Randomised trial comparing hysterectomy and transcervical endometrial resection: effect on health related quality of life and costs two years after surgery

Abstract

Objective: To compare the impact of endometrial resection and abdominal hysterectomy on a range of health outcomes and health service costs, based on longer term follow up of patients randomised to a clinical trial.

Design: A parallel group of randomised control trial.

Setting: The gynaecology department of a teaching hospital.

Participants: 196 women requiring surgical treatment for menorrhagia were randomised and received surgery (88 underwent resection and 97 hysterectomy). Longer term follow up was undertaken using a postal questionnaire sent to all 196 women.

Main outcome measures: Longer term assessment was on the basis of menstrual symptoms, health related quality of life using the Short Form 36 (SF36) and the EuroQol visual analogue scale, patient satisfaction and health service resource cost.

Results: Of 196 women who were sent a questionnaire, 155 (79%) responded at an average interval of 2.8 years after initial surgery. All aspects of health outcomes were as good or better in patients randomised to hysterectomy. Among patients randomised to resection, 57% had experienced no improvement in premenstrual symptoms following surgery and 23% had taken time off work due to menstrual problems; among hysterectomy patients, these rates were 23% and 4%, respectively. Women randomised to hysterectomy had better mean scores on seven of the eight dimensions of the SF36 health related quality of life instrument, with the greatest difference being on the pain dimension (P = 0.01). Women randomised to hysterectomy were generally more satisfied with treatment (P = 0.002). By two years after initial surgery, women randomised to resection had a 12% probability of having had a repeat resection and a 16% chance of having had a hysterectomy. As a percentage of the mean total cost associated with women randomised to hysterectomy, the mean total cost of resection was 53% based on four months follow up; this proportion had increased to 71%, based on an average overall follow up of 2.2 years.

Conclusions: These results show that, at an average follow up of 2.8 years among responders to a questionnaire, women randomised to hysterectomy experienced more of an improvement in menstrual symptoms and higher rates of satisfaction with treatment. There is also some evidence of superior health related quality of life amongst hysterectomy patients. However, the health service cost of endometrial resection remains lower than that of hysterectomy. An assessment of the relative cost effectiveness of the two procedures awaits further research.