had negative endocervical and urethral and intrapelvic cultures, two were positive for antichlamydial IgG at equivalent titers on both sera, had negative cultures, and were negative for antichlamydial IgA and IgM. These 10 patients were considered as having no current chlamydial infection. In this group, the sensitivity of antichlamydial IgM in the detection of current chlamydial acute salpingitis was 27 of 32 (84.4%), the sensitivity of antichlamydial IgA was 29 of 32 (90.6%), the sensitivity of antichlamydial IgG was 100% with 2 of 34 (5.9%) false positives. In group B, 30 women were considered as having a current C. trachomatis infection, 28 because of positive antiHenry-Suchet et al.
Chlamydia serology in gynecology
299
Table 2
Evolution of Chlamydia Antibody Titers on 12 months Follow-up of Groups A and B (n
o to 8 mo
o to 4 mo Antibody toCT
Total number
No. with significant decrease
IgG IgA IgM
47 44 42
1 (2)* 1 (2) 13 (31)t
No. with decrease to zero 0(0) 1 (2) 11 (26)t
* Values in parentheses are percentages.
t IgM
significantly different from IgG and IgA decrease, P < 0.001.
=
47)
o to 12 mo
No. with significant decrease
No. with decrease to zero
No. with significant decrease
No. with decrease to zero
9 (19) 4 (9) 26 (62)+
1 (2) 4 (9) 26 (62)+
15 (22) 13 (29) 32 (76)+
2 (4) 7 (16) 32 (76)+
+
IgM significantly different from IgG and IgA decrease to zero, P < 0.0001.
chlamydial IgM and two because of positive intrapelvic cultures and evolutive antichlamydial IgA and IgG or evolutive IgG only.
seronegative at 8 and 12 months, respectively. The difference is significant versus antichlamydial IgG and IgA seronegativation (P < 0.0001).
Antibodies to C. trachomatis in Long-term Follow-up Sera
Detection of C. trachomatis by Culture
Changes in antibody titers at 4, 8, and 12 months were not significantly different between the 22 women with acute salpingitis and the 25 women with tubal infertility. Therefore, the data for the two groups (47 patients in total) were combined for analysis (Table 2). IgG Decrease
The antichlamydial IgG antibody titer decreased in 1 of 47 (2%) patients at 4 months, 9 of 47 (19%) at 8 months, and 15 of 47 (31.9%) by 12 months. However, only 2 of 47 women (4%) reverted from IgG seropositive to seronegative over this 1-year interval. IgA Decrease
Similarly, antichlamydial IgA antibody titers decreased in 1 of 44 (2%) women at 4 months, 4 of 44 (9%) at 8 months, and 13 of 44 (29%) at 12 months. Only 7 of 44 women (16%) became antichlamydial IgA negative after 12 months. IgM Decrease
The titer of IgM antibodies to C. trachomatis, in contrast to the other antibody isotypes, was significantly decreased in 13 of 42 women (30.9%) at 4 months (P < 0.001), 26 of 42 women (61.9%) at 8 months, and 32 of 42 women (76%) at 12 months (P < 0.001). Eleven of 42 women (26%) converted to antichlamydial IgM seronegative at 4 months, 26 of 42 (61.9%) and 32 of 42 women (76%) became 300
Henry-Sucbet et al.
