Dynamics of Dwarf Galaxies Disfavor Stellar-Mass Black Holes as Dark Matter.

We study the effects of black hole dark matter on the dynamical evolution of stars in dwarf galaxies. We find that mass segregation leads to a depletion of stars in the center of dwarf galaxies and the appearance of a ring in the projected stellar surface density profile. Using Segue 1 as an example we show that current observations of the projected surface stellar density rule out at the 99.9% confidence level the possibility that more than 6% of the dark matter is composed of black holes with a mass of few tens of solar masses.

Maximum Redshift of Gravitational Wave Merger Events.

Future generations of gravitational wave detectors will have the sensitivity to detect gravitational wave events at redshifts far beyond any detectable electromagnetic sources. We show that if the observed event rate is greater than one event per year at redshifts z≥40, then the probability distribution of primordial density fluctuations must be significantly non-Gaussian or the events originate from primordial black holes. The nature of the excess events can be determined from the redshift distribution of the merger rate.

Indication of Gamma-Ray Emission from the Newly Discovered Dwarf Galaxy Reticulum II.

We present a search for γ-ray emission from the direction of the newly discovered dwarf galaxy Reticulum II. Using Fermi-LAT Collaboration data, we detect a signal that exceeds expected backgrounds between ∼2-10 GeV and is consistent with annihilation of dark matter for particle masses less than a few ×10^{2} GeV. Modeling the background as a Poisson process based on Fermi-LAT diffuse models, and taking into account trial factors, we detect emission with p value less than 9.8×10^{-5} (>3.7σ). An alternative, model-independent treatment of the background reduces the significance, raising the p value to 9.7×10^{-3} (2.3σ). Even in this case, however, Reticulum II has the most significant γ-ray signal of any known dwarf galaxy. If Reticulum II has a dark-matter halo that is similar to those inferred for other nearby dwarfs, the signal is consistent with the s-wave relic abundance cross section for annihilation.

Sperm donor recruitment, attitudes and provider practices--5 years after the removal of donor anonymity.

STUDY QUESTION: Has the change in donor anonymity legislation in UK affected the recruitment of men wanting to be sperm donors and also affected the attitudes of the practitioners who provide donor sperm treatment? SUMMARY ANSWER: We have performed fewer IUI and IVF treatments using donor sperm following the change in legislation in April 2005 than before. However, we have seen an overall increase in men wanting to donate their sperm, including a small increase in men from ethnic minorities. WHAT IS KNOWN ALREADY: Sweden, which removed donor anonymity in 1985, had an initial drop in men wanting to donate and then 10 years later started to have an increase. The Human Fertilisation and Embryology Authority (HFEA) and other studies in the UK have shown an overall downward trend, but have not been able to compare large time scales either side of the change in legislation. STUDY DESIGN, SIZE, DURATION: This was a retrospective descriptive study that looked at all men who approached the clinic between the years 2000 and 2010, i.e. 5 years either side of the change in legislation (April 2005). Overall, we had 24 men wanting to be donors prior to the rule change and 65 men after the rule change. We also investigated the total number of all treatments with donor sperm, and this included a total of 1004 donor sperm treatments prior to the change in legislation and 403 donor sperm treatments after the change in legislation. PARTICIPANTS, SETTING, METHODS: The study was set in an NHS IVF clinic in South East London. We compared the indicators of service provision, provider practices and donor attitudes, in the period between April 2000 and March 2005 (Group A) with those between April 2005 and March 2010 (Group B), i.e. 5 years either side of the change in legislation. MAIN RESULTS: There were 875 IUI treatments and 129 IVF or ICSI treatments in Group A and 325 IUI and 78 IVF/ICSI treatments in Group B with the use of donor sperm, of which, 11.9% (119 out of 1004) in Group A and 39.5% (159 out of 403) in Group B were with donor sperm recruited by our unit. The clinical pregnancy rate per cycle of treatment in Group A was (86 out of 875) 9.8% for IUI and (27 out of 129) 20.9% for IVF/ICSI and in Group B (32/325) 9.8% and (28 out of 78) 35.9%, respectively. There was a sharp yearly fall in donor sperm treatments from 2004. Twenty-four men were screened in Group A, of which 18 (75.0%) were recruited for long-term storage and 12 (50%) were registered as donors with the HFEA when the sperm was used, whereas in Group B, 65 men were screened, 53 (82.0%) were recruited and 24 (36.92%) were registered as donors. Six (24.0%) men in Group A failed in screening because of poor semen analysis when compared with 9 (13.8%) men in Group B. The majority of post-recruitment dropouts were because of loss of follow-up or withdrawal of consent. More donors in Group A were white (92.0 versus 77.0%) and born in UK (92.0 versus 68.0%) when compared with those in Group B. Donors in Group B were more likely to be single (46.0 versus 4.0%) and to have informed their relevant partner of donation (71.0 versus 54.0%) when compared with those in Group A. 83.0% of donors in Group A were heterosexual when compared with 69.0% in Group B. The primary reason for donating in both groups of potential donors was 'wanting to help' (46.0% 'altruistic donors' in Group A versus 72.0% in Group B). Fewer donors in Group B (37%) had specific restrictions about the use of their sperm when compared with 46.0% in Group A. LIMITATIONS, REASONS FOR CAUTION: As this was a retrospective study, there is a chance for the introduction of bias. WIDER IMPLICATIONS OF THE FINDINGS: We have shown that despite no active in-house recruitment procedures, we are managing to recruit more potential sperm donors after the change in UK legislation, and we are able to meet the demand for treatments with in-house recruited donor sperm that is a reassuring finding for donor sperm treatment services in the wider UK. FUNDING/COMPETING INTERESTS: No external funds were sought for this work. None of the authors have any competing interests to declare.

