Objective To test the hypothesis that at two years bladder and ovarian function function are no different following either simple hysterectomy or endometrial ablation (transcervical resectiodlaser ablation). Design Randomised controlled trial comparing hysterectomy with endometrial ablation. Two years after trial entry bladder and ovarian function were evaluated subjectively by means of questionnaires and objectively by means of cystometry and estimation of serum follicle stimulating hormone respectively. Setting Aberdeen Royal Infirmary. Participants Two hundred and four women with dysfunctional uterine bleeding who, when recruited to the initial study two years previously, were aged less than 50 years, weighed less than 100 kg, and who would otherwise have undergone hysterectomy. Results Of the 204 women originally recruited, 101 re‐attended the clinic and underwent cystometry and follicle stimulating hormone estimation. These, together with a further 59 women, completed postal questionnaires (79% of original cohort). Rates of stress incontinence (44%vs 44%, 95% CI of difference −16% to +15%), urge incontinence (21% vs 19% 95% CI of difference −11% to +14%), and hot flushes (30%vs 44%, 95% CI of difference −25% to +7%) were similar in the hysterectomy and endometrial ablation groups, respectively. Cystometry revealed 14 (31%) cases of bladder dysfunction after hysterectomy and 17 (35%) after hysteroscopic surgery (95% CI of difference −23% to +15%). Serum follicle stimulating hormone levels > 40 mIu/L were found in three (6%) women following hysterectomy and five (10%) of women after endometrial ablation. Conclusion This study suggests that in comparison with endometrial ablation, simple hysterectomy for dysfunctional uterine bleeding does not compromise bladder or ovarian function, at least at two years after the operation. Due to lack of power the estimates of any differences are imprecise, and clinically significant effects cannot be ruled out.

A retrospective study of the complications of cone biopsy showed that among 9 15 women examined between t h e years 1976 and 1982, 121 (13%) had primary or secondary haemorrhage, 153 (17%) cervical stenosis and 39 (4%) subsequent infertility or an abnormal pregnancy. Cervical stenosis was commonest among women who had had long cones removed. Stenosis occurred more often in the group of women who had been assessed by colposcopy before operation but this was due t o the fact that prior colposcopy selected a favourable group of patients with lesions of limited extent that were susceptible to treatment by local destructive therapy, so that prior colposcopic assessment resulted in the removal of longer cones.

Perinatal deaths and major lethal and non-lethal congenital malformations occurring in this hospital from 1979-82 inclusive were related to the ethnic group of the 15 438 mothers. The highest crude perinatal mortality rates occurred in Indian and Pakistani populations (18-3 per 1000 and 24-1 per 1000 respectively). The highest incidence of congenital abnormality also occurred in these groups (13-3 per 1000 and 12*8 per 1000 respectively), but there was considerable variation in the distribution of different malformations.

were analysed for their ethnic origins. Social classes IV and V predominated in all groups. A high proportion of Indian mothers fell into the low-risk group based on age and parity but had the highest stillbirth and perinatal mortality rates (15-1 and 27-5/1000 respectively) and infants of low mean birth weight (2986 g). Elderly and multiparous mothers were characteristic of the Pakistani and Bangladeshi groups. Young, primiparous mothers were more common among the West Indians and Europeans, in whom the stillbirth and perinatal mortality rates were low; infants in the European group had a mean birth weight higher than in any other group (3231 g). From these findings ethnic origin of the mother is apparently an important factor in perinatal mortality.

Summary. The reported incidence of ectopic pregnancy in Aberdeen City and suburbs (1950–1985), using as denominators maternities, pregnancies and women aged 15–44 years, has increased threefold since 1970 to 6·4/1000 pregnancies. This increased incidence persisted after the exclusion of previously sterilized women. A total of 11128 women were sterilized in Aberdeen City and suburbs between 1960 and 1982; 36 ectopic pregnancies occurred in this sterilized population. The prevalence of ectopic pregnancy was 3·55/1000 sterilizations. This did not alter significantly over the period of study despite changes in the method of sterilization. However, due to the increased popularity of sterilization, the proportion of ectopic pregnancies in women who had been sterilized increased from 0% in the 1950s to 21% in the quinquennium 1975–1979.

No abstract

To assess the clinical potential of serial serum CA125 measurements in the follow-up of patients with epithelial ovarian cancer, 74 consecutive unselected patients with histologically confirmed ovarian carcinoma were studied prospectively. There was an 83% concordance between clinical assessment and CA125 assessment of response. The positive predictive values of a rising CA125 for disease progression and a falling CA125 for disease regression were 0.93 and 0.94, respectively. The absolute CA125 values during observations of complete response (mean 96 U/ml; 95% confidence interval; 33 to 128 U/ml), partial response (mean 134 U/ml; 95% confidence interval; 98 to 159 U/ml) and stable or progressive disease (mean 391 U/ml; 95% confidence interval; 282 to 545 U/ml) were significantly different. A randomized study is required to determine whether CA125 monitoring has any benefit in terms of outcome, and particularly survival, in epithelial ovarian cancer.

Gonadotrophin dynamics in women receiving immediate or delayed transdermal estradiol after oophorectomv. Obstet Gwiecol 1991;78:98-102. AUTHORS' REPLY,-Our study showed significantly higher concentrations of gonadotrophin in oophorectomised women treated with the 0 05 mg oestradiol patch than in those given a 50 mg oestradiol implant. As stated in our paper, we think that the lower gonadotrophin concentrations in the implant group were due to the higher oestradiol concentrations observed. Unpublished data from this group of patients at 12 months show significantly different oestradiol concentrations: a mean (SE) in the patch group of 224 (43) pmol/l and in the implant group 544 (42) pmol/l (95% confidence interval for difference between means -443 to -197, p<0-0001). As stated by Stevenson et al, patches have been shown to prevent postmenopausal bone loss, as have implants,' but it has also been observed that percentage increase in bone density correlates with plasma concentrations of oestradiol,' and so treatments resulting in higher oestradiol concentrations may be more effective in preventing this long term consequence of the menopause.This suggests that Eliot et al may not be correct in their assertion that serum oestradiol levels greater than 120 pmoUl provide no additional effect. We agree that implants should not be forced on postmenopausal women who are keen to have patches and vice versa.Reid and Ganger question the ethics of our study. Delaying treatment ensured that all women had equivalent baseline hormone profiles before starting oestrogen replacement; given that one of the aims was to compare hormone profiles, this was a necessary part of the study design. They also question the doses used in this study: we did not set out to compare equivalent doses, but rather, the recommended starting doses. Careful reading of Chetkowski et ars findings reveals that transdermal oestradiol significantly decreased gonadotrophin levels in a dose dependent manner.' Kamel's paper referred to administration of the 0-2 mg oestradiol patch immediately after oophorectomy; suppression of gonadotrophins was not maintained.4 Our study showed that the 005 mg patch did not suppress gonadotrophin release after oophorectomy (mean concentration of follicle stimulating hormone after oophorectomy, 37-3 IU/l; after four months' treatment, 53 4 IU/l). Although there is no evidence from studies on the long term benefits of the 0 05 mg patch compared with the 50 mg implant, our study shows that there are differences in gonadotrophin concentrations which we suggest are due to differences in oestradiol concentrations and which may be reflected in the long term benefits of oestrogen replacement therapy.