Recent advances in hormone replacement therapy

Roger NJ Simth/John WW Studd, Kings College Hospital, London

I n v i e w o f t h e benefits o f h o r m o n e replacement t h e r a p y ( H R T ) a n d t h e experience t h a t o n l y 1 0 % of w o m e n m a i n t a i n t h e r a p y , i t IS v i t a l t h a t a l l doctors are f a m i l i a r w i t h t h e pros a n d cons of H R T , so t h a t w o m e n m a y be correc t ly advised H e r e we discuss new issues a n d developments m H R T a n d c u r r e n t views o n i n d i c a t i o n s , compl i cat ions , a n d regimens of t r e a t m e n t

M r Smith- is- correntiy Research Fellow and Mr Studd I S currently Consultant Gynaecolopst m the Department of Obstetrics and Gynaecology at The Chelsea and Westmrnster Hospital, London

Correspondence to Mr John Studd, Consultant Gynaecologist, Chelsea and •Westminster Hospital, London SW10 9 N H

I > n England and Wales there are approx i ­mately 10 m i l l i o n climacteric or postmeno­pausal' women ( C e n t r a l Statistical' Office, 1991), therefore one i n five o f the t o t a l populat ion IS a potent ia l candidate for hor ­mone replacement therapy ( H R T ) {Ftg 1)

Therapy has medical implications m several areas, mcludmg depression, ischaemtfr heart dtsease- (fHE>->, stroke- and-thrombosis, osteoporosis, premature meno­pause, and breast cancer

Depression I t IS generally accepted t h a t the climacteric frequently leads to- -^^anous unpleasant symptoms, which are opt imal ly treated w i t h oestrogen replacement (Studd et al, 1977) T h e best charactenzed o f these are h o t flushes and vagmal dryness I n addit ion , a variety o f other symptoms, such as m -sommar headaches,, d-yspareuniar loss, o f l i b ido , generalized aches and pams, poor concentration, poor memory , i r r i t a b i l i t y and depression, are often associated -with the menopause and constitute the less we l l -defmed menopausal syndrome T h e emer­gence of data supporting: the theory that fluctuations m oestrogen levels may p r o ­foundly influence m o o d supports the existence of such a climacteric syndrome before the fmal cessation of penods

Fig 1 Number of climacteric and postmenopausal women in England and Wales (Central Statistical Office, 1991)

Changes m the pattern of ovanan hor ­mone secretion are often associated w i t h the development o f depression I n premenst^ua^ syndrome (PMS), the m o o d change is an abnormal response t o the physiological changes o f the menstrual cycle, b u t m b o t h postnatal depression and clunactenc depres­sion m o o d change is associated w i t h a major upheavat o f ovanan f o n c t t o n These obser­vations suggest that ovanan hormones may modulate mood , a concept m keepmg w i t h the generally greater mcidence of depressive disorders m women compared w i t h men

T h e pubhshed data all suggest that oestrogen elevates m o o d Oestrogen-therapy has been shown to accelerate recovery f r o m postnatal depression (Henderson et al, 1991), t o be an effective treatment for PMS, mcludmg premenstrual depression (Magos et al , 1986a), and t o signi£<anjly ekva-te m o o d m. clunactenc. depression (Montgomery et al, 1987) (Fig 2) Oestrogen also elevates m o o d m asymp­tomat ic non-depressed postmenopausal women ( D i t k o f f et al , 1991), and m h igh doses IS an effective treatment for severe depression (Klaiber et al^ 1979) By compan-son there is no evidence that either pro ­gestogens or progesterone elevate mood They are no t effective treatments for PMS and Magos et al (1986b) have shown that progestogens can cause many of the symp­toms o f PMS, mcludmg depression

W h i l e much collaborative w o r k o n the relationship between oestrogen, proges­terone and depression remains t o be done, i t is now t ime for a wider acceptance by psychiatrists that oestrogen has a role i n the treatment of depression m women

Ischaemic heart disease, stroke and thrombosis

I t IS regrettable that I H D is s t i l l widely regarded as a contraindication t o H R T , as this IS completely at odds w i t h the pubhshed data

I n one o f the rare circumstances m

Bncish Joumil of Hospiod Mediane 1993, Vol 49, No 11 799

i i „ L _ I I h I I 1 H I ^ I L | p i I i l L L I U

2 -

» - * i 1 i 1 i-.—

Ftg. 2 Effect of oestradiQl 5Q mg, implaata o a perimeaopausaL deprea-sion (Montgomery et al, 1987} * = significantly different from placebo (P<0 01)

Fig 3 Potential impact of hormone replacement therapy (HRT) upon annual deaths from ischaemtc heart disease

wiuch the data are i n ahnost complete agreement, i t is clear t h a t H R T markedly reduces the nsk of I H D A recent meta­analysis of 31 pubhshed studies f o u n d an overall nsk rat io for I H D o f 0 56 (95% confidence mterval. 5.-0.61.). foe oestrogen users (Stampfer and C o l d i t z , 1991) Even when only the 13 most rigorous studies employmg a prospective cohor t design were mcluded, the nsk rat io f o r oestrogen users was stiU 0 58 (95 % confidence mterval 0 48-0 69). T h i s reduct i on m n s k o f I H D 14 numencally the most significant l ong - term effect of H R T , because I H D is the greatest cause o f death among postmenopausal women U s m g the 1990 m o r t a h t y statistics and populat ion data f r o m the 1981 census, a reduced relative r isk o f death f r o m I H D among oestrogen users of 0 56 w o u l d be translated i n t o a savmg o f approximately 24 000 lives per annum m England and

800

Myocartisi krfafctwfl nsk

Strokarek

Fig 4 Differential effects of the piit and hormone replacement therapy upon cardio­vascular disease

