Polycystic ovary syndrome — from gynaecologicalcuriosity to multisystem endocrinopathy
R.Homburg
Fertility Unit, Golda Medical Center, Petah Tikva and SacklerSchool of Medicine, Tel Aviv University, Israel
Recent progress in the diagnosis, pathophysiology, long-term ramifications and treatment of polycystic ovary syn-drome has been rapid but the pathogenesis remains achallenging enigma and the treatment symptomatic.Objective ultrasound criteria for diagnosis are being formu-lated and have enabled an appreciation of the true preval-ence and the associated clinical and biochemicalmanifestations. Although a heterogeneous syndrome, thefinal common pathway seems to involve a dysregulation ofenzymes responsible for ovarian androgen biosynthesis,possibly influenced by insulin, growth factors and lutein-izing hormone. A single gene defect, inherited in anautosomal dominant pattern, has been proposed. The treat-ment is necessarily symptomatic, depending on the needsof the patient Long-term deleterious sequelae now emer-ging may demand suppression of the syndrome earlier inlife. As the most prevalent cause of anovulatory infertility,a further elucidation of the basic pathogenesis is neededto allow the application of more specific and successfulmodalities of treatmentKey words: GiiRHa/hyperandrogenisrn/insulin/ovulation induc-tion/polycystic ovary syndrome
Introduction
Polycystic ovary syndrome (PCOS) is probably the mostprevalent endocrinopathy in women and by far the mostcommon cause of anovulatory infertility, yet its pathogenesisstill remains an enigma. This situation has produced anabundance of theories and a plethora of treatment regimes.The intellectual challenge which the solving of the puzzlepresents has been taken up seriously of late and will hopefullyproduce more specific and successful treatment in the nearfuture. The story is complicated by the complexity of thepathophysiological interactions, the heterogeneity of the clin-ical expression and even by the failure of investigators toagree on a common definition. This syndrome is no longer amere rare gynaecological curiosity but an endocrinopathy withmultisystem sequelae. This paper will attempt the well nighimpossible task of crystallizing the state of the art withoutintroducing too much personal opinion or bias.
History
As early as 1844, Chereau described sclerocystic changes inthe human ovary (Chereau, 1844) some 90 years before the
classic paper of Stein and Leventhal (1935). Elevated lutein-izing hormone (LH) concentrations were first reported in 1958,creating a criterion for diagnosis, and the introduction ofradioimmunoassays in 1971 stimulated reliance on a biochem-ical diagnosis. Although it was clear as early as 1962 thatthere was a wide variety of clinical presentation, it wasonly in 1976 that the concept of PCOS with normal LHconcentrations was conceived (Rebar et al, 1976). A furthermilestone was the discovery of the association of PCOS andinsulin resistance by Khan et al (1976) and Burghen et al.(1980). Swanson et al. (1981) first described the ultrasoundfindings of women with PCOS in 1981 but only after Adamset al. (1985) refined and critically defined diagnostic criteriain their seminal paper did the ultrasound diagnosis of PCOSbecome accepted throughout the world.
Diagnosis
The progress in our ability to make an accurate diagnosis ofPCOS has an historical perspective. For many years followingthe original description by Stein and Leventhal (1935), thediagnosis was purely clinical and relied on the presence ofobesity, hirsutism, amenorrhoea and enlarged ovaries. It isclear today that this description represents extreme or advancedcases and if judgement was made purely on these clinicalsymptoms alone, many less blatant but nevertheless clinicallysignificant cases would be missed. With the advent of radio-immunoassays, it was revealed that a large proportion of thewomen answering the clinical criteria had raised concentrationsof LH. Since then, a raised concentration of LH or anLH:follicle stimulating hormone (FSH) ratio of 2:1 or morehas become for some an essential criterion for diagnosis. Thepossibility of estimating testosterone in its total and free state,as well as many basic androgens, has added a further dimensionto die diagnostic possibilities. The typical appearance ofpolycystic ovaries at laparoscopy, with or without confirmationby ovarian biopsy, is still regarded as a sine qua non fordiagnosis by a number of clinicians.
It is my personal opinion that the advent of high resolution,transvaginal ultrasound examination of the ovaries has providedthe biggest single contribution to the diagnosis of polycysticovaries and that this examination, performed by an experiencedoperator, should be the basis of the diagnosis. It is a non-invasive technique for die characterization of ovarian morpho-logy which has a high concordance rate with laparoscopicexamination (Eden, 1988) and histological examination (Saxtonet al, 1990). With ultrasound we are looking directly at thehub of the problem and at the one feature which is commonto every case of polycystic ovaries (PCO) as opposed to
29
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
R-Homburg
clinical or endocrinological features which are notoriouslyvariable, heterogenic and inconsistent. Arguments that theultrasonic features of PCO found in apparently normal womenare merely a morphological variant and are of no clinicalsignificance have been largely dispelled by several studies. Ina series of apparently normal, eumenorrhoeic women, Poisonet al. (1988) found that >90% of the group with PCO had aclinical or biochemical feature consistent with the ultrasounddiagnosis. In similar groups of women, all ovulatory, thosewith PCO were more likely to be troubled by subfertility(Eden et al., 1989) and recurrent miscarriage (Sagle et al.,1988). In a very large series, Conway et al. (1989) demonstratedoverall, the typical endocrinological and classical clinicalfindings of PCOS in women in whom the diagnosis was madesolely on the basis of the ultrasound findings.
Ultrasound assessment of the ovarian morphology shouldthus be the gold standard for the diagnosis of PCO. Based onthe criteria of Adams et al. (1985), PCO should be diagnosedwhen more than eight discrete follicles of <10 nun diameterare seen in the ovary, usually peripherally arrayed around anenlarged, hyperechogenic, central stroma. Objective estimatesof the size of the stroma are difficult but stroma occupyingmore than 25% of the ovarian volume has been suggested asthe criterion. Polycystic ovaries are usually but not invariablyenlarged and normal size ovaries with typical features do notpreclude diagnosis. Recently, Dewailly et al. (1994) havedescribed an objective quantitative method of ovarian stromalassessment by using a computerized ultrasonic technique tomeasure stromal and cyst areas. Reassuringly, they foundthat the stromal area in hyperandrogenaemic women wassignificantly larger than in those with normal androgen levels.In these women, the stromal area correlated with serumandrostenedione and 17-hydroxyprogesterone concentrationsbut not with basal serum testosterone, LH or insulin concentra-tions. This suggests that by using this computerized ultrasonictechnique, a much more objective, standardized diagnosis ofPCO can be obtained.
Definitions
Based on the above and for the sake of clarity, the followingdefinitions are suggested.
Polycystic ovaries: ovaries displaying the typical ultrasoundfeatures described above.
Polycystic ovary syndrome: the typical ultrasound featuresof die polycystic ovary in the presence of oligo/amenorrhoeaand/or clinical symptoms of hyperandrogenism such as hirsut-ism or acne.
Polycystic ovaries have been associated with 75% of caseswith anovulatory infertility (Adams et al, 1986; Hull, 1987).Using ultrasound criteria, PCO has been found in 87% ofwomen with oligomenorrhoea and in a similar proportion ofwomen with hirsutism and regular, ovulatory menstruation(Adams et al., 1986). These significant findings have helpedexplain many of the cases previously designated as 'idiopathic'and have facilitated specific treatment. Similarly, Adams et al.(1986) discovered PCO in 3(M0% of women with amenor-rhoea.
The ultrasonic examination of the ovaries of 82 womenpresenting at a dermatology clinic complaining of acne revealedthat 83% had PCO and this finding has revolutionized thetreatment of this complaint (Bunker et al, 1989).
Clinical manifestations
Women with PCO may display a wide range of clinicalsymptoms. At the extreme ends of this range are the womenwho are phenotypically normal, have no signs of hyperandro-genism and who are eumenorrhoeic but have PCO on ultra-sound examination and the women who have the full blownexpression of the classical Stein-Leventhal syndrome.
Several concepts are important in order to understand thisheterogeneity of clinical presentation. Polycystic ovaries arethere for life and only the clinical expression may change,while the typical ultrasonic findings will remain identifiable.If accepting that PCOS is basically a disorder of ovarianfunction, symptomatology is dictated by extra-ovarian factors.A typical illustrative example is that of an asymptomatic,eumenorrhoeic and ovulatory woman with demonstrable PCOand of normal body weight who gradually becomes obese.She will almost invariably develop menstrual disturbances,anovulation and signs of hyperandrogenism such as hirsutismand acne. These symptoms may be reversed by returning toher previous body weight probably by reducing her exposureto hyperinsulinaemia (Kiddy et al., 1992). Treatment withcombined oral contraceptive pills or long-term gonadotrophin-releasing hormone agonists (GnRHa), by reducing exposure toendogenous gonadotrophins, is capable of severely decreasingsymptoms of hyperandrogenism (Givens et al., 1976; Shaw,1989).
Finally, hirsutism and menstrual cycle disturbances havebeen found to be significantly more frequent in obese thannon-obese women with PCOS (Conway et al., 1989). Alopeciamay occur in severe PCOS in untreated women and acanthosisnigricans is a rare but pathognomonic finding associated withobesity and insulin resistance.
Prevalence
Ultrasound diagnosis of PCO has enabled us to appreciate thetrue prevalence of this phenomenon. An attempt was made byPoison et al. (1988) to assess the prevalence of PCO in thegeneral, healthy female population of fertile age using 257volunteers. Polycystic ovaries were found in 22%. In a similarstudy on 100 normal Arab women, Abdel Gadir et al (1992)found a prevalence of 16%.
