Objectives Chlamydia trachomatis, which is asymptomatic in most women, causes significant adverse effects for pregnant women and neonates. No programmes conduct antenatal screening in Latin America. We determined chlamydia prevalence, feasibility and acceptability of chlamydia screening, and adherence to treatment in pregnant women in two urban public hospitals in Lima, Peru.
Methods We offered chlamydia screening using self-collected vaginal swabs to pregnant women ≥16 years of age during their first antenatal visit. Chlamydia-infected women were contacted within 14 days and asked to bring partners for counselling and directly observed therapy with oral azithromycin. Unaccompanied women received counselling, directly observed therapy, and azithromycin to take to partners. Test of cure was performed ≥3 weeks after treatment.
Results We approached 640 women for the study and enrolled 600 (93.8%). Median age was 27.3 years (range 16–47), median lifetime partners 2.3 (range 1–50), and median gestational age 26.1 weeks (range 4–41). Chlamydia prevalence was 10% (95% CI 7.7% to 12.7%). Of 60 infected patients, 59 (98%) were treated with one dose of azithromycin. Fifty-two of 59 (88%) returned for test of cure, all of whom were treated successfully, with 46 (86%) achieving negative test of cure with one dose of azithromycin, and 6 (12%) after retreatment with a second dose.
Conclusions C. trachomatis screening and treatment in pregnancy was feasible and highly acceptable in two urban hospitals in Peru. Chlamydia prevalence was high. Clinical trials to evaluate efficacy and cost-effectiveness of chlamydia screening, and treatment of pregnant women to prevent adverse pregnancy outcomes in low-resource settings, are warranted.
- Chlamydia Trachomatis
Statistics from Altmetric.com
Chlamydia trachomatis, the most common sexually transmitted bacterial infection worldwide, can cause significant adverse outcomes in pregnancy, including preterm birth, low birth weight, premature rupture of membranes, stillbirth, and miscarriage, as well as inclusion conjunctivitis and pneumonia in neonates.1 No programmes routinely conduct C. trachomatis screening in antenatal care in Latin America, and there are no WHO recommendations for routine C. trachomatis screening and treatment in pregnant women. To prepare for a trial of C. trachomatis screening and treatment in pregnancy to reduce adverse pregnancy outcomes, we explored the feasibility and acceptability of C. trachomatis screening in pregnant women during the first antenatal visit, and determined C. trachomatis prevalence and patient and partner outcomes to treatment in Lima, Peru.
We conducted a prospective study in two large urban hospitals in Lima, Peru: Instituto Nacional Materno Perinatal (INMP) and Hospital Nacional Arzobispo Loayza (HNAL). INMP participants were recruited in January 2013; HNAL participants were recruited December 2012–January 2013.
During the recruitment period, all pregnant women attending their first antenatal visit were given a brief explanation by hospital midwives about risks of chlamydia infection during pregnancy and were told about the study. We focused on the first antenatal visit since women routinely have antenatal counselling and HIV/syphilis screening at this time. Women ≥16 years old who were interested in participating were screened for eligibility by research midwives and enrolled after providing informed consent. Consecutive women were recruited at the HNAL. Even-numbered women were recruited at the INMP (odd-numbered patients were recruited to another concurrently running research study). Women not mentally competent to understand informed consent were excluded; minors were required to have consent from parent or guardian to participate. The study protocol was approved by the institutional review or ethical boards at the University of California, Los Angeles Universidad Peruana Cayetano Heredia, and each participating hospital.
Women provided self-collected vaginal swabs for chlamydial testing after being instructed on collection technique by the study midwife, and then completed a face-to-face questionnaire regarding demographic data, reproductive and medical history, and number of sexual partners.
Women who tested positive for chlamydial infection were asked to return to the hospital for counseling, and directly observed treatment with 1 g of oral azithromycin. They were given the option of bringing their partner(s) with them for counselling and directly observed concurrent treatment at the hospital, or of delivering 1 g azithromycin to the partner at home. Approximately 3 weeks after treatment, infected women were contacted to provide a second self-collected vaginal specimen to perform a test of cure to document clearance of infection.
Testing was free, and we reimbursed women their transportation costs to return for treatment and for test of cure.
Specimens were tested for C. trachomatis infection using the Aptima Combo2 system (Hologic, Gen Probe Incorporated, San Diego, California, USA) at the Universidad Peruana Cayetano Heredia Laboratory of Sexual Health in Lima, Peru.
Data management and statistical analysis
Screening acceptability and C. trachomatis prevalence were calculated with 95% CIs. To test the association between categorical variables and C. trachomatis positivity, we used either χ2 or Fisher's exact tests. All other numerical variables were assessed using the Mann–Whitney test. Individuals with missing data were excluded only from the affected analysis. We conducted all analysis using Stata V.12.1 (Stata Corporation, College Station, Texas, USA).
Of 640 pregnant women during the recruitment period who were informed about the study, three were excluded (due to high-risk pregnancy, unaccompanied minor status, or lack of intention to return to hospital). Of the remaining 637 eligible women, 600 (93.8%) enrolled: 333 (55.5%) from INMP and 267 (44.5%) from HNAL. The most common reasons given for not participating were lack of time (n=15), fear of being tested (n= 7), and not considering the study important (n=7). Five women did not give any reason, and three wanted to consult with family/friends before enrolling but never enrolled.
Table 1 shows participant characteristics.
C. trachomatis prevalence
C. trachomatis was identified in 60 study participants (10.0%; 95% CI 7.7% to 12.7%). Prevalence decreased with age; the youngest women (16–23 years) had the highest prevalence (15.6%), and older women (≥31 years) had the lowest (5.2%). Prevalence was higher for single women than for women who were married or cohabiting, but was unrelated to lifetime number of sex partners, education level, or current vaginal symptoms.