Wales alone i f all postmenopausal women used H R T (Central Statistical Office, 1991, Office o f Population Censuses and Surveys, 1991) (Fig 3)

T h e one theoretical flaw m this epidemio­logical consensus is that for the last t w o or three decades H R T may n o t have been given t o patients w i t h a h igh nsk o f heart disease Thus i t is possible t h a t these encou-ragmg results are due t o selection bias, a l though the L i p i d Research Climcs study f o u n d the greatest reduction m cardiovas­cular mor tahty among high-nsfe patients w i t h elevated b lood h p i d levels (Bush et al, 1987)

U n t i l recently i t was t h o u g h t that oestro­gen acted mamly t h r o u g h h p i d changes, 1 e b y mcreasmg high density l ipoprotem and r e d u c i r ^ low density- hpoprotem choles­t e r o l Recent data mdicate that oestrogen also has other desirable physiological actions A n i m a l data have shown that oestrogen exerts a direct atheroma-m h i b i t m g action o n the artenal wal l mde-pendently^of b l o o d cholesterol-(Hoagfc and 2dversmit , 1986) Co l our flow D o p p l e f imagmg has demonstrated that oestrogen mcreases penpheral b l o o d flow, b o t h m the uterme artery and m the m t e m a l carotid artery (Bourne et al, 1990, Gangar et al, 1991). Oestrogen-may also-work b y modi fy -m g carbohydrate metabohsm and body fat d i s t n b u t i o n , b o t h o f which are related t o I H D nsk

A history o f thrombosis is also often regarded as an absolute contramdication t o H B T v presumably because, i t is. assumed t h a t postmenopausal H R T has the same thrombogemc potent ia l as the contracep­t ive p i l l T h i s is n o t the case ( f j g 4) There is n o w wor ldwide expenence o f H R T , pre-dommant ly oral conjugated eqmne oestro-gens (Premarm), which has n o t sho-wn any mcreased mcidence of thrombosis , mdeed, H R T actually reduces the r isk o f stroke (Paganmi-Hi l l et a l , 1988)

Thnm H R T

t »

Fig 5 Increase in bone density achieved with different routes of oestrogen administration (Chnstiansen and Christiansen, 1981, Lindsay et al, 1976, Stevenson et al, 1990, Studd et al, 1990) Implant = 75 mg, patch = 50 ng, oral = oestradiol 2 mg

Ttere - are. ^ezeraL -differences, between H R T and the combined oral contraceptive p d l ( C O C P ) wh i ch account for this apparent discrepancy M o s t i m p o r t a n t l y , the synthetic oestrogens m the C O C P are f our t o eight times as potent as the natural oestrogens m H R T at mducing. the hver enzyme systems that produce c l o t t m g fac­tors (Campbell , 1982) I n addiuon , the dose o f oestrogen t h a t is usually employed m H R T is approximately one s ix th o f that commonly used m the C O C P T h e avoid­ance of first-pass hepatic effects by employ­mg a non-oral route of delivery reduces any potent ia l thrombogemc nsk even further W h i l e conventional oral H R T results m some nunor changes m laboratory measures of c l o t t m g funct ion , such as a shght reduc-tton m a n t i t h r o m b m I I I , no such changes m coagulation, fibrmolysis or platelet funct ion are seen, even w i t h a potent non-oral pre­parat ion, e g a 50 m g subcutaneous oestra­d io l implant (Studd et al, 1978)

Thus I H D , or the presence of nsk factors for I H D , should be regarded, no t as con­traindications, b u t as positive mdications f o r postmenopausal H R T A history of previous thrombosis need only constitute a contramdication m those rare cases where there is a permanent mcrease m thrombot i c nsk , e g due t o a n t i t h r o m b m I I I , p ro tem C o r p r o t e m S def faeuqr Where concern exists, non-oral oestrogens such as percuta­neous patches or subcutaneous implants

1 - \f Vol 49 No 11

should be used m preference t o oral therapy

Osteoporosis I n osteoporosis, bone is normal ly m m e r -ahzed b u t reduced m quant i ty such t h a t mass per u m t volume is reduced Osteo­porosis IS a major cause of pam, disabihty, suffenng and death m elderfy women One quarter of white women over the age o f 60 years have radiological evidence of vertebral crush fractures, whde the combmed m a -dence of fractures of vertebra, femoral neck and forearm m the over-65s is estimated t o b e 55-4Q%- (-Stevenson a n d Whttei iead, 1982) I n t u r n , the fmancial burden on the health service is considerable, -mth fracture o f the femur bemg the t h i r d commonest reason for occupancy o f a non-psychiatnc hospital bed m England and Wales

The- trs^edy. is that postmenopausal osteoporosis is preventable First , i t is wel l established t h a t oestrogen replacement therapy b o t h prevents the acceleranon o f bone loss which normally accompanies the menopause (Lmdsay et al , 1978) and reduces tke. sihsfiiquent n s k o f feictiice ( K e i l e t ai,. 1987) Second, oestrogen actually mcreases bone density, so that oestrogen replacement can be used t o correct l ow bone density m thosealready at mcreased nsk of fracture I n this respect oestrogen seems to exhib i t a dose-response effect For example,, ora l oestrogen m general mcreases vertebral bone density by approximately 1-2% per annum (Lmdsay et al , 1976, Christiansen and Christiansen, 1981), whereas the 75 m g oestradiol implant , which achieves oestra­d i o l levels m general at least double those of oral therapy, has been shown to mcrease vertebral bone density b y 8 3% per annum (Studd et al, 1990) T h e comparable figure for the 50 ^g transdermal patch is a 3 5% mcrease over 12 months (Stevenson et a l , 1990) (Fig 5) T h e f m d m g o f a significant correfation between oestradiol level and mcrease m vertebral bone density by Studd et al (1990) confirms this dose-response relationship