Chlamydia serology in gynecology
Before treatment, C. trachomatis was detected in 16 of 34 endocervical or urethral cultures (47%) and 4 of 14 pelvic cultures (28.6%) in women with acute salpingitis (group A). Of the 16 patients with positive endocervical or urethral cultures, only 2 had pelvic cultures before treatment and they were positive. In total, 18 acute salpingitis patients had either endocervical, urethral, or pelvic positive cultures. Among the women with tubal infertility (group B), 4 of 112 endocervical and urethral (3.6%) and 9 of 89 pelvic (10.1 %) cultures were positive; there was a significant difference between the two groups in endocervical and urethral cultures (P < 0.01). In group A, of 12 usual partners who were sampled, 6 had positive cultures for C. trachomatis, 4 of them having no clinical signs. Of 110 usual partners in group B who were sampled, 4 had positive cultures, all of them being asymptomatic. One month after treatment, all endocervical and urethral cultures were negative with the exception of one acute salpingitis case whose cervix remained positive after a first course of treatment and became neg~tive after a second course. Her husband had positive pretreatment cultures, which remained positive despite three different courses of synthetic tetracycline treatment for 3 weeks. The isolate was kept and its sensitivity to antibiotics was tested in vitro, without showing any resistance to usual minimal inhibitory concentrations. Intrapelvic cultures performed during follow-up laparoscopy were still positive for C. trachoma tis in 6 of 25 women after acute salpingitis (24%) and 10 of 59 women with tubal infertility (16.9%), without significant difference between the two groups. Fertility and Sterility
Table 3
Chlamydia Cervical-Urethral and Intrapelvic Cultures Before Treatment Correlated to Serology Group A
Group B
Serology
Cervix -urethra
Intrapelvic
Cervix -urethra
IgG IgA IgM
N+/N*
N+/N* Cul-de-sac Adhesion Tube
N+/N*
n+/N*
o ofS o of3
10fS o of2 10fl
o of 20 o of2S o of47
o of 30 o of 19
+ + + +
+ + Total
+ +
1of3 o of 1t 15 of 19t 16 of 34
2of3 4 of 14
o of5 o of2
o of 1
10fl 1 of 1 2of9
o of1
* N + /N, number of cases with positive cultures/total number of cases sampled.
Relation Between Before Treatment Cultures and Serology Cervical- Urethral Cultures
In group A, of six women with acute salpingitis who were positive for antichlamydial IgG and/or IgA antibodies and negative for antichlamydial IgM, only one (16.7%) had a positive endocervical or urethral culture. In significant contrast, 15 of 20 women (75%) with antichlamydial IgG, IgA, and IgM (19 cases) or IgG and IgM (1 case) antibodies were endocervical or urethral culture positive (P < 0.001). In group B, among the women with tubal infertility, the small number of positive cervical cultures, 4 of 112, precluded statistical analysis. However, each of the culture-positive women had antichlamydial IgG, IgA, and IgM antibodies (Table 3). Intrapelvic Cultures
In group A, among the four women with acute salpingitis whose intrapelvic cultures were positive for C. trachomatis, one was negative on first serum and had a seroconversion to antichlamydial IgG antibodies in her second serum, one had antichlamydial IgG plus IgA with significant evolution on second serum and two were positive for anti C. trachomatis IgG, IgA, and IgM (Table 3). In Group B, positive intrapelvic cultures were obtained in nine women with tubal infertility, of which two had antichlamydial IgG and IgA without IgM; one had IgG and IgM without IgA (the IgA became positive on second serum); and six had IgG, IgA, and IgM antibodies. The incidence of positive intrapelvic cultures in this group was significantly higher in Vol. 62, No.2, August 1994
1of2
o of2
2 of 3 30f7
4of27 4 of 112
Intrapelvic
2of26 1 of It 6 of 13t 90fS9
Cul-de-sac Adhesion
o of 25 o of 14 o of 21 lofl 5 of 12 6of73
Tube
o of5 o of2
o of5 o ofS
2 of 15
1 of 12
2of5 4of27
lof2 2of27
t Cases with positive IgM versus cases without IgM, P < 0.001.
