Objective: To compare single dose systemic methotrexate (50 mg/m2) with laparoscopic surgery for the treatment of unruptured tubal pregnancy.
Design: An open, pragmatic, prospective randomised trial.
Setting: Departments of obstetrics and gynaecology at three hospitals in Auckland, New Zealand.
Participants: Clinically stable women with an unruptured tubal pregnancy diagnosed by transvaginal ultrasound and quantitative serum beta-hCG measurement. Inclusion criteria included a serum beta-hCG concentration < 5,000 IU/L, and a tubal pregnancy of < 3.5 cm diameter.
Main outcome measures: Treatment success, physical and psychological functioning, side effects, and subsequent ipsilateral tubal patency.
Results: Two hundred and eighteen women with ectopic pregnancies were seen at the three hospitals. 79 women (36% eligibility rate) were eligible for trial entry and 62 women (78% recruitment rate) were recruited. Twenty-six of the 28 women (93%) randomised to laparoscopic surgery required no further treatment, compared with 22 of the 34 women (65%) randomised to methotrexate (95% CI of difference in success rate 10 - 47%; P < 0.01). Two women (7%) in the laparoscopic surgery group had persistent trophoblast. Nine women (26%) in the methotrexate group required more than one dose of methotrexate and five women (15%) underwent laparoscopy during follow up. In the laparoscopy group three women (11%) had negative laparoscopies and two women (7%) had were found to have a ruptured fallopian tube at the time of surgery. Women treated with methotrexate had significantly better objective physical functioning scores but there were no differences in any other psychological outcomes. Women treated with methotrexate experienced greater and more prolonged vaginal bleeding. The likelihood of methotrexate treatment failure was greater at higher serum beta-hCG concentrations. Ipsilateral tubal patency rates were similar in each group.
Conclusion: This trial shows that in the treatment of tubal pregnancy single dose systemic methotrexate is a less effective treatment than laparoscopic salpingotomy. It is well tolerated, but should only be offered as an alternative to surgery to women who have mild symptoms and present at low serum beta-hCG concentrations. In our population this likely to be no more than a quarter of women presenting with a tubal pregnancy.