Chapter 10

Premature Ovarian Failure

Premature ovarian failure is a term used to indicate a syndrome of amenorrhoea, low oestradiol, and high gonadotrophins levels in women under the age of 40 years. A statistical definition has been suggested which related these clinical and endocrine parameters to an age group two standard deviations below the mean age of the menopause, estimated for the reference population (1). This is rather a complicated definition, and the fixed age-related one is more popular in clinical practice.

It is not only the definition of the syndrome which drew some debate. The name itself has been a subject of some controversy. The terms premature menopause, premature ovarian dysfunction and hypergonadotrophic hypogonadism have all been used. Each of these terms has its drawbacks, and may not represent all aspects of the syndrome. Premature menopause is definitely not a suitable option, as the term menopause indicates permanent cessation of menstruation. Ovarian activity can resume in many patients, at different times. In fact 50% of patients with POF may menstruate (2, 3), but the exact timing or pattern of such intermittent recovery can not be predicted. Furthermore, the term itself has devastating psychological effects on young women in their early 20s or 30s. It is a term usually related to mothers and grandmothers and not to young women in the prime of their youth. Furthermore, many women with POF may have primary amenorrhoea, and the term premature menopause will not be a valid one to use. Hypergonadotrophic hypogonadism and premature ovarian insufficiency sound softer on the patients. The later term was used first by Albright et al in 1942 (4), and seems to be the most appropriate. It is also suitable to cover many women who had regular menstruation, polymenorrhoea or oligomenorrhoea and a hypergonadotropic state. Though they are not amenorrhoeic, they still have reduced fertility potential and can be hypo-oestrogenic at different times. Few of them may even have hot flushes. Nonetheless, premature ovarian failure stood the test of time, and remained the most popular term to use, though the word failure also gives a bad psychological impact of insufficiency and personal inadequacy.

To complicate matters yet further, there is no general agreement on the criteria necessary to make a diagnosis. Different authors have documented different durations of amenorrhoea (3-6 months), and levels of follicle stimulating hormone (FSH) from 10 – 40 IU/L, before making a diagnosis. All these points lead to hesitancy on the part of clinicians to make a diagnosis. It can take a patient many years, and seeing many doctors before a diagnosis is finally made. Alzubaidi et al (5) reported that 50% of young patients who had secondary amenorrhoea needed to see three or more clinicians before investigations were started. Accordingly, it is imperative that any young patients with history of 3 months secondary amenorrhoea or irregular menstruation should be investigated (6), especially those with significant personal or family history, as will be discussed later on in this chapter.

The incidence of premature ovarian failure has been reported differently in different societies and in relation to age as well. A total prevalence of 0.3 – 1.0% has been frequently quoted. It affects 1.0% of women younger than 40 years, 0.1% under the age of 30 years, and 0.01% of women under the age of 20 years (7). These figures can be affected by all the points discussed before in relation to the definition, as well as the delays in diagnosis. Furthermore, more young women are expected to be diagnosed with POF, due to the higher success rates for treating childhood malignancies (8). Also ethnic differences may have an impact as reported by Luborsky et al in 2002 (9). An incidence of 1.4% was quoted for African-Americans and Hispanics, 1.0% for Caucasians, 0.5% for Chinese, and 0.1% for Japanese. Both mono and dizygotic twins are more liable to develop POF. A combined Australian and British study involving 832 sets of twins showed 3- to 5-folds increased prevalence of POF in twins than the general population at age thresholds 40 and 45 years. No specific factors were found to account for this higher risk of early menopause. In spite of significant differences within monozygotic twins, the age range at the menopause was more harmonious than for dizygotic twins, confirming a hereditary component to the timing of the menopause (10).

Patients with POF can present with primary amenorrhoea, secondary amenorrhoea, or even irregular and delayed menstruation. About 10-28% of patients with primary amenorrhoea may have POF vs. a prevalence of 4-18% in women with secondary amenorrhoea (11).

