- Email: [email protected]
Rates of bacterial vaginosis in women undergoing in vitro fertilisation for different types of infertility
BJOG: an International Journal of Obstetrics and Gynaecology June 2002, Vol. 109, pp. 714– 717
Rates of bacterial vaginosis in women undergoing in vitro fertilisation for different types of infertility Janet D. Wilson*, Susan G. Ralph, Anthony J. Rutherford Objective To assess whether the rate of bacterial vaginosis (BV) is higher in women with tubal factor infertility compared with those with other causes of infertility. Design Cross-sectional study. Setting Assisted conception unit of a teaching hospital in Leeds. Population Consecutive women undergoing in vitro fertilisation. Methods Women undergoing in vitro fertilisation (IVF) had a vaginal smear taken at the time of their egg collection. The smear was Gram-stained and graded as normal, intermediate or BV. Main outcome measures The presence of bacterial vaginosis and the causes of infertility. Results A total of 749 women were included. The vaginal smears were normal in 63.6%, intermediate in 12.1%, and BV in 24.3%. The rates of BV in women with different types of infertility were 36.4% in tubal factor, 15.6% in male factor, 33.3% in anovulation, 12.5% in endometriosis and 18.9% in unexplained infertility. After controlling for the effects of age and smoking using a multivariate logistic regression model, women with tubal infertility were significantly more likely to have BV than women with endometriosis OR 3.63 (95% CI 1.52 – 8.67); male factor OR 2.98 (95% CI 1.80 –4.90); and unexplained infertility OR 2.20 (95% CI 1.35 –3.59). The adjusted figures for the increase of BV in women with anovulation were: endometriosis OR 3.77 (95% CI 1.28 –11.08); male factor OR 3.09 (95% CI 1.37– 6.96); and unexplained infertility OR 2.29 (95% CI 1.02 –5.12). Conclusions Women with tubal infertility were three times more likely to have BV than women with endometriosis, male factor or unexplained infertility. These findings support the association between BV, pelvic inflammatory disease (PID) and tubal damage but do not help distinguish between cause and effect. Women with anovulation were also three times more likely to have BV than women with endometriosis or male factor infertility, supporting suggestions of hormonal influence on vaginal flora. INTRODUCTION The main causes of pelvic inflammatory disease (PID), and hence tubal infertility, are Chlamydia trachomatis and Neisseria gonorrhoeae1. However, bacterial vaginosis (BV) is being increasingly implicated in upper genital tract infections in women. It can cause ascending infection in pregnant women, the chorioamnionitis, then predisposing to preterm delivery2. It also increases the rate of PID following surgical termination of pregnancy 3,4. Nonpregnant women with BV, but without gonorrhoea or chlamydia and without any symptoms or signs of upper genital tract inflammation, have been found to have plasma cell endometritis5. Symptoms of abdominal pain6 and intermenstrual bleeding7 are more common in women with BV than those with normal flora. BV is usually present in women with acute salpingitis, and BV-associated organisms are commonly isolated from the upper genital tract of
The General Infirmary at Leeds, Leeds, LS1 3EX, UK * Correspondence: Dr J. D. Wilson, The General Infirmary at Leeds, Leeds, LS1 3EX, UK. D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology PII: S 1 4 7 0 - 0 3 2 8 ( 0 2 ) 0 1 2 9 7 - 1
patients with PID. They are usually found as co-infections with C. trachomatis and N. gonorrhoeae but they have sometimes been isolated from the upper genital tract in the absence of these, with the suggestion that they increase the risk of upper genital tract infection in women with and without sexually transmitted infections8. There is therefore mounting evidence that BV may cause PID in non-pregnant women in the absence of gonorrhoea and chlamydia. Consequently, BV could be associated with tubal factor infertility. The aim of this study was to assess whether the rate of BV is higher in women with tubal factor infertility compared with those with other causes of infertility.