Large loop excision of transformation zone and cervical length in the prediction of spontaneous preterm delivery.

OBJECTIVES: To investigate the association between previous large loop excision of transformation zone (LLETZ) and risk for subsequent spontaneous preterm delivery (sPD) and whether this effect is reflected in the measurement of cervical length at mid-gestation. DESIGN AND SETTING: A secondary analysis of data from women recruited for clinical trials of interventions to prevent preterm labour. POPULATION: A total of 26,867 women with singleton pregnancies attending for routine antenatal care. METHODS: Transvaginal sonographic measurement of cervical length was carried out at 20(+0) to 24(+6) weeks. Logistic regression analysis was used to determine the significant predictors of sPD among maternal characteristics, obstetric history, previous history of LLETZ and cervical length. MAIN OUTCOME MEASURES: Spontaneous preterm delivery. RESULTS: In the 473 women who had undergone LLETZ, compared with the 25,772 without a history of LLETZ, the rate of sPD before 34 weeks of gestation was higher (3.4 versus 1.3%, P = 0.0002) and the median cervical length was shorter (32 mm versus 34 mm, P < 0.0001). Regression analysis demonstrated that in the prediction of sPD there were significant contributions from racial origin, cigarette smoking, previous preterm delivery and LLETZ and the detection rate of sPD was 29.8%, at a false-positive rate of 10%. However, after addition of cervical length, LLETZ did not remain a significant predictor in the model, which detected 52.6% of sPD, at a false-positive rate of 10%. CONCLUSIONS: LLETZ increases the risk of sPD, even after adjustment for maternal risk factors. The effect of a previous LLETZ on sPD in a subsequent pregnancy is reflected in the measurement of cervical length at mid-gestation.

Exclusion of canonical weakly interacting massive particles by joint analysis of Milky Way dwarf galaxies with data from the Fermi Gamma-Ray Space Telescope.

Dwarf spheroidal galaxies are known to be excellent targets for the detection of annihilating dark matter. We present new limits on the annihilation cross section of weakly interacting massive particles based on the joint analysis of seven Milky Way dwarfs using a frequentist Neyman construction and Pass 7 data from the Fermi Gamma-Ray Space Telescope. We exclude generic weakly interacting massive particle candidates annihilating into bb with a mass less than 40 GeV that reproduce the observed relic abundance. To within 95% systematic errors on the dark matter distribution within the dwarfs, the mass lower limit can be as low as 19 GeV or as high as 240 GeV. For annihilation into τ+ τ-, these limits become 19, 13, and 80 GeV, respectively.

A direct comparison of women's perceptions and acceptability of micronised progesterone and medroxyprogesterone acetate in combination with transdermal oestradiol in the management of young postmenopausal women, under 45 years of age.