Causes of POF

Premature ovarian failure can occur spontaneously, or may follow certain medical interventions. No specific cause may be found in more than 90% of spontaneous cases, but the following explanations were suggested:

1. There was reduced differentiation of the cohort of oocytes in the ovaries during fetal life.
2. There was rapid exhaustion of the oocytes pool during fetal life, childhood or early adult life. Up to the stage of sex cords development within the ovaries, one X chromosome is needed, but two X chromosomes are necessary for further development beyond the stage of primary oocytes. In the absence of all or part of the second X chromosome, rapid atresia of the follicles occurs, leading to the formation of streak gonads or premature ovarian failure, depending on the atresia rate.
3. Resistance of the follicles to gonadotrophins stimulus is mostly related to mutations in FSH and LH receptors genes.

The first two points can be related to problems with the oocytes themselves, and have been discussed in great details in Chapters 2 and 3, especially in relation to gonadal dysgenesis and different chromosomal combinations. This chapter has been written with this information in mind, and further repetition will be avoided. On the other hand, the third point relates to granulosa cells dysfunction.

The aetiology of spontaneous POF will be discussed in relation to the following causes:

1. Genetic or familial causes;
2. Autoimmune factors;
3. Infections.

On the other hand, iatrogenic causes are easier to diagnose, and may follow one or more of the following interventions:

1. Pelvic surgery;
2. Pelvic irradiation;
3. Use of cytotoxic drugs.

Spontaneous POF

There are differences between sporadic and familial cases of spontaneous POF, which should be taken into consideration when counselling or investigating these patients. Accordingly, thorough history should be taken to prevent excessive and unnecessary investigations and exhaustion of resources, as only 20-30% of the spontaneous cases are familial (12).

Chromosomal and genetic factors

Most of the associated genetic disorders are related to the X chromosome (13), though autosomal dominant and recessive abnormalities have been documented as well (14). Such chromosomal abnormalities and other genetic causes were more common in patients with similar family history, and in patients presenting with primary amenorrhoea. Most patients usually have normal 46XX karyotyping.  Abnormal results were reported in 50% and 13% of the patients with primary and secondary amenorrhoea respectively (15). A critical region in the long arm of the X chromosome, Xq13 to 26, excluding section Xq22, is critical for ovarian development and function (16). Accordingly, translocations in this region, and X chromosome deletions, and duplications, as well as balanced translocations are likely in patients with spontaneous POF. The most commonly known chromosomal abnormality in this condition is 45XO (X chromosome monsomy or Turner’s syndrome), or a mosaic combination including such cell line. Patients with 45XO premature ovarian failure usually have significant physical characteristics which may even obviate the need for chromosomal analysis. They have short stature, webbing of the neck, increased carrying angle of the elbows, widely separated nipples, low set ears and hair line, high arched palate and coarctation of the aorta. Another specific condition is trisomy X syndrome (47XXX) which is caused by non-disjunction of the X chromosome, but its prevalence as a cause of POF is unknown. Other conditions include deletions in either the short or long arms of the X chromosome, and balanced X chromosome translocations. The last two authors (16) suggested that all patients with POF should have their peripheral karyotyping tested. This advice is valid in cases with familial POF. It may not help with the management of sporadic individual patients, except in the presence of a Y chromosome which increases the risk of gonadal malignancy. Furthermore, patients may carry two cell lines, with the abnormal one being at the ovarian level. This has been shown by a report published by Abdel-Gadir and Ramadan in 1990 (17) regarding a 32 year old nulliparous patient who presented with secondary amenorrhoea of 7 years duration. This was preceded by 12 years of oligomenorrhoea dating back to menarche, which she attained at the age of 13 years. She had normal 46XX peripheral lymphocytic karyotyping. Ovarian tissues karyotyping revealed reciprocal translocation 45XOt(3:20)(q22:q12)t(13:15)(q14:q13). Accordingly, normal peripheral karyotyping may not reflect the exact abnormality, especially in mosaics with two cell lines. Therefore, it is not indicated as a routine test in sporadic POF cases. Despite history of some ovarian function, almost 90% of the patients were nulliparous at the time of POF diagnosis (16).

Growth differentiation factor 9 (GDF9) and bone morphogenetic protein 15 (BMP15) genes are necessary for normal human reproduction. This subject was reviewed by Hoekstra et al in 2008 (18). Higher mutation frequency in both genes was reported in patients with POF, when compared to controls. A slight increase in POF was also noted in mothers of dizygotic compared to monozygotic twins (19, 20). This was related to increased frequency of mutations in GDF9 and BMP15, though different variants of mutations were detected in patients with POF and dizygotic twins (18).