METHODS Women undergoing in vitro fertilisation (IVF) at the assisted conception unit of The General Infirmary at Leeds, during the period April 1996 to June 1997, were recruited. All participants gave written consent and ethical approval was obtained. Baseline demographic data, including age, smoking habit, ethnic origin, previous obstetric history and www.bjog-elsevier.com
BV RATES IN WOMEN UNDERGOING IVF FOR DIFFERENT TYPES OF INFERTILITY
current gynaecological history were included in standardised records. All patients recruited had been thoroughly investigated according to an established protocol, which included a baseline early follicular phase endocrine profile, mid-luteal progesterone, detailed semen analysis, and either a hysterosalpingogram or laparoscopy to assess tubal patency dependent on the patient’s history. Those patients with symptoms suggestive of pelvic pathology or a past history of established disease were investigated by laparoscopy, with the remainder having a hysterosalpingogram. If pelvic pathology was identified on the hysterosalpingogram, a laparoscopy was performed to confirm the extent of the disease. On completion of the investigations, the presumptive cause of the infertility was recorded. In cases where there was more than one pathology, the most significant condition was identified as the primary cause of infertility. During the participants’ IVF cycle, immediately prior to egg collection, a non-lubricated bivalve speculum was inserted and the vaginal walls and posterior fornix were sampled with a sterile cotton swab. This was smeared onto a glass slide, air-dried and then stained according to Gram’s method. The women were simultaneously screened for coexistent sexually transmitted infections. The full methodology and results of the IVF have been described elsewhere9. Two independent observers (JDW and SGR) read the slides. They were blinded to each other’s results and to the patient details. The criteria used to diagnose BV on the basis of Gram’s stain alone were the modified Spiegel’s criteria2. Thus, flora were graded as either normal (predominantly Lactobacilli), intermediate (reduced Lactobacilli mixed with other morphotypes) or BV (few or absent Lactobacilli with greatly increased numbers of Gardnerella vaginalis, other morphotypes or both). Where the observers disagreed, the slide was restained and reviewed until consensus was reached.
Statistical methods The rates of BV were calculated for the different types of infertility. Any differences between the BV rates were compared by m2 test. Associations between BV and age, ethnicity, and smoking were assessed by m2 test, as were associations between type of infertility, age and smoking. Finally, all the factors were included in a multiple logistic regression model to adjust for the other confounding variables.
RESULTS Full data are available on 749 women. The age range was 21– 44 (mean 33.1) years, 711 (94.9%) were white and 140 (18.7%) were smokers. No sexually transmitted infections were detected in the first 200 women enrolled so D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
715
Table 1. The reasons for infertility and rates of bacterial vaginosis (BV) with the different types of infertility. Cause of infertility Tubal Male factor Unexplained Endometriosis Anovulation
No. of women (%) 258 211 185 56 39
(34.4) (28.2) (24.7) (7.5) (5.2)
Rates of BV (%) 94/258 33/211 35/185 7/56 13/39
(36.4) (15.6) (18.9) (12.5) (33.3)
screening was thereafter abandoned. The results of the vaginal smears were normal in 476 (63.6%), intermediate in 91 (12.1%) and BV in 182 (24.3%). The kappa value of inter-observer variation was 0.94. The causes of infertility and the rates of BV with the different types of infertility are shown in Table 1. There was no significant difference in rates of BV between women with endometriosis, male factor infertility and those with unexplained infertility. However, women with tubal infertility were significantly more likely to have BV than women with all other causes OR 2.70 (95% CI 1.87– 3.9); endometriosis OR 3.98 (95% CI 1.62 –10.94); male factor OR 3.23 (95% CI 1.99 – 5.25); and unexplained infertility OR 2.49 (95% CI 1.54 – 4.04). In addition, women with anovulation were also significantly more likely to have BV than women with endometriosis OR 3.23 (95% CI 1.01 –10.59), and male factor infertility OR 2.62 (95% CI 1.12– 6.09). There was no significant difference in rates of BV between women with anovulation