Recognising, understanding and managing endometriosis

Endometriosis is defined as the presence of tissue lesions or nodules, histologically similar to the endometrium, at sites outside the uterus. It is a highly variable condition that has a wide spectrum of symptoms. The aetiology of endometriosis is probably multifactorial, with a strong familial component recognised. Women with endometriosis have multiple disturbances of function in the eutopic endometrium that women without the disease do not have. A firm diagnosis of endometriosis is rarely possible in general practice. The ‘gold standard’ for the diagnosis of pelvic endometriosis is currently a diagnostic laparoscopy.

Endometriosis is a chronic recurring disease that is oft en left undiagnosed. A high level of suspicion is required in women who present with pelvic pain.
'One of the most famous sufferers from endometriosis was Marilyn Monroe. The condition was so severe that it destroyed her marriages, her wish for children, her career and ultimately her life. In days before eff ective conservative surgery or effective medical therapies, it led to progressively increasing use of strong analgesics, tranquilisers and hypnotics -and drug dependency.' [1] Fortunately, few cases nowadays end like poor Marilyn's, but a minority of cases are still almost as diffi cult to manage. Endometriosis is a word that is now widely recognised by Australian women (indeed, widely called 'endo' in the community), yet it is a disease that is still not accorded as much serious att ention as it ought to be by doctors. It is extremely common. [2] Specialist gynaecologists see numerous patients in whom there has been a substantial delay in diagnosis, oft en with classic symptoms overlooked for many years. In many developed countries the average delay between the onset of symptoms and diagnosis is still six to seven years. This can be greatly extended to up to 10 years when pain symptoms begin in the teenage years.

DEFINITION
Endometriosis is still conventionally defi ned as the presence of tissue lesions or nodules that are histologically similar to the endometrium, but are present at sites outside the uterus. The incidence and prevalence of endometriosis cannot be accurately determined because of the uncertainties in making a defi nite diagnosis without laparoscopy. It is thought to aff ect up to 5 to 10% of women of reproductive age, although in those presenting with pelvic pain and/or infertility its frequency may reach 50%.

Aetiology
The aetiology of endometriosis is still controversial and is almost certainly multifactorial. [3] A strong familial component has been recognised for many years, with a young woman who has a sister or mother with proven disease having an eightfold greater risk of developing endometriosis than a woman with no family history. This very important underlying familial predisposition is probably infl uenced by a multiplicity of genes with different functions, which have proven diffi cult to identify. [4,5] Other factors that appear to play important roles in determining if a woman will develop the clinical condition include: • reproductive lifestyle, especially a delay in child-bearing • poorly understood immunological factors • some environmental factors, probably including exposure to a range of environmental toxins • reproductive tract occlusion, such as an imperforate hymen.
The theory that toxic environmental substances with a weak oestrogenic eff ect play an important role is gradually gaining experimental support. [6] Mechanisms of development The most widely accepted mechanism by which endometriosis develops is through retrograde menstruation of viable fragments of endometrium, which are then able to implant on the peritoneal surface. [7] However, this cannot be the only mechanism of development, because retrograde menstruation occurs in almost all women and most do not develop the clinical features of endometriosis. It seems probable that there needs to be abnormalities of function, either within the eutopic endometrium lining the uterus of women predisposed to endometriosis and/or impaired immune surveillance mechanisms normally responsible for the recognition and removal of endometrial fragments, which fi nd their way into the peritoneal cavity.
Evidence to support this theory includes the observed distribution of the most common peritoneal endometriotic lesions close to the fi mbrial ends of the fallopian tubes on the broad ligaments, the uterosacral ligaments, the pouch of Douglas and the surface of the ovaries [ Table 1]. Also, endometrial cells recovered from the pelvic cavity at the end of menstruation are viable and capable of proliferating. These same endometrial cells express adhesion molecules that allow them to att ach to the peritoneal surface. They may also express the aromatase enzyme (allowing them to synthesise active local oestrogen) and have the capacity to secrete angiogenic and neurogenic molecules (encouraging tissue growth and potential pain-generation mechanisms within the developing lesions).
A novel concept developed by Australian researchers proposes that women who develop endometriosis are shedding endometrial stem cells from basal endometrium during menstruation. These rare cells (only present in very small numbers in the endometrium) have the unique ability to produce progenitor cells, which in turn have a great capacity for proliferation and perhaps the potential to initiate a lesion from a single cell.

