Objective To assess the relative efficacy of three disposable plastic instruments in the retrieval of ‘missing’ IUCD threads.

Design A prospective randomised comparative single centre study.

Setting Family Planning Clinic in London, UK.

Subjects 217 of 350 IUCD users referred to the research team with ‘missing’ IUCD threads entered the study.

Intervention All women initially underwent exploration of the endocervical canal with Spencer Wells forceps. When this procedure did not retrieve the threads, the patients were entered into the study. A maximum of two randomly chosen plastic IUCD thread retrievers were then used in any one patient to explore the uterine cavity and capture the ‘missing’ threads. The order in which the two retrievers were employed was also determined at random. Four separate entries into the uterine cavity were permitted with each instrument, the endocervical canal being explored with Spencer Wells forceps after each retraction of the instrument to identify the possible descent of the threads.

Main outcome measures Threads brought down beyond the external cervical os, or threads brought to within the endocervical canal and then grasped by Spencer Wells forceps.

Results In approximately 40% of all patients, the threads were retrieved with Spencer Wells forceps alone and a further 40% with the disposable plastic retrievers. About 5% had no retrievable threads, and only 2.5% of the referred patients required general anaesthesia for removal of their IUCD. The analysis of the comparative trial was confined to the 197 patients with retrievable threads which could not be brought below the external os with Spencer Wells forceps. The first plastic retriever used was successful in 50% of patients. The Retrievette (59%) and the Emmett (53%) performed better than the Mi-Mark Helix (37%) in this study. The difference was statistically significant (P=0.03) and the 95% confidence interval for the difference of the Mi-Mark Helix from the other two retrievers was 4% to 33%. This retrieval rate for the Mi-Mark Helix was much worse than in previously reported studies, though one doctor did have a better success rate with this retriever. The success rates, using a second plastic retriever randomly chosen from the two not used in the first attempt, were almost identical to those observed with the first retrievers: 63%, 56% and 36%. The success rate did not appear to be influenced by the length of thread, day of cycle, device type or parity. The success of the second retriever tried did not seem to be influenced by the retriever that had failed previously.

Conclusions Based on our experience, the initial exploration of the endocervical canal with Spencer Wells forceps is invaluable. If this simple manoeuvre fails to retrieve the ‘missing’ threads, either the Retrievette or the Emmett thread retrievers are useful tools in general practice or in the family planning clinic setting.