and tubal infertility. There was an association between age and BV, younger women were more likely to have BV (m2, P ¼ 0.009), and between smoking and BV, smokers were more likely to have BV (OR 3.41; 95% CI 2.22– 5.25). There was no significant association between ethnicity and vaginal flora, but there were small numbers of non-white women. There was an association between type of infertility and age, women with tubal infertility were younger (m2, P ¼ 0.03), and those with unexplained infertility were older (m2, P ¼ 0.002); and between type of infertility and smoking, women with tubal infertility were more likely to smoke (OR 3.35; 95% CI 2.26 –4.98). The significant differences in rates of BV with different types of infertility remained after adjusting for all the other confounding variables by multiple logistic regression. The adjusted figures for the increase of BV in women with tubal infertility were: adjusted OR 3.63 (95% CI 1.52– 8.67) for endometriosis, adjusted OR 2.98 (95% CI 1.80– 4.90) for male factor and adjusted OR 2.20 (95% CI 1.35 –3.59) for unexplained infertility. The adjusted figures for the increase of BV in women with anovulation were: adjusted OR 3.77 (95% CI 1.28– 11.08) for endometriosis, adjusted OR 3.09 (95% CI 1.37 – 6.96) for male factor and adjusted OR 2.29 (95% CI 1.02– 5.12) for unexplained infertility. The adjusted increased risk of BV in smokers was OR 2.82 (95% CI 1.81 – 4.40).
716
J.D. WILSON ET AL.
DISCUSSION The overall BV rate of 24.3% was much higher than the 11 – 15% reported from obstetric and gynaecology units within the UK2,10, but comparable to the rate of 25.6% found in a similar IVF population11. We had initially speculated that the high rate of BV could be due to hormonal manipulation during the IVF, however, we found no difference between the BV rate prior to IVF and at the egg collection12. There was great variation between the BV rates in women with different causes of infertility, with rates of 12 – 15% in those with endometriosis and male factor infertility and 33– 36% in those with anovulation and tubal infertility. Women with tubal factor infertility and anovulation were two to three times more likely to have BV than women with other types of infertility. There was an inter-relationship between younger age, tubal infertility, smoking and BV. Other groups have reported an association between age and BV2, and smoking and BV13. The main cause of tubal infertility is previous salpingitis. Smoking and young age have both been positively associated with PID14 . This inter-relationship is therefore not surprising, but importantly, the significantly raised rate of BV in women with tubal infertility remained after adjusting for these other confounding factors. Two other groups have reported this association between BV and tubal infertility, but due to the small number of women in these studies, comparisons could only be made between tubal infertility and all other causes11,15. From our study, it is obvious that not all women with infertility have the same rates of BV. A weakness of our study is that we do not have a control group of women from obstetric and gynaecology clinics, or with infertility but not undergoing IVF. However, it does appear that those with endometriosis, male factor and unexplained infertility have rates comparable with those published from UK obstetric and gynaecology populations2,10, whereas women with tubal infertility and anovulation are two to three times more likely to have BV. The significantly raised BV rate in women with anovulation is interesting. There is mounting evidence of a hormonal influence on vaginal flora. Three longitudinal studies of vaginal flora showed an association with changing flora and the menstrual cycle suggestive of hormonal influence16 – 18. Our finding of a significantly increased rate of BV in women with chronic abnormal oestrogen and progesterone levels warrants further investigation.
CONCLUSION The finding of a significantly raised rate of BV in women with tubal factor infertility supports the association
between BV, PID and tubal damage, but this cross sectional study does not help distinguish whether the increased BV rate is secondary to previous tubal damage initiated by C. trachomatis or N. gonorrhoeae, or whether BV-associated organisms compromise host immunity and facilitate spread of these sexually transmitted infections into the upper genital tract, or whether BV itself can cause damage to the fallopian tube.