OBJECTIVE: To assess the acceptability and perception of postmenopausal women, to two different hormone replacement therapy regimens, in relation to the control of their symptoms and development of adverse effects. STUDY DESIGN: Prospectively recruited postmenopausal women, <45 years, were randomised to one of two treatment arms for 12-months: cyclical micronised progesterone or medroxyprogesterone acetate in combination with transdermal oestradiol. A self-reported questionnaire with matrix rating scales was completed and repeated after 3, 6 and 12-months. MAIN OUTCOME MEASURES: Symptom control and development of adverse effects. RESULTS: Seventy-one individuals were screened, with baseline data available for 67 subjects. A total of 190 questionnaires were returned. The most commonly reported symptoms were low energy levels, vasomotor symptoms and sexual dysfunction. The prevalence of adverse effects ranged between 57.89 and 87.50%, with a reduction seen in the transdermal oestradiol + micronised progesterone arm (73.91% at 3-months, decreasing to 57.89% at 12-months; p = 0.33), compared to the transdermal oestradiol + medroxyprogesterone acetate arm (76.92% at 3-months, increasing to 87.50% at 12-months; p = 0.69). The main reported adverse effects were bloating, weight change and psychological symptoms. A significant difference was documented between the groups after set intervals, with a greater proportion reporting breast tenderness after 3-months (p = 0.01), lower numbers reporting mood swings at 6-months (p = 0.01) and irritability at 12-months (p = 0.03) in the transdermal oestradiol + micronised progesterone arm compared to the transdermal oestradiol + medroxyprogesterone acetate arm. CONCLUSIONS: The acceptability of both regimens was high despite adverse effects, but tolerability of transdermal oestradiol combined with micronised progesterone appeared to be better with fewer women reporting psychological concerns.

Contraception and medical gynaecology for HIV-positive women in a one-stop clinic.

HIV-positive women may be reluctant to attend gynaecology or family planning clinics for fear of divulging their condition. Therefore, a referral clinic was opened within the HIV clinic. Retrospective case-note reviews of 197 new patients revealed 109 with a variety of medical gynaecology conditions (menorrhagia being the commonest) and 88 sought contraception. The full range of contraceptives was used, including Mirena for the treatment of menorrhagia as well as contraception and the combined pill adjusted for interaction with liver enzyme-inducing antiretroviral drugs. The acceptance of contraceptive advice and gynaecological evaluation by the patients has resulted in improved reproductive health services for these HIV-positive women. In centres with large cohorts of HIV-positive women, this type of one-stop specialist clinic will be very effective in providing high-quality reproductive health care and hence, this type of clinic is recommended for such centres.

Outcome analysis of 4 years' follow-up of patients referred for colposcopy with one smear showing mild dyskaryosis.

OBJECTIVE: To assess the 4-year outcome of patients after one smear showing mild dyskaryosis with respect to smear regression rate, prevalence of cervical intraepithelial neoplasia (CIN) and the effect of age. METHODS: Retrospective analysis of patients diagnosed with initial mildly dyskaryotic smear during the year 2000 with a follow-up period of 48 months. These women had not had any previous abnormal smears. SETTINGS: Cytopathology Department and Colposcopy Unit, King's College Hospital, London, UK. RESULTS: We identified 524 patients of whom 375 patients with complete follow-up data are included. The age range was 19-67 years with a median of 29 years. There were 207 patients aged 35 years or less (55%). At 6 months, 258 smears were performed and 47.8% of them were negative (95% CI: 41.6-54.0%). The total number of negative follow-up smears in the first year was 198 out of a total of 397 smears performed (50%). This proportion has significantly increased between 1 and 4 years' follow-up to 67.5% (RR: 1.24; 95% CI: 1.14-1.35). Over the 4-year period, 791 smears were performed and 477 were negative (60.3%; 95% CI: 56.9-63.7%). Of the 477 negative smears there were only 61 smears (12.8%; 95% CI: 10-16%), in 54 patients (14%; 95% CI: 11-18%) that reverted back to low-grade cytological abnormality. In only one case the repeat smear showed high-grade abnormality after initial negative follow-up; however, on biopsy, histology showed CIN I. Out of the 375 patients, 70 required treatment with excisional biopsy (19%; 95% CI: 15.0-22.9%). Histology confirmed high-grade CIN in only 41 cases giving a prevalence of 11% (95% CI: 8.1-14.5%). There were no cases of microinvasive or invasive cancer detected. Age (< or =35 years versus >35 years) did not significantly affect either cytological or histological outcome. CONCLUSION: Sixty per cent of follow-up smears after initial mild dyskaryosis subsequently became negative; of them 87.2% remained negative over the 4 year follow-up. Treatment was only required in 19% of patients, with 11% prevalence of high-grade CIN. Age did not affect the outcome. These results are reassuring and indicate that colposcopic referral may not be necessary after only one mildly dyskaryotic smear.