Oestrogen should be regarded as the benchmark agamst which al l other potent ia l treatments, b o t h prophylactic andremedial , should be judged Calc ium supplements are prescnbed as a non-hormonal alternative t o oestrogen, b u t although calcium may be impor tant m early life t o estabksh adequate peak bone mass, there is no evidence t h a t c r f c r a m i a t e r r n h f e e i ther prevents- ortreats-osteoporosis Sinularly, while undoubtedly able t o mfluence bone mass, physical exer-

801

Recent advances m hormone replacement therapy

I M n M l l l d l B l

oiiii.»nattmit KH*MM«ltf 2 «

i f c i M i I i M t o a n •akiiuitnD]

- r 05

r r r

10 I S 20 25

Beiaiivanslcan<195%«>n6denc8inferval

30 3S

Fig 6 Disparity of data concerning oestrogen and breast cancer incidence (from Dupont and Page, 1991, references from tfie figure can be found with origmaf article)

CISC alone i s n o t adequate t o prevent meno­pausal osteoporosis F luonde mcreases bone mass, b u t has been f o u n d t o mcrease fracture madence because o f an adverse effect upon bone quahty (Lindsay and Cosman, t99£9" Ca i c i t onm is able t o pre­vent bone loss b u t has the major disadvan­tage o f admimstrat ion b y either mject ion o r nasal spray, and i t is very expensive T h e most mterestmg non-hormonal approach t o date IS the bisphosphonate, disodium e t i ­dronate- (Didronely T h » preparat ion has the convemence o f oral admimstrat ion , is able n o t on ly t o prevent bone loss b u t also t o mcrease bone density by approximately 2% per year, and has been shown t o reduce the mcidence o f radiologically detected vertebral fracture (Storm-et a l , 19.9&, Watts -et al , 1990)

T h e mechamsm b y wh i ch oestrogen exerts i t s beneficial effect upon bone mmeral and m a t r i x is at present u n k n o w n I t does n o t seem t o influence c a l a u m or -o-tamm D metabolism,, h u t may a.ct. m -direct ly by mcreasmg the act iv i ty o f caici­t o n m A l b n g h t et al (1941) beheved that postmenopausal osteoporosis was essen­t ia l ly a generahzed deficiency of collagen, v n t h loss o f the collagenous bone m a t n x bemg the prmcipa l mechamsm I t is k n o w n t h a t postmenopausal oestrogen therapy results m a 3 0 % mcrease m skm thickness and a 3 4 % mcrease m s k m collagen content (Bnncat et al , 1985) I t is possible t h a t i n a similar manner oestrogen mcreases the co l ­lagenous m a t n x o f bone (Studd et a l , 1990) T h i s IS current ly bemg mvestigated by means of serial h is tomorphometnc studies o n bone biopsy specimens T h e possibihty t h a t oestrogen may act t h r o u g h collagen

holds ou t the possibihty that oestrogen therapy may be able t o promote healmg of bone m which trabecular d isrupt ion has already occurred, and thus be used t o treat severe osteoporosis m which bones have already fractured

Premature menopause and infertility

Cases of premature menopause are usually id iopathic b u t may also be secondary t o surgery, chemotherapy or chromosomal dis­orders such as Turner ' s syndrome W h i l e the adverse sequelae o f premature w i t h ­drawal of oestrogen such as premature osteoporosis and mcreased nsk o f I H D should be prevented w i t h appropnate H R T , the personal tragedy o f in f e r td i ty m such women has u n t d recently remamed untreatable T h e advent o f o v u m donation m the context of assisted conception has changed that depressmg out look Abdalla et al (1989) treated 29 women w i t h a mean age o f 36 years Donated oocytes were fert i l ized w i t h spermatozoa f r o m the recipient's p a r t n e r and then frozen u n t i l either i n t r a -uterme embryo transfer or zygote m t r a -faHopian transfer was undertaken Preg­nancy rates o f approximately 3 0 % were achieved, and this success rate has been mamtamed w i t h 116 pregnancies m 371 couples- (AbdaH* a n d Studd, mrpubhshed data) I t must be remembered that these are n o t agemg women t r y m g t o cheat nature, b u t rather women w h o m nature has cheated

Ereast caacer T h e hfet ime mcidence o f breast cancer m the developed w o r l d is approximately 10% and any factor that may mcrease this figure IS a larmmg Epidemiological e-vidence that early menarche, late f i rst pregnancy and late menopause are assoaated widbtan mcreased n s k of breast cancer has been mterpreted as mdicatmg that ovanan hormones, part i cu­lar ly oestrogen, mcrease the n s k of breast cancer

U n l i k e the s i tuat ion w i t h I H D and H R T ^ where the data are m agreement^ there is great dispanty among the numerous studies that have sought t o mvestigate the relationship between H R T and breast cancer mcidence (Fig 6) O f the 28 studies pubhshed smce 1972, 19 faded t o achieve statistical significance O f the significant studies, SIX showed an increased nsk of breast cancer w i t h oestrogen use and three a reduced n s k Such data are d i f f i cu l t t o interpret Meta-analysis of al l 28 studies

802 I<M1 Vnl if, 11

Recent advances m hormone replacement therapy

* A difficult qu&ticm. IS whether HRT may be given to women who have already had breast cancer There are no specthcally relevant data to give an answer, but m light of the above it seems appropriate to judge each case on its merits and to prescribe HRT where a strong indication exists '

yieLis. a. relative n s k o£ 1 QJ f o r oestrogen users, b u t the 9 5 % confidence mterval o f 0 96 -1 2 mdicates that this difference is n o t staustically significant ( D u p o n t and Page, 1991) A simdar study covermg only 16 studies found a relative risk of 1 3 (95% confidence mterval 1 2 - 1 6) b u t must be criticized because of its selective nature (Steinberg et al , 1991) Perhaps the most obvious conclusion i s t h a t there is no evidence of a major i n c r e a s e m the mcidence of breast cancer m oestrogen users