Acknowledgements The authors would like to thank Wendy Parsons for her help with the statistical analysis.
References 1. Cates Jr W, Rolf Jr RT, Aral SO. Sexually transmitted disease and infertility: an epidemiologic update. Epidemiol Rev 1990;12: 199 – 220. 2. Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson J. Abnormal bacterial colonisation of the genital tract and subsequent preterm delivery and late miscarriage. BMJ 1994; 308:295 – 298. 3. Larsson P-G, Berman B, Forsum U, Platz-Christensen JJ, Pahlson C. Mobiluncus and clue cells as predictors of PID after first trimester abortion. Acta Obstet Gynecol Scand 1989;68:217 – 220. 4. Hamark B, Forssman L. Postabortal endometritis in Chlamydianegative women—association with preoperative clinical signs of infection. Gynecol Obstet Invest 1991;31:102 – 105. 5. Korn AP, Bolan G, Padian N, Ohm-Smith M, Schachter J, Landers DV. Plasma cell endometritis in women with symptomatic bacterial vaginosis. Obstet Gynecol 1995;85:387 – 390. 6. Eschenbach DA, Hillier S, Critchlow C, Stevens C, De Rouen T, Holmes KK. Diagnosis and clinical manifestations of bacterial vaginosis. Am J Obstet Gynecol 1988;158:819 – 828. 7. Larsson P-G, Bergman B, Forsum U, Pahlson C. Treatment of bacterial vaginosis in women with vaginal bleeding complications or discharge and harboring Mobiluncus. Gynecol Obstet Invest 1990;29: 296 – 300. 8. Hillier SL, Kiviat NB, Hawes SE, Hasselquist MB, Wollner Hanssen P, Eschenback DA, et al. Role of bacterial vaginosis-associated microorganisms in endometritis. Am J Obstet Gynecol 1996;175: 435 – 441. 9. Ralph SG, Rutherford AJ, Wilson JD. Influence of bacterial vaginosis on conception and miscarriage in the first trimester: cohort study. BMJ 1999;319:220 – 223. 10. Hay PE, Morgan DJ, Ison CA, et al. A longitudinal study of bacterial vaginosis during pregnancy. Br J Obstet Gynaecol 1994;101: 1048 – 1053. 11. Liversedge NH, Turner A, Horner PJ, Keay SD, Jenkins JM, Hull MGR. The influence of bacterial vaginosis on in-vitro fertilisation and embryo implantation during assisted reproduction treatment. Hum Reprod 1999;14:2411 – 2415. 12. Wilson JD, Ralph SG, Jackson F, Rutherford AJ. Infertility, preterm birth and bacterial vaginosis. Lancet 1999;354:511. 13. Hellber D, Nilsson S, Mardh P-A. Bacterial vaginosis and smoking. Int J STD AIDS 2000;11:603 – 606. 14. Marks C, Tideman RL, Estcourt CS, Berry G, Mindel A. Assessment of risk for pelvic inflammatory disease in an urban sexual health population. Sex Transm Inf 2000;76:470 – 473.