Endometriosis lesions
Peritoneal lesions mature through a series of stages from clear surface papules to 'red fl are' lesions with active new blood vessel formation [ Figure 1A and B]. They then progress to white scarred lesions and mature 'black' lesions containing inspissated old blood deposits under the peritoneal surface. This progression is calculated to take about 10 years. Endometriomas containing large volumes of altered menstrual blood products are usually only seen within the ovaries [ Figure 2] and can grow to more than 10 cm in diameter. These are actually pseudocysts, which invaginate into the surface of the ovary from adherent lesions on the pelvic sidewall. Deep pelvic nodules of endometriosis may develop in the rectovaginal septum. These tend to contain a great deal of fi brous tissue and have similar histological appearances to adenomyosis. There is controversy as to whether these diff erent types of endometriosis actually constitute diff erent forms of the disease or whether they are merely the result of accidental development of the ectopic deposits in diff erent pelvic sites.

The eutopic endometrium
It is now clear that women with endometriosis have multiple disturbances of function in the eutopic endometrium (the normally situated endometrium inside the uterus) that women without the disease do not have. [8] These functional disturbances are currently being studied using a range of powerful laboratory techniques including genomics, proteomics and immunohistochemistry.
This research has so far revealed that there are major changes in the presence and function of: • many types of endometrial immune cells capable of regulating multiple key metabolic processes • a wide range of endometrial proteins [ Table 2] including structural cell proteins -for example, cytokeratins, vimentin and actin • apoptosis (the process of organised or 'programmed' cell death) • cellular adhesion molecules, growth factors, signalling molecules and aromatase enzymes • angiogenesis (the process of new blood vessel formation) • lymphangiogenesis (new lymph vessel formation) • neurogenesis (the growth of new nerve fi bres).
There also seems to be a general reduction in response of the tissue to stimulation by progesterone.
Our latest research from the University of Sydney has revealed that fi ne sensory nerve fi bres are present throughout the functional layer of endometrium in all women with endometriosis, but not found in those without endometriosis. [9] The density of nerve fi bres in the myometrium is increased and a high concentration of nerve fi bres are also present in the ectopic peritoneal endometriosis lesions. These nerve fi bres must play some role in pain-generating mechanisms.
Following demonstration of these multiple functional and microanatomical abnormalities within eutopic endometrium, with no obvious single common underlying molecular abnormality, it is hypothesised that endometriosis is an example of a disease where multigene activation occurs. The presence of multiple genes modulated by many co-activators and co-repressors could help to explain the great variability of endometriosis.
This theory does not explain the cause of all cases of endometriosis or fully explain the implantation theory. In some situations there is evidence to support the involvement of coelomic metaplasia, development from embryonic rests (a fragment of embryonic tissue that has been retained after the period of embryonic development), lymphatic and vascular metastases and indeed a combination of some or all of these theories. Variations in pathogenesis may explain the development Table 2: Probable complex pathological mechanisms underlying endometriosis Probable primary anomalies of endometrial function • Increased intrinsic aromatase enzyme activity, resulting in local oestrogen production • Reduced apoptosis, resulting in greater viability of menstrually shed tissue fragments • Increased production of intercellular adhesion molecules, resulting in an increased likelihood of cells attaching to the peritoneum • Increased secretion of angiogenic molecules, such as vascular endothelial growth factors, which stimulate the growth of new blood vessels and lymph vessels • Decreased response of endometrial mechanisms to progesterone stimulation • Greatly increased secretion of neurogenic molecules, such as nerve growth factor and nerve growth factor receptor, which stimulate the development of new sensory nerves within the endometrium, myometrium and endometriotic lesions • Several changes in immune cell populations, such as macrophages, dendritic cells, plasma cells and their associated secretory products, including cytokines and many related molecules • Anomalies have been shown in many other molecular systems, such as the generation of free oxygen radicals, structural proteins, many growth factors and signalling molecules Probable anomalies within the peritoneum, peritoneal fl uid and ectopic lesions • Increased numbers of macrophages, but they appear to have reduced capability to 'mop up' cell fragments • Disturbance of other immune cells, such as dendritic cells (decreased) and plasma cells (increased) • Increased expression of many molecules, such as prostaglandins, which have the potential to stimulate sensory nerve fi bres • Many molecular systems refl ecting the functional anomalies seen in eutopic endometrium Differences between endometriotic lesion and eutopic endometrium functions • There are many differences between endometrium from women without endometriosis and the function of endometriotic lesions, as could be predicted from the list of functional anomalies seen in endometrium from women with endometriosis. However there also appear to be subtle differences between the function of lesions and function of the eutopic endometrium from which they arose • These subtleties of functional difference are currently under intense study, but one obvious difference is that hormonal therapy almost completely eliminates nerve fi bres from the functional layer of eutopic endometrium, but it does not eliminate them from ectopic lesions of endometriosis in other anatomical sites [ Table 1].
There is also increasing evidence to support the hypothesis that endometriosis is primarily an 'endometrial' disease. The many endometrial functional disturbances described appear to occur before the development of ectopic lesions and these key changes in the endometrium of women with endometriosis appear essential for att achment, viability, growth and innervation of fragments of menstrually-shed endometrium in ectopic sites. The key functional anomalies appear to be the expression of intercellular adhesion molecules, the presence of local aromatase enzyme activity, decreased apoptosis, increased angiogenesis and increased neurogenesis [ Table 2].