women who were antichlamydial IgM seropositive (7/14, 50%) versus women who were IgG or IgA seropositive and IgM seronegative (2/45, 4.4%; P < 0.001). Relation Between After Treatment Intrapelvic Cultures and Serology
In group A, among the six women who had acute salpingitis and still were chlamydia culture positive 3 months after treatment, five remained seropositive for IgG, IgA, and IgM antibodies to C. trachomatis; the sixth woman was only IgG positive. In group B, similarly, ofthe 10 women with tubal infertility who remained culture positive, 8 had IgG, IgA, and IgM antichlamydial antibodies and two had IgG and IgA without IgM antibody (Table 4). Women in group B who were positive for IgM antibody to C. trachomatis 3 months after treatment had a significantly higher incidence of positive intrapelvic cultures for this organism after treatment (8/15, 53.3%) than women who were chlamydia IgM negative (2/44, 4.5%; P < 0.01). In group B, there was no significant difference in the rate of positive cultures before and after treatment among women with positive antichlamydial IgM. Of the 15 tubal factor infertility cases who had intrapelvic cultures taken before and after treatment, 5 of 8 who initially vyere antichlamydial IgG, IgA, and IgM positive had positive cultures before treatment. In contrast, the other seven patients who were antichlamydial IgM negative and IgG or IgG and IgA positive had negative initial cultures. After treatment, antichlamydial IgM persisted in four cases; this was associated with positive cultures after treatment in all cases; of the 11 IgM-negative women, only 2 (18.2%) had positive cultures after treatment. Henry-Suchet et al. Chlamydia serology in gynecology
301
r Table 4
Chlamydia Intrapelvic Cultures After Treatment at Follow-up Laparoscopy Correlated to Serology Group A
Serology IgG
IgA
IgM
N+/N*
Cul-de-sac
Adhesion
Tube
N+/N*
Cul-de-sac
1 of 2 o of 7 3 of 14 4 of 23
o of2 o of 5
o of2
o of 2 o of 7
o of2 o of4
2of9 2 of 16
2 of 5 2 of 7
2 of 35 8 of 15t 10 of 59
1 of 30 5 of 12 6 of 48
o of 0 + + +
+ + Total
+
Group B
1 of 3 o of 7 5 of 15 6 of 25
Adhesion
Tube
o of 3 o of 12
o of 2
2of5 20f20
2 of 11 2 of 5 4 of 18
* N+/N, number of cases with positive cultures/total number of cases sampled.
t Cases with positive IgM versus cases without IgM, P < 0.001.
Sequelae in Group A at Follow-up Laparoscopy
peated questions, it was presumed that cases with positive cultures after treatment were not due to recontamination. A serotyping of the strains should have given better proof of the absence of recontamination but was not done in this series. Before treatment, the detection of IgM antibodies to C. trachomatis, in contrast to IgG and IgA antibodies, was correlated with culture-positive chlamydial infections in endocervical and urethral samples of women with acute salpingitis and pelvic samples of women with tubal infertility. We may regret not having collected endocervical and urethral samples in the control series of pregnant women but, according to another series done in our unit during the same period (15), endocervical and urethral prevalence of C. trachoma tis in pregnant women was 2%. The absence of antichlamydial IgM in this control series is in total contrast to the two other groups; the 16.3% rate of antichlamydial IgA in pregnant women possibly is due to a previous silent infection, without tubal sequelae. The persistence of IgM antibody to C. trachomatis 3 months after the completion of treatment was correlated with continued positive intrapelvic cultures in tubal factor infertility cases and to tubal obstruction or adhesions in acute salpingitis cases. However, as only 3 of 11 women with sequelae had pretreatment laparoscopy, we cannot exclude a preexisting silent disease in the 8 other cases. Immunoglobulin G and IgA antibodies to C. trachomatis persisted in most seropositive women and were unrelated to antibiotic failure. Thus, in women with upper genital tract infections, IgM antibodies to C. trachomatis appear to indicate the current presence of this organism. However, the presence of IgA and/or IgG antibodies to C. trachomatis and the absence of IgM antibodies was observed in a minority of our culture-positive subjects and five acute salpingitis cases were considered of chlamydial origin for having significantly evolutive IgG and/or IgA
Among the 25 women of the acute salpingitis group who had follow-up laparoscopies, 3 had previous lesions that precluded the evaluation of sequelae. In the 22 other cases, the presence of tubal lesions (tubal obturation and/or peritubal adhesions) was correlated with the persistence of IgM antibody to C. trachomatis: 11 of 15 with persistent antichlamydial IgM had sequelae compared with 1 of 10 without IgM (P < 0.01). In contrast, no similar relationship was observed between tubal lesions and antichlamydial IgA antibodies in this series.