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
BV RATES IN WOMEN UNDERGOING IVF FOR DIFFERENT TYPES OF INFERTILITY 15. Gaudoin M, Rekha P, Morris A, Lynch J, Acharya U. Bacterial vaginosis and past chlamydial infection are strongly and independently associated with tubal infertility but do not affect in vitro fertilisation success rates. Fertil Steril 1999;72:730 – 732. 16. Priestley CJF, Jones BM, Dhar J, Goodwin L. What is normal vaginal flora? Genitourin Med 1997;73:23 – 28. 17. Schwebke JR, Morgan SC, Weiss HL. The use of sequential self-
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
717
obtained vaginal smears for detecting changes in the vaginal flora. Sex Transm Dis 1997;24:236 – 239. 18. Keane FEA, Ison CA, Taylor-Robinson D. A longitudinal study of the vaginal flora over a menstrual cycle. Int J STD AIDS 1997;8: 489 – 494. Accepted 20 February 2002
Rates of bacterial vaginosis in women undergoing in vitro fertilisation for different types of infertility Janet D. Wilson*, Susan G. Ralph, Anthony J. Rutherford Objective To assess whether the rate of bacterial vaginosis (BV) is higher in women with tubal factor infertility compared with those with other causes of infertility. Design Cross-sectional study. Setting Assisted conception unit of a teaching hospital in Leeds. Population Consecutive women undergoing in vitro fertilisation. Methods Women undergoing in vitro fertilisation (IVF) had a vaginal smear taken at the time of their egg collection. The smear was Gram-stained and graded as normal, intermediate or BV. Main outcome measures The presence of bacterial vaginosis and the causes of infertility. Results A total of 749 women were included. The vaginal smears were normal in 63.6%, intermediate in 12.1%, and BV in 24.3%. The rates of BV in women with different types of infertility were 36.4% in tubal factor, 15.6% in male factor, 33.3% in anovulation, 12.5% in endometriosis and 18.9% in unexplained infertility. After controlling for the effects of age and smoking using a multivariate logistic regression model, women with tubal infertility were significantly more likely to have BV than women with endometriosis OR 3.63 (95% CI 1.52 – 8.67); male factor OR 2.98 (95% CI 1.80 –4.90); and unexplained infertility OR 2.20 (95% CI 1.35 –3.59). The adjusted figures for the increase of BV in women with anovulation were: endometriosis OR 3.77 (95% CI 1.28 –11.08); male factor OR 3.09 (95% CI 1.37– 6.96); and unexplained infertility OR 2.29 (95% CI 1.02 –5.12). Conclusions Women with tubal infertility were three times more likely to have BV than women with endometriosis, male factor or unexplained infertility. These findings support the association between BV, pelvic inflammatory disease (PID) and tubal damage but do not help distinguish between cause and effect. Women with anovulation were also three times more likely to have BV than women with endometriosis or male factor infertility, supporting suggestions of hormonal influence on vaginal flora. INTRODUCTION The main causes of pelvic inflammatory disease (PID), and hence tubal infertility, are Chlamydia trachomatis and Neisseria gonorrhoeae1. However, bacterial vaginosis (BV) is being increasingly implicated in upper genital tract infections in women. It can cause ascending infection in pregnant women, the chorioamnionitis, then predisposing to preterm delivery2. It also increases the rate of PID following surgical termination of pregnancy 3,4. Nonpregnant women with BV, but without gonorrhoea or chlamydia and without any symptoms or signs of upper genital tract inflammation, have been found to have plasma cell endometritis5. Symptoms of abdominal pain6 and intermenstrual bleeding7 are more common in women with BV than those with normal flora. BV is usually present in women with acute salpingitis, and BV-associated organisms are commonly isolated from the upper genital tract of
The General Infirmary at Leeds, Leeds, LS1 3EX, UK * Correspondence: Dr J. D. Wilson, The General Infirmary at Leeds, Leeds, LS1 3EX, UK. D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology PII: S 1 4 7 0 - 0 3 2 8 ( 0 2 ) 0 1 2 9 7 - 1
patients with PID. They are usually found as co-infections with C. trachomatis and N. gonorrhoeae but they have sometimes been isolated from the upper genital tract in the absence of these, with the suggestion that they increase the risk of upper genital tract infection in women with and without sexually transmitted infections8. There is therefore mounting evidence that BV may cause PID in non-pregnant women in the absence of gonorrhoea and chlamydia. Consequently, BV could be associated with tubal factor infertility. The aim of this study was to assess whether the rate of BV is higher in women with tubal factor infertility compared with those with other causes of infertility.