Symptom variability
Endometriosis is a highly variable condition that has a wide spectrum in the experience of symptoms, age of presentation, anatomical extent of disease, incidence of infertility, response to treatment, likelihood of recurrence, rate of progression and natural long-term history of the condition. Indeed, a small minority of women will exhibit this condition without experiencing any symptoms. It is unclear if this great variability is genetically predetermined or is infl uenced by exposure to lifestyle or environmental factors.
It seems that the great variability of the condition is partly responsible for the considerable delay in diagnosis that oft en occurs. Sadly, this delay still occurs in many women who present with the classic symptoms of considerable dysmenorrhoea or chronic pelvic pain. A high index of suspicion should be maintained in all women with pelvic pain, even when the presentation may be atypical, because although the spectrum of symptoms is highly variable, the great majority of women with signifi cant endometriosis will present with some form of pelvic pain [ Table 3]. A minority will not be aware of any substantial symptoms or may complain solely of infertility.
The majority of women with some form of pelvic pain will also report disturbances of uterine bleeding, multiple bowel symptoms (oft en misdiagnosed as irritable bowel syndrome), painful abdominal bloating, cyclical urinary symptoms, extreme cyclical lethargy or occasionally very unusual symptoms [ Table 3]. The progress of the condition is also unpredictable.

Natural history of the condition
The natural history of endometriosis is poorly understood, because few patients will progress through their reproductive lifespan without att empts at eff ective therapy. [10] In a small proportion of patients symptoms will resolve spontaneously, but in most the symptoms will persist or worsen. Some women appear to be 'cured' of the condition aft er eff ective surgery, but most will experience a recurrence of symptoms. Endometriosis usually becomes inactive aft er the menopause and few women will continue to experience symptoms beyond this time. The very occasional woman who experiences ongoing symptoms aft er the menopause appears to have nodules of endometriosis that are capable of producing local oestrogen through activity of the aromatase enzyme.

Special features of adolescent endometriosis
Endometriosis appears to begin in adolescence more frequently than generally recognised and there is oft en an inordinate delay before a diagnosis is made (10 years or more). These girls typically have a strong family history. Menstrual pain oft en begins at or soon aft er the menarche and may initially be typical of severe primary spasmodic dysmenorrhoea beginning on about day one of menses and then gradually developing the more typical combinations of endometriosis symptoms.
Our belief is that these young women already have the genetically determined abnormalities of endometrial function that lead to the generation of pain within the endometrium and myometrium. As the typical peritoneal or ovarian lesions of endometriosis develop, innervation of these lesions rapidly follows and pain generation then begins from more distant parts of the pelvic cavity, such as the uterus. This leads to the more 'congestive', dragging and gnawing diff use pelvic pain and backache of classic endometriosis, which typically begins well before menstruation.

