Methods

Top Abstract Methods References

Clinical Evidence search and appraisal for systematic reviews, January 1999. We sought RCTs by electronic searching of databases, handsearching of 30 key journals, searching the reference lists of other RCTs, and identifying unpublished studies from abstracts, proceedings, and pharmaceutical companies. We used the search strategy and database of the Cochrane Menstrual Disorders and Subfertility Group to identify RCTs on Medline and Embase. We included RCTs that used adequate diagnostic criteria for inclusion of participants (endometriosis diagnosed either by laparoscopy or laparotomy in association with dysmenorrhoea, dyspareunia, other pelvic pain, or infertility) and clinical outcomes. Studies of assisted reproductive technologies were not included.

Question: What are the effects of hormonal treatments?

Four small systematic reviews found that all hormonal treatments except dydrogesterone were equally effective, compared with placebo, in reducing pain attributed to endometriosis. One systematic review of small RCTs found no evidence that hormonal treatments improved fertility. RCTs have found that six months' postoperative treatment with gonadotrophin releasing hormone analogues or a combination of danazol plus medroxyprogesterone reduces pain and delays the recurrence of pain significantly better than placebo. They found no evidence of an effect of postoperative gonadotrophin releasing hormone analogues on fertility.

Benefits
In women with pain attributed to endometriosis: We found four systematic reviews (search dates 1998, 1996, 1997, 1997) comparing six months' continuous suppression of ovulation (using danazol, gestrinone, medroxyprogesterone acetate, dydrogesterone, oral contraceptives, or gonadotrophin releasing hormone analogues) and placebo.^14-17 None of the reviews included more than 200 women. They found that except for dydrogesterone, which was given at two different dosages in the luteal phase, with no evidence of effect, all treatments were equally effective at reducing severe and moderate pain at six months. We found no placebo controlled RCTs of oral contraceptives. One RCT compared oral contraceptives and gonadotrophin releasing hormone analogues in 49 women.¹⁸ It found no differences in rate of relief for all types of pain except menstrual pain, for which oral contraceptive was better. In women with subfertility attributed to endometriosis: We found one systematic review (search date 1996), which identified four RCTs in a total of 244 women.¹⁹ The trials evaluated six months' treatment with danazol, medroxyprogesterone, or gonadotrophin releasing hormone analogues compared with placebo. They found no evidence of an effect on the probability of pregnancy (odds ratio for pregnancy compared with placebo 0.83; 95% confidence interval 0.50 to 1.39). In women who have undergone surgery: We found no systematic review. We found three placebo controlled RCTs of gonadotrophin releasing hormone analogues in a total of 443 women who had undergone surgery for endometriosis.^20-22 The smallest trial (n=65) evaluated three months' treatment and found no difference in pain relief; the two larger trials (n=109, n=269) evaluated six months' treatment and found that gonadotrophin releasing hormone analogues significantly reduced pain scores (P=0.008)²¹ and delayed the recurrence of pain by more than 12 months. ^{21 22} Two of the trials evaluated fertility (n=65, n=269) and found no difference in pregnancy rates or time to conception. ^{20 22} One RCT in 60 women found that postoperative treatment with a combination of danazol (600 mg/day) and medroxyprogesterone (100 mg/day) reduced pain more than placebo six months after surgery.²³

Harms
Gonadotrophin releasing hormone analogues: Adverse effects occurred in 11% of women taking gonadotrophin releasing hormone analogues,¹⁴ which are associated with hypo-oestrogenic symptoms such as hot flushes and vaginal dryness. RCTs have found that adding oestrogen, progesterone, or tibolone significantly relieves hot flushes caused by gonadotrophin releasing hormone analogues (by 50% or more on symptom scores). ^{13 24 25} Danazol: Adverse effects occurred in 15% of women taking danazol.¹⁵ Danazol is associated with androgenic symptoms of acne, weight gain, and hirsutism and with decreased breast size, muscle cramps, and hunger. Gestrinone is associated with a higher frequency of hot flushes than are gonadotrophin releasing hormone analogues, and also with greasy skin and hirsutism.¹⁵ Medroxyprogesterone: The trials gave no information on adverse effects of medroxyprogesterone.

Comment
The trials were mainly small, with no long term follow up. Most trials compared gonadotrophin releasing hormone analogues with danazol. No summary statistics could be calculated because the trials compared different drugs with placebo or with no treatment.

Benefits
We found one systematic review (search date 1998), which identified two small RCTs in women with endometriosis. These found no difference in pain relief between women treated with laparascopic uterine nerve ablation and those who were not.²⁶

Harms
The trials gave no information on adverse effects. Potential harms include denervation of pelvic structures and uterine prolapse.²⁶

Comment
The trials may have been too small to rule out a beneficial effect.²⁶

Benefits
In women with pain attributed to endometriosis: We found no systematic review and no RCTs evaluating laparoscopic ablation of deposits alone. We found one RCT comparing the combination of ablation of deposits plus laparascopic uterine nerve ablation with diagnostic laparoscopy in 63 women. ^{27 28} This found that combined ablation reduced pain at six months (median decrease in pain score 2.85 for ablation, 0.05 for diagnostic laparoscopy; P=0.01). In women with subfertility attributed to endometriosis: We found no systematic review. We found one RCT comparing laparoscopic ablation or excision of endometriotic deposits and diagnostic laparoscopy in 341 women with subfertility attributed to mild or moderate endometriosis.²⁹ This found that laparoscopic surgery increased cumulative pregnancy rates (relative risk of pregnancy after 36 weeks 1.7, 1.2 to 2.6; NNT 8). Laser versus diathermy ablation: We found no RCTs.

Harms
The trials gave no information on adverse effects.^27-29 Potential harms include adhesions, reduced fertility, and damage to other pelvic structures.

Comment
Further RCTs comparing ablation alone and ablation plus laparascopic uterine nerve ablation are under way (C Sutton and R Dover, personal communication). A systematic review of laser versus diathermy ablation is planned (C Farquhar and N Johnson, personal communication).

Benefits
We found no systematic review. We found one RCT comparing three months' preoperative treatment with a gonadotrophin releasing hormone analogue and no treatment in 75 women with moderate or severe endometriosis.³⁰ There was no difference in ease of surgery between the two groups.

Harms
See above under hormonal treatments.

Comment
The trial may have been too small to rule out a clinically important effect.

Benefits
We found no systematic review. We found one RCT comparing laparoscopic cystectomy and laparoscopic drainage in 64 women.³¹ In women with pain attributed to endometrioma: The trial found that cystectomy reduced recurrence of pain at two years (odds ratio 0.2, 0.05 to 0.77) and increased the pain free interval after operation (median interval 19 months v 9.5 months; P<0.05). In women with subfertility attributed to endometrioma: Cystectomy increased the pregnancy rate (66.7% v 23.5%; odds ratio 8.25, 1.15 to 59; P<0.05).

Harms
The trial reported no intraoperative or postoperative complications in either group.

Comment
None.