METHODS Women undergoing in vitro fertilisation (IVF) at the assisted conception unit of The General Infirmary at Leeds, during the period April 1996 to June 1997, were recruited. All participants gave written consent and ethical approval was obtained. Baseline demographic data, including age, smoking habit, ethnic origin, previous obstetric history and www.bjog-elsevier.com
BV RATES IN WOMEN UNDERGOING IVF FOR DIFFERENT TYPES OF INFERTILITY
current gynaecological history were included in standardised records. All patients recruited had been thoroughly investigated according to an established protocol, which included a baseline early follicular phase endocrine profile, mid-luteal progesterone, detailed semen analysis, and either a hysterosalpingogram or laparoscopy to assess tubal patency dependent on the patient’s history. Those patients with symptoms suggestive of pelvic pathology or a past history of established disease were investigated by laparoscopy, with the remainder having a hysterosalpingogram. If pelvic pathology was identified on the hysterosalpingogram, a laparoscopy was performed to confirm the extent of the disease. On completion of the investigations, the presumptive cause of the infertility was recorded. In cases where there was more than one pathology, the most significant condition was identified as the primary cause of infertility. During the participants’ IVF cycle, immediately prior to egg collection, a non-lubricated bivalve speculum was inserted and the vaginal walls and posterior fornix were sampled with a sterile cotton swab. This was smeared onto a glass slide, air-dried and then stained according to Gram’s method. The women were simultaneously screened for coexistent sexually transmitted infections. The full methodology and results of the IVF have been described elsewhere9. Two independent observers (JDW and SGR) read the slides. They were blinded to each other’s results and to the patient details. The criteria used to diagnose BV on the basis of Gram’s stain alone were the modified Spiegel’s criteria2. Thus, flora were graded as either normal (predominantly Lactobacilli), intermediate (reduced Lactobacilli mixed with other morphotypes) or BV (few or absent Lactobacilli with greatly increased numbers of Gardnerella vaginalis, other morphotypes or both). Where the observers disagreed, the slide was restained and reviewed until consensus was reached.
Statistical methods The rates of BV were calculated for the different types of infertility. Any differences between the BV rates were compared by m2 test. Associations between BV and age, ethnicity, and smoking were assessed by m2 test, as were associations between type of infertility, age and smoking. Finally, all the factors were included in a multiple logistic regression model to adjust for the other confounding variables.
RESULTS Full data are available on 749 women. The age range was 21– 44 (mean 33.1) years, 711 (94.9%) were white and 140 (18.7%) were smokers. No sexually transmitted infections were detected in the first 200 women enrolled so D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
715
Table 1. The reasons for infertility and rates of bacterial vaginosis (BV) with the different types of infertility. Cause of infertility Tubal Male factor Unexplained Endometriosis Anovulation
No. of women (%) 258 211 185 56 39
(34.4) (28.2) (24.7) (7.5) (5.2)
Rates of BV (%) 94/258 33/211 35/185 7/56 13/39
(36.4) (15.6) (18.9) (12.5) (33.3)
screening was thereafter abandoned. The results of the vaginal smears were normal in 476 (63.6%), intermediate in 91 (12.1%) and BV in 182 (24.3%). The kappa value of inter-observer variation was 0.94. The causes of infertility and the rates of BV with the different types of infertility are shown in Table 1. There was no significant difference in rates of BV between women with endometriosis, male factor infertility and those with unexplained infertility. However, women with tubal infertility were significantly more likely to have BV than women with all other causes OR 2.70 (95% CI 1.87– 3.9); endometriosis OR 3.98 (95% CI 1.62 –10.94); male factor OR 3.23 (95% CI 1.99 – 5.25); and unexplained infertility OR 2.49 (95% CI 1.54 – 4.04). In addition, women with anovulation were also significantly more likely to have BV than women with endometriosis OR 3.23 (95% CI 1.01 –10.59), and male factor infertility OR 2.62 (95% CI 1.12– 6.09). There was no significant difference in rates of BV between women with anovulation and tubal infertility. There was an association