ABSTRACT Objective To study the influence of the position of the threads of an intrauterine contraceptive device (IUCD) on the development of genital tract infection. Design A multicentre randomized controlled trial. Subjects Women requesting an IUCD. Interventions The women were randomized to be fitted with an IUCD either with the threads contained in the uterine cavity (threads‐up group) ( n = 208 ) or passing through the cervix to the vagina in the usual way (threads‐down group) ( n = 237 ). Multiple centre study with follow‐up at three months, 1 and 2 years. At the final visit ‘missing’ threads were retrieved using a disposable instrument (Retrievette R ). Main outcome measures The occurrence of infection in the lower or upper genital tract. Results 63 women in the threads‐up group and 78 in the threads‐down groups dropped out. Previous gynaecological infection was reported by 21 and 48 women in the threads‐up and threads‐down group, respectively (odds ratio 0.44, 95% CI 0.24 to 0.79), 21 and 53 subjects had signs of infection at gynaecological examination (odds ratio 0.39,95% CI 0.21 to 0.69) and a wet‐smear was pathological in 33 and 79 (odds ratio 0.38, 95% CI 0.23 to 0.61). In the threads‐up group the vaginal pH was also lower at the final check up after 2 years. Spontaneous descent of the threads occurred in 11% of the threads‐up group and in six women in the threads‐down group the threads were in the cervix. In 93 women the threads were easily retrieved by means of the Retrievette R , four women insisted on the threads remaining in the uterus and in 18 thread removal was performed under local or general anaesthesia. Conclusions Infectious complications in women using an IUCD are more frequent if the threads lead from the uterine cavity to the vagina. This problem can be reduced by inserting the threads so that they remain entirely within the uterine cavity, a feasible procedure now that an effective instrument for IUCD thread retrieval is available.

The role of the threads of an intrauterine contraceptive device (IUD) in the development of genital infection was studied in a total of 380 women at family planning clinics. Symptoms and signs of genital tract infections were recorded before and 3 months after insertion of a Multiload CU 250 IUD. In randomized order 1 group of women had the threads inserted into the uterine cavity together with the device and 1 group was fitted the usual way with threads in the vagina. 8 drop outs from the study for medical or social reasons occurred in the former and 11 in the latter group leaving 175 and 186 women in the 2 respective groups for comparison. No differences were observed between the groups before IUD insertion whereas after 3 months the relation between the threads up and threads down groups were as follows: history of genital tract infection or other gynecological disease 17/32 (p=0.045) signs of infection at gynecological examination 4/12 (p=0.048) pathological wet smear of vaginal content 11/27 (p=0.008). The pH of the vaginal content showed no difference. An ultrasonic examination confirmed the intrauterine position of the IUD in all but one of the women with threads not visible at the 3 months check-up. These results suggest that infectious complications due to the use of IUD could be reduced by inserting the threads together with the IUD a feasible procedure since an effective disposable instrument is available for retrieval of the threads before IUD removal. (Authors).

Summary. In a double blind controlled study of the inhibition of lactation 13 women received 300 mg of cyclofenil and 11 women 2.5 mg of bromocriptine twice daily for 14 days. Lactation was effectively inhibited by both drugs, but with bromocriptine there was a significantly higher frequency of relapse. The plasma concentration of prolactin, which decreased rapidly with bromocriptine, returned to the pretreatment level the day after drug treatment stopped, but with cyclofenil it remained low. Plasma oestradiol followed a similar pattern. Plasma FSH increased more rapidly with bromocriptine than with cyclofenil. There was no significant difference between the treatment groups at any stage for haematology, coagulation or liver function tests. The more sustained effect of cyclofenil on prolactin secretion with a reduced frequency of relapse, and the lower oestradiol level, which might indicate a reduced risk of thromboembolism, suggest that this drug has some advantage over bromocriptine in the inhibition of postpartum lactation.