The challenges of diagnosis
A fi rm diagnosis of endometriosis is rarely possible in general practice, but a high index of suspicion can be applied.
Features that may lead to specialist referral are summarised in the Annex 1. Several international research groups are currently attempting to develop simple symptomatic screening algorithms or diagnostic decision tools that will assist in the diagnosis of endometriosis in general practice. The 'gold standard' for the diagnosis of pelvic endometriosis is currently a diagnostic laparoscopy (and oft en a peritoneal biopsy) by an experienced gynaecologist for confi rmation or exclusion. [11] Even this presents diffi culties such as the atypical appearances of peritoneal lesions (particularly in adolescents) or deep and 'partially hidden' lesions, especially for the inexperienced surgeon.
Transvaginal ultrasound may make a strong presumptive diagnosis of an ovarian endometrioma, but not of other types of endometriosis and other imaging techniques are rarely helpful.
Much recent energy has been invested into research to fi nd a specifi c blood test for endometriosis but with limited success. Genomic chip technologies and proteomic techniques hold considerable future promise, although they are likely to be expensive at fi rst. We have developed a highly promising endometrial biopsy technique looking for unmyelinated nerve fi bres, which has an extremely high specifi city and sensitivity (almost 100%), equivalent to that of diagnostic laparoscopy. [12] Its potential use in routine clinical practice requires much more research. • Perimenstrual diarrhoea and/or constipation and troublesome painful abdominal bloating, often mimicking irritable bowel syndrome but not relieved by a bowel motion • Generalised pelvic discomfort; erratic, sudden acute, sharp or dull pelvic pain; peri-ovulatory pain • Urinary urgency or bladder discomfort, pain with urination, rarely haematuria • Other vicarious menstruation (e.g. haemoptysis); rare • Infertility *Many of these symptoms are 'classic' and well recognised and should not be overlooked. However, the combination of presenting symptoms can be highly variable and a high index of suspicion is needed.

Annex 1: When to refer a patient with endometriosis
• 'Typical' symptoms of congestive, dragging dysmenorrhoea with other pelvic pain symptoms (deep dyspareunia, pain with a bowel motion or 'ovulation' pain) which interfere with lifestyle. • Severe dysmenorrhoea of any pattern unresponsive or poorly responsive to several courses of different standard analgesics (paracetamol or prostaglandin inhibitors). • Severe or increasing localised or positional deep dyspareunia or tenderness at the vaginal vault or on cervical excitation at vaginal examination). • Pelvic pain in a woman with a moderate or strong family history of endometriosis. • Infertility of more than one year duration, especially when the woman is over 30 years and sooner when some dysmenorrhoea is present.
• The presence of the following symptoms in someone with pelvic pain: menstrual bleeding prolonged to more than eight days; painful, cyclical abdominal bloating (with or without other gastrointestinal or 'irritable bowel' type symptoms, including cyclical diarrhoea and constipation); pain with urination.

The role of imaging
Transvaginal ultrasound carried out by operators with specialised experience in the assessment of female pelvic disorders is now the mainstay of female pelvic imaging and remarkable structural and vascular detail can oft en be detected. Nevertheless, it is important to understand that ultrasound will generally only demonstrate endometriotic lesions, which form cysts or endometriomas in the ovaries. Peritoneal lesions cannot be demonstrated and, in most cases, neither can deep pelvic nodules in the rectovaginal septum without specialised techniques.
Ovarian endometriomas usually give a typical 'echogenic' signature on ultrasound but haemorrhagic corpus luteum cysts, which are very common, can mimic endometriomas. A repeat ultrasound in two or six weeks to avoid the next luteal phase may be necessary to confi rm that the 'haemorrhagic cyst' is indeed a persistent endometrioma.

Active management
Increasingly, the effective management of endometriosis requires an individual approach for each aff ected woman, because of the great variability of the disease. [11] The Cochrane Collaboration's approach to systematically reviewing the highest quality randomised trials is important when determining the primary management of women with endometriosis, but may be diffi cult to apply in the many women in whom initial therapies are ineff ective or in whom recurrences occur.
Initial eff ective management generally requires the major pain symptoms to be controlled using standard analgesics, followed by surgical excision of the lesions or medical (usually hormonal) suppression of the lesion's activity. Some of the many possible therapies are listed in Tables 4 and 5.
Endometriosis is such a variable condition that individualisation of the approach to management is important in all but the initial approaches to active treatment. In certain circumstances, almost any surgical or medical approach may be the optimal initial management, although options always need to be discussed with the patient and their partner or family, if appropriate. This is a challenging and ongoing condition and it is always valuable to include a second person who the patient trusts in the consultations.

Analgesic management
Analgesic management of endometriosis requires gradual progression from paracetamol to the nonsteroidal prostaglandin inhibitors (generally little to choose between the alternatives), followed by compound codeine alternatives and then intermitt ent tramadol (Durotram XR, Tramahexal SR, Tramal, Tramedo, Zydol) in adequate doses as dictated by the exacerbations of this condition. For exceptionally severe episodes, oxycodone (Endone, OxyContin, OxyNorm, Proladone), pethidine and morphine (Anamorph, Kapanol, MS Contin, MS Mono ordine, Sevredol) may be considered, but frequently repeated dosage should be avoided.