Premutations in the fragile site mental retardation 1 (FMR1) gene, which is responsible for fragile X syndrome, have been diagnosed in 6% of women with 46XX karyotyping and POF (21, 22). Such risk was found to be higher in cases with similar family history (14%), in comparison to sporadic cases (2%), as reported by Sherman in 2000 (22). About 5-10% of these patients may conceive spontaneously following resumption of ovarian function, but carry a high risk of having mentally retarded children (23). Accordingly, they should be referred for genetic counselling as well as those with known family history of fragile X syndrome, unexplained mental retardation, tremor / ataxia syndrome and dementia (23). Frequency of miscarriages is also increased in females with fragile X syndrome.

Other rare enzymatic causes and signal defects have been described, but are not frequent enough to warrant routine testing in sporadic cases. The most known conditions in this group are gene mutations of the FSH receptor (FSHR), luteinising hormone receptor, and galactose-1-phospate uridylytransferase (GALT) in patients with galactosaemia. Despite good dietary control of galactosaemia, 70-80% of the patients develop POF (24). Patients with complete or partial FSHR mutation are more likely to have primary or early secondary amenorrhoea respectively. The later group are more likely to show follicles during ultrasound examinations of the ovaries than other patients with POF (25). On the other hand, LH receptors gene mutation is usually associated with primary amenorrhoea. One further problem is partial or complete deficiency of 17a-hydroxylase which occurs in 1:50,000 – 100,000 newborns, usually in combination with 17/20 lyase deficiency. Due to failure of androgens and oestrogens production, these patients usually fail to develop secondary sexual characteristics and present with primary amenorrhoea and hypokalaemic hypertension. More related information has already been given in Chapter 3.

Autoimmune factors

Patients with POF are at risk of having other autoimmune disorders (26). A figure of 20% has been quoted for such an association by Goswami and Conway in 2005 (27). The syndrome has been associated mainly with autoimmune thyroid dysfunction, adrenal insufficiency, and diabetes. Other associations included systemic lupus erythematosus, Sjogren’s syndrome, pernicious anaemia, vitiligo, Crohn’s disease and rheumatoid arthritis. Autoimmune lymphocytic oophoritis has been suggested as the cause of ovarian failure, related to steroidogenic cell autoimmunity. Nonetheless, ovarian antibodies have not proved to be useful for the screening or diagnosis of POF. They were reported in a wide range of 10 - 69% of patients with POF, but also in a significant number of controls (28). A study by Yan et al in 2000 (29) showed significant increase in CD8 density in T lymphocytes. Though ovarian biopsy can be more diagnostic of autoimmune oophoritis, it is not justifiable because of the lack of definitive treatment as suggested by Khastgir et al in 1994 (30). On the other hand, identification of patients with adrenal antibodies allows proper follow up of patients at risk of developing adrenal failure or even adrenal crisis. About 4% of women with spontaneous POF tested positive for adrenal antibodies (31), and asymptomatic adrenal insufficiency was reported in 2% of patients with spontaneous POF. This last group would have been especially at risk of adrenal crisis had they conceived spontaneously, or through ovum donation. In contrast, Turkington and Lebovitz (32) reported that 23% of patients who already had adrenal insufficiency had POF at the same time. In these cases, POF preceded the development of clinical Addison's disease by five to fourteen years. This fact confirms the need for regular follow up of these patients. Autoimmune hypothyroidism is another common risk, and a figure of 20% has been reported by Kim et al in 1997 (33). Accordingly, all patients diagnosed with spontaneous POF should have their thyroid stimulating hormone (TSH), free thyroxine (T4), and thyroid peroxidase antibodies tested at regular intervals. Screening for the other autoimmune problems mentioned previously is not necessary, unless clinically indicated.

