Gonadotrophin-releasing hormone analogues for endometriosis: bone mineral density.

BACKGROUND: Gonadotrophin-releasing hormone analogues (GnRHas) are generally well tolerated, and are effective in relieving the symptoms of endometriosis (Prentice 2003). Unfortunately the low oestrogen state that they induce is associated with adverse effects including an acceleration in bone mineral density (BMD) loss.

OBJECTIVES: To determine the effect of treatment with gonadotrophin-releasing hormone analogues (GnRHas) on the bone mineral density of women with endometriosis, compared to placebo, no treatment, or other treatments for endometriosis, including GnRHas with add-back therapy.

SEARCH STRATEGY: We searched the Cochrane Menstrual Disorders and Subfertility Group's specialised register of controlled trials (23rd October 2002) and the Cochrane Central Register of Controlled Trials (Cochrane Library, issue 4, 2002). We also carried out electronic searches of MEDLINE (1966 - March Week 2 2003) and EMBASE (1980 - March Week 2 2003). We also searched the reference lists of articles and contacted researchers in the field.

SELECTION CRITERIA: Prospective, randomised controlled studies of the use of GnRHas for the treatment of women with endometriosis were considered, where bone density measurements were an end point. The control arm of the studies was either placebo, no treatment, another medical therapy for endometriosis, or GnRHas with add-back therapy.

DATA COLLECTION AND ANALYSIS: Two reviewers (JF and MS) independently assessed trial quality and extracted data. Study authors were contacted for additional information.

MAIN RESULTS: Thirty studies involving 2,391 women were included, however only 15, involving 910 women, could be included in the meta-analysis. The meta-analysis showed that danazol and progesterone + oestrogen add-back are protective of BMD at the lumbar spine both during treatment and for up to six and twelve months after treatment, respectively. Between the groups receiving GnRHa and the groups receiving danazol/gestrinone, there was a significant difference in percentage change of BMD after six months of treatment, the GnRH analogue producing a reduction in BMD from baseline and danazol producing an increase in BMD (SMD -3.43, 95 % CI -3.91 to -2.95). Progesterone only add-back is not protective; after six months of treatment absolute value BMD measurements of the lumbar spine did not differ significantly from the group receiving GnRH analogues (SMD 0.15, 95 % CI -0.21 to 0.52). In the comparison of GnRHa versus GnRHa + HRT add-back, that is oestrogen + progesterone or oestrogen only, there was a significantly bigger BMD loss in the GnRHa only group (SMD -0.49, 95 % CI -0.77 to -0.21). These numbers reflect the absolute value measurements at the lumbar spine after six months of treatment. Due to the small number of studies in the comparison we are unable to conclude whether calcium-regulating agents are protective. No difference was found between low and high dose add-back regimes but again only one study was identified for this comparison. Only one study comparing GnRH analogues with placebo was identified, but the study gave no data. No studies comparing GnRH with the oral contraceptive pill (OCP) or progestagens were identified.

REVIEWER'S CONCLUSIONS: Both danazol and progesterone + oestrogen add-back have been shown to be protective of BMD, while on treatment and up to six and 12 months later, respectively. However, by 24 months of follow-up there was no difference in BMD in those women who had HRT add-back. Studies of danazol versus GnRHa did not report long-term follow-up. The significant side effects associated with danazol limit its use.