A comparative study of glycodelin concentrations in uterine flushings in women with subseptate uteri, history of unexplained recurrent miscarriage and healthy controls.

OBJECTIVE: To compare the concentration of glycodelin in uterine flushing at the implantation window obtained from women with subseptate uteri, women with a history of recurrent first trimester miscarriage and fertile controls. STUDY DESIGN: Glycodelin concentration was assessed using Enzyme Linked Immunohistochemistry (ELISA) at The Early Pregnancy & Gynaecology Assessment Unit, King's College Hospital, London, England. Eight women with a subseptate uterus, 20 women with a history of unexplained recurrent first trimester miscarriage and 16 fertile controls had uterine cavity flushing, for glycodelin concentration, done 7 days after the luteinising hormone surge. RESULTS: Glycodelin concentrations in uterine flushing obtained from women with subseptate uteri (n=8) (median 32.9 ng/ml, range 17.1-52.4 ng/ml) and recurrent miscarriage (n=20) (median 26.8 ng/ml, range 9.7-78.5 ng/ml) were significantly lower than in the control group (n=16) (median 67.7 ng/ml, range 59.0-77.6 ng/ml) (chi(2)=19.565, p<0.001). CONCLUSION: Peri-implantation levels of glycodelin are lower in women at high risk of early pregnancy failure.

Abnormal steroid excretion in gestational trophoblastic disease complicated by ovarian theca-lutein cysts.

Serum and urine steroids were examined in two subjects with trophoblastic disease accompanied by large ovarian theca-lutein cysts and compared with those from 10 patients with trophoblastic disease but without palpable cysts. In the patients without cysts normal values were obtained for serum oestradiol, progesterone, 17 alpha-hydroxyprogesterone and androstenedione, and for urinary total oestrogens, pregnanediol, pregnanetriol, and 17-oxosteroids. Nineteen urinary steroid metabolites, quantified by capillary gas-liquid chromatography, were either within reference limits or marginally raised. In several cases relatively minor increases in serum testosterone and cortisol and urinary free cortisol were observed. In contrast, the subjects with cysts showed pronounced excesses of androgen metabolites, 17 alpha-hydroxypregnenolone, pregnanediol, and pregnanetriol, and both exhibited a similar pattern of unusual additional metabolites. The profiles superficially resembled those seen in 21-hydroxylase deficiency adrenogenital syndrome, but there were important discrepancies reflecting known differences in ovarian and adrenal steroid metabolism. Chemotherapy led to decline of human chorionic gonadotrophin concentrations, regression of the cysts, and return to normal of the steroid profile. Excess steroids in the patients with cysts may have originated in the ovary rather than in the trophoblastic tissue.

A cross-sectional study of the effects of long-term percutaneous hormone replacement therapy on bone density.

The effect of hormone implants on the bone density of postmenopausal women was studied in 110 patients (mean age 54.7 years; mean menopausal age 8.6 years, range 2-30) who had received hormone replacement in the form of estradiol (50-75 mg) and testosterone (100 mg) pellets at 6-month intervals for 2-24 years (mean 5.2). They were compared with 254 untreated women (mean age 55.0 years; mean menopausal age 6.8 years, range 1-37). The bone density at the spine, measured by quantitative digital radiography, was 1.123 grams hydroxyapatite (gHa)/cm2 in the treated group and 0.951 gHa/cm2 in the controls (P less than .0001). The total bone density at the proximal femur was 1.002 gHa/cm2 in the treated group, compared with 0.914 gHa/cm2 in the controls (P less than .0001). There were significant differences in the density of the trochanteric, intertrochanteric, and neck areas of the proximal femur as well as the Ward triangle (all P less than .0001). These differences became significant from the age of 55 at the neck of the femur, Ward triangle, and lumbar spine, and from age 60 for all other values. Subcutaneous estradiol and testosterone prevent postmenopausal osteoporosis and maintain normal bone density for as long as treatment is continued.

Endometrial histology and bleeding patterns after 8 years of continuous combined estrogen and progestogen therapy in postmenopausal women.