30
Late sequelae
Untreated PCOS may be regarded as a progressive syndromeup to the time of the menopause. Much interest of late hasconcerned the long-term sequelae of PCOS. In a seminal study,Dahlgren et al. (1992) examined 33 women aged 40-59 yearswho had undergone wedge resection 22-31 years previously.Compared with age-matched controls, they had a significantlygreater incidence of hypertension and diabetes mellitus. Using
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
Polycystlc ovary syndrome
multiple logistic regression analysis, the authors went on topredict that women with PCOS are at a much greater risk ofdeveloping cardiovascular disease. Hyperinsulinaemia is anadverse risk factor for coronary artery disease and is associatedwith reduced high density lipoprotein (HDL2) concentrations(Conway and Jacobs, 1993). Dahlgren et al. (1991) studiedwomen with endometrial carcinoma in whom the incidenceof hirsutism, obesity, infertility and hypertension was highcompared with controls. While much more work is needed toclarify the association of PCOS with these possible latesequelae, it is highly likely that prolonged treatment withcombined oral contraceptives or ethinyl oestradiol and cypro-terone acetate may prove to be important in their preventionby improving endocrine and lipid profiles and reducing hyper-androgenaemia in women with PCOS.
Laboratory manifestations
In concert with the symptomatology, the biochemical findingsin PCOS also lack uniformity. However, excessive ovarianandrogen production is characteristic of most women with thetypical ovarian morphology. Overall, comparisons of largegroups of women witii PCOS with healthy controls havedemonstrated elevated concentrations of testosterone, freetestosterone, androstenedione, LH, LH:FSH ratio, free oestra-diol, oestrone and fasting insulin and reduced concentrationsof sex hormone binding globulin (SHBG) (Conway et al,1989; Franks, 1989).
These overall results, however, are more informative whenthe groups of women with PCOS are separated into subgroupsof obese and non-obese. Comparisons of these groups revealhigher concentrations of insulin, free testosterone and oestronebut lower concentrations of SHBG, LH, insulin-like growthfactor binding protein-1 (IGFBP-1) and growth hormone (GH)in the obese compared with the non-obese (Conway et al.,1990; Homburg et al., 1992; Insler et al., 1993).
The endocrine profile of women with PCOS tends tovary with the symptomatology. For example, those who areovulating with regular menstrual cycles will have lower LH,FSH, testosterone, androstenedione and fasting insulin concen-trations than those with menstrual irregularities (Franks, 1989).Insulin insensitivity in PCOS occurs when there is oligo/amenorrhoea but not when menses are regular (Robinson et al.,1993). Hirsute patients have higher androgen and lower SHBGlevels than the non-hirsute (Franks, 1989) and LH concentra-tions are higher in infertile patients and in those who miscarry(Homburg et al., 1988a).
Pathophysiology and pathogenesis
The exact aetiology of PCO remains unknown. While PCOmay be regarded as a multifactorial syndrome, the finalcommon pathway of the typical changes in ovarian morphology(a sine qua non for diagnosis) and excessive ovarian androgenproduction are, in my opinion, the hub of the problem in thevast majority of cases. The key question in this premiseis therefore the source of the excessive ovarian androgenproduction. The enzyme cytochrome P450cl7a, which cata-
lyses 17-hydroxylase and 17/20 lyase activities, the rate limitingstep in androgen biosynthesis (Barnes et al., 1989), is disor-dered in PCOS (Rosenfield et al, 1990). In response to asingle dose of GnRHa, women with PCOS have an exaggeratedincrease in both 17-hydroxyprogesterone and androstendione,more typical of a male response (Rosenfield et al., 1990). Thishas been confirmed by the Franks group who, in addition,have evidence of an abnormal regulation of cytochromeP450cl7a in ovulatory women with PCOS (Franks andWhite, 1993).
Several pathways which may cause increased activity ofcytochrome P450cl7a have been suggested. Luteinizing hor-mone is an obvious candidate, as this hormone is presumablythe main regulator of androgen production from theca cellsduring a normal ovulatory cycle. It is therefore reasonable tosuggest that persistently high concentrations of LH will produceexcessive amounts of androstendione by causing a hyperactivityof P450cl7a. However, a proportion of women with typicalfeatures of PCOS do not have elevated concentrations of LH,particularly those who are obese (Homburg et al, 1992; Insleret al, 1993), so that an alternative mechanism must be soughtto explain the source of the hyperandrogenism. One suggestion(Rosenfield et al, 1990) is that, whereas in the normalovulatory cycle the surge of LH in mid-cycle downgrades itsown receptors on theca cells and so limits androgen production,this does not happen for some reason in PCOS. A morethoroughly investigated suggestion is that a larger than usualnumber of LH receptors are maintained by a faulty fine tuningmechanism. Insulin-like growth factor-1 (IGF-1) potentiatesthe expression of LH receptors (Adashi et al., 1985) andstimulates LH-induced androgen production and the accumula-tion of androgens in the ovary (Barbieri et al., 1986; Cara andRosenfield, 1988). Although circulating concentrations of IGF-1 have generally been reported to be not significantly highercompared with those in non-PCOS (Homburg et al, 1992),the biological activity of IGF-1 may be raised by a decreasein concentrations of IGFBP-1 which serves as an acutemodulator of IGF-1 action (Suikkari et al, 1988). As IGFBP-1 has been shown to inhibit the action of IGF-1 in the ovary(Angervo et al, 1991), suppression of IGFBP-1 may increasethe potential for IGF-1 to act synergistically with LH tostimulate the theca and interstitial cells of the ovary to produceandrogens (Cara et al., 1990). Our group (Homburg et al.,1992) and others (Suikkari et al., 1988; Insler et al., 1993)have found that in obese PCOS patients (i.e. predominatelyhyperinsulinaemic with normal LH) serum IGFBP-1 concentra-tions were inversely correlated with insulin concentrations andwere significantly lower than those found in non-obese PCOS.It is thus postulated that the synergism of free IGF-1 and LHis capable of increasing ovarian androgen output by inducinghyperactivity of cytochrome P450cl7a.
Since the association of insulin resistance, the consequenthyperinsulinaemia and PCOS was first described (Khan et al,1976; Burghen etal., 1980), it has become clear that hyperinsul-inaemia plays a key role in the pathophysiology and probablythe pathogenesis of PCOS. Several facts have prompted thisstatement. The type of insulin resistance associated with PCOSseems to be unique to this syndrome and is probably a decrease
31
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
R-Horoburg
in insulin sensitivity due to a defect in post-receptor signaltransduction between the receptor kinase and the glucosetransporter (Dunaif et al, 1993). The tyrosine kinase domainof the insulin receptor gene, however, appears to be normal(Conway et al, 1994). Insulin receptors are present in thehuman ovary (Poretsky et al, 1985) and insulin may also bindto IGF-1 receptors and acts as a gonadotrophic hormone,enhancing induction of ovarian LH receptors and LH bindingcapacity (Cara et al., 1990). Hyperandrogenaemia has beenshown to be a consequence of hyperinsulinaemia and thereduction of insulin concentrations with diazoxide (Nestleret al, 1989) or weight loss (Kiddy et al, 1992) decreasesovarian androgen secretion. As already mentioned, the directcontrol and inverse relationship of insulin and IGFBP-1 mayplay a cardinal role in ovarian hyperandrogenism. In addition,hyperinsulinaemia is a probable cause of decreased levels ofSHBG (Plymate et al., 1988) and so also plays a role in thedisturbance of androgen transport and raised concentrations offree testosterone. A study of Lanzone et al. (1994) suggeststhat insulin affects the responsiveness of the adrenal glands toadrenocorticotrophic hormone (ACTH) so that hyperinsulinae-mia may lead to a chronic adrenal hyperfunction. In obesewomen with PCOS, hyperinsulinaemia is extremely common(Franks, 1989) with an estimated incidence of about 75%(Conway et al., 1990). Although insulin resistance and hyperin-sulinaemia correlate with body mass index, elevated concentra-tions of insulin can be found in >30% of non-obese patientswith PCOS (Dunaif et al, 1989; Conway et al, 1990; Buyaloset al, 1992; Grulet et al, 1993) who overall, have higherinsulin levels compared with normal ovulatory women.
The kinetics of growth hormone (GH) are also disturbed inPCOS. Despite initial disparate results, when GH concentra-tions are controlled for obese and non-obese PCOS andcompared with weight matched controls, it has become apparentthat they are significantly reduced in obese and significantlyraised in non-obese PCOS (Prelevic et al, 1992; Insler et al,1993). In obese PCOS, the low levels and reduced pituitaryresponse on stimulation of GH release with L-dopa (Acar andKadanli, 1993), arginine (Ovesen et al, 1992) or exercise(Jaatinen et al, 1993) may be accounted for by a negativefeedback response to increased levels of biologically activeIGF-1 (see above). The higher than normal concentrations ofGH found in non-obese (predominately not hyperinsulinaemic)PCOS are harder to explain and have given rise to speculationthat GH may be directly involved in the pathogenesis of thesecases (Insler et al, 1993; Homburg, unpublished data). Ourgroup has gone one step further and shown that GH adminis-tered to women with PCOS acutely decreases IGFBP-1 andSHBG and increases IGF-1 and insulin concentrations. Thedecrease in IGFBP-1 concentrations may well be a direct actionof GH rather than mediated through insulin as Tapanainen et al(1991) have shown that this response precedes any rise ofinsulin or IGF-1. Thus, the high concentrations of GH in non-obese PCOS may be partly responsible for ovarian hyper-androgenism.