T h e above data are all concerned w i t h the mcidence of breast cancer W h i l e there are few data concemmg m o r t a h t y f r o m breast cancer m oestrogen users, those that exist are o f great mterest Even m a study i n which the mcidence o f breast cancer m oestrogen users was mcreased, H u n t et al (1987, 1990) and Bergkvist et a l (1989) reported a reduced m o r t a h t y f r o m the dis­ease There are several possible explananons for this apparent discrepancy First , breast cancer occurring in oestrogen users may have a better prognosis, this idea is sup­p o r t e d by the high p r o p o r t i o n o f stage I tumours occurrmg among oestrogen users ( H u n t et al, 1987, 1990) and the sigmficantly improved 5-year survival f ound by Bergkvist et al (1989)

T h e imphcat ion that oestrogen improves prognosis- is n o t necessardy- m conf l ict -wi th k n o w n data regardmg tamoxifen, as t a m o x i ­fen has b o t h oestrogen antagomst and agonist properties I n postmenopausal women i t may induce endometrial prohfera-t i o n and vagmal bleedmg, and m some cases hyperplas i i a n d even eaFemoma l a a d d i ­t i o n , I t has oestrogen-like favourable effects upon h p i d fractions (Love et al, 1991) and, hke oestrogen, reduces the risk of myocar­dial infarct ion (McDonald and Stewart, 1991)

Asecondexplanaiion. is .thatthemcreased level of breast screening by means of palpa­t i o n and mammography t h a t oestrogen users are subjected t o may result m the overdiagnosis of histologically ambiguous lesions (Studd, 1989), or the detection of tumours of very low grade malignancy which otherwise wou ld never have become chnicaUy apparent or w o u l d no t have done so for many years I n support o f the lat ter concept is the presence of chnically imsus-pected m-s i tu and invasive breast carcmoma at medicolegal autopsy (Nielsen et al, 1?87) and the f m d m g by K i e m i et al (1992) t h a t , e v e n after adjustment f o r s i z e , tumours detected by screenmg were of lower h is to ­logical and cytological mahgnancy

T h e r e is. a. suggestion from, epidemio^ logical and cell culture studies that proges­terone may mcrease the nsk of breast cancer These data are confused, b u t the chmcal imphcanons are senous However , the conclusions remam unconvmcmg and further w o r k is required t o clarify the issue

I t i s possible t h a t , i n spite of the selection and survival bias of these studies, H R T may be associated w i t h a shght mcrease m i n c i ­dence of breast cancer, b u t t h a t this is balanced by a correspondmg reduction m m o r t a h t y Whi l e the explanation of th is paradox remams unknown , the result is t h a t the effect o f H R T upon breast cancer is probably neutral

A di f f i cult question is whether H R T may be given t o women who have already had breast cancer There are no specifically relevant data t o give an answer, b u t m hght o f the above i t seems appropnate t o judge each case on its merits and t o prescribe H R T where a strong mdication exists Use of oestrogen replacement does n o t preclude the concomitant use of tamoxi fen

H R T regimens Progestogens I t has been established practice t o give cyclical progestogen w i t h postmenopausal H R T smce i t was demonstrated that a course o f tO—15 days each m o n t h co tdd prevent the development o f endometnal hyperplasia (Sturdee et a l , 1978) and atypical hyperplasia (Paterson et al, 1980) T h i s pohcy has proved effective m removmg the excess nsk of endometnal carcmoma asso­c iated W i t h - unopposed oestrogen- therapy (Persson et al , 1989) Unfor tunate ly i t has produced unwanted side effects, i e cychcal bleeding and progestogemc psychological and physical symptoms

W i t h d r a w a l bleedmg is currently the greatest p m b k m w i t h esrabiished cegunens. of H R T T h e bleeding may be heavy, p r o ­longed and/or pa infu l , and is probably the major reason f o r lack of comphance There are several approaches t o this problem A lower dose o f progestogen may be given continuously every day w i t h the aim o f rendering the endometnum atrophic — so-called contmuous combmed therapy I n the l ong term this type o f regimen is very effective, -with 95 % o f women amenorrhoeic at 12 months I n the short t e r m , however, approximately one t h i r d o f women w d l discontmue therapy w i t h i n the f i rst 6 months , mostly because of unacceptable irregular bleedmg (Magos et a l , 1985) A suitable startmg regimen is conjugated

I

Fig 7 Short-term vanability m oestradiol levels v»fith oral, patch and implant therapy

eqmne oestrogen 0 625 m g daily together w i t h n o r e t h i s t e r o n e l 05 m g Onceamenor-rhoea has been achieved, the dose o f oestrogen may be mcreased t o 1 25 m g and the dose o f norethisterone reduced T h e progestogen is convemently prescribed as the progestogen-only contracepuve N o n -day, containmg 0 35 m g norethisterone per tablet A s l ong as the dose o f progestogen is l o w and the t o t a l m o n t h l y dose is no greater than w i t h cychcal therapy, there appear t o be no adverse metabohc sequelae (Chnstiansen and Rus, 1990) A n y bleedmg after prolonged amenorrhoea requires endo-m e t r ^ biopsy because o f the possibihty o f adverse endometnal pathology (Leather et al ,1991)

A n o t h e r o p t i o n is the new synthetic denvauve of norethynodrel , nbolone , wh i ch possesses weak oestrogemc, andro­genic a n d pTog^Stogefirc properties I t has been f o u n d t o prevent postmenopausal bone loss and t o reheve menopausal symp­toms (Crona et al , 1988) T h e claim is t h a t i t I S n o t u tero t roph ic , therefore progesterone I S imnecessary and there w i l l be no bleedmg Al though- t r u e f o r mos t woms^, approx i ­mately 15% do expenence irregular bleed­i n g A possible argument against i ts l ong -t e r m use I S tha t i t may n o t af ford the same pro tec t i on agamst I H D as oestrogen because i t on ly possesses weak oestrogemc effects.and.may reduce hjgh.densiEy hpopro­t e m cholesterol (Bendek-Jaszmaim, 1987, De Aloys io et a l , 1987) I t s use shoidd probably therefore be reserved f o r m e d i u m -t e r m rehef o f clunactenc symptoms where the avoidance o f bleeding is paramount '