Other changes in the immune system which reflect its association with POF have been reviewed by Anasti (11). Increased number of activated T cells and reduced number of natural killer (NK) cells are two examples. Increased numbers of activated T cells have been described in other autoimmune endocrine disorders, which suggested a similar pathology in POF. Furthermore, reduced number of NK cells has also been shown in patients with Grave’s disease. This can affect the function of B and T cells, leading to increased production of autoantibodies, which could lead ultimately to tissue damage. The same author reviewed reports of resumed ovarian function after immunotherapy in patient with such diseases as myasthenia gravis, systemic lupus erythematosus, adrenal failure and poly glandular failure. No clinical parameters or biochemical criteria could be identified to predict who would respond or not, including the presence or absence of ovarian autoantibodies. Accordingly, the current evidence in the literature is not strong enough to support the use of immunotherapy for the treatment of patients with POF.

Infections

POF has also been related to various infections, but the incidence or the magnitude of the problem is not known. Mumps has been associated with 3-7% risk of viral oophoritis, during epidemics (34). Other infections included cytomegalovirus infection in patients with compromised immune systems, secondary to lymphoma or acquired immunodeficiency syndrome (AIDS), and in organ transplant patients on anti-rejection treatment (34). There are no specific tests to relate POF to these infections and patients’ history is the only clue to such infections. A figure of 3.5% for such infections has been reported by Rebar and Connolly in 1990 (2), and they included varicella, shigellosis and malaria, as possible causes as well.

Iatrogenic causes of POF

Pelvic surgery

As mentioned previously, surgery, pelvic irradiation, and the use of cytotoxic drugs are the main causes of iatrogenic or induced POF. Removal of the ovaries leads to abrupt loss of ovarian function, and causes severe oestrogen deprivation. This can follow bilateral oophorectomy performed because of ovarian tumours or severe pelvic endometriosis. Busacca et al in 2006 (35) reported 2.4% incidence of POF following laparoscopic excision of bilateral endometriomas. Furthermore, early loss of ovarian function may follow a hysterectomy (36, 37), as almost 30% of the blood flow into the ovaries is provided by the uterine arteries. Ovarian damage following ovarian drilling has been suggested as a possibility, but has not materialised as a reality with increased incidence of POF after the procedure, despite the widespread use of the technique. Nevertheless, care should be taken to avoid excessive use of the energy during this procedure, and to restrict the punctures to a reasonable number, depending on the size of the ovaries. More information can be found about this subject in Chapter 7.

Cytotoxic and pelvic radiation therapy

With the recent success in the treatment of childhood malignancies, many of these children are expected to reach adult life. The risk of POF is high in this group of women, secondary to the use of chemotherapeutic agents and radiotherapy. This is on top of those who have already had surgical treatment. A recent study addressed the risk factors involved with the development of non surgical POF in these patients (38). These factors included patient’s age, ovarian exposure to increasing doses of radiation, the number of cytotoxic drugs used and their cumulative dose, and the diagnosis of Hodgkin lymphoma. The authors reported a figure of 30% incidence of POF for patients who were treated with alkylating agents plus abdominopelvic radiation. Older patients can be affected more, as the risk of POF following such treatment increases with the patient’s age after puberty. It is a known fact that chemotherapy reduces the number of oocytes, but also affects the structure of the granulosa cells and oocytes at the same time. These effects depend on the type of drug and dosage used (39).

Adjuvant chemotherapy for the treatment of breast cancer is becoming an important factor in the development of POF in older women within their reproductive years. Rates of 21-71% and 49-100% have been quoted for women younger and older than 40 years respectively (40). These wide variations in incidence are most likely related to the type of drug and the dosage used. Some reports showed that simultaneous medication of gonadotrophin releasing hormone agonist can reduce the toxic effects of these drugs, but this is not well verified yet, and is not routinely used in these cases.

As for chemotherapy, the age of the patient and the dose of radiotherapy used affect the risks of POF, with prepubertal ovaries being more resistant to the damage. The older the patient is at the time of exposure to radiotherapy, the more her chances will be to develop irreversible ovarian damage. Exposure to 800 rads can lead to permanent ovarian failure, irrespective of the age group. Lower doses may be coupled with partial or complete recovery in younger patients. Transposition of the ovaries before radiation can save some ovarian function, and preserve fertility in many patients. Furthermore, freezing oocytes and ovarian tissue are two other available options to preserve future fertility, before using cytotoxic drugs or radiotherapy.