Epidermal growth factor may also be involved in thepathological meshwork, as it has been reported to negativelyinfluence the FSH-induced aromatase reaction and has been
found in higher than normal amounts in the follicular fluid ofwomen with PCOS (Mason et al, 1990).
There is a high prevalence of PCOS within families, sug-gesting a genetic component, but the demonstration of a clearmode of inheritance has proved difficult (Hague et al, 1988).However, Carey et al (1993) have lately produced convincingevidence for a single gene defect by examining, in detail, thefamilies of 10 probands with PCO and finding that each familyshowed autosomal dominant inheritance for PCO with >90%penetrance. Interestingly, they also described early onset malepattern baldness as an accurate phenotype for obligate malecarriers. Men with premature male pattern baldness have anelevation of their serum androgens when compared to age andweight-matched controls (Stephens et al, 1993). The group ofCarey et al (1994) have gone a step further by identifying anew single base change in the 5' promoter region of CYP17,the gene encoding for P450cl7 on chromosome 10q24,3 in 81individuals with PCOS/male pattern baldness. While thisvariation in the A2 allele of CYP17 may cause modification ofexpression, it has been excluded as the primary genetic defect.
The neuroendocrine abnormalities observed in PCOS haveengendered support for the hypothesis that the primary lesioncausing the syndrome lies within the hypothalamus or highercentres. Increases in mean LH concentrations have been welldocumented and they are apparently induced by the increasedamplitude of LH (GnRH-generated) pulses. However, thepathophysiological mechanism underlying this change is notyet clear. A congenital lack of dopamine has been suggested(Vaitukaitis, 1983) but it seems likely that decreases in dopa-mine levels in PCOS are secondary to hyperoestrogenism.Diminished opioid output is also a possibility (Quigley et al,1981). Raised oestrone levels, due to increased conversion ofandrogens in fatty tissue, have been purported to exert apositive feedback effect on the hypophysis, so increasing LHsecretion. This is contradicted somewhat by the fact thatnon-obese women with PCOS have predominantly higherconcentrations of LH than the obese (Homburg et al, 1992;Insler et al, 1993). Increased concentrations of free 17-oestradiol are also evident (Franks, 1989) and may increase LHsecretion through a similar mechanism. However, a significantincrease in LH concentrations was observed before any increasein serum oestradiol could be detected during ovulation induc-tion with pulsatile GnRH, suggesting that an ovarian steroidwas not responsible for exaggerated secretion of LH (Schachteret al, 1995). In-vitro evidence that insulin facilitates LHsecretion (Adashi et al, 1981) introduces a further possibility,supported by Prelevic et al. (1990). However, we and others(Homburg et al, 1992; Insler et al, 1993), found a negativecorrelation between these hormonal concentrations. An attract-ive, new theory proposes that the hypersecretion of LH inPCOS is mediated by a deficiency of the putative gonadotrophinsurge attenuating factor which is produced by small andmedium sized follicles and is thought to curtail the LH surgeby reducing the sensitivity of the hypophysis to GnRH (Sopelekand Hodgen, 1984; Messinis and Templeton, 1990; Balenet al., 1993a).
Most investigators have also reported an increased pulsefrequency of LH in addition to the increased amplitude (Kazer
32
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
Polycystic ovary syndrome
et al., 1987) and Crowley et al. (1993) argue that this increasedfrequency of GnRH pulsatility desensitizes pituitary FSHsecretion and explains the finding that FSH concentrations inPCOS are some 30% below those in the early follicular phaseof normal women. In their opinion, this may be the cause ofthe arrested folliculogenesis of PCOS. As serum inhibinconcentrations are not elevated in PCOS compared with normalwomen (McLachlan et al., 1987), inhibin does not appear tobe involved in this process. The discussion as to whether theneuroendocrinological manifestations in PCOS are the primarycause of the syndrome or whether these changes are secondaryto abnormalities of secretion of ovarian steroids, growthfactors or other substances is still unresolved. The fact thatnormalization of GnRH pulsatility by progesterone administra-tion (Homburg et al., 1988b) or the imposition of 'physiologi-cal' doses and timing of GnRH (Christman et al., 1991),improves endocrinological and clinical parameters, does notcontribute to the solution of the problem. It is my personalopinion that the abnormalities in the setting of the GnRHpulsatility clock are a secondary phenomenon and are not atthe hub of the basic problem. This statement is prompted bythe following reasons: PCO may be identified before menarche(Bridges et al., 1993) in women with hypogonadotrophichypogonadism (Homburg and Jacobs, 1990; Shoham et al.,1992) and in a single ovary (Poison et al., 1986). In addition,the administration of long-term GnRHa or combined oralcontraceptives does not alter the typical ovarian morphologyof PCOS.
In summary, I would suggest that PCOS is basically apredominantly genetic disorder of ovarian androgen productioninduced by extraovarian factors, in particular, insulin resistance,hyperinsulinaemia and their biochemical sequelae. The clinicaland endocrinological expression of the syndrome is determinedby the extraovarian factors present.
Treatment
As the detailed basic cause of PCOS is still unknown, thereis no single effective treatment regimen, merely a plethora ofprotocols which do not address the real cause but aim to breakthe vicious circle of pathological chain reactions. The treatmentschedules outlined here are therefore symptomatic therapy, thechoice of which is made according to the principal complaintof the patient.
However, there is one overriding principle for the treatmentof obese women with PCOS, whether the main problem iscosmetically unacceptable signs of hyperandrogenism such ashirsutism or acne, infertility or prevention of long-term seque-lae. Loss of 5% or more of total body weight is capableof reversing or severely reducing these symptoms anoVorfacilitating treatment of infertility (Kiddy et al., 1992). Thisis apparently achieved by reducing insulin and increasingSHBG and IGFBP-1 concentrations with a consequent reduc-tion of ovarian androgen production and circulating freetestosterone. It is safe to predict that a reduction in body massindex, maintained within normal range, may prevent thedevelopment of long-term sequelae such as diabetes mellitus,hypertension, cardiovascular and hyperlipidaemic disorders.
More sophisticated methods of reduction of insulin concen-trations with diazoxide (Nestler et al., 1989), metformin(Velaquez et al., 1994) or a somatostatin analogue (octreotide)(Prelevic et al., 1990) have proved to be effective in improvingthe endocrine milieu. Prelevic et al., 1995 utilized the propertyof octreotide to decrease insulin, LH and testosterone concen-trations by administering 100 \ig twice daily in parallel withHMG for 58 cycles in 28 anovulatory women with PCOS.Compared with a control group of 29 cycles using humanmenopausal gonadotrophin (HMG) alone, a more orderlyfollicular growth with a more appropriate endocrine milieu atthe time of human chorionic gonadotrophin (HCG) administra-tion and a significantly decreased incidence of ovarian hyper-stimulation syndrome (OHSS) were recorded. Moreinvestigation is needed to examine the clinical feasibility ofthis mode of treatment.
For women with PCOS who are not interested in conceiving,the treatment of choice is a combination of cyproterone acetateand ethinyl oestradiol. Cyproterone acetate is a syntheticprogesterone with both antigonadotrophic and antiandrogeniceffects, giving it an advantage for the treatment of PCOSover the majority of progestins in oral contraceptives. Itscombination with ethinyl oestradiol has been successful insuppressing hyperandrogenism, resulting in improvement ofclinical signs and normalization of the hormonal disturbanceswhich characterize PCOS (Prelevic et al., 1989). The efficacyof this treatment is due to the antiandrogenic activity atandrogen receptor level, suppression of serum LH and ovarianandrogen concentrations, reduction in 5-reductase concentra-tions, increased metabolic clearance of testosterone andincrease in SHBG concentration. The effect on acne andseborrhoea is usually rapid but due to the physiological cycleof the hair follicle, at least 6 months treatment are requiredbefore the reduction of hirsutism becomes clinically evident.As previously mentioned, untreated PCOS may be regardedas a progressive syndrome. It is therefore my opim'on that earlytherapeutic intervention will not only temporarily alleviatesymptoms but will place the progress of the syndrome 'onhold'. This would appear to be beneficial for future fertilityprospects and possible delay or prevention of long-termsequelae.
The chronic anovulation and infertility associated withPCOS can often be successfully treated with clomiphene citrate(CC). Some 75% of those treated with CC in doses of 50-200mg/day from day 4 or 5 of the cycle for 5 days will respondby ovulating. However, the overall pregnancy rate is only 30-40%. Several reasons have been forwarded to explain thisapparent discrepancy. The main action of CC is through thehypothalamus, stimulating GnRH secretion and increasing FSHrelease from the hypophysis. This is often accompanied by astriking increase in serum LH concentrations (Van den Bergand Yen, 1973) and this may seriously compromise pregnancyrates in these patients (Shoham et al., 1990a). Pretreatmentwith progesterone is capable of modulating LH pulsatility,reducing LH concentrations and inducing more FSH synthesisand storage, so creating a more favourable environment forovulation induction with CC (Homburg et aL, 1988b). Thistreatment improved response to CC and consequent pregnancy
33
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
R-Homburg
rates (Homburg et al, 1988b). An additional causative factorin those ovulating but not conceiving is the anti-oestrogeniceffect of CC on the cervical mucus and, possibly, on theendometrium, hypothetically disturbing normal mechanismsof sperm transport and implantation. In some cases, the additionof ethinyl oestradiol to CC will enhance the pregnancy ratewithout interfering with the induction of ovulation.