There may y e t be a. place m. carefully selected women f o r the use of unopposed cychcal oestrogens, as approximately one

quarter o f patients o n such therapy do n o t bleed I t may be specifically smtable for the o lder postmenopausal woman w i t h - osteo­porosis who refuses t o accept any vagmal bleedmg, b u t she must be counselled care­f u l l y about the small b u t real n s k of endo­m e t n a l pathology and be prepared t o undergo an endometnal biopsy approxi ­mately every -Z years. (Leather a n d Studd, 1990)

Endometna l ablation may have a place m the management o f the woman m w h o m the on ly progestogemc problem i s bleeding I t wiU render roughly one t h i r d o f women amenorrhoeic and w i l l reduce the amount o f bleeding m almost all ( M ^ o s e i a l , 1991) I t does n o t obviate the need fo r progestogen as there is hkely t o be residual endometnum even m those women becoming amenor­rhoeic, and m a disqmetmg number o f w o m e n troublesome dysmenorrhoea may ensue (Slade et al , 1991) There is probably a m u c h greater place for hysterectomy m the postmenopausal woman tak ing H R T than IS current ly realized I n expenenced hands the procedure is safe, i t guarantees amenorr­hoea and completely avoids the need for progestogen

Unpleasant PMS-hke symptoms such as breast tenderness, nausea, headaches, i m t -abdity and water re tent ion may occur pre-menstruaUy m up t o 20% o f oestrogen users t a k m g cychcal progestogen (Studd and Magos, 1988) T h a t these symptoms are caused by progestogen and no t oestrogen was proved by Magos et al (1986b) ImnaHy a different progestogen such as norethis­terone, dydrogesterone o r medroxyproges­terone acetate should be used I f symptoms persist I t may be acceptable t o reduce the number o f days f o r which progestogen is taken, b u t even a 7-day course is assoaated w i t h a 3% mcidence o f endometnal hyperplasia, and some irregular bleedmg Despite such manoeuvres, there w i l l be someworaenm'whctopifogeStogfenic-symp­toms persist W i t h such seventy that either t reatment is discontmued or a hysterectomy IS the preferred o p t i o n W h e n usmg oestra­d i o l implants and cychcal progestogen t o treat PMS, Watson et al (1990) f o u n d t h a t , m t h e l ong t e r m , tO-%- o f women- had- a-hysterectomy because o f adverse progesto­gemc symptoms

N o n - o r a l oestrogen dehvery One o f the main benefits o f non-oral oestro­gen dekvery i s t h a t any p o t e n t i a l t h r o m h o -t i c tendency is minmuzed by the avoidance o f first-pass hepatic metabolism There are

804

••Jie. L U E I i l l i 1 I I J ; L * . I) i n IlM I ' II UiU 1(_ t £ | | j l i l l U l ! - l l l HJ l l I ' l i l . i H'l

Fig 8 Different routes of oestrogen delivery Mean hormone levels achieved (Englund and Johansson, 1977,1978, Thom et al, 1981, Studd et al, 1990, Stanczyck, 1991) F = follicular, L = luteal, M = menopausal, 0 = ovarian

other advantages m t h a t oral oestradiol administrat ion results m a ' ro l ler coaster' 24-hour prof i le of oestradiol concentration H i g h levels o f oestrogen ^ p e a r m the systemic c irculat ion w i t h m a few hours o f mgestion^foEowed hy a. steady dedme such, t h a t levels have fallen t o near baselme before the next dose is due I n addiUon, there is a complete reversal o f the physiological 2 1 oestradiol oestrone rat io so that the latter predommates I n comparison, b o t h trans­dermal patches and subcutaneous oestradiol implants mamtam the 2 1 rat io o f oestradiol oestrone and produce far more stable hormone profiles (Smith and Studd , 1992) (Fzg 7)

T h e f trst transdermal oestradiol dehvery system, Estraderm T T S incorporates oes­t rad io l m an alcohol reservoir contamed w i t h m an adhesive patch (Padwick et a l , 1985, Whitehead et al , 1985) W h i l e gener­ally very w e l l tolerated, there are some cluneal problems Y o u n g women, par t i cu ­lar ly , may f m d the contmual wearmg o f an adhesive patch an embarrassment, a n d there may also be adhesion problems, espeaaUy m h o t weather PeAaps the most c ommon complamt is skm reactions M m o r topica l reactions were f ound m 2 0 % o f study weeks by Place et al (1985), whde Bel lantom et al (Wt) found some tmtation at the patch site m 20 o f 28 women These data almost certainly overestunate the size o f the p r o b ­

lem, b u t nonetheless skm i r r i t a t i o n is a significant problem fo r a m m o n t y of women and m a few may necessitate withdrawal of treatment M o s t such skm reactions are t h o u g h t t o be an i rr i tat ive response t o the alcohol m the drug, reservoir N e w trans­dermal patches wh i ch ut ihze different tech­nology m order t o mimmize such skm problems are now undergomg clmical tnals , and should be available m the near future

The admmistrat ion o f cychcal norethis­terone transdermally m a combmed oestrogen/progestogen patch has recently been reported (Whitehead et a l , 1990) Whether this w i l l have any advantage over oral progestogen remams t o be seen