Clinical implications

The major clinical effects of POF result from the hypoestrogenic state and deficiency of androgen production by the ovaries. This is compounded by substantial psychological problems caused by loss of menstrual function, and more importantly a substantial reduction in fertility chances. These changes are usually perceived as loss of femininity with strong feelings of grief and anger, as well as being worthless and insecure. These are the immediate effects following the diagnosis, which should be handled with great care. As for natural menopause, hypoestrogenism exposes patients with POF to immediate, intermediate and delayed problems, but these effects will be worse in the long term as the patients have lost their ovarian function at a younger age. Vasomotor symptoms can disrupt the patients’ lives, and may start even before cessation of menstruation, mainly premenstrually. This may be compounded by night sweats, headaches, and loss of concentration and mood swings, as for the natural menopause. There is a chance for these symptoms to be worse to start with, due to the severe psychological trauma inflicted on the patient by the diagnosis itself. Sexual difficulties due to vaginal dryness and aging skin changes also take their toll. Nevertheless, the main long term problems are CVD and osteoporosis. The increased risk of CVD may follow accelerated endothelial dysfunction, which precedes atherosclerosis. This process was reversed by cyclical HRT use (41, 42). Hu et al in 1999 (43), observed a significant association between younger age at the menopause and higher risk of coronary heart disease only in women who never used hormone replacement therapy. Since the increased risk was seen only in current smokers, but not in women who never smoked before, it was suggested that such increased risk reflected only a confounding effect of smoking. The risk of osteoporosis may be worse in women with POF than those who go through a natural menopause, because of the prolonged duration of oestrogen deficiency, if HRT had not been used. This can be even worse in patients who develop POF at a very young age, as the maximum bone density is usually set around the age of 20 years.

Management of POF

This is a very difficult subject that needs to be addressed with great care and compassion. Young women with 3 or more months of irregular menstrual function should be investigated. It is important to elicit any family or personal history of autoimmune problems. Family history of POF, mental retardation, ataxia or dementia should be sought. Symptoms of oestrogen withdrawal, even in patients with regular menstrual cycles, can give a clue to the hypoestrogenic state. Usually physical and pelvic examinations are not remarkable during the early stages. Signs of autoimmune disorders may be seen in few patients including vitiligo, thinning of the axillary and pubic hair, butterfly facial skin rash etc. At the same time, vaginal skin atrophy can be seen in long standing cases. Transvaginal scan examination may show a small uterus with thin endometrium and small non active ovaries, but may not be helpful in many cases. A hormone test for FSH, LH, oestradiol, prolactin, TSH and T4 should be done. It has been recommended that women with a high FSH level ≥40 IU/L should have the blood test repeated after one month time, before making a definitive diagnosis of POF. Such rise in FSH level can be episodic and is usually late, in comparison to the decline in the level of inhibin B and antimullerian hormone (AMH). In fact the decline in AMH starts first while women still have regular menstrual cycles, and has a strong positive correlation to the antral follicles count (44). Adrenal and thyroid peroxidase antibodies should be tested at regular intervals, even if reported negative during the initial endocrine assessment.

The adverse psychological effects have already been mentioned before, and counselling should be an integral part of the management plan. Patients should be seen more frequently to answer all their doubts and queries. The differences between natural menopause as a permanent process and POF should be explained. It is very important to choose the right words when addressing these patients. Words such as menopause and failure should be avoided. Chances of intermittent resumption of ovarian function should be addressed, as 50% of the patients have intermittent menstrual cycles, as mentioned before. Nevertheless, the exact timing of these cycles can not be predicted before hand, and patients who wish to conceive should have regular intercourse. Excluding women with familial POF due to fragile X chromosome syndrome, there is no increased risk of congenital abnormalities when these patients manage to conceive. Furthermore, the course of the pregnancy is usually not different to other women with regular menstrual cycles. Nonetheless, the presence of autoimmune problems should be taken into consideration, especially for the adverse effects of adrenal insufficiency. Postnatal adrenal crisis is also a risk factor. These risks stress the importance of the point raised before regarding screening patients with POF for adrenal autoantibodies before they get pregnant. Any medication to increase the chances of pregnancy will be ineffective, and can interfere with the intermittent chances of spontaneous resumption of ovarian function. This is especially so for attempts to induce ovulation with antioestrogens or gonadotrophins. A note about immunosuppressants has already been mentioned before. The normal 5 -10% chance of spontaneous pregnancy rate (6) should be conveyed to the patients.