Monitoring, by ultrasound examination of the ovaries, ofthose who fail to ovulate on maximal doses of CC, will revealthat the majority of these women will show inadequatefollicular development, demanding an 'upgrade' of treatment.However, a minority of the non-ovulators will develop largefollicle(s) but have an isolated defect of oestrogen-mediatedpositive feedback and may respond to a well-timed injectionof HCG. Finally, the addition of dexamethasone in the fewwho have high dihydroepiandrosterone sulphate concentrationshas been reported to improve results (Daly et al, 1984).
In our practice, 'clomiphene failure' is regarded as a failureto ovulate on maximal doses of CC or a failure to conceive,despite apparent ovulation in six courses of treatment. In thelatter case, male and mechanical factors are re-evaluated.Several treatment modes are now available to women withPCOS who are 'clomiphene failures', most of which arereasonably successful in breaking the vicious circle of chainreactions and inducing ovulation and pregnancy without theability to attack the unknown source. The administration ofnative GnRH in a pulsatile fashion, either subcutaneously ori.v. in a suggested dose of 15 u.g/pulse every 90 min using apump apparatus, is apparently capable of superimposing theabnormal pattern of secretion with a more physiological patternso producing a more balanced output of gonadotrophins. Thisform of treatment is, however, associated with low ovulationrates (50%/cycle) and pregnancy rates/ovulatory cycle of 29%(Shoham et al, 1990b). It is particularly unsuitable for obese,hyperandrogenaemic women and the high LH concentrationsobserved during induction present a further obstacle (Homburget al., 1989). The advantages of this mode of treatment arethat it will yield a monofollicular response in a large proportionand therefore ovarian hyperstimulation (OHSS) is not encoun-tered and multiple pregnancies are rarely seen if HCG adminis-tration is avoided.
Stimulation of the ovaries with exogenous gonadotrophinsis more acceptable treatment for CC-resistant patients. Aconventional 'step-up', individually adjusted dose regimenemploying HMG in our hands (Farhi et al., 1993) yields acumulative conception rate of 82% after six cycles. However,due to the high sensitivity of the PCO to gonadotrophinstimulation and its propensity to multiple follicular develop-ment, this treatment regimen is plagued by a high frequencyof multiple pregnancies and OHSS (Wang and Gemzell, 1980).The use of urinary human FSH has done little to remedy thissituation (Garcea et al, 1985; Homburg et al, 1990; McFaulet al., 1990). Following the initial publication of Seibel et aL(1984), several investigators have examined the utility of achronic low dose regimen of FSH in an attempt to reduce thecomplication rate (Poison et al., 1987; Hamilton-Fairley et al,1991; Shoham et al, 1991; Dale et al, 1993; Scheele et al,1993a; Homburg et al., 1995a). The basic thinking behind this
34
regimen is the 'threshold theory' which demands the attainmentand maintenance of follicular development with exogenousFSH, without exceeding the threshold requirement of the ovary(Brown, 1978) which, with supraphysiological doses of FSH,provokes an initial development of a large cohort, stimulatesadditional follicles and even rescues those destined for atresia(Insler and Lunenfeld, 1991). This is what tends to occur inPCOS with its peculiar hypersensitivity to gonadotrophinswhen concentrations of FSH well above the threshold areinduced during conventional treatment. This threshold has ahigh inter-individual variability (5.7-12 IU/1) and higher FSHconcentrations are associated with multifollicular, as opposedto monofollicular, growth (Van der Meer et al, 1994). Theprinciple of the chronic low dose regimen is, therefore, toemploy smaller incremental dose rises (e.g. 37.5 IU or less)at intervals of 7 days until follicular development is initiated.A compilation of reported results using this or similar regimens,reveals that 218 patients who completed 471 cycles achieved83 pregnancies (18%/cycle). This rate compares favourablywith that of more conventional regimens but has the advantagethat it is accompanied by a very low prevalence of OHSS (onecycle) and multiple pregnancies (5%). A prospective studyfrom our centre (Homburg et al, 1995a), comparing conven-tional with low-dose protocols employing 50 women with CC-resistant PCOS, revealed that the low-dose protocol slightlyimproved pregnancy rates (40 versus 24%) while completelyavoiding OHSS and multiple pregnancies which were prevalent(11 and 33% respectively) using conventional incrementaldoses of FSH. We concluded that this treatment modalityhas distinct advantages and could well replace conventionalgonadotrophin therapy for these patients.
The original report of Fleming et al. (1985) on cotreatmentwith GnRHa and HMG for anovulatory PCOS encouraged itsuse for this indication. While hopes that the state of nearhypogonadotrophic hypogonadism induced by GnRHa wouldproduce results with HMG as successful as those of WHOgroup I cases have not been fulfilled, there is a definite placefor the incorporation of GnRHa into stimulation protocols forwomen with PCOS. The main contribution is the reduction ofLH concentrations throughout the follicular phase of the cycle.This almost completely eliminates the troublesome problemof premature luteinization and the need to abandon cycles forthis reason, so increasing treatment efficiency. There is nowmounting evidence that the ability of GnRHa to reduce theinordinately elevated concentrations of LH prevalent in PCOS,serves to increase ovulation and pregnancy rates and, mostimportantly, to reduce the prevalence of early spontaneousmiscarriages, which are notoriously high in PCOS accompaniedby raised LH concentrations.
In a retrospective analysis from our centre (Homburg et al.,1993a), 239 women with PCOS received HMG with or withoutGnRHa for ovulation induction or superovulation for in-vitrofertilization (IVF)/embryo transfer. Of pregnancies achievedwith GnRHa, 17.6% miscarried compared with 39.1% of thoseachieved with gonadotrophins alone. Cumulative live birthrates for GnRHa after four cycles were 64% compared with26% for gonadotrophins only. Reduced miscarriage rates withGnRHa have also been reported by Balen et al. (1993b). While
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
PoJycystic ovary syndrome
a randomized, prospective study is sadly lacking to confirmthese data, it is our policy to administer GnRHa to womenwith high LH concentrations in the follicular phase on thebasis that there is little point in inducing a pregnancy whichhas a very high chance of being aborted. A further study fromour centre (Homburg et al, 1993b) looked at the performanceof women with PCOS undergoing IVF/embryo transfer whohad high mean LH concentrations, compared with a controlgroup of normally cycling women with mechanical infertility.Pregnancy rates were similar in the groups but whereas GnRHareduced the miscarriage rate compared with gonadotrophinsalone in the PCOS group, its administration to the controlgroup had no such effect.
Administration of GnRHa to women with PCOS reducesLH and androgen concentrations. It has no effect on insulinresistance and hyperinsulinaemia, IGF-1 or IGFBP-1 concen-trations (Dale et al, 1992; Homburg et al., 1995b). Thereduction by GnRHa of intrafollicular concentrations of andro-gens which would normally induce atresia may be responsiblefor the increased number of follicles induced by consequentstimulation by GnRHa/HMG compared with HMG alone(Homburg et al., 1990). The increased number of developingfollicles induced by cotreatment with GnRHa/HMG is accom-panied by a raised incidence of OHSS and multiple pregnancies(Homburg et al., 1990).
Two interesting attempts have been made to utilize thebeneficial effects of GnRHa, increase its efficiency and reduceits undesired influence on multiple follicular developmentFilicori et al. (1988) followed pretreatment by GnRHa withpulsatile GnRH. While seemingly paradoxical, this combina-tion produced good ovulation and acceptable pregnancy ratesand no multiple pregnancies, but was limited by the longduration of each cycle and a relatively high incidence ofabortion and luteal phase abnormalities. The logical idea ofcombining GnRHa with low dose FSH therapy was studiedby Scheele et al. (1993b) but failed to reduce multiplefolliculogenesis and its consequences. In the opinion of theseinvestigators, the extreme sensitivity of the follicles to FSHonce growth is initiated may be tempered by using evensmaller incremental dose rises than those employed.
Surgical treatment by bilateral wedge resection, althoughrelatively successful in restoring ovulation, has fallen fromgrace due to its propensity to adhesion formation. Laparoscopicovarian diathermy was introduced by Gjonnaess (1984),working on the principle that similar results could be achievedwhile avoiding the introduction of a mechanical factor. In hissmall study and in larger studies since (Aakvaag, 1985; Kovacset al, 1991; Naether et al., 1993), ovulation rates of 70-90%and pregnancy rates of 40-70% have been achieved. Post-operative laparoscopy has revealed the presence of intraperi-toneal adhesions in about 20% of these cases but they arereported to be mild and unilateral (Dabirashrafi et al., 1991;Naether et al., 1993) and did not apparently affect the highpregnancy rate (Greenblatt and Casper, 1993). Significantfalls of LH, androstenedione, dihydroepiandrosterone sulphate(DHEAS) and testosterone have been uniformly noted afterovarian diathermy. Laparoscopic argon laser capsule drillinghas also been used with good results (Heylen et al., 1994).
Laser photodiathermy may lessen adhesion formation and,following the operation, stimulation with HMG gives anenhanced ovarian response compared with the pre-diathermyresponse (Dabirashrafi et al., 1991; Kovacs et al, 1991).Although the majority of authors reported that the benefit ofovarian laparoscopic laser treatment is limited to approximately6 months (reviewed by Gurgan et al, 1994), Naether et al.(1994) are of the opinion that the benefit of monopolar currentis greater and that the effects may last for several years. Theideal protocol following this procedure to yield optimal resultshas yet to be firmly established. An excellent review byDonesky and Adashi (1995) on laparoscopic ovulation induc-tion called for carefully constructed controlled trials beforethe procedure can be viewed as efficacious and safe.