Subcutaneous ©estradiol implants, while havmg the convenience o f 6-monthly admmistrat ion and the improved com­phance resulting therefrom, have the added advantage o f bemg able t o elevate oestradiol levels higher i n t o the normal premenopausal range than any other currently available route o f oestrogen admimstrat ion, such that the mean oestradioTTevelof767 p m o F htre achieved w i t h long-term use is roughly midway between physiological midluteal phase and preovulatory levels (Gamet t et al, 1990) (Fig 8) Such oestradiol levels are an advantage m women w i t h l o w bone density because they achieve a greater increase m bone density over 12 months than do the levels achieved w i t h either oral or patch

806

Recent advances m hormone replacement therapy

therapy T h e po tent ia l disadvantage is. t h a t some women may develop supraphysiolo-gical oestradiol levels G a m e t t et al (1990) f o u n d this so-called tachyphylaxis m only 3% o f long- term implant users M o s t o f these women had been receivmg repeat irnplants at b o t h a higher dose and a shorter mterval than recommended, and had a h igh mcidence of psychiatnc disorders I t seems that such women require frequent h igh doses of oestradiol t o keep symptoms at bay, this may wel l be a manifestation of a dose-response effect between oestrogen and m o o d I n practice, a l though startmg doses may be 75 or 100 m g , eventual maintenance doses should m general be 25 or 50 m g every 5-6 months I f used correctly m this manner b y a chnician alert t o the potent ia l problem, tachyphylaxis should n o t occur "Where supraphysiotogical levels already exist, the correct management i s t o contmue therapy b u t at reduced dosage u n t i l levels faU t o w i t h i n the physiological range Complete wi thdrawal o f oestrogen wou ld be b o t h unnecessanly harsh and potential ly

erous

KEY POINTS

One^trievery^livaof ̂ pioputetion inEngfa^ ctimacterte or postmenopausal womm

• Oestrogen therapy fnci^ses bone density m a dose-dependent manner

• O^trogen therapy reduces the risk of Ischaemic heart d&ease iiy over 40%

• ConfraindicMtons to the oraf eontraceptiye piil such as heart disease, hypertension and thrombosis do not apply to hormone replacemeni therapy (HRT); Indeed, these are indications for HRT

• The overaii effect ot HRT upon breast cancer Is probatiiy neutral

• Oestrogen can be effective treatment for ctimacteric depression and cyclical depression of premenstniat syndronw when approaching ttie menopause.

• Ovum donation is an effective bvatment for fiie InfertUIty of premature menopause

• Non-oral oestrogen administraflon residts in more physiofogfcaf ieveis and is potentially safer

• AX prmntihe uw (ff tajnttiraou* ©>mW»<rregla»iw oimt or liysterectomy are the best means of avoiding withdrawai t)ieeding.

Conclusion. T h e past decade has produced data demon-stratmg that H R T has even greater benefits t h a n previously thought by those who f i rs t recognized the chmactenc syndrome as just h o t flushes, sweats and vagmal dryness I t i s n o w apparent that the role o f H R T m osteoporosis is no t merely preventive b u t also remedial Evidence is emergmg t h a t oestrogen may have a major place m the treatment of depression, and i ts potent ia l t o reduce the mcidence of I H D by the order o f 4 0 % IS probably o f greater importance t o the prevention o f heart disease m women t h a n any other discovery

Undoubted ly the resumption o f mens­t r u a t i o n and the cychcal symptoms of m o n t h l y progestogen are strong factors, reducmg uptake and compliance, b u t use o f contmuous combmed regimens can avo id this I n view o f the potent ia l bene&ts o f H R T , hysterectomy should be more readdy offered Overall the data do n o t mdicate t h a t the nsks f r o m breast cancer are mcreased by H R T

I t IS tune that gynaecology ceased t o be the on ly branch o f the medical profession aware of the importance o f H R T I t is tune for cardiologists, psychiatrists, neurologists and general physicians t o accept that oestro­gen replacement has an impor tant role m preVeWtJfve 'medtcme w i th in - theSf own- prac­tice Every menopausal woman should be offered the potenttal beneffts o f H R T , b u t particularly those w i t h heart disease, depression and osteoporosis These are the very patients who are often demed H R T fo r illogical, reasons

Abdalla H I , Baber R J , Kirkland A , Leonard T . Studd JW W (1989) Pregnancy m -women with premature ovanan failure usmg tubal and mtrauterme transfer of cryopreserved zygotes Br ] Ohtet Gynaecol 96 1071-5

Albngfat F , Smith P H , Richardson A M (1941) Post menopausal osteoRQCQSK,. its. 4:linieal features JAMA 116 2465-74

Belkntom M F , Harman SM, Cullms V E , Engelhardt SM, Blackman M R (1991) Transdermal estradiol -snth oral progestin biological and clinical effects m younger and older postmenopausal women J Geron­tol 4(, 216-22

Bendek-Jaszmann L J (1987) Long-term placebo-controlled efficacy and safety study of Org O D 1 in clunactenc women Maturitas (Sugpl 1) 25-33

Bergkvist L , Adami H O , Persson 1, Bergstrom R, ;S:usemo U B (1989) Prognosis after breast cara-noma diagnosis m women exposed to estrogen and estrogen-progestogen replacement therapy Am J EpidemtolliO 221-8

Bourne T , Hdlard, Whitehead M, Crook D, Campbell S (1990) Oestrogens, artenal status and postmeno­pausal women Lancet 355 1470-1

Briicat M, Moniz C I , Studd J W W et al (1985) Long-term effects or the menopause and sex hor mones on skm thickness Br J Ohstet Gynaecol 92 256-9

Bush T L , Barrett Connor E , Cowan L D et al (1987) Cardiovascular mortahty and non-contraceptive use of estrogen in women results form the Lipid