between age and BV, younger women were more likely to have BV (m2, P ¼ 0.009), and between smoking and BV, smokers were more likely to have BV (OR 3.41; 95% CI 2.22– 5.25). There was no significant association between ethnicity and vaginal flora, but there were small numbers of non-white women. There was an association between type of infertility and age, women with tubal infertility were younger (m2, P ¼ 0.03), and those with unexplained infertility were older (m2, P ¼ 0.002); and between type of infertility and smoking, women with tubal infertility were more likely to smoke (OR 3.35; 95% CI 2.26 –4.98). The significant differences in rates of BV with different types of infertility remained after adjusting for all the other confounding variables by multiple logistic regression. The adjusted figures for the increase of BV in women with tubal infertility were: adjusted OR 3.63 (95% CI 1.52– 8.67) for endometriosis, adjusted OR 2.98 (95% CI 1.80– 4.90) for male factor and adjusted OR 2.20 (95% CI 1.35 –3.59) for unexplained infertility. The adjusted figures for the increase of BV in women with anovulation were: adjusted OR 3.77 (95% CI 1.28– 11.08) for endometriosis, adjusted OR 3.09 (95% CI 1.37 – 6.96) for male factor and adjusted OR 2.29 (95% CI 1.02– 5.12) for unexplained infertility. The adjusted increased risk of BV in smokers was OR 2.82 (95% CI 1.81 – 4.40).
716
J.D. WILSON ET AL.
DISCUSSION The overall BV rate of 24.3% was much higher than the 11 – 15% reported from obstetric and gynaecology units within the UK2,10, but comparable to the rate of 25.6% found in a similar IVF population11. We had initially speculated that the high rate of BV could be due to hormonal manipulation during the IVF, however, we found no difference between the BV rate prior to IVF and at the egg collection12. There was great variation between the BV rates in women with different causes of infertility, with rates of 12 – 15% in those with endometriosis and male factor infertility and 33– 36% in those with anovulation and tubal infertility. Women with tubal factor infertility and anovulation were two to three times more likely to have BV than women with other types of infertility. There was an inter-relationship between younger age, tubal infertility, smoking and BV. Other groups have reported an association between age and BV2, and smoking and BV13. The main cause of tubal infertility is previous salpingitis. Smoking and young age have both been positively associated with PID14 . This inter-relationship is therefore not surprising, but importantly, the significantly raised rate of BV in women with tubal infertility remained after adjusting for these other confounding factors. Two other groups have reported this association between BV and tubal infertility, but due to the small number of women in these studies, comparisons could only be made between tubal infertility and all other causes11,15. From our study, it is obvious that not all women with infertility have the same rates of BV. A weakness of our study is that we do not have a control group of women from obstetric and gynaecology clinics, or with infertility but not undergoing IVF. However, it does appear that those with endometriosis, male factor and unexplained infertility have rates comparable with those published from UK obstetric and gynaecology populations2,10, whereas women with tubal infertility and anovulation are two to three times more likely to have BV. The significantly raised BV rate in women with anovulation is interesting. There is mounting evidence of a hormonal influence on vaginal flora. Three longitudinal studies of vaginal flora showed an association with changing flora and the menstrual cycle suggestive of hormonal influence16 – 18. Our finding of a significantly increased rate of BV in women with chronic abnormal oestrogen and progesterone levels warrants further investigation.
CONCLUSION The finding of a significantly raised rate of BV in women with tubal factor infertility supports the association
between BV, PID and tubal damage, but this cross sectional study does not help distinguish whether the increased BV rate is secondary to previous tubal damage initiated by C. trachomatis or N. gonorrhoeae, or whether BV-associated organisms compromise host immunity and facilitate spread of these sexually transmitted infections into the upper genital tract, or whether BV itself can cause damage to the fallopian tube.
Acknowledgements The authors would like to thank Wendy Parsons for her help with the statistical analysis.