Of the 60 C. trachomatis-positive patients, 59 (98.3%) received treatment. Fifty-five of 59 partners (93%) received treatment, 21 of them (36%) at the hospital, concurrently with the women, and 34 (58%) with medication brought home by the women.
Fifty-two (91%) treated women returned for test of cure. Forty-six tested negative (infection cured). Of the six who tested positive, indicating continuing infection, three had received concurrent therapy with their partners, two had brought treatment home, and one denied partner contact after treatment. All six were retreated, and subsequent tests of cure were negative.
Chlamydial screening in pregnant women at two large urban hospitals in Lima was feasible and highly acceptable. All women who tested positive for chlamydia and returned for treatment and test of cure were successfully treated.
Our data regarding prevalence were consistent with previous research in Peru and globally, showing that the youngest women are most likely to be infected.1 ,2 It is worth noting that in our study, prevalence was also high (5.1%) among women ≥31 years, an age category not generally included in screening programmes.
The participation rate for screening with self-administered vaginal swabs was high (93.8%), which is consistent with previous studies in high-income countries.3 Use of self-administered vaginal swabs for nucleic acid amplification testing is a non-invasive diagnostic method that has sensitivity and specificity equivalent to provider-collected samples,4 an advantage in resource-limited settings where there may be shortages of health personnel. Additionally, vaginal swabs are easier to transport, and are equally or more sensitive and specific for diagnosis of chlamydial infection than urine samples.4
One potential disadvantage of such molecular-based testing is the absence of laboratory testing capacity in low and middle-income settings. However, with the advent of HIV/AIDS RNA testing and the widespread introduction of molecular testing for tuberculosis in low-income and middle-income settings, that capacity is rapidly increasing.5 Since results are not available at point-of-care, another drawback of such testing is the potential loss to follow-up, but as yet there exists no point-of-care testing with adequate sensitivity and specificity for chlamydia screening.6
C. trachomatis positivity at test of cure was 12%, similar to levels in previous studies of recurrent or persistent infections in women treated for genital chlamydia.7 No significant resistance of C. trachomatis to azithromycin has been reported in the literature, but pharmacologic treatment failure, defined as persistent infection despite antibiotic use, may result from variations in drug absorption, metabolism, or host immune response.8 False positive results in the test of cure may occur from persistence of residual DNA from non-viable chlamydia.3 To avoid this problem, current treatment guidelines recommend waiting at least 3 weeks before repeating nucleic acid amplification testing,3 although residual DNA may persist for longer periods.9
Although women were encouraged to bring partners to the hospital for treatment and counselling, nearly 60% chose to bring medication home to partners. This practice, known as patient-delivered partner therapy, is recommended by the US Centers for Disease Control as an alternative therapy for certain sexually transmitted diseases. This is an important consideration, since several studies in developing countries suggest that reliance on patient referral of partners for therapy is often ineffective.10 More research is needed on the use of patient-delivered partner therapy for partner treatment in low-resource settings.
Our study had several limitations. Women were recruited only from large public hospitals in Lima, so results might not be generalisable to other settings, such as rural areas and mid-sized cities. The educational level in our sample was somewhat higher than average for metropolitan Lima, and since we have no demographic data on the women who chose not to participate, we cannot rule out the possibility that there may have been a selection bias such that women who are more educated were more likely to participate in the study. Despite these limitations, however, we believe that because our study was carried out in two national hospitals with large antenatal services, and because most women in Lima give birth in hospitals, our prevalence estimates, treatment acceptability, and risk factors are similar to those for the larger population of low-risk pregnant women in Lima.
C. trachomatis screening and treatment in pregnancy was feasible and acceptable in two large urban maternity hospitals in Lima, Peru. Partner treatment was also readily accepted. The prevalence of C. trachomatis infection was high. Given the strong associations between C. trachomatis in pregnancy and adverse pregnancy outcomes, a clinical trial to demonstrate the efficacy and cost-effectiveness of C. trachomatis screening and treatment in low-income and middle-income countries is urgently needed.
We are grateful to Hologic Genprobe, San Diego, CA for donation of the Aptima Combo 2 assay kits used in the study. For their contribution to the study, we are indebted to research midwives Paola Pfluker, Carla Obregón, Vivian Mendoza, and Verónica Dioses and counsellor midwives Nancy Acosta and Soledad Rodriguez at the Instituto Nacional Materno Perinatal and Bertha Zavaleta at Hospital Nacional Arzobispo Loayza, as well as the staff at both hospitals and at Universidad Peruana Cayetano Heredia. Jeanne Cabeza was a research fellow in the South American Program in HIV Prevention Research (SAPHIR- funded by NIH R 25 MHO87222) through the programme in Global Health at the David Geffen School of Medicine, UCLA, Los Angeles, CA and gratefully acknowledges the support she received from the staff in both Lima and Los Angeles. Finally, we thank the patients themselves for their participation, as without them, this study would not have been possible.
Abstract in Spanish
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Abstract in Spanish - Online abstract
Handling editor Jackie A Cassell
Contributors JC had full access to all the data in the study and takes responsibility for the integrity of the data analysis. Study concept and design: JC, PJG, JDK. Study supervision: JC, PJG, PG, FE, JDK. Critical revision of the manuscript for important intellectual content: JC, PJG, JDK, PG, FE. Drafting of the manuscript: JC, PJG, JDK. Administrative technical or material support: ES, SLR, SL. Statistical analysis ES.
Funding This research was supported by a grant (NIH R 25 MHO87222) from the National Institute of Mental Health.
Competing interests None.
Ethics approval UCLA (IRB #12-000914), UPCH (#60048), HNAL (105 CIEI), and INMP (DG No 959).
Provenance and peer review Not commissioned; externally peer reviewed.