British Journal of Obstetrics and Gynaecology zyxwvut June 1998, Vol. 105, pp. zyxwvuts 599-604 A randomised comparison of strategies for reducing infective complications of induced abortion *GiUian C. Penney zyxwvutsr Clinical Research Fellow, **Margaret Thomson Consultant Gynaecologist, tJane Norman Senior Lecturer, TTHamish McKenzie Senior Lecturer, #Luke Vale Research Assistant, **Robert Smith Consultant Gynaecologist, **Moira Imrie Research Nurse *Department zyxwvutsr of Obstetrics and Gynaecology, Aberdeen Maternity Hospital; **Ninewells Hospital, Dundee, ?Department of Obstetrics and Gynaecology, Royal Infirmary, Glasgow; zyxwvuts f t Department of Medical Microbiology and §the Health Economics Research Unit, Medical School, Foresterhill, Aberdeen Objectives To determine lower genital tract carriage rates of C. trachomatis, A! gonorrhoeae and bacterial vaginosis among women seeking termination of pregnancy. To compare two clinical management strategies for minimising the risks of infective morbidity after induced abortion. Design Prevalence of infections was assessed by screening women undergoing abortion. Clinical management strategies were compared by a randomised trial. Setting The gynaecology departments of four hospitals in Scotland. Participants 1672 women undergoing induced abortion. Interventions Women randomised to prophylaxis received metronidazole 1 g rectally before abortion plus doxycycline 100mg twice daily for seven days. Women randomised to screen-and- treat received appropriate antibiotics only if screening proved positive for one or more infection. Main outcome measures Prevalences of infections; morbidity in the eight weeks following abortion as assessed by reported symptoms, general practitioner consultation and prescription rates and hospital re-attendances;costs to the NHS of alternative managements. Results Prevalence rates: C. trachomatis 5.6%; A! gonorrhoeae 0.19%; bacterial vaginosis 17.5%. Overall, women allocated to receive prophylaxis had lower rates of measures of short term infective morbidity than those allocated to screen-and-treat. These differences only reached statistical significance for women who were reported negative on screening. The direct costs to the NHS of prophylaxis and screen-and-treat were calculated to be E8.17 and f18.34 per woman, respectively. Conclusions Prevalences of lower genital tract infections which have been implicated in increased rates of infective morbidity after abortion are similar to those reported elsewhere. Universal antibiotic prophylaxis is at least as effective as a policy of screen-and-treat in minimising the risk of short term infective morbidity and is far more cost efficient. INTRODUCTION Induced abortion is one of the most commonly performed gynaecological procedures, with around 154,000 terminations performed annually in England and Wales (based on data from 1995). Post-abortion pelvic inflammatory disease is a recognised compli- cation of the procedure and is associated with short term morbidity and long term sequelae in the form of tuba1 infertility and ectopic pregnancy. Reported Correspondence: Dr G. C. Penney, Department of Obstetrics and Gynaecology, Aberdeen Maternity Hospital, Cornhill Road, Aberdeen AB25 2ZL. rates of post-abortion pelvic inflammatory disease (variously defined) range from 5% to 10%'. Risk fac- tors for pos t-abortion pelvic inflammatory disease include the presence of infection in the lower genital tract at the time of abortion. Three infections have been particularly implicated: N gonorrhoeae*, C. trachomati,?' and bacterial vagino~is~.~. The first aim of the present study was to determine the preva- lence of these three lower genital tract infections among women seeking abortion in Scotland. We also compared two clinical management strategies, uni- versal antibiotic prophylaxis and a screen-and-treat policy, each of which has been shown to reduce the z 0 RCOG 1998 British Journal of Obstetrics and Gynaecology zyxwvu 599