Continuous combined estrogen and progestogen preparations enable the postmenopausal woman to enjoy the benefits of estrogen replacement without the inconvenience of regular progestogen-induced withdrawal bleeding. The endometrium appears to be adequately protected in the short term, but no published data are available on the bleeding patterns or endometrial response after more than 18 months of therapy. Therefore, we reviewed 41 patients who continued on such preparations for up to 10 years (mean duration of use 8.0 years). Six women had experienced episodes of breakthrough bleeding after achieving amenorrhea, two of whom had benign endometrial polyps and two with adenocarcinoma of the endometrium. The remaining 35 women each had prolonged amenorrhea and were found to have an atrophic inactive endometrium. It is too early to comment on the long-term endometrial effects of these preparations because the numbers are too small; however, any breakthrough bleeding occurring after a period of prolonged amenorrhea must be investigated by means of endometrial biopsy.

The effects of plasma estradiol levels on increases in vertebral and femoral bone density following therapy with estradiol and estradiol with testosterone implants.

Percutaneous estradiol (E2) implants effectively preserve bone density in postmenopausal women. However, these implants are often given with testosterone, which may itself have an anabolic effect on bone. To determine whether testosterone confers any additional bone-sparing effect, we studied 50 postmenopausal women randomly allocated to receive E2 (75 mg) alone or with testosterone (100 mg) every 6 months for 1 year. Women with an intact uterus received cyclic norethindrone (5 mg) for 10 days of each calendar month. Twenty-five untreated women were recruited to act as a reference group. Bone density was measured at the lumbar spine and proximal femur by dual x-ray densitometry. By 1 year, bone density at the lumbar spine had fallen by 1.8% in the reference group. In the women treated with E2 alone, it increased significantly by 7.8% (P less than .0001) and in those receiving E2 with testosterone, it increased by 6.3% (P less than .0001). At the femoral neck, bone density decreased by 3% in the controls and increased by approximately 4% in both treated groups (P less than .0001). The increase in bone density at these sites was unrelated to the woman's chronological age, menopausal age, or initial bone density. However, it correlated significantly with the serum E2 levels attained after 1 year of therapy. In no treated patients did bone density decrease significantly. These data show that testosterone confers no additional bone-sparing effect in postmenopausal women.

Prospective evaluation of logistic regression models for the diagnosis of ovarian cancer.

OBJECTIVE: To test the accuracy of three logistic regression models in diagnosing malignancy in women with adnexal masses. METHODS: This was a prospective collaborative study. Women were recruited from three hospitals and all assessments were performed at the Gynaecology Ultrasound Unit, King's College Hospital. One hundred women with known adnexal masses were examined preoperatively. The demographic, biochemical, and sonographic data recorded for each patient included age, menopausal status, CA 125 levels, ultrasound morphology, and Doppler blood flow analysis. The diagnosis of malignancy was made for each woman using three logistic regression models previously described by Alcazar et al, Tailor et al, and Timmerman et al. Variables used in these models were then combined to form a new model. The results were compared with the final histopathologic diagnosis. RESULTS: Sixty-seven women had benign tumors and 33 had ovarian cancer. Women with malignant tumors were older than those with benign masses. There were significant differences in CA 125 levels, presence of papillary proliferations, and ascites between the two groups. The sensitivities and specificities achieved respectively by the models were as follows: 45% and 93% with Tailor et al's model, 9% and 99% with Alcazar et al's model, and 73% and 91% with Timmerman et al's model. There was no significant improvement over the performance of Timmerman et al's model and the new combined model. CONCLUSION: All models performed less well than originally reported. Combining the models did not lead to a significant improvement in performance. Larger sample sizes that incorporate all types of ovarian tumors are necessary to design more accurate diagnostic models.

Bone loss after hysterectomy with ovarian conservation.

OBJECTIVE: To examine the long-term effects of hysterectomy with conservation of the ovaries on bone density of the lumbar spine and proximal femur. METHODS: A cross-sectional study of the bone density of 40 postmenopausal women who had undergone hysterectomy with ovarian conservation before menopause compared with a matched group of 40 women who had not had hysterectomy. The 40 women who had undergone hysterectomy were also compared with a control population, using multiple linear regression analysis. Bone density of the femoral neck and lumbar spine was measured by quantitative digital radiography. RESULTS: Bone density in the hysterectomy group was significantly reduced at the spine (P < .05) and at the femoral neck (P < .05) compared with the matched group. Comparisons of the hysterectomy group with the reference group demonstrated that in addition to significant reductions in bone density at the spine (P < .05) and hip (P < .05), bone density at the femoral neck (P < .05), trochanter (P < .05), Ward's triangle (P < .05), and the second (P < .05) and fourth (P < .05) lumbar vertebrae was also significantly less in the hysterectomy group. CONCLUSION: Premenopausal women who have hysterectomy will have significantly lower bone density than controls, despite conservation of both ovaries at the time of surgery.