Surgical excision
Initial surgery will almost always involve gynaecological laparoscopy for diagnosis, assessment and/or active surgical excision. Thorough surgical excision can be diffi cult and relatively few practitioners have the advanced laparoscopic • Analgesic, antipyretic (no anti-infl ammatory activity); mainly central action • Inhibition of the pain-mediating effects of peripheral prostaglandins; anti-infl ammatory • Analgesic, antipyretic • Codeine potentiates the action of paracetamol • Weak narcotic; has other actions; very rarely addictive • Narcotic alternative to morphine • Narcotics; act on central opioid receptors • Complex gamma-aminobutyric acid uptake inhibitor surgical skills necessary for some of the more diffi cult excisions, although more young trainees are being mentored through these diffi cult learning curves. Surgery may merely involve excision of small and easily visible peritoneal lesions, but many of these lesions extend deeply under the peritoneal surface and a minority involve the rectum and structures adjacent to the rectovaginal septum. Therefore, these women may require the surgical att ention of a team of experts, including colorectal or urological specialists, for safe and eff ective excision.
Surgical excision of ovarian endometriomas is one of the most controversial surgical approaches to the management of endometriosis. Ovarian endometriomas can be excised at laparoscopy or laparotomy, depending on the size of the pseudocyst and the skill of the surgeon, but the controversy focuses on the amount of tissue that is removed. 'Stripping' of the cyst wall removes some normal underlying ovarian cortex and only small areas of this tissue contain endometriosis. The main area of active endometriosis is around the site of att achment to the pelvic sidewall [ Figure  2] and this needs to be fully excised. The best approach to identify and limit excision of the residual active areas inside the pseudocyst remains to be defi ned.
There is clear evidence that women with infertility

Mode of action
Rapid, short-term hormonal suppression of disease symptoms Gonadotrophin-releasing hormone analogues Goserelin implants (Zoladex) and nafarelin nasal spray (Synarel) for a maximum six-month course Danazol (Azol) and gestrinone (Dimetriose) 'Pseudo-menopause' to very effectively reduce ovarian oestrogen secretion for the duration of therapy. These therapies may be combined with oral progestogen as add-back therapy to alleviate side effects of hot fl ushes (e.g. 10 mg/day medroxyprogesterone acetate [Provera, Ralovera]; 5 mg/day norethisterone [Primolut N]) Weak, impeded androgenic hormonal environment with inhibition of endometrial growth and function; very effective but less popular now because of mild, dose-related androgenic side effects in a minority of women and other progesterone-receptor modulators are effective at suppressing endometrial/endometriotic proliferation and infl ammatory functions; new developments are underway associated with endometriosis benefi t greatly from surgical excision of the disease, [13] whereas medical therapy has litt le benefi cial eff ect. [14] Assisted reproductive technologies, such as IVF, are successful at overcoming infertility associated with endometriosis, although perhaps less successful than with other causes of infertility.

Medical therapy
Medical therapy involves sound analgesic management followed by hormonal suppression at one of two levels. Highly eff ective suppression is provided by the gonadotrophin-releasing hormone (GnRH) analogues, especially the goserelin implants (Zoladex) given over six months, but danazol (Azol) provides very similar suppression. The side eff ects of both these medications are well known and include vasomotor symptoms for the GnRH analogues and mild androgenic changes for danazol and need to be addressed with counselling and regular follow up. Hirsutism or voice changes with danazol should dictate cessation of the medication.
Longer-term suppression is becoming more of an issue with the recognition that surgery does not usually off er such a long-term benefi t as once believed. The progestogens off er the most eff ective levels of suppression, although not all patients respond well. Dosages may need to be individualised. The oral contraceptive pill may work satisfactorily as an initial approach to medical therapy, but only a minority do well. There is a gradual move to use the intrauterine (Mirena) and subdermal (Depo-Provera, Depo-Ralovera, Implanon Implant) progestogen delivery systems for their convenience and effi cacy. Diffi cult cases may do even bett er with a combination of these two delivery systems, [15] but this is specialised therapy for women who fail to respond to other therapies.