For patients with PCOS who have failed to conceive duringsix ovulatory cycles of gonadotrophin therapy, we have foundthat IVF/embryo transfer is a very viable alternative. In astudy from our centre (Homburg et al, 1993b), 68 such womenunderwent 208 cycles of IVF/embryo transfer with a cumulativeconception rate of 82% at six cycles, almost identical to thatof a control group of women with a pure mechanical factorundergoing similar treatment. MacDougall et al. (1993)reported very similar results. There are two possible explana-tions for the fact that these results were achieved with IVF/ET but not with gonadotrophin therapy. Either an overtmechanical factor was present or, more likely, this procedureallowed a more liberal approach to superovulation rather thanconcentrating on monofollicular development. Both of thesecomparative studies (Homburg et al., 1993b; MacDougallet al, 1993) reported that women with PCOS required lessHMG, but produced more follicles and oocytes. However,fertilization rates were lower, probably a reflection of thenumber of oocytes retrieved from relatively immature follicles(<14 mm diameter) from women with PCOS. A fascinatingrecent development by Trounson et al. (1994) reports thesuccessful recovery of immature oocytes from the ovaries ofuntreated PCOS patients and their maturation, fertilization anddevelopment in vitro. If this method can be successfullyadopted, the problems of ovulation induction may well bebypassed.
In summarizing the treatment of infertility associated withPCOS, basically, the induction of ovulation may be achievedby boosting FSH stimulation of the ovaries either indirectlywith clomiphene or native pulsatile GnRH, or directly withgonadotrophin preparations. The selection of treatment couldnot be guided by basal clinical or endocrine features in a seriesof 306 treatment cycles whose outcome was reported by Farhiet al. (1993). However, there seem to be two main determinantsof the success of this treatment in achieving a live birth:the degree of hyperinsulinaemia and the concentrations ofcirculating LH. Either of these, when in excess, not only makeinduction of ovulation and conception relatively difficult butare associated with high rates of early miscarriage (Homburget al, 1988a; Hamilton-Fairley et al., 1992). Their correction,particularly in obstinate cases, should be a major considerationin the attempt to achieve optimal results. With such a rangeof reasonably successful treatments for the induction of ovula-tion in PCOS, the emphasis in the selection of merapy should
35
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
R.Homburg
now be placed on minimizing the prevalence of undesired sideeffects while retaining acceptable efficiency.
The LH hypothesis
Women with PCOS have a very high prevalence of high LHconcentrations. A number of clinical reports have linked highconcentrations of LH in the follicular phase to decreasedreproductive function. When high concentrations were foundin the first few days before oocytes were collected fromwomen for FVF (Howies et al, 1986) or on the day ofHCG administration (Stanger and Yovich, 1985), they wereassociated with impaired rates of fertilization and conception.
In a prospective study of almost 200 women with regularmenstrual cycles receiving no treatment, Regan et al. (1990)reported a striking association of raised follicular phase concen-trations of LH with infertility and miscarriage. When the mid-follicular phase concentrations of LH were >10 IU/1, theconception rate was 61% at 12 months compared with 80%in the women with normal LH concentrations. In those whodid conceive, the miscarriage rate in the women with normalconcentrations was 12% compared with 64% in those withelevated concentrations. These differences were very signific-ant, both statistically and clinically.
The deleterious effect of raised LH concentrations has alsobeen reported during treatment cycles. In women receivingCC, a reduced conception rate was found in those in whomserum LH concentrations were high in the follicular phase(Shoham et al., 1990a). In a large series of women with PCOSfrom our unit, treated with pulsatile, native GnRH, both failureto conceive despite ovulation and the occurrence of miscarriagerather than continuation of pregnancy were associated withelevated follicular phase concentrations of LH (Homburg et al,1988a). There was no correlation between follicular phaseconcentrations of either androgens or progesterone and concep-tion or miscarriage rates in these women.
Following laparoscopic ovulation induction, patients withthe highest preoperative LH concentrations who experiencedthe fall of LH were the most likely to ovulate spontaneouslyfollowing the operation (Abdel Gadir et al., 1993). Earlyresumption of the anovulatory state was associated with areturn to pretreatment concentrations of LH.
In addition, it has been reported that 80% of women withrecurrent spontaneous miscarriage had ultrasonically diagnosedPCO (Sagle et al., 1988). It may be argued that the verypresence of PCOS in some way influences the rates of fertilityand miscarriage. Both our series (Homburg et al, 1988a) anda study of more than 500 women with PCOS (Conwayet al., 1989), in which those complaining of infertility hadsignificantly higher LH concentrations than those with provenfertility, suggest that hypersecretion of LH is the true culpritrather than the mere presence of PCO.
Regarding the mechanism of these deleterious effects of LH,we have suggested (Homburg etal., 1988a) that, hypothetically,when concentrations of LH are high throughout the follicularphase rather than merely at the time of the physiological surge,the hormone penetrates the follicle and allows the oocyte tomature prematurely, resulting in the ovulation of an oocyte
that is physiologically 'aged'. Such oocytes are unlikely to befertilized or will tend to produce embryos that survive poorlyand therefore abort. In the normal ovulatory cycle, it is onlywith the appearance of the mid-cycle LH surge that thecompletion of the first meiotic division occurs and this timelymaturation is a prerequisite for successful fertilization anddevelopment of the embryo.
There is still some discussion as to whether the effects ofhigh LH concentrations are a direct influence on the oocyteor affect the endometrium. In an ongoing study, we haveexamined, in recipients, the performance of embryos derivedfrom oocytes donated by women with PCOS who underwentovarian stimulation with gonadotrophins with or withoutGnRHa. Thus, the endometrium of the recipients was aconstant, while the donated oocytes were exposed (no GnRHa)or not exposed (with GnRHa) to high concentrations of LHduring superovulation. In those not receiving GnRHa, theimplantation rate/embryo in the recipients was 6.6% (14/212)compared to 11.7% (26/222) in those who received GnRHa{P < 0.05, Fisher's exact test). This insinuates that the effectof high LH concentrations is directed at the oocyte and notthe endometrium.
A possible partial solution to the problem, referred toabove, is to administer GnRHa in order to block endogenousgonadotrophin excretion before and during gonadotrophintherapy. In this way, we succeeded in halving the miscarriagerate in women with PCOS treated for induction of ovulationand for superovulation for FVF (Homburg et al, 1993a). Othermethods which lower raised LH follicular phase concentrations,such as progesterone pretreatment (Homburg et al, 1988b)and ovarian wedge resection or cauterization (Abdel Gadiret al, 1993), facilitate ovulation and improve results oftreatment.
Conclusions
This summary of the state of the art of our knowledge ofPCOS has also served to indicate the gaps in this knowledge.When the jigsaw puzzle of the pathogenesis and pathophysi-ology is solved, the treatment modalities will become morespecific and, presumably, more successful regarding the therapyof the cosmetic, fertility or long-term ramifications of thisfascinating syndrome.
ReferencesAakvaag, A. (1985) Hormonal response to electrocautery of the ovary in
patients with polycystic ovarian disease. Br. J. Obstet. Gynaecol., 92,1258-1264.
Abdel Gadir, A., Khatim, M.S., Mowafi, R.S., Alnaser, H.M.I., Muharib, N.S.and Shaw W. (1992) Implications of ultrasonically diagnosed polycysticovaries. I. Correlations with basal hormonal profiles. Hum. Reprod., 4,453-457.
Abdel Gadir, A., Khatim, M.S., Alnaser, H.M.I., Muwafi, R.S. and Shaw,R.W. (1993) Ovarian electrocautery: responders versus non-responders.GynecoL Endocrinol., 7, 43-48.
Acar, B. and Kadanli, S. (1993) Diminished growth hormone responses to L-dopa in polycystic ovarian disease. FertiL Steril., 60, 984-987.
Adams, J., Franks, S., Poison, D.W., Mason, H.D., Abdulwahid, N. et al.(1985) Multifolhcular ovaries: clinical and endocrine features and responseto pulsatile gonadotrophin releasing hormone. Lancet, ii, 1375—1378.
Adams, J., Poison, D.W. and Franks, S. (1986) Prevalence of polycystic
36
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
Polycystic ovary syndrome
ovaries in women with anovulation and idiopathic hirsutism. Br Med. ].,293, 355-359.
Adashi, E., Hsueh, AJ.W. and Yen, S.S.C. (1981) Insulin enhancement ofluteinizing hormone and follicle stimulating hormone release by culturedpituitary cells. Endocrinology, 108, 1441-1449.
Adashi, E.Y., Resnick, C.E., D'Ercole, J., Svoboda, M.E. and Van Wyk, JJ.(1985) Insulin-like growth factors as intraovarian regulators of granulosacell growth and function. Endocr. Rev., 6, 400-420.
Angervo, M., Koistinen, R., Suikkari, A.M. and Seppala, M. (1991) Insulin-like growth factor binding protein-1 inhibits the DNA amplification inducedby insulin-like growth factor-1 in human granulosa-luteal cells. Hum.Reprod., 6, 770-773.
Balen, A.H., Tan, S.L. and Jacobs, H.S. (1993a) Hypersecretion of luteinisinghormone — a significant cause of infertility and miscarriage. Br. J. Obstet.Gynaecol., 100, 1082-1089.