DISCUSSION
The present study documents the persistence of C. trachoma tis in the upper genital tract, but not in the cervix, several months after antibiotic therapy for chlamydia-associated acute salpingitis or tubal infertility. The pretreatment discrepancy between the rate of positive endocervical and urethral cultures in acute salpingitis versus tubal infertility cases is a possible consequence of a long-term contamination in the latter group. Other investigators also have reported evidence of intrapelvic infection in chronic cases (11) and after antibiotic treatment (12). We previously have observed the persistence of C. trachomatis-positive cultures in pelvic samples after treatment in 12% of women with acute salpingitis and 45% of women with tubal infertility who had, before treatment, positive intrapelvic cultures (13), and suggested that this was a cause for tuboplasty failure (14). Together, these reports strongly indicate that, in the absence of reinfection due to new sexual contact, C. trachomatis can persist as a chronic infection in the upper genital tract of some women despite standard treatment. On the basis of instructions given to patients, the fact that the partners were treated, and the answers to re302
Henry-Suchet et a1.
Chlamydia serology in gynecology
Fertility and Sterility
antibody titers in the absence of antichlamydial IgM. This lack of antichlamydial IgM detection in five cases possibly is due to a low response in these women. Previous efforts to detect IgM antibodies to C. trachomatis have met with variable success. Anestad et al. (9), using an assay similar to ours, reported a comparable incidence of antichlamydial IgM antibodies in their study population. Miettinen et al. (16), using enzyme immunoassay (EIA) and immunoblot analysis, observed antibodies to the major outer membrane protein of C. trachomatis elementary bodies in all patients who had C. trachomatis isolated. They concluded that acute salpingitis patients with and without upper genital tract C. trachomatis infection cannot be differentiated by reactivity of sera to specific chlamydial polypeptide antigens. The effective single test in their series was the determination of specific serum IgA antibody response by EIA. In a I-year followup of 69 antigen-positive patients, Clad et al. (17) found IgG and IgA antichlamydial antibodies that were correlated with positive cervical cultures and persisted in most cases despite treatment, suggesting the persistence of inflammation. Piura et al. (18) have emphasized the persistence of antichlamydial IgG and IgA antibodies without significant change after treatment in 63.6% of acute salpingitis cases and regretted not having correlated these findings to intrapelvic cultures. In our experience, antichlamydial IgM antibody determination by immunofluorescence using C. trachomatis-infected cells is easier to read and yields a higher incidence of positives than does immunofluorescence on purified elementary bodies. The use of infected cells allows detection of antibodies to a larger repertoire of chlamydial antigens than does analysis using elementary bodies, more particularly lipopolysaccharide and heat-shock proteins, which are genus antigens, related to deep-seated infection and delayed hypersensitivity. Inclusion of an absorption step to remove IgG and rheumatoid factor is effective in reducing the rate of false negatives due to competition between IgG and IgM antibodies for binding sites and false positives due to the binding of IgM rheumatoid factor to IgG that ha~ reacted with chlamydial antigens. Additional testing using C. pneumoniae reduces the incidence of cross-antigen reaction even further; in the present study, 4 of 28 women (14.3%) positive for the C. trachomatis IgM assay also were positive for C. pneumoniae IgM. Further studies are required to Vol. 62, No.2, August 1994
clarify the value of C. trachomatis antibody determinations in different populations. Our findings also reinforce previous observations that C. trachomatis is a major pathogen in both acute salpingitis and tubal infertility (1-3) and that analysis of cervical samples for C. trachomatis in tubal infertility cases does not reflect its possible occurrence in the upper genital tract (2). Because early detection and treatment of intrapelvic chlamydial infection is crucial to the preservation of fertility, the continued analysis and modification of serologic methods to detect asymptomatic C. trachomatis infections should receive the highest priority.
Acknowledgments. We thank Yvonne Perol, M.D., Ph.D. (Laboratoire de microbiologie, Hopital Saint Louis, Paris, France; Steven S. Witkin, Ph.D., Department of Obstetrics and Gynecology, Cornell University Medical College, New York, New York; and Michael P. Diamond, M.D., Department of Obstetrics, Gynecology, and Surgery, Vanderbilt University, Nashville, Tennessee for scientific advice.