References 1. Cates Jr W, Rolf Jr RT, Aral SO. Sexually transmitted disease and infertility: an epidemiologic update. Epidemiol Rev 1990;12: 199 – 220. 2. Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson J. Abnormal bacterial colonisation of the genital tract and subsequent preterm delivery and late miscarriage. BMJ 1994; 308:295 – 298. 3. Larsson P-G, Berman B, Forsum U, Platz-Christensen JJ, Pahlson C. Mobiluncus and clue cells as predictors of PID after first trimester abortion. Acta Obstet Gynecol Scand 1989;68:217 – 220. 4. Hamark B, Forssman L. Postabortal endometritis in Chlamydianegative women—association with preoperative clinical signs of infection. Gynecol Obstet Invest 1991;31:102 – 105. 5. Korn AP, Bolan G, Padian N, Ohm-Smith M, Schachter J, Landers DV. Plasma cell endometritis in women with symptomatic bacterial vaginosis. Obstet Gynecol 1995;85:387 – 390. 6. Eschenbach DA, Hillier S, Critchlow C, Stevens C, De Rouen T, Holmes KK. Diagnosis and clinical manifestations of bacterial vaginosis. Am J Obstet Gynecol 1988;158:819 – 828. 7. Larsson P-G, Bergman B, Forsum U, Pahlson C. Treatment of bacterial vaginosis in women with vaginal bleeding complications or discharge and harboring Mobiluncus. Gynecol Obstet Invest 1990;29: 296 – 300. 8. Hillier SL, Kiviat NB, Hawes SE, Hasselquist MB, Wollner Hanssen P, Eschenback DA, et al. Role of bacterial vaginosis-associated microorganisms in endometritis. Am J Obstet Gynecol 1996;175: 435 – 441. 9. Ralph SG, Rutherford AJ, Wilson JD. Influence of bacterial vaginosis on conception and miscarriage in the first trimester: cohort study. BMJ 1999;319:220 – 223. 10. Hay PE, Morgan DJ, Ison CA, et al. A longitudinal study of bacterial vaginosis during pregnancy. Br J Obstet Gynaecol 1994;101: 1048 – 1053. 11. Liversedge NH, Turner A, Horner PJ, Keay SD, Jenkins JM, Hull MGR. The influence of bacterial vaginosis on in-vitro fertilisation and embryo implantation during assisted reproduction treatment. Hum Reprod 1999;14:2411 – 2415. 12. Wilson JD, Ralph SG, Jackson F, Rutherford AJ. Infertility, preterm birth and bacterial vaginosis. Lancet 1999;354:511. 13. Hellber D, Nilsson S, Mardh P-A. Bacterial vaginosis and smoking. Int J STD AIDS 2000;11:603 – 606. 14. Marks C, Tideman RL, Estcourt CS, Berry G, Mindel A. Assessment of risk for pelvic inflammatory disease in an urban sexual health population. Sex Transm Inf 2000;76:470 – 473.
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
BV RATES IN WOMEN UNDERGOING IVF FOR DIFFERENT TYPES OF INFERTILITY 15. Gaudoin M, Rekha P, Morris A, Lynch J, Acharya U. Bacterial vaginosis and past chlamydial infection are strongly and independently associated with tubal infertility but do not affect in vitro fertilisation success rates. Fertil Steril 1999;72:730 – 732. 16. Priestley CJF, Jones BM, Dhar J, Goodwin L. What is normal vaginal flora? Genitourin Med 1997;73:23 – 28. 17. Schwebke JR, Morgan SC, Weiss HL. The use of sequential self-
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 714 – 717
717
obtained vaginal smears for detecting changes in the vaginal flora. Sex Transm Dis 1997;24:236 – 239. 18. Keane FEA, Ison CA, Taylor-Robinson D. A longitudinal study of the vaginal flora over a menstrual cycle. Int J STD AIDS 1997;8: 489 – 494. Accepted 20 February 2002