Balen, A.H., Tan, S.L., MacDougall, J. and Jacobs, H.S. (1993b) Miscarriagerates following in-vitro fertilisation are increased in women with polycysticovaries and are reduced by pituitary desensitization with buserelin. Hum.Reprod, 8, 959-964.
Barbieri, R.L., Makris, A., Randall, R.W., Daniels, G., Kistner, R.W. andRyan, K_J. (1986) Insulin stimulates androgen accumulation in incubationsof ovarian stroma obtained from women with hyperandrogenism. J. Clin.EndocrinoL Metab., 62, 904-910.
Barnes, R.B., Rosenfield, R.L., Burstein, S. and Ehrmann, D.A. (1989)Pituitary-ovarian responses to nafarelin testing in the polycystic ovarysyndrome. N. Engl. J. Med., 320, 559-565.
Bridges, N.A., Cooke, A., Healy, Mj.R., Hindmarsh, PC. and Brook, C.G.D.(1993) Standards for ovarian volume in childhood and puberty. Fertil.Steril., 60, 456-^60.
Brown, J.B. (1978) Pituitary control of ovarian function — concepts derivedfrom gonadotrophin therapy. Aust. N.Z J. Obstet. Gynaecol. 18, 47-54.
Bunker, C.B., Newton, J., Kilborn, J., Patel, A., Conway, G.S., et al. (1989)Most women with acne have polycystic ovaries Br. J. Dermatol., 121,675-680.
Burghen, G.A., Givens, J.R. and Kitabchi, A.E. (1980) Correlation ofhyperandrogenism with hyperinsulinism in polycystic ovarian disease.J. Clin, EndocrinoL Metab. 50, 113-116.
Buyalos, R.P., Geffner, M.E., Bersch, N., Judd, H I . , Watanabe, R.M.,Bergman, R.N. and Golde, D.W. (1992) Fertil Steril, 57, 796-803.
Cara, JJ . , Fan, J., Azzraello, J. and Rosenfield, L.R. (1990) Insulin-likegrowth factor-1 enhances luteinizing hormone binding capacity to ratovarian theca-interstitial cells. J. Clin. Invest., 86, 560-565.
Cara, J.F. and Rosenfield, R. (1988) Insulin-like growth factor-1 and insulinpotentiate luteinizing hormone-induced androgen synthesis by rat ovarianthecal-interstitial cells. Endocrinology, 123, 733-739.
Carey, A.H., Chan, K.L., Short, D., White, D., Williamson, R. and Franks, S.(1993) Evidence for a single gene defect causing polycystic ovaries andmale pattern baldness. Clin. Endocrinol, 38, 653-658.
Carey, A.H., Waterworth, D., Patel, K., White, D., Little, J. et al. (1994)Polycystic ovaries and premature male pattern baldness are associated withone allele of the steroid metabolism gene CYP17. Hum. MoL Gen., 3,1873-1876.
Chereau, A. (1844) Memoires Pour Servir a I 'Etude des Maladies des Ovaries.Fortin, Masson and Cie, Paris.
Christman, G.M., Randolf, J.F., Keich, R.P. and Marshall, J.C. (1991)Reduction of gonadotropin-releasing hormone pulse frequency is associatedwith subsequent selective follicle stimulating hormone secretion in womenwith polycystic ovarian disease. J. Clin. Endocrinol. Metab., 72, 1278-1281.
Conway, G.S. and Jacobs, H.S. (1993) Clinical implications ofhyperinsulinaemia in women. Clin. Endocrinol., 39, 623-632.
Conway, G.S., Honour, J.W. and Jacobs, H.S. (1989) Heterogeneity of thepolycystic ovary syndrome: clinical, endocrine and ultrasound features in556 patients. Clin. Endocrinol., 30, 459-470.
Conway, G.S., Jacobs, H.S., Holly, J.M.P. and Wass, J.A- (1990) Effects ofluteinizing hormone, insulin, insulin-like growth factor-1 and insulin-likegrowth factor small binding protein-1 in the polycystic ovary syndrome.Clin. EndocrinoL, 33, 593-603.
Conway, G.S., Avey, C. and Rumsby, G. (1994) The tyrosine kinase domainof the insulin receptor gene is normal in women with hyperinsulinaemiaand polycystic ovary syndrome. Hum. Reprod., 9, 1681-1683.
Crowley, W.F., Hall, J.E., Martin, K.A., Adams, J. and Taylor, A.E (1993)An overview of the diagnostic considerations in polycystic ovariansyndrome. Ann. NX Acad. Sci., 687, 235-241.
Dabirashrafi, H., Mohamad, K., Behjatnia, Y. and Moghadami-Tabrizi, N.(1991) Adhesion formation after ovarian electrocauterization on patientswith polycystic ovarian syndrome. FertiL SteriL, 55, 1200-1201.
Dahlgren, E., Friberg, L-G., Johanasson, S., Lindstrom, B., Oden, A., Samsioe,
G. (1991) Endometrial carcinoma. Ovarian dysfunction — a risk factor inyoung women. Eur. J. Obstet. GynecoL Reprod. Biol., 41, 143-150.
Dahlgren, E., Johansson, S., Lindstedt, G., Knuttson, F., Oden, A. et al. (1992)Women with polycystic ovary syndrome wedge resected in 1956-65. Along-term follow-up focusing on natural history and circulating hormones.FertiL Steril., 57, 505-513.
Dale, P.O., Tanbo, T., Djoseland, O., Jervell, J. and Abyholm, T. (1992)Persistence of hyperinsulinemia in polycystic ovary syndrome after ovariansuppression by gonadotropin-releasing hormone agonist Ada. Endocrinol.,126, 132-136.
Dale, P.O., Tanbo, T., Lunde, O. and Abyholm, T. (1993) Ovulation inductionwith low dose follicle stimulating hormone in women with the polycysticovary syndrome. Acta. Obstet. GynecoL Scand., 72, 43—46.
Daly, D.C., Walters, C.A., Soto-Albors, C.E., Tohan, N. and Riddick, D.H.(1984) A randomized study of dexamethasone in ovulation induction withclomiphene citrate. FertiL Steril., 41, 844-848.
Dewailly, D., Robert, Y, Helin, I., Ardaens, Y., Thomas-Desrousseaiix, P.et al. (1994) Ovarian stroma] hypertrophy in hyperandrogenic women. Clin.EndocrinoL, 41, 557-562.
Donesky, B.W. and Adashi, E.Y. 1995) Surgically induced ovulation inthe polycystic ovary syndrome: wedge resection revisited in the age oflaparoscopy. FertiL Steril., 63, 439-463.
Dunaif, A., Segal, K., Futterweit, W. and Dobrjansky, A. (1989) Profoundperipheral resistance independent of obesity in polycystic ovary syndrome.Diabetes, 38, 1165-1174.
Dunaif, A., Segal, K.R., Green, G., Dobjransky, A. and Licholai, T. (1993)Evidence for distinctive and intrinsic defects in insulin action in polycysticovary syndrome. Diabetes, 41, 1257-1266.
Eden, J.A. (1988) Which is the best way to detect the polycystic ovary? Aust.N.Z J. Obstet Gynaecol., 28, 221-224.
Eden, J.A., Place, J., Carter, G.D., Alaghband-Zadeh, J. and Pawson, M.(1989) Is the polycystic ovary a cause of infertility in the ovulatory woman?Clin. EndocrinoL, 30, 77-82.
Farhi, J., Homburg, R., Lemer, A. and Ben-Rafael, Z. (1993) The choice oftreatment for anovulation following failure to conceive with clomiphene.Hum. Reprod, 8, 1367-1371.
Filicori, M., Campaiello, E., Michelacci, L., Pareschi, A., Ferrari, P. et al.(1988) Gonadotropin-releasing hormone (GnRH) analog suppression renderspolycystic ovarian disease patients more susceptible to ovulation inductionwith pulsatile GnRH. J. Clin. EndocrinoL Metab., 66, 327-333.
Fleming, R., Haxton, MJ., Hamilton, M.P.R., McCune, G.S., Black, W.P.et al. (1985) Successful treatment of infertile women with oligomenorrhoeausing a combination of an LHRH agonist and exogenous gonadotrophins.Br. J. Obstet. GynaecoL, 92, 369-373.
Franks, S. (1989) Polycystic ovary syndrome: a changing perspective. Clin.EndocrinoL, 31, 87-120.
Franks, S. and White, D.M. (1993) Prevalence of and aetiological factors inpolycystic ovary syndrome. In Franks, S. and Neuman, F (eds), PolycysticOvary Syndrome. A New Approach to Treatment. Chester, UK, pp. 19—21.
Garcea, N., Campo, S., Panetta, V., Venneri, M., Siccardi, P. et al. (1985)Induction of ovulation with purified urinary follicle stimulating hormone inpatients with polycystic ovarian syndrome. Am. J. Obstet. GynecoL, 151,635-640.
Givens, J.R., Anderson, R.N., Wiser, W.L., Umstat, E.S. and Fash, S.A.(1976) The effectiveness of two oral contraceptives in suppressing plasmaandrostendione, testosterone, LH and FSH, and in stimulating plasmatestosterone-binding capacity in hirsute women. Am. J. Obstet. GynecoL,124, 333-339.
Gjonnaess, H. (1984) Polycystic ovarian syndrome treated by ovarianelectrocautery through the laparoscope. FertiL Steril, 41, 20-25.
Greenblatt, E.M. and Casper, R.F. (1993) Adhesion formation afterlaparoscopic ovarian cautery for polycystic ovary syndrome — lack ofcorrelation with pregnancy rate. FertiL Steril., 60, 766-770.
