ORIGINAL STUDY Screening for Chlamydia trachomatis Using Self-Collected Vaginal Swabs at a Public Pregnancy Termination Clinic in France: Results of a Screen-and-Treat Policy Vincent Lavoue ´, MD,*† Laurent Vandenbroucke,*† Sophie Lorand, MD,*† Patrick Pincemin,* Estelle Bauville, MD,† Laurence Boyer, MD,† Daniel Martin-Meriadec, MD,† Jacques Minet, PhD,‡ Patrice Poulain, MD,*† and Karine Morcel, MD, PhD*† Objective: To assess the prevalence of Chlamydia trachomatis (CT) infection and the risk factors for CT infection among women presenting for abortion at a clinic in France. Methods: Women seeking surgically induced abortions were sys- tematically screened by PCR on self-collected vaginal swabs between January 1, 2010, and September 30, 2010. CT-positive women were treated with oral azithromycin (1 g) before the surgical procedure. Results: Of the 978 women included in the study, 66 were CT positive. The prevalence was 6.7% (95% confidence intervals [CI] 5.1%– 8.3%). The risk factors for CT infection were the following: age 30 years (Odds ratio [OR]: 2.0 [95% CI: 1.2–3.5]), a relationship status of single (OR: 2.2 [95% CI: 1.2– 4.0]), having 0 or 1 child (OR: 5.2 [95% CI: 2.0 –13.0]), not using contraception (OR: 2.4 [95% CI: 1.4 – 4.1]), and completing 11 weeks or more of gestation (OR: 2.1 [95% CI: 1.3–3.6]). Multiple logistic regression indicated that 4 fac- tors— having 0 or 1 child, a single relationship status, no contraceptive use, and a gestation of 11 weeks or more—were independently asso- ciated with CT infection. The rate of postabortion infection among all patients was 0.4% (4/978). Conclusions: These results reveal a high prevalence (6.7%) of CT-positive patients among French women seeking induced abortions. Because it is not common practice to screen the general population for CT, screening before induced abortions seems relevant. A cost-effec- tiveness study is required to evaluate this screen-and-treat policy. I nduced abortion is one of the most common surgical proce- dures worldwide, with approximately 189,000 procedures per year in England and Wales, 1 1 million in the United States, 2 and 220,000 in France. 3 Postabortion infections occur in 0.6% to 4.7% of cases. 4–6 Some studies gave wide range of post- abortion infections largely because of inconsistent definitions of postabortal infection. 6 Postabortion infection can cause, when untreated, long-term serious sequelae, including tubal infertility, chronic pain, and ectopic pregnancy. 7 When a single organism is identified in postabortal pel- vic inflammatory disease (PID), it is most often Chlamydia trachomatis (CT). Studies show that as many as 20% to 63% of women who had CT-positive test results at the time of their surgically induced abortions developed PID, compared with only 1.6% of CT-negative women. 8,9 Early first-trimester med- ical abortion is a noninvasive procedure with a low infection rate (approximately 0.3%), requiring no systematic antibiotic prophylaxis. 6,10,11 Appropriate preoperative antibiotic prophylaxis is likely to reduce the incidence of postabortion PID to 1%. 6,12,13 To prevent postabortion infection, both a screen-and-treat policy and universal antibiotic prophylaxis are in use. Recently, the Society of Family Planning recommended the routine use of antibiotic prophylaxis before surgical abortion. 6 Universal an- tibiotic prophylaxis is more cost effective than a screen-and- treat strategy, 14,15 but the benefits and costs of each strategy (universal antibiotic prophylaxis or screen-and-treat policy) depend on the prevalence of CT among those seeking induced abortions. For universal antibiotic prophylaxis, the more the prev- alence rate decreases, the more the infectious risk decreases, and the more the number of women who need to receive antibi- otics increases, while the risks of side effects and adverse reactions from the antibiotics persist. 6 However, the addi- tional cost of a screen-and-treat strategy, compared with the universal antibiotic prophylaxis, is mainly because of the cost of screening tests to detect CT. Thus, a risk-based algorithm may be a more cost-effective alternative. In France, the CT prevalence in people aged 18 to 44 years was estimated at 1.6% (95% confidence interval [CI]: 1.0%– 2.5%) for women, 16 but fewer data on the prevalence of CT are available for the population of women requesting in- duced abortion. Thus, the aim of this study was to determine the preva- lence of CT infection among women who underwent induced abortions in France and to identify the factors associated with a positive CT test so as to identify those for whom the screen- ing test will be most relevant. MATERIALS AND METHODS Setting We conducted a retrospective cohort study on patients who underwent an elective first-trimester surgical abortion. From the *Faculte ´ de Me ´decine, Universite ´ de Rennes 1, Rennes, France; †De ´partement de Gyne ´cologie et d’Obste ´trique, CHU Rennes, Rennes, France; and ‡De ´partement de microbiologie, CHU Rennes, Rennes, France The authors thank American Journal Experts for editing the manuscript. V.L. and L.V. have both contributed equally to his work. Correspondence: Vincent Lavoue ´, MD, Service de gyne ´cologie et d’obste ´trique, CHU Ho ˆpital sud, 16 Bd de Bulgarie, BP 90347, F-35 203 Rennes Cedex 2, France. E-mail: vincent.lavoue@chu-rennes.fr. Received for publication November 3, 2011, and accepted March 8, 2012. DOI: 10.1097/OLQ.0b013e318254ca6f Copyright © 2012 American Sexually Transmitted Diseases Association All rights reserved. 622 Sexually Transmitted Diseases ● Volume 39, Number 8, August 2012