Screening for chlamydia related bacteria to prevent miscarriage. Proof of principle study

Lead Research Organisation: St George's University of London
Department Name: Community Health Sciences


One in six women will suffer a miscarriage during their lifetime. Some women have repeated miscarriages which can be devastating both emotionally and physically. Caring for women with miscarriage is also very expensive and estimated to cost the NHS over £250 million per year in England and Wales.

Some miscarriages may be due to vaginal or genitourinary infections spreading up into the womb. New research suggests that infection with recently discovered chlamydia related bacteria which cause cows to miscarry might do the same in humans. But we do not know how common these infections are in healthy pregnant women, nor whether women who are infected really are more likely to suffer a miscarriage.

Thirteen years ago, 1200 newly pregnant women attending general practices and family planning clinics in south London agreed to help in our research looking at infections and miscarriage. They provided self-taken vaginal samples and urine specimens when they were less than 10 weeks pregnant, and completed questionnaires both at 16 weeks and after the birth. The questionnaires asked about miscarriages or the baby being born early - before 37 weeks of pregnancy. We found that a vaginal infection called bacterial vaginosis was associated with late miscarriage at 13-15 weeks. This important finding was published in the British Medical Journal. As a result some women with bacterial vaginosis may now be treated with antibiotics to try to prevent them from miscarrying. We now want to see if the same may be true for chlamydia related bacteria.

We still have stored frozen urine samples provided by 915 of the women. We have checked a few of the samples to make sure they are adequately preserved and that we can detect chlamydia related bacteria. We now want to test all the samples to see if women infected with chlamydia related bacteria were more likely to suffer a miscarriage before 16 weeks than those who weren't infected. The study will also provide important information on how common these infections are during pregnancy, which women are more likely to be infected, and whether the infections may also be associated with babies being born too early.

The strengths of the study are that it is based in the community rather than in hospital, the women were only just pregnant-average 7 weeks, over 99% of the women followed up at 16 weeks, and we will be using exciting new tests. As we already have the samples and have shown that the tests work well, the study is also relatively low cost and low risk.

If we find that chlamydia related bacteria are associated with miscarriage, further research might show whether testing and treating women for these infections may prevent some miscarriages. Ultimately this might lead to changes in clinical practice both nationally and internationally, benefit the health of women and their babies, and save healthcare costs.

Technical Summary

Chlamydiales species in the genitourinary tract may be associated with miscarriage, but no prospective studies have been done and we do not know how common these organisms are in healthy pregnant women.We have 915 stored urine samples from a carefully characterised cohort of pregnant women (mean 7 weeks gestation) recruited in 1998-2000 to a prospective study to investigate the association of bacterial vaginosis or chlamydial infection with miscarriage before 16 weeks. Follow up was 99.8% with a miscarriage rate of 10%. Unfunded preparatory work in 2011-12 confirmed bacterial DNA integrity in stored urines and showed Chlamydiales species including Waddlia chondrophila were detectable.

Objectives: To find the prevalence of Chlamydiales species in first void urine samples before 10 weeks gestation, and to see if infection is associated with miscarriage before 16 weeks.

Design: Community-based cohort study

Setting: 37 general practices and five family planning clinics in south London

Participants: 915 pregnant women, mean age 31 years (range 16 to 46), presenting in the community at less than 10 weeks gestation. 22% were from ethnic minorities; and 15% had bacterial vaginosis, 2% C.trachomatis and 1% M.genitalium at baseline. Of these women, 92 miscarried before 16 weeks: 41 at <10 weeks, 40 at 10-12 weeks, and 11 at 13-15 weeks. Thirty nine women (4%) had a preterm birth before 37 weeks gestation.

Methods: In 2013-14, samples stored at Statens Serum Institut, Denmark, will be tested using validated Chlamydiales and W.chondrophila specific real-time PCRs.

Main outcome measures:
Prevalence and predictors of Chlamydiales in 915 women
Relative risk of miscarriage before 16 weeks in 894 women with and without Chlamydiales infection at <10 weeks gestation.

Planned Impact

Patients and women
For some women, miscarriages may be devastating; and a third of women attending specialist clinics may be depressed. The proposed study will raise public awareness that some miscarriages may be due to genitourinary infections. If Chlamydiales are found to be a cause of miscarriage, this could have public health implications. Depending on the results of possible future trials, women with recurrent miscarriage might be advised to have screening and treatment for Chlamydiales and other potentially pathogenic vaginal infections as well as bacterial vaginosis.They might also be advised to avoid unnecessary contact with animals. Changes in clinical practice which reduce the rate of miscarriage, particularly recurrent miscarriage, would greatly benefit the health of many women and their babies.

Becton, Dickinson and Co is an industrial partner of the eSTI2 consortium and is interested in tests for Chlamydiales species. If screening and treatment for Chlamydiales is found to be beneficial, this could lead to the development of reliable and cheap commercial diagnostic tests.

Academic benefits
The study will increase our understanding ofthe role, prevalence and predictors of genitourinary Chlamydiales in pregnancy.The proposal is relatively low risk and low cost as the laboratory samples are already collected and have been shown to be suitable for testing. Our unfunded preliminary work found positive results using both the Chlamydiales and the W.chondrophila assay. The study will show if the organisms are likely to be associated with early or late miscarriage or preterm birth and how common they are in normal pregnancy. This could lead to a trial of screening and treatment. Findings should have wide applicability and thus be of importance both nationally and internationally.

Potential benefits to the NHS
Miscarriage and preterm birth are associated with considerable morbidity and costs. If future research suggests that screening and treating some women with Chlamydiales reduces the risks of adverse pregnancy outcome, this could ultimately save NHS costs.

Voluntary organisations and charities
Patient and public involvement is crucial in understanding people's priorities and helping to explain research findings. We have links with Dr Mary Newburn, head of R and D at the National Childbirth Trust and Ruth Bender Atik, National Director of the Miscarriage Association. They are supportive and will publicise the study and results among their membership. This will facilitate public engagement with research.


10 25 50
publication icon
Reid F (2017) Chlamydia related bacteria (Chlamydiales) in early pregnancy: community-based cohort study. in Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases

Description nihr rfpb
Amount £245,000 (GBP)
Funding ID PB-PG-1014-35007 
Organisation National Institute for Health Research 
Department Research for Patient Benefit
Sector Public
Country United Kingdom
Start 08/2016 
End 08/2018