Grulet, H., Hecart, A.C., Delemer, B., Gross, A., Sulmont, V. et al. (1993)Roles of luteinizing hormone and insulin resistance in lean and obesepolycystic ovary syndrome. Clin. EndocrinoL, 38, 621-626.
Gurgan, T., Yarali, H. and Urman, B. (1994) Laparoscopic treatment ofpolycystic ovarian disease. Hum. Reprod., 9, 573-577.
Hague, W.M., Adams, J., Reeders, ST., Peto, T.E. and Jacobs, H.S. (1988)Familial polycystic ovaries: a genetic disease? Clin. Endocrinol., 29,593-605.
Hamilton-Fairley, D., Kiddy, D., Watson, H., Paterson, C. and Franks, S.(1992) Association of moderate obesity with a poor pregnancy outcome inwomen with polycystic ovary syndrome treated with low dosegonadotrophin. Br. J. Obstet Gynaecol. 99, 128-131.
Hamilton-Fairley, D., Kiddy, D., Watson, H., Sagle, M. and Franks, S. (1991)
37
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
R-Homburg
Low dose gonadotrophin therapy for induction of ovulation in 100 womenwith polycystic ovary syndrome. Hum. Reprod., 6, 1095-1099.
Heylen, S.M., Punemans, PJ. and Brosens, LA. (1994) Polycystic ovariandisease treated by laparoscopic argon capsule drilling: comparison ofvaporization versus performation technique. Hum. Reprod., 9, 1038-1042.
Homburg, R. and Jacobs, H.S. (1990) Clinical features of growth hormoneaction on human ovarian function. Isr. J. Obstet. GynecoL 1, 114-129.
Homburg, R., Armar, N.A., Eshel, A., Adams, J. and Jacobs, H.S. (1988a)Influence of serum luteinizing hormone concentrations on ovulation,conception and early pregnancy loss in polycystic ovary syndrome. Br.Med. J. 297, 1024-1026.
Homburg, R., Weissglass, L. and Goldman, J. (1988b) Improved treatment foranovulation in polycystic ovarian disease on the inappropriate gonadotrophinrelease and clomiphene response. Hum. Reprod., 3, 285-288.
Homburg, R., Eshel, A., Tucker, M., Mason, P., Adams, J. et al. (1989)One hundred pregnancies following pulsatile luteinising hormone-releasinghormone therapy for ovulation induction. Br. Med. J. 298, 809—812.
Homburg, R., Eshel, A., Kilborn, J., Adams, J. and Jacobs, H.S. (1990)Combined luteinising hormone-releasing hormone analogue and exogenousgonadotrophins for the treatment of infertility associated with polycysticovary syndrome. Hum. Reprod., 5, 32-35.
Homburg, R., Pariente, C , Lunenfeld, B. and Jacobs, H.S. (1992) The roleof insulin-like growth factor-1 (1GF-1) and IGF binding protein-1 (IGF-BP-1) in the pathogenesis of polycystic ovary syndrome. Hum. Reprod., 7,1379-1383.
Homburg, R., Levy, T., Berkovitz, D., Farhi, J., Feldberg, D. et al. (1993a)Gonadotropin-releasing hormone agonist reduces the miscarriage rate forpregnancies achieved in women with polycystic ovary syndrome. Fertil.SteriL, 59, 527-531.
Homburg, R,, Berkovitz, D., Levy, T., Feldberg, D., Ashkenazi, J. and Ben-Rafael, Z. (1993b) In vitro fertilization and embryo transfer for the treatmentof infertility associated with polycystic ovary syndrome. Fertil. SteriL, 60,858-863.
Homburg, R., Levy, T. and Ben-Rafael, Z. (1995a) A comparative prospectivestudy of conventional regimen with chronic low dose administration offollicle stimulating hormone for anovulation associated with polycysticovary syndrome. Fertil. SteriL, 63, 729-733.
Homburg, R., Levy, T. and Ben-Rafael, Z. (1995b) Adjuvant growth hormonefor induction of ovulation with gonadotropin releasing-hormone agonistand gonadotropins in polycystic ovary syndrome: a randomized, double-blind, placebo controlled trial. Hum. Reprod., 10, 2550-2553.
Howies, CM., MacNaroee, M.C., Edwards, R.G., Goswamy, R. and Steptoe,P.C. (1986) Effect of high tonic concentrations of luteinising hormone onoutcome of in vitro fertilisation. Lancet, U, 521-523.
Hull, M.G. (1987) Epidemiology of infertility and polycystic ovarian disease:endocrinological and demographic studies. Cynaecol. Endocrinol. 1, 235-245.
Insler, V. and Lunenfeld, B. (1991) Pathophysiology of polycystic ovariandisease: new insights. Hum. Reprod., 6, 1025—1029.
Insler, V., Shoham, Z., Barash, A., Koistenen, R., Seppala, M., Hen, M.,Lunenfeld, B. and Zadik, Z. (1993) Polycystic ovaries in obese and non-obese patients: possible pathophysiological mechanism based on newinterpretation of facts and findings. Hum. Reprod., 8, 379-384.
Jaatinen, T.A., Antilla, L., Erkkola, R., Koskinen, P., Laippola, P. et al. (1993)Hormonal responses to physical exercise in patients with polycystic ovarysyndrome. FertiL SteriL, 60, 262-267.
Kazer, R.R., Kessel, B. and Yen, S.S. (1987) Circulating luteinizing hormonepulse frequency in women with polycystic ovary syndrome. J. Clin.EndocrinoL Metab., 65, 233-6.
Khan, C.R., Flier, J.S., Bar, R.S., Archal, J.A. and Gorden, P. (1976) Thesynehomes of insulin resistance and acanthosis nigricans: Insulin receptordisorders in man. New. EngL J. Med., 294, 739-745.
Kiddy, D.S., Hamilton-Fairiey, D., Bush, A., Anyaoku, V, Reed, MJ. andFranks, S. (1992) Improvement in endocrine and ovarian function duringdietary treatment of obese women with polycystic ovary syndrome. Clin.Endocrinol., 36, 105-111.
Kovacs, G., Buckler, H., Bangah, M., Outch, K., Burger, H., Healy, D.(1991) Treatment of anovulation due to polycystic ovarian syndrome bylaparoscopic ovarian electrocautery. Br. J. ObsteL Gynaecol., 98, 30-35.
Lanzone, A., Fulghesu, A.M., Guido, M., Ciampelli, M., Caruso, A. andMancuso, S. (1994) Differentia] androgen response to adrenocorticotrophinhormone stimulation and effect of opioid antagonist on insulin secretion inpolycystic ovary syndrome. Hum. Reprod., 9, 2342-2349.
MacDougall, MJ., Tan, S I . , Balen, A. and Jacobs, H.S. (1993) A controlledstudy comparing patients with and without polycystic ovaries undergoingin-vitro fertilization. Hum. Reprod., 8, 233-237.
Mason, H.D., Margara, R., Winston, R.M., Beard, R.W., Reed, MJ. and
Franks, S. (1990) Inhibition of oestradiol production by epidermal growthfactor in human granulosa cells of normal and polycystic ovaries. Clin.EndocrinoL, 33, 511-517.
McFaul, P.B., Traub, A.I. and Thompson, W. (1990) Treatment of clomiphenecitrate resistant polycystic ovary syndrome with pure follicle stimulatinghormone or human menopausal gonadotropin. Fertil. SteriL, 53, 792-797.
McLachlan, R.I., Robertson, D.M., Healy, D.L., Burger, H.G. and deKretser,DM. (1987) Circulating immunoreactive inhibin concentrations during thenormal human menstrual cycle. J. Clin. EndocrinoL Metab., 65, 954-957.
Messinis, I.E. and Templeton, A. (1990) In vivo bioactivity of gonadotrophinsurge attenuating factor (GnSAF). Clin. EndocrinoL, 33, 213-218.
Naether, O.G., Fischer, R., Weise, H.C., Geiger Kotzler, L., Delfs, T. andRudolf, K. (1993) Laparoscopic electrocoagulation of the ovarian surfacein infertile patients with polycystic ovarian disease. FertiL Steril., 60, 88—94.
Naether, O.GJ., Baukloh, V., Fishcher, R. and Kowalczyk, T. (1994) Longterm follow-up in 206 infertility patients with polycystic ovary syndromeafter laparoscopic electrocautery of the ovarian surface. Hum. Reprod., 9,2342-2349.
Nestler, J.E., Barlascini, CO., Matt, D.W., Steingold, K.A., Plymate, S.R.,Clore, J.N. and Blackard, W.G. (1989) Suppression of serum insulin bydiazoxide reduces serum testosterone levels in obese women with polycysticovary syndrome. J. Clin. EndocrinoL Metab., 68, 1027-1032.
Ovesen, P., Moller, J., Moller, N., Christiansen, J.S., Jorgenson, J.O.L. andOrskov, H. (1992) Growth hormone secretory capacity and serum insulin-like growth factor-1 levels in primary infertile anovulatory women withregular menses. FertiL SteriL, 57, 97-101.
Plymate, S.R., Matej, L.A., Jones, R.E. and Fnedl, K.E. (1988) Inhibition ofsex hormone binding globulin in the human hepatoma (HepG2) cell lineby insulin and prolactin. J. Clin. EndocrinoL Metab., 67, 460—464.
Poison, D.W., Adams, J., Steer, PJ. and Franks, S. (1986) Unilateral polycysticovary. Case report Br. J. Obstet. Gynaecol., 93, 1100-1113.
