We thank Mr Hutchon for his interest in our work. Like him we are trying to reduce the morbidity of women undergoing surgery for dysfunctional uterine bleeding and we agree that for many women the ideal operation is a vaginal hysterectomy. It is clear, however, that the majority of women who undergo surgery still have abdominal hysterectomies. It is important to address the reasons for this. The majority of us select women for vaginal hysterectomy on a number of criteria, one of which is uterine size. Clearly uteri can reach a considerable size even though individual fibroids have not been clinically detected, and this can be a major factor in influencing the proportion of women undergoing vaginal hysterectomy. If we are to produce a shift in practice from abdominal to vaginal surgery, then a drug that would make surgery easier for those less confident in conducting vaginal surgery would be of great benefit. Although our study did not set out to detect this, we would expect that a reduction in uterine volume by one third would influence many to conduct a vaginal hysterectomy on a uterus of ‘borderline size’. A study with greater power than ours is needed to address this.

Secondly, Mr Hutchon appears to believe that the only way to reduce operative morbidity is to train gynaecologists so that all women undergoing surgery for dysfunctional uterine bleeding have vaginal hysterectomy. However, the reality of current practice is that for many gynaecologists an abdominal hysterectomy is the “default” operation. If we had made a principled stand and refused to conduct any research which aimed to reduce abdominal hysterectomy morbidity then the trial would have excluded large numbers of women who undergo this operation.

Finally we agree that vaginal hysterectomy appears to have a lower morbidity than abdominal hysterectomy. Trials suggest, however, that the true difference is not as great as it may seem. Randomised trials comparing abdominal hysterectomy and laparoscopically assisted vaginal hysterectomy show that the mean recovery time, even in the hands of enthusiastic vaginal surgeons, was just 10–20 days shorter in the laparoscopically assisted vaginal hysterectomy groups1,2. It may be that the introduction of novel techniques to reduce abdominal hysterectomy morbidity will narrow this gap further. If so, then an abdominal hysterectomy could become the favoured operation, especially as the 5%–10% whose operations which start vaginally and end with laparotomy appear to suffer the highest morbidity of all.