Prevention
Physicians are beginning to think about how they might be able to identify those women in whom recurrence aft er surgery is likely and, therefore, how they might provide eff ective therapy to prevent recurrence. The natural extension of this is to consider how to identify those women who are most likely to develop endometriosis and then produce and apply preventive lifestyles and therapies that may obviate or at least delay the onset of the disease process. [16] Although there are few data in the literature to address this issue directly, it is likely that any prolonged exposure to Fraser: Managing endometriosis Annex 2: Endometriosis: Key messages • Endometriosis is a highly variable condition in its age and mode of presentation, range of symptoms, anatomical sites, response to treatment and likelihood of recurrence. • Endometriosis is a chronic, recurring disease in many women and can have a devastating effect on lifestyle in a minority of them. • The aetiology of endometriosis is complex and unclear. Although a strong genetic component is common, reproductive and environmental factors are often important. It may be primarily an 'endometrial' disease. • Endometriosis can be asymptomatic, but usually presents as pelvic pain with or without infertility. • The gold standard for the diagnosis of endometriosis is still laparoscopy, but other less invasive approaches through endometrial biopsy and blood tests are being developed. • Medical and surgical treatments of endometriosis need to be individualised and many women will require both approaches at different times during the course of their disease. • A substantial proportion of women with endometriosis may gain long-term relief after thorough excisional laparoscopic surgery by an expert. • Effective long-term medical suppression and prevention of recurrence of endometriosis is becoming a reality. • Endometriosis is usually spontaneously cured at menopause, but may occasionally persist, especially during oestrogen-only hormone replacement therapy.

Information for patients
The overall quality of medical information available through the internet has improved substantially in recent years, but is still of highly variable individual quality. Many websites, both professional and consumer, offer specifi c services and may therefore present a degree of bias. The following websites provide sound consumer information: • The Endometriosis Association (USA) invests a substantial amount of effort and money into good-quality information and consumerorientated research -www.endometriosisassn.org • The Endometriosis Association of Victoria also provides information relevant to Australian women -www.qvwc.org.au/infohub/ health_and_wellbeing (joint website with the Queen Victoria Women's Centre; type 'endometriosis' in the search box) • The Global Forum for Information about Endometriosis -www.endometriosis.org • The National Women's Health Resource Center provides a wide range of sound information -www.healthywomen.org (type 'endometriosis corner' in the search box) • The Jean Hailes Foundation for Women's Health, based in Melbourne, provides excellent, sound consumer information and regular consumer forums -www.endometriosis.org.au • Some women respond well to the opportunity to exchange information in a 'blog' forum such as www.chronicbabe.com or www.
howtocopewithpain.org progestogens alone or even to combined oral contraceptive preparations, may be eff ective in delaying, minimising or preventing the onset of symptomatic endometriosis in many women. It is hypothesised that because of its highly focused local pro-gestogen eff ect the levonorgestrel intrauterine system will have a dramatic role in the prevention of endometriosis and there are limited published data to support this concept. [17] A large scale multicentre trial is needed to address this concept. There is considerable optimism that in the future progesterone-receptor modulators will have an even bett er preventive eff ect, both of primary disease and recurrences.

New research and information access on endometriosis
Endometriosis is one of the most actively researched of all benign gynaecological diseases, yet there are many unanswered questions about it. It is still an 'enigmatic' disease.
The World Endometriosis Society has been formed and runs a triennial World Congress on Endometriosis, the most recent of which was held in Melbourne in March 2008. [18] A remarkable number of lectures, presentations, videos, posters and live operating sessions were presented, indicative of the vast amount of research that is currently ongoing. As in most areas of medicine, Australia is contributing productively at the forefront of this research in several fi elds. A collaborative research group called the World Endometriosis Research Foundation is a joint initiative of the World Endometriosis Society, the American Society for Reproductive Medicine and the European Society for Human Reproduction and Embryology. [19] Active progress is being made in understanding the aetiology and pathogenesis of endometriosis, in improving diagnostic and assessment techniques, in developing new approaches to medical therapy, in gaining confi dence in the use of longterm medical therapy for the suppression and prevention of recurrence and in the application of new endoscopic excisional skills. We still have litt le understanding of the mechanisms of generation of pain and other symptoms and have so far been unsuccessful in developing successful screening algorithms for reliable earlier diagnosis.

CONCLUSION
Endometriosis is a complex and highly variable disease that still challenges modern science in every aspect of presentation, symptoms, aetiology and mechanisms, diagnosis and management. Key aspects of understanding and managing endometriosis are highlighted in the Annex 2. Resources for patients are given in the box on this page.