807

Research Chmcs Program Foflow-up Study CiTCulatwn 75 1102-9 Campbell S (1982) Potency and hepatocellular effects of oestrogens

afier oral, percutaneous and subcutaneous admmistranon In Van Keep,PA>.UtianWH»VenneulenA,eds TTieCoiitroversialChmac­tenc M T P Press Lancaster 103

Central Statistical Of&ce (1991) Aitmial Abstract of Statistics 1991 No 127 H M S O , London

Christiansen C , Christiansen MS (1981) Bone mass m post mencpausal women after withdrawal of oestrogen/gestagen therapy Lancet i 459-61

Chnstiansen C, Rus B (1990) Frre years with continuous combined oestrogen/progestogen therapy Effects on calcium metabohsm, hpoproteinx and bleeding patterns Br J Ohitet Gynaecol 97 1087-92

Crona N , Samsioe G , Lmdferg U B , Siffverstolpe G (1988) Treatment of chmactenc complaints with Org O D 14 a comparaove study with oestradiol valerate and placebo Mtstmitas3 303-8

D c Aloysio D , Fabiam A G , Maulom M, Botughom F (1987) Use of Org O D 14 foi the treatment of clunactenc symptoms Maturitas (SuppI 1) 49-65

Ditkoff E C , Crary W G , Cnsto M, Lobo R A (1991) Estrogen improves psychological function m asymptomatic postmenopausal women OfatetGwco/78 991-5

Dupont W D . Page DIL (1991) Menopausal estrogen replacement therapy and breast cancer Arch Intern Med 151 67-72

Englund D E , Johansson E D B (1977) Pharmacokmetics and pharmaco ^mamic studies on oestradiol valerate admmistered orally to postmenopausal women Acta Ohstet Gynecol Scand 65 (Suppl) 27-31

Englund D E , Johansson E D B (1978) The plasma levels of oestrone, oestradiol and gonadotrophins m postmenopausal women after oral and vaginal administration of conjugated equme ostrogens (Pre •marrn)' Bt-J- Ohstet Gynaecol 85- W - 6 4

Jangar K, Gust M, Whitehead Gangar K, Cust M, whitehead MI (1989) Symptoms of oestrogen defaaency assoaated with supraphysioiogicaf plasma oestradiol concentrauons m women with oestradiol implants Br Med J 299 601-2

Gangar K F , Vyas S, Whitehead M, Crook D , Meire H , Campbefl S (1991) Pulsatihty mdex m mtemal carotid artery m relation to trans­dermal oestradiol and time smce menopause Lancet 338 839-42

Gamett T , Studd JWW, Henderson A , Watson N , Sawas M, Leather A{1.9.9Q)^Hormoneimplants-asdtachyphylaxis. Br}OhstetGynaecoL 97 917-21

Henderson A F Gregoire AJP, Kumar R, Studd TWW (1991) The treatment of severe postnatal depression with oestradiol skm patches Lancet u 816

Hough J L , Zdversmit D B (1986) Effect of 17p-oestradioI on aortic cholesterol content and metabohsm m cholesterol fed rabbits Arteriosclerosis 6 57-63

Hunt K, Vessey M. McPherson K , Coleman M (1987) Long-term sucveillance of mortahCT and, canser madence in. women recstving. hormone replacement therapy Br J Ohstet Gynaecol 94 620-35

Hunt K, Vessey M, McPherson K (1990) Mortality m a cohort of long-term users of hormone replacement therapy an updated analysts Br J Ohstet Gynaecol 97 1080-6

Ked DP, Felsoa D T , Anderson J J , Wdson D W F , Moskowitz MA (1987) Hip fractures and use of estrogens m postmenopausal women NEnd!Med m 1169-74

Klaiber E L , Broverman D M , Vogel W , Kobayashi Y (1979) Estrogen replacement therapy for severe persistent depression m women Arch Gen FsycStatry if, 550-4

Klenu PJ, Toensuu H , Toikkanen S et al (1992) Aggressiveness of breast cancers round with and without screenmg Br Med J i04 467-9

Leather A T , Studd J W W (1990) Can the withdrawal bleed followmg oestrogen therapy be avoided Br J Ohstet Gynaecol 97 1071-9

Leather A T , Sawas M, Studd J W (1991) Endometnal histology and bleeding patterns after 8 years of contmuous combmed estrogen and progestogen therapy m postmenopausal women Ohstet Gynecol 78 1008-10

Lmdsay R, Cosman F (f990) Estrogen m prevention and treatment of osteoporosis Ann NY Acad Sa USA 592 326-33

Lmdsay R, Hart D M , Aitken JM, McDonald E B , Anderson J B , Clare A C (1976) Long term prevention of postmenopausal osteoporosis by oestrogen Lancet i 1038-40

Lmdsay R, Hart D M , Maclean A et al (1978) Pathogenesis and prevention of postmenopausal osteoporosis In Cooke I D , ed The Role ofOestrogen/Progestogen Management of the Menopause M T P , Lancaster 9-25

Love-RR, Wiebfr D A , f̂ewcomb•P-A-etal•(•199•l•> Effects isf-tim&rfen: on cardiovascular nsk factors m postmenopausal women Arm Intern Med 1X5 860-4

McDonald C C , Stewart H J (1991) Fatal myocardial infarction m the Scotnsh adjuvant tamoxifen tnal Br Med J 303 435-7

Magos A L , Brmcat M, Studd JWW, Wardle P, Schlesmger P, O'Dowd T (1985) Amenorrhea and endometnal atrophy with contmuous oral estrogen and progestogen therapy m postmenopausal women Ohstet Gynecol b5 496-9

Mag.QS. A L , , Bcuica.t Studd. J W W (1986a). Treatment o£ -tbt premenstrual syndiome by subcutaneous oestradiol implants and cychcal norethisterone Br Med J i 1629-33

