SEARCH

SEARCH BY CITATION

Author’s Reply

  1. Top of page
  2. Author’s Reply
  3. References

Sir,

We thank Burch and Avasarala1 for their interest in our study.2 Regarding the selection of trial participants, routine gynaecological check-ups in symptom-free women are common in many countries, including Norway. If an ultrasound examination raises suspicion of endometrial pathology, women are usually referred for an outpatient consultation at our hospital. At this consultation, saline intrauterine sonography is performed to visualise polyps or fibroids in the uterine cavity. Women with asymptomatic endometrial polyps are usually treated by hysteroscopic resection. None of the women in our trial had previously undergone hysteroscopy, but 29/68 (43%) of the premenopausal women and 8/23 (35%) of the postmenopausal women had at some point during their lives undergone a dilatation and curettage. We do not think that this will have influenced the effect of misoprostol on cervical ripening. The trial was not a random population sample as these women were referred for a particular reason. However, all pre- and postmenopausal women referred to our hospital were eligible for trial participation. Whether all women with symptom-free polyps should be offered hysteroscopy is controversial, and there are currently continuing trials at our hospital to determine the spontaneous regression rate and the risk of malignancy in symptom-free endometrial polyps.

Regarding the outcome measures, the use of preoperative cervical dilatation as a primary outcome for assessing cervical ripening is open for discussion. However, it is a very widely used method in randomised trials.3,4 Using outcomes such as ‘difficulty in dilating the cervix’ or ‘pain experienced by the women’ might be open to even greater subjectivity and much wider variation in results. A secondary outcome measure of cervical dilatation of 5 mm was also intended to re-enforce the impression of the cervical ripening effect caused by misoprostol. In the premenopausal misoprostol group, 88% of women experienced cervical dilatation of 5 mm, compared with 65% in the placebo group. Miniature hysteroscopes do not necessarily require cervical dilatation, but a 10-mm resectoscope almost always does. A 1 mm difference in the cervical dilatation might be a questionable outcome if one were only comparing a few women against each other. Used in a trial with a sufficient number of trial participants, this can be a statistically significant difference.

Regarding the choice of dose, misoprostol is a safe and well-tolerated drug, even in high doses. The dosages chosen for previous trials on nonpregnant women have been similar to those used prior to termination of pregnancy (the majority using 400 micrograms of misoprostol). The only previous trial we found to use 1000 micrograms of misoprostol5 did not show any effect if given vaginally 2–4 hours before surgery. We chose a higher dosage of misoprostol and a longer dosage interval because there was no clear consensus from previous trials as to whether misoprostol has a significant ripening effect in nonpregnant women. Failure in previous trials to have any effect on cervical ripening may have been due to inadequate dosage or time from taking the tablets until the operation. In pregnant women at term, a dosage as small as 25 micrograms of misoprostol may be enough to induce labour, while we have observed that 1000 micrograms of misoprostol had no effect on postmenopausal women, compared with placebo, after 12 hours. Furthermore, the individual vaginal absorption of misoprostol varies greatly. We believe that our trial design may give a clear indication whether misoprostol is effective for cervical ripening in nonpregnant women or not. We refer to our previous correspondence regarding the flow chart error.

References

  1. Top of page
  2. Author’s Reply
  3. References