Response of skin thickness and metacarpal index to estradiol therapy in postmenopausal women.

A radiologic method for measuring skin thickness and metacarpal index was used to investigate 41 postmenopausal women treated with estradiol (100-mg) subcutaneous implants (Organon, UK). All the women completed the first six months of the study, and 33 completed one year. Both skin thickness and metacarpal index increased to a statistically significant degree over the one-year period, with most of the increase occurring in the first six months of therapy. Skin thickness showed the largest increases, from a mean of 0.86 mm at the start of the study to 0.97 mm at six months and 1 mm at one year. The metacarpal index increased from a mean of 0.77 at the start of the study to a mean of 0.799 and 0.8 at six months and one year, respectively.

Early experience with gamete intrafallopian transfer (GIFT) and direct intraperitoneal insemination (DIPI).

We present our early experience with gamete intrafallopian transfer (GIFT) and direct intraperitoneal insemination (DIPI) combined with intrauterine insemination (IUI), two recently described methods of assisting conception in patients with patent fallopian tubes. Sixty-nine patients (93 cycles) were entered into the study. Thirty-three patients (51 cycles) entered the DIPI/IUI programme and 36 patients (42 cycles) entered the GIFT programme. The mean age, duration and aetiology of infertility were similar in both groups. In the GIFT programme 12 pregnancies occurred, which is a 29% pregnancy rate per cycle and a 33% pregnancy rate per patient. In the DIPI/IUI programme only 3 pregnancies occurred, being a 6% pregnancy rate per cycle and a 9% pregnancy rate per patient. With the live birth rate of in vitro fertilization (IVF) being 12% per embryo transfer, we conclude that GIFT is more successful than either DIPI/IUI or IVF in patients with patent fallopian tubes. Further controlled studies are required to assess the future role of DIPI/IUI in clinical practice.

Skeletal effects of oral oestrogen compared with subcutaneous oestrogen and testosterone in postmenopausal women.

STUDY OBJECTIVE: To compare oral and implanted oestrogens for their effects in preventing postmenopausal osteoporosis. DESIGN: Non-randomised cohort study of postmenopausal women treated with oral or depot oestrogens and postmenopausal controls. SETTING: Gynaecological endocrine clinic in tertiary referral centre. PATIENTS: Oral treatment group of 37 postmenopausal women (mean age 57.5 years, median 8.75 years from last menstrual period), compared with 41 women given oestrogen implants (mean age 56.2 years, median 9.5 years from last menstrual period) and 36 controls (mean age 51.8 years, median 2.0 years from last menstrual period). Weight was not significantly different among the groups. INTERVENTIONS: Oral treatment group was given continuous treatment with cyclic oestrogen and progesterone preparations (Prempak C or Cycloprogynova) for a median of 8.0 years. Implant group was given subcutaneous implants of oestradiol 50 mg combined with testosterone 100 mg, on average six monthly for a median of 8.5 years. Controls were not treated. END POINT: Significant increase in bone density. MEASUREMENTS AND MAIN RESULTS: Bone density measured by dual beam photon absorptiometry was 1.02 (SD 0.13) g hydroxyapatite/cm2 in implant group versus 0.89 (0.11) in oral group (p less than 0.01) and 0.87 (0.14) in controls (p less than 0.01). Serum oestradiol concentration in implant group was (median) 725 pmol/l versus 170 pmol/l in oral group (p less than 0.01) and 99 pmol/l in controls (p less than 0.01). Serum follicular stimulating hormone was median 1 IU/l (range 1-11) in implant group (equivalent to premenopausal values) versus 43 (4-94) IU/l in oral group (p less than 0.01) and 72 (28-99) IU/l in controls (p less than 0.01). CONCLUSIONS: Subcutaneous oestrogen is more effective than oral oestrogen in preventing osteoporosis, probably owing to the more physiological (premenopausal) serum oestradiol concentrations achieved. It also avoids problems of compliance that occur with oral treatment.