Magos A L , Bresvster E , Smgh R, O Dowd T , Brmcat M, Studd JWW

(1986b) The effects of norethisterone m postmenopausal women on oestrogen replacement therapy a mtwel for the premenstrual syndrome Br J Ohstet Gymtead 9i 1290-6

Magos A L , Bajimann R , Locfcwood G M , Tnmbull A C (1991) E x penence with the fnrst 250 endometnal resecuons for menoniagia LmcEt 337 1074-8

Montgomery J C , Appld>y L,&ii icttMetal (1987) Effects of oestrogen and testosterone imptmts on psydiological disorders of the diinkc-tenc Lancet! 297-9

Nielsen M, Thomson J L , Pnmdahl S, Dyreboi^ U , Andersson JA (1987) Breast cancer and atypia among young and middle a^sd women a study of 110 medico legal autopsies Br J Cancer 56 814-19

Office of Populanon Censuses and Surveys (1991) Mortality Stattsucs Cause 1990senesDH2No 17 H M S O , London

Padwick M L , Endacott J , Whitehead MI (1985) Efficacy, acceptabiLcy and metabohc effects of transdermal estradiol m the man^ement of postmenopausal women Am J Ohstet Gynecol 152 1085-91

Pagamm-Hill A , Ross R K , Henderson B E (1988) Postmenopausal oestrogen treatment and stroke a prospective study Br Med J 297 519-22

Paterson M E L , Wade-Evans T , Sturdee DW, Thom M H , Studd JWW (1980) Endometnal disease aftM" tneatment ymb oestrogens- and progestogens m the clunactenc Br Med J i 822-4

Persson O , Adami H O , Bergkvist L et al (1989) Risk of endometnal cancer after treatment with oestrogen alone or m combination -widi progestogens, results of a prospective survey Br Med J 298 147-51

Place V A , Powers M, Darley P E , Schenkel L , Good W R (1985) A double-bhnd comparative study of Estraderm and Premarm m the amehoranon of postmenopausal symptoms Am J Ohstet Gynecol 1-52 1592-9-

Slade R J , Hasib Ahmed A l . Gdhner M D G (1991) Problems with endometnal resecnon Lancet 337 1473

Smith R N J , Studd JWW (1992) Honnone replacement therapy a review jDrugDevA 235-44

Stampfer MT, Oilditz G A (1991) Estrogen replacement therapy and coronary heart disease a quanutative assessment of the epidemi­ological evidence Prev Med 20 47-63

Stanczyk F Z (1991) Steroid hormones hx Mishell D R , Davajan V , Lobo,RA^eds. Ztrfaiiluy^ GQntrM:ept^OJ^.and£iej^cc^cU^ct^m.Endor crmology 3rd edn Blackwell Scientihc, Cambndge, MA 53-74

Sternberg K K , ThackerSB, SnuthSJetal('1991) Ameta -anaWof the effect of estrogen therapy on the nsk of breast cancer J AM A 265 1985-90

Stevenson J C , Whitehead MI (1982) Postmenopausal osteoporosis Br Med J 2S5 585-8

Stevenson J C , Cust MP, Gangar K F , Hilkrd T C , Lees B, Whitehead MI (1990) Effects of transdermal versus oral hormone replacement therapy on. bone density m spme and proximal femur in postmeno­pausal women Lancet 336 265-9

Storm T , Thamsborg G , Stemiche T , Genant H K , Sorensen O H (1990) Effect of mtermittent cychcal etidronate therapy on bone mass and fracture rate m women with postmenopausal osteoporosis

,N Engl J Med 322 1265-72 Studd J (1989) Hormone replacement therapy and breast cancer Lancet

u 1164 Studd JWW, Magos A L (1988) Oestrogen therapy and endometnal

pathology In Studd J , Whitehead M,eds The Menopause Blackwell foennffc, Oxford 197-212

StuddJ,ChakravartiS,OramD(1977)Thechmactenc I n Greenblatt R B , StuddJ, eds Cltmcsm Obstetrics and Gynaecology vol 4(1) The Menopause W B Saunders, Philadelphia 3

Studd JWW. Dubiel M, Kakkar W . Thom M, Wbite P (1978) The effect of hormone replacement therapy on glucose tolerance, clottmg factors, fibnnolysis andplatelet function m post menopausal women In Cooke I D , ed The Role of Oestrogen Progestogen m the Management of the Menopause M T P Press, Lancaster 41-59

Studd JWW, S a m s M , Watson N-, GarnsttT, Foeetman I, CooperD-(1990) The relationship between plasma estradiol and the mcrease m bone density m postmenopausal women after treatment with sub­cutaneous hormone implants Am J Ohstet Gynecol 163 1474-9

Sturdee DW, Wade-Evans T , Paterson M E L , Thom M, Studd JWW (1978) Relations between bleeding pattern, endometnal histology and oestrogen treatment m menopausal women Br Med J i 1575

Thom M H , Colhns WP, Studd JWW (1981) Hormonal profdes m postmenopausal women after therapy with subcutaneous unplants Br-Ohstet Gynaecol 426?-3-3-

Watson N R , Studd JWW, Sawas M, Baber R J (1990) The long-term effects of estradiol implant therapy for the treatment of premenstrual syndrome Gynecol Endocrmol 4 99-107

Watts MB, Hams S T , Genant H K (1990) Intermittent cychcal etidronate treatment of post menopausal osteoporosis N Engl J Med 323 73-9

Whitehead M L Padwick ML, Endacott J , Pryse-Davies J (1985 Endometnal responses to transdermal estradiol m postmenopausa w<Bafi!i .Axn. J. Ohstet Qymcql-152. 1079-4i

Whitehead MI, Fraser D, Schenkel L , Crook D , Stevenson J C (1990) Transdemiil admmistraaon of oestrogen/progestogen hormone replacement therapy Lancet 335 310-12

808 Bmish Journal of Hospital Medicine 1993 -Vol 49 N o l l