REFERENCES 1. Mlhdh PA, Ripa T, Svenson L, Westrom L. Chlamydia trachomatis infection in patients with acute salpingitis. N Engl J Med 1977;296:1377-9. 2. Henry-Suchet J, Catalan F, Loffredo V, Serfaty D, Siboulet A, Perol Y, et al. Microbiology of specimens obtained by laparoscopy from controls and from patients with pelvic inflammatory disease or infertility with tubal obstruction. Am J Obstet Gynecol 1980;138:1022-5. 3. Henry-Suchet J, Utzmann C, De Brux J, Ardoin P, Catalan F. Microbiologic study of chronic inflammation associated with tubal factor infertility: role of Chlamydia trachomatis. Fertil Steril1987;47:274-7. 4. Sarov I, Sarov B, Lunenfeld E, Zion H, Chaim W, Piura B. The significance of Chlamydia specific serum IgA antibodies in Chlamydia trachomatis infections. In: Esculapio, editor. Proceedings of the European Society for Chlamydia Research Bologna: Italy, 1988:234-8. 5. Sarov I, Lunenfeld E, Sarov B, Hanuka N, Rosenzweig R, Potashnik G, et al. Chlamydia specific IgG and IgA antibodies in women with obstructive infertility as determined by immunoblotting and immunoperoxidase assays. Eur J Epidemiol 1998;4:216-23. . 6. Sellors JW, Mahony JB, Chernesky MA, Rath DJ. Tubal factor infertility: an association with prior chlamydial infection and asymptomatic salpingitis. Fertil Steril 1988; 49:451-7. 7. Scieux C, Colimon R, Perol Y. Detection des anticorps antichlamydiens par immunofluorescence indirecte sur inclusions de Chlamydia trachomatis. Med Mal Infect 1982; 8:466-9. 8. Scieux C, Colimon R, Bianchi A, Felten A, Perol Y. Interet diagnostique de la recherche des anticorps anti-chlamydiens au cours des salpingites. Presse MEld 1987;16:715-8.
Henry-Suchet et al.
Chlamydia serology in gynecology
303
9. Anestad G, Lunde 0, Moen M, Dalaker K. Infertility and chlamydial infection. Fertil Steril1987;48:787-90. 10. Wang SP, Grayston JT, Alexander ER, Holmes KK. Simplified micro-immunofluorescence test with trachoma lymphogranuloma venerum (Chlamydia trachomatis) antigens for use as screening test for antibody. J Clin Microbiol 1975;1:250-6. 11. Keilani A, Boulieu D, Raudrant D, Carraz M, Quenin P. Role de Chlamydia trachomatis dans les pathologies tubaires (salpingites aigues et srerilite tubaire). J Gynecol Obstet BioI Reprod (Paris) 1989;18:167-72. 12. Shepard MK, Jones RB. Recovery of Chlamydia trachomatis from endometrial and fallopian tube biopsies in women with infertility of tubal origin. Fertil Steril 1989;52:232-8. 13. Henry-Suchet J, Douieb P. Treatment of culture documented intrapelvic Chlamydia trachomatis infection in women selected for tuboplasties: preliminary studies. In: Oriel D, Ridgway G, Schachter J, Taylor-Robinson D, Ward M, editors. Chlamydial infections. New York: Cambridge University Press, 1986:201-4.
304
Henry-Suchet et al.
Chlamydia serology in gynecology
14. Henry-Suchet J, Veluyre M, Pia P. Etude statistique des facteurs influen~ant Ie pronostic des plasties tubaires. J Gynecol Obstet BioI Reprod (Paris) 1989;18:571-80. 15. Shafer MA, Pessione F, Scieux C, Bianchi A, Casin M, Athea N, et a1. Chlamydia trachomatis: facteurs de risques chez les femmes de la region parisienne. J Gynecol Obstet BioI Reprod (Paris) 1993;22:163-8. 16. Miettinen A, Heinonen PK, Teisala K, Punnonen R, Paavonen J. Antigen specific serum antibody response to Chlamydia trachomatis in patients with acute pelvic inflammatory disease. J Clin Pathol 1990;43:758-61. 17. Clad A, Baumgarten B, Petersen E. Long term persistence of disappearance of Chlamydial IgA and IgG in gynecologic patients. In: Mardh P A, La Placa M, Ward M, editors. Proceedings of the European Society for Chlamydia Research. Uppsala, Sweden: Uppsala University Center for STD Research, 1992:119. 18. Piura B, Sarov B, Sarov I. Persistence of antichlamydial antibodies after treatment of acute salpingitis with doxycycline. Eur J Obstet Gynecol Reprod BioI 1993;48:117-21.
Fertility and Sterility