Poison, D.W., Mason, H.D., Saldahna, M.B. and Franks, S. (1987) Ovulationof a single dominant follicle during treatment with low dose pulsatilefollicle stimulating hormone in women with polycystic ovary syndrome.Clin. EndocrinoL, 26, 205-212.
Poison, D.W., Adams, J., Wadsworth, J. and Franks, S. (1988) Polycysticovaries — a common finding in normal women. Lancet 1, 870-872.
Poretsky, L., Grigorescu, E, Seibel, M., Moses, A.C and Flier, S. (1985)Distribution and characterization of insulin and insulin-like growth factor-1 receptors in normal human ovary. J. Clin. EndocrinoL Metab., 61,728-734.
Prelevic, G.M., Ginsburg, J., Maletic, D., Hardiman, P., Okolo, S. et al. (1985)The effects of somatostatin analogue ocrreotide on ovulatory performancein women with polycystic ovaries. Hum. Reprod., 10, 28-32.
Prelevic, G.M., Wurzburger, M.I. and Balint-Peric, L. (1989) Effects of a lowdose estrogen-antiandrogen combination (Diane-35) on clinical signs ofandrogenization, hormone profile and ovarian size in patients with polycysticovary syndrome. GynecoL EndocrinoL, 3, 269-280.
Prelevic, G.M., Wurzburger, M.I., Balint-Peric, L. and Nesic, J.S. (1990)Inhibitory effect of sandostatin on secretion of luteinising hormone andovarian steroids in polycystic ovary syndrome. Lancet, 336, 900-903.
Prelevic, G.M., Wurzburger, M.I., Balaint-Peric, L. and Ginsburg, J. (1992)Twenty-four h serum growth hormone, insulin, c-peptide, and blood glucoseprofiles and serum insulin-like growth factor-1 concentrations in womenwith polycystic ovaries. Horn. Res., 37, 125—131.
Quigley, M.E., Rakoff, J.S. and Yen, S.S. (1981) Increased luteinizing hormonesensitivity to dopamine inhibition in the polycystic ovary syndrome. J. Clin.EndocrinoL Metab., 52, 231-234.
Rebar, R., Judd, HX., Yen, S.C.C, Rakoff, J., Vandenberg, G. and Nafiolin,F. (1976) Characterization of the inappropriate gonadotropin secretion inpolycystic ovary syndrome. / . Clin. Invest., 57, 1320-1329.
Regan, L., Owen, EJ. and Jacobs, HS. (1990) Hypersecretion of luteinisinghormone, infertility and miscarriage. Lancet 336, 1141-1144.
Robinson, S., Kiddy, D., Gelding, S.V., Willis, D., Niththyanathan, R. et al.(1993) The relationship of insulin insensitivity to menstrual pattern inwomen with hyperandrogenaemia and polycystic ovaries. Clin. EndocrinoL,39, 351-355.
Rosenfield, RX., Barnes, R.B., Cara, J.F. and Lucky, AW. (1990) Dysregulationof cytochrome P450c 17 alpha as the cause of polycystic ovary syndrome.FertiL Steril., 53, 785-791.
Sagle, M., Bishop, K., Ridley, N., Alexander, F.M., Michel, M. et al. (1988)Recurrent early miscarriage and polycystic ovaries. Br. Med. J., 297,1027-1028.
Saxton, D.W., Farquhar, CM., Rae, T., Beard, R.W., Anderson, M.C. andWadsworth, J. (1990) Accuracy of ultrasound measurements of femalepelvic organs. Br. J. ObsteL Gynaecol., 97, 695-699.
Schachter, M., Balen, A.H., Patel, A. and Jacobs, H.S. (1995). Hypo-gonadotrophic patients with ultrasonographically detected polycystic ovaries
38
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
Polycystic ovary syndrome
have aberrant gonadotropin secretion when treated with pulsatilegonadotropin releasing hormone — a new insight to the pathophysiologyof polycystic ovary syndrome. Fertil Steril. (in press).
Scheele, E, Hompes, P.G.A., Van der Meer, M., Schoute, E. and Schoemaker, J.(1993a) The effects of a gonadotrophin-releasing hormone-agonist ontreatment with low dose follicle stimulating hormone in polycystic ovarysyndrome. Hum. Reprod., 8, 699-704.
Scheele, F., Hompes, P.G.A., Van der Meer, M., Schoute, E. and Schoemaker, J.(1993b) The relationship between FSH dose and FSH level and its relevancefor ovulation induction with adjuvant gonadotropin-releasing hormone-agonist treatment. Fertil. Steril., 60, 620-625.
Seibel, M.M., Kamrava, M.M., McArdle, C. and Taymor, M.L. (1984)Treatment of polycystic ovarian disease with chronic low dose folliclestimulating hormone: biochemical changes and ultrasound correlation. Int.J. Fertil., 29, 39-43.
Shaw, R.W. (1989) Use of nafarelin to investigate die pathophysiology of thepolycystic ovary syndrome. J. Reprod. Med., 34, 1039-1043.
Shoham, Z., Borenstein, R., Lunenfeld, B. and Pariente, C. (1990a) Hormonalprofiles following clomiphene citrate therapy in conception and non-conception cycles. Clin. Endocrinol., 33, 271-278.
Shoham, Z., Homburg, R. and Jacobs, H.S. (1990b) Induction of ovulationwith pulsatile LHRH. Ball. Clin. Obstet. GynecoL, 4, 589-608.
Shoham, Z., Patel, A. and Jacobs, HS. (1991) Polycystic ovary syndrome:safety and effectiveness of stepwise and low dose administration of purifiedfollicle stimulating hormone. Fertil Steril., 55, 1051-1056.
Shoham, Z., Conway, G.S., Patel, A. and Jacobs, H.S. (1992) Polycysticovaries in patients with hypogonadotropic hypogonadism: similarity ofovarian response to gonadotropin stimulation in patients with polycysticovary syndrome. Fertil. Steril., 58, 37-45.
Sopelak, V.M. and Hodgen, G.D. (1984) Blockade of the estrogen-inducedluteinizing hormone surge in monkeys: a nonsteroidal, antigenic factor inporcine follicular fluid. Fertil Steril, 41, 108-113.
Stanger, J.D. and Yovich, J.L. (1985) Reduced in-vitro fertilisation of humanoocytes from patients with raised basal luteinising hormone levels duringthe follicular phase. Br. J. Obstet. Cynaecol., 92, 385-393.
Stein, I.F. and Leventhal, M.L. (1935) Amenorrhea associated with bilateralpolycystic ovaries. Am. J. Obstet. Gynecol., 29, 181-191.
Stephens, J., Carey, A.H., Short, F., Williamson, R. and Franks, S. (1993)Premature balding, the male phenotype of polycystic ovaries. J. Endocrinol,139,31.
Suikkari, A.M., Koivisto, V.A., Rutanen, E.M., Yki-Larvinen, H., Karonen,S.L. and Seppala, M. (1988) Insulin regulates the serum levels of lowmolecular weight insulin-like growth factor-binding protein. / . Clin.Endocrinol. Metab., 66, 266-272.
Swanson, M., Sauerbrie, E.E. and Cooperberg, P.L. (1981) Medicalimplications of ultrasonically detected polycystic ovaries. / . Clin.Ultrasound, 9, 219-222.
Tapanainen, J., Ronnberg, L., Martikainen, H., Reinila, M., Koistenen, R. andSeppala, M. (1991) Short and long-term effects of growth hormone oncirculating levels of insulin-like growth factor-1 (IGF-1), IGF-1 bindingprotein-1 and insulin: a placebo controlled study. J. Clin. EndocrinolMetab., 73, 7\-14.
Trounson, A., Wood, C. and Kausche, A. (1994) In vitro maturation and thefertilization and developmental competence of oocytes recovered fromuntreated polycystic ovarian patients. Fertil Steril., 62, 353-362.
Vaitukaitis, J.L. (1983) Polycystic ovary syndrome: what is it? N. Engl J.Med.,309, 1245-1246.
Van den Berg, G. and Yen, S.S.C. (1973) Effect of anti-estrogenic action ofclomiphene during the menstrual cycle: evidence for a change in thefeedback sensitivity. J. Clin. Endocrinol Metab., 37, 356-365.
Van der Meer, M., Hompes, P.G.A., Scheele, F., Schoute, E., Veersema, S.and Schoemaker, J. (1994) Follicle stimulating hormone (FSH) dynamicsof low dose step-up ovulation induction with FSH in patients with polycysticovary syndrome. Hum. Reprod., 9, 1612-1617.
Velaquez, E.M., Mendoza, S., Hamer, T., Sosa, F. and Glueck, CJ. (1994)Metformin therapy in polycystic ovary syndrome reduces hyperinsulinemia,insulin resistance, hyperandrogenemia and systolic blood pressure whilefacilitating normal menses and pregnancy. Metabolism, 43, 647-654.
Wang, C.F. and Gemzell, C. (1980) The use of human gonadotropin for theinduction of ovulation in women with polycystic ovarian disease. FertilSteril, 33, 479-486.
Received on April 28, 1995; accepted on October 12, 1995
39
Dow
nloaded from https://academ
ic.oup.com/hum
rep/article/11/1/29/595893 by guest on 28 July 2022
![[Use of metformin in polycystic ovary syndrome]](https://static.documents.pub/doc/80x56/6348558ade40dd034d092622/use-of-metformin-in-polycystic-ovary-syndrome.jpg)