Hormone replacement therapy (HRT) makes the cornerstone for the management of patients with POF. There are no controlled studies as yet for the best HRT to be used in these young women. They usually need double the dose of oestrogen to control their vasomotor symptoms, compared to other postmenopausal women (20). Both the oral and transdermal routes can be used effectively. Progestogens will be needed for at least 12 days every month to prevent endometrial hyperplasia. This can be used in a cyclic HRT form, which is preferred by many patients as it gives the reassurance of the monthly withdrawal bleeding. Continuous HRT can also be used, but there is a higher risk of breakthrough bleeding with this type of medication. The oral contraceptive pill can be used instead, in areas where dedicated HRT drugs are not available, despite the high dose of the synthetic oestrogen used. Some authorities advocate its use, as it can provide contraception for patients who are not keen to get pregnant. Others are against its use, as it can mask detection of natural remission of ovarian function. Moreover, it does not offer very good contraception, and pregnancies did occur while patients with POF were taking the pill regularly.

The debate regarding the use of HRT has been addressed in Chapter 9, and will not be repeated here. It is enough to say that young women with POF should be encouraged to use HRT till the age of the natural menopause, without any significant risk of breast cancer or CVD. On the contrary, it can improve their quality of life and prevent osteoporosis, CVD, sexual difficulties and premature aging. The same advice regarding diet and weight bearing exercises should be given, as for postmenopausal women. Furthermore, they should be treated the same way as other postmenopausal women once they reach the age of 51 years.

Summary

Premature ovarian failure is a distressing problem which can not be reversed medically. Patients usually feel neglected, and 71% of women with POF were not satisfied with the manner in which the diagnosis had been relayed to them (45). It is always important to avoid using the words menopause or ovarian failure. Furthermore, patients should be informed about the possible intermittent resumption of ovarian function, and the risk or chance of spontaneous pregnancy, even while on the pill or HRT. The services of a counsellor are most important, and patients should be directed to join a support group of patients with a similar diagnosis. It is important to ascertain similar family history, as it has some bearing on the management plan of the concerned patient. In addition, regular follow up of patients with spontaneous POF should include regular screening for adrenal autoantibodies and thyroid function. Women with family history of POF should be referred for genetic counselling, and those with positive adrenal antibodies should see a medical endocrinologist for further assessment of their adrenal glands function. It is evident that a multidisciplinary team is needed for the proper management of patients with premature ovarian failure who need support and compassion, as well as HRT. The adverse effects of HRT reported in older women do not apply in this young age group.

References

1.   Rees M and Purdie D. Premature menopause. In: Management of the menopause: The Handbook, 4^th ed. London: Royal Society of Medicine Press Ltd; 2006; pp 142 -1 49.

2.  Rebar RW and Connolly HV. Clinical feature s of young women with hypergonadotropic amenorrhoea. Fertil Steril 1990; 53: 804 - 810.

3.  Nelson m, Anasti JN, Kimzey LM, Defensor RA, Lipetz KJ, White BJ, Shawker TH and Merino MJ. Development of luteinized graafian follicles in patients with karyotypically normal spontaneous premature ovarian failure. J Clin Endocrinol Metab 1994; 79: 1470 - 1475.

4.  Albright F, Smith P and Fraser A. A syndrome characterised by primary ovarian insufficiency and decreased stature. Am J Med Sci 1942; 204: 625 - 648.

5.  Alzubaidi NH, Chapin HL, Vanderhoof VH, Calis KA and Nelson LM. Meeting the needs of young women with secondary amenorrhoea and spontaneous premature ovarian failure. Obstet Gynecol 2002; 99: 720 - 725.

6. van Karseren YM, Schoemaker J. Premature ovarian failure: a systemic review of therapeutic interventions to restore ovarian function and achieve pregnancy. Hum Reprod Update 1999; 5: 483 – 492.

7.   Panay N and Fenton A. Premature ovarian failure: a growing concern. Climacteric 2008; 11: 1 - 3.

8.  Sklar CA, Mertens AC, Mirby P Whitton J, Stovall M, Kasper C, Mulder J, Green D, Nicholson HS, Yasui Y and Robinson LL. Premature menopause in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study; J Natl Cancer Inst 2006; 98: 890 - 896.

9.  Luborsky JL, Meyer P, Sowers MF, Gold EB and Santoro N. Premature menopause in a multiethnic population study of the menopause transition. Hum Reprod 2002; 18: 199-206.

10. Gosden RG, Treloar SA, Martin NG, Cherkas LF, Spector TD, Faddy MJ, Silber SJ.Prevalence of premature ovarian failure in monozygotic and dizygotic twins. Hum Reprod. 2007; 22(2): 610 - 615.

11. Anasti JN. Premature ovarian failure: an update. Fertil Steril 1998; 70: 1-15

12. Woad KJ, Watkins WJ, Prendergast D and Shelling AN. The genetic basis of premature ovarian failure. Aus N Z J Obstet Gynaecol 2006. 46: 242 - 244.

13. Shelling AN. X chromosome defects and premature ovarian failure. Aust N Z J MED 2000; 30: 5 - 7.

14. Toniolo D. X linked premature ovarian failure: a complex disease. Curr Opin Genet Dev 2006; 16: 293 - 300.

15. Rebar RW, Erickson GF and Yen SSC. Idiopathic premature ovarian failure: clinical and endocrine characteristics. Fertil Steril 1982; 37: 35 - 41.

16. Kalu E and Panay N. Spontaneous premature ovarian failure: management challenges. Gyn Endocrinol 2008; 24(5): 273 - 279.

17. Abdel Gadir A and Ramadan AA. Abnormal ovarian cell line: a cause for ovarian failure. Br J Obstet Gynaecol 1990; 97: 446 - 448.

18. Hoekstra C, Zhao ZZ, Lambalk CB, Willemsen G, Martin NG, BDI and Montgomery GW. Dizygotic twinning. Hum Reprod Update 2008; 14(1): 37 – 47.

19. Martin JA, Park MM. Trends in twin and triplet birth: 1980 – 1997. Natl Vital Stat Rep 1999; 47: 1 – 16.

20. Gosden RG, Treloar SA. Martin NG, Cherkas LF, Spector TD. Faddy MJ and Silber SJ. Prevalence of premature ovarian failure in monozygotic and dizygotic twins. Hum Reprod 2007; 22: 610 -615.

21. Marozzi A, Vegetti W, Manfredini E, Tibiletti MG, Testa G, Crosignani PG, Ginelli E, Meneveri R and Dalpra L. Association between idiopathic premature ovarian failure and fragile X permutation. Hum Reprod 2000; 15: 197 - 202.

22. Sherman SL. Premature ovarian failure in fragile X syndrome. Am J Med Genet 2000; 97: 189 - 194.

23. Nelson LM, Covington SN and Rebar RW. An update: spontaneous premature ovarian failure is not an early menopause. Fertil Steril 2005; 83(5): 1327 - 1332.

24. Laml T, Preyer O, Umek W, Hengstschläger M and Hanzal E. Genetic disorders in premature ovarian failure. Hum Reprod Update 2002; 8: 483 - 491.

25. Aittomaki K, Herva R, Stenman UH, Juntunen K, Ylostalo P, Hovatta O and de la Chapelle A. Clinical features of premature ovarian failure caused by a point mutation in the follicle stimulating hormone receptor gene. J Clin Endocrinol Metab 1996; 81: 3722 - 3776.

26. LaBarbera AR, Miller MM, Ober C and Rebar RW. Autoimmune aetiology in premature ovarian failure. Am J Reprod Immunol 1988; 16: 115 - 122.

27. Goswami D and Conway GS. Premature ovarian failure. Hum Reprod Update 2005; 11: 391 - 410.

28.Wheatcroft NJ, Salt C, Milford-Ward A, Cook ID and Weetman AP. Identification of ovarian antibodies by immunofluorescence, enzyme-linked immunosorbent assay or immunoblotting in premature ovarian failure. Hum Reprod 1997; 12: 2617 - 2622.

29. Yan G, Schoenfeld D, Penny C, Hurxthal K, Taylor AE and Faustman D. Identification of premature ovarian failure patients with underlying autoimmunity. J Women Health Gend Based Med 2000. : 275 - 287.

30. Khastgir G, Abdalla H and Studd JW. The case against ovarian biopsy for the diagnosis of premature menopause. Br J Obstet Gynaecol 1994; 101: 6 - 98.

31. Bakalov VK, Vanderhoof VH, Bondy CA, Nelson LM. Adrenal antibodies detect asymptomatic auto-immune adrenal insufficiency in young women with spontaneous premature ovarian failure. Hum Reprod 2002; 17: 2096 - 2100.

32. Turkington RW and Lebovitz HE. Extra-adrenal endocrine deficiencies in Addison’s disease. Am J Med 1967; 43: 499 - 507.

33. Kim TJ, Anasti JN, Flack MR, Kimzey LM, Defensor RA and Nelson LM. Routine endocrine screening for patients with karyotypically normal spontaneous premature ovarian failure. Obstet Gynecol 1997; 89: 777 - 779.

34. Moncayo R and Moncayo HE. Premature ovarian failure: evidence of the immunologic component. In: Ovarian autoimmunity: clinical and experimental data. 1st ed. Austin, TX: RG Landes Co., 1995: 27 - 75.

35. Busacca M, Riparini JR, Somigliana E, Giulia O, Stefano I Michele V,  Massimo C Post surgical ovarian failure after laparoscopic excision of bilateral endometriomas. Am J Obstet Gynecol 2006; 195: 421 – 425.

36. Siddle N, Sarrel P and Whitehead M. The effect of hysterectomy on the age of ovarian failure: identification of a subgroup of women with premature ovarian function and literature review. Fertil Steril 187; 47: 94 - 100.

37. Khastgir G and Studd J. Hysterectomy, ovarian failure and depression. Menopause 1998; 5(2): 113 - 122.

38. Sklar CA, Mertens AC, Mitby P, Whitton J, Stovall M, Kasper C, Mulder J, Green D, Nicholson HS, Yasui Y and Robison LL. Premature menopause in survivors of childhood cancer: a report from the childhood cancer survivor study. J Natl Cancer Inst 2006; 98(13): 890 - 896.

39. Nippita TA and Baber RJ. Premature ovarian failure: a review. Climacteric 2007; 10: 11 - 22.

40. Bines J, Oleske DM and Cobleigh MA. Ovarian function in premenopausal women treated with adjuvant chemotherapy for breast cancer. J Clin Oncol 1996; 14: 1718 - 1729.

41. Kalantaridou SN, Naka KK, Bechlioulis A, Makrigiannakis A, Michalis L and Chrousos GP. Premature ovarian failure, endothelial dysfunction and oestrogen-progesterone replacement. Trends Endocrinol Metab 2006; 17: 101 - 109.

42. Kalantaridou A, Naka KK, Papanikolaou E, Kazakos N, Kravariti M, Calis KA, Paraskevaidis EA, Sideris DA, Tsatsoulis A, Chrousos GP and Michalis LK. Impaired endothelial function in young women with premature ovarian failure: normalisation with hormonal therapy. J Clin Endocrinol Metab 2004; 89: 3907.

43. Hu FB, Gradstein F, Hennekens CH, Colditz GA, Johnson m, Manson JE, Rosner B, Stamper MJ. Age at natural menopause and risk of cardiovascular disease. Arch Intern Med 1999; 159: 1061 - 1066.

44. de Vet A, Laven JS, de Jong FH, Themmen APN, Fauser BC. Antimullerian hormone serum levels: a putative marker for ovarian aging. Fertil Steril 2002; 77: 357 - 362.

45. Groff AA. Covington SN, Halverson LR Fitzgerald Or, Vanderhoof V, Calis K and Nelson LM. Assessing the emotional needs of women with spontaneous premature ovarian failure. Fertil Steril 200; 3: 1734 - 1741.