obesity and female reproductive function

8/7/2019 Obesity and female reproductive function

1/18

Obesity and fem ale reproductive functionR Pettigrew and D Hamilton-FairleyDepartm ent of Obstetrics and G ynaecology, St Thom as' Ho spital, Londo n, UK

Obe sity has consistently been demonstrated to have a d etrimental effect upon thefem ale reproductive system.This review explores the common association of obesitywith polycystic ovary syndrome (PCOS), the effect of obesity on the clinical andendocrinological parameters, and the role of insulin resistance in the expression ofthis disorder. An improvement in menstrual function, a decrea se in the clinicalandrogenic profile, and significant increase in spontaneous pregna ncy rates havebeen rep orted following weight loss.Obesity is associated with poor pregnancyoutcome a nd miscarriage in both women w ith PCOS, and in those with normalovarian m orphology.The optimal weight ga in during pregn ancy remainscontrovers ial, but obesity is a risk factor for both m aternal and fetal complications,and d ietary advice should be o ffered on an individual basis according to the p re -pregnancy BMI.We ight ga in at the time of menopause is common, and d ietaryadvice is p aramou nt as obesity is an independent risk factor for thrombosis,coronary heart disease (C HD ), and breast and endom etrial cancer. Effectivenutritional counselling should be offered at all stages of the female reproductivelifecycle.

Corretpondence to:Mits D Hamilton-Fairley,

Department of Obstetricsand Gynaecology,

St Thomas' Hos pital,Lambeth Palace Road,

London SEI 7EH, UK

Obesity is an increasingly common health problem which is animportant risk factor in obstetrics and gynaecology, and has been foundto affect female reproductive function in several respects. Complicationsare more prevalent during pregnancy and childbirth, and menstrualdisturbance and infertility are frequent consequences of female obesity.One of the most common gynaecological disorders associated withobesity is polycystic ovary syndrome (PCOS). Insulin resistance isthought to play a central role in the expression of this syndrome, and therelationship between PCOS, obesity and insulin will be discussed in thisreview. Weight reduction in these women has been shown to improvemenstrual function and reduce hyperinsulinaemia, hyperandrogenaemiaand peripheral insulin resistance. Women with normal ovarianmorphology also are affected adversely by obesity throughout theirreproductive life, and this will also be discussed.There has been much interest and controversy surrounding the idealweight gain for pregnant women, and the effects of excess weight uponpregnancy outcome in terms of both maternal and fetal morbidity andmo rtality. The recomm endations concerning gestational weight gain andthe adverse outcomes associated with obesity in pregnancy are reviewed

British M.d.ca / Bulletin 1997;53 (No. 2 1:341-358 Th British Council 1997

y

g

j

g
http://bmb.oxfordjournals.org/


2/18

Obesity

in this chapter. Obesity has consistently been found to have adetrimental effect upon reproductive endocrine function, and weightloss should be encouraged pre-conceptually with nutritional advice beingoffered during pregnancy.

Polycystic o v a ry syn drom e (PCOS)Polycystic ovary (PCO) syndrome was initially described by Stein andLeventhal in 1935 as a syndrome of obesity, hirsuitism, anovulation andinfertility, associated with enlarged and polycystic ovaries1. This is acomm on heterogeneous disorder; Poison et al. reported the prevalence ofPCO morphology in 23% of a large group of asymptomatic womenvolunteers of reproductive age2. There still exists much controversysurrounding the diagnosis, pathophysiology and aetiology of thisdisorder, as a broad spectrum of symptoms is encompassed by thisdiagnosis3.PCOS may be an incidental finding on a routine ultrasound scan in awoman of normal weight with regular menstrual cycles who does notexhibit signs of hyperandrogenism or, conversely, these women maypresent with a classical appearance of hirsuitism, obesity and oligo oramenorrhoea. The principal abnorm ality in women w ith PCOS is one ofanovulation manifested by oligomenorrhoea or secondary amenorrhoea.The diagnostic criteria which are generally agreed upon are based uponthe morphological appearance of the ovaries on ultrasound scan, whichappears to be a more sensitive marker for PCOS compared with theclassical endocrinological features. Accepted criteria describe ovarianmorphology with more than 10 peripheral cysts in a single plane, sizedbetween 2-8 mm, arranged in an echodense stroma 4. This sonographicappearance of PCO may occur together, or in isolation with abiochemical imbalance, which classically involves metabolic andhormonal changes. Concentrations of luteinizing hormone (LH) areelevated in 45-75% of cases in a reported series, and a raisedtestosterone is seen in 80% of patients 5; these are the usual biochemicalindicators of the presence of this syndrome.The aetiology of polycystic ovaries has long remained elusive, and thepathophysiology of PCOS is still uncertain. There is increasing evidencethat the primary abnormality is excessive ovarian androgen production.It is a minority of women with PCOS who are obese, studies suggesting35-50% 6-21, but obesity has consistently been shown to increase theprevalence of hirsuitism and anovu lation when compared to lean w omenwith PCOS.

3 4 2 Britith Medical Bulletin 199 7^3 (No. 2)

byguestonMarch7,2011

bmb.oxfordjournals.org

Downloa

dedfrom
http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/http://bmb.oxfordjournals.org/


3/18

Obesity and female reproductive function

Over the last decade, increasing evidence has implicated the role ofinsulin in the pathophysiology of PCOS. It was first noticed in 1980 byBurghen et al. that insulin resistance and PCOS were associated, andthere was a correlation between androgen concentration and seruminsulin7. It was, therefore, hypothesized that obesity and the consequenthyperinsulinaemia might lead to the genesis of PCOS in susceptibleindividuals8.The ovary possesses receptors for bo th insulin and IGF-1 and there is across reactivity between these hormones and their receptors. Thus,hyperinsulinaemia will cause insulin receptors in the ovary to bestimulated directly, and there may be a spill-over effect involvingstimulation of IGF-1 receptors. IGF-1 has been shown to potentiate thestimulating effect of follicle stimulating hormone (FSH) on ovarianaromatase activity, on oestradiol and progesterone secretion, andexpression of LH receptors9. Fasting insulin levels correlate positivelywith serum IGF-1 and inversely with IGFBP-1 concentrations in leanwomen with PCO S. IGFBP-1 is a low molecular weight binding p roteinwhich is thought to play a central role in regulating the bioavailability ofIGF-2, and to a lesser extent IGF-1, and is known to be regulated byinsulin. Insulin will cause a decrease in the circulating IGFBP-1 and soincrease the potential for IGF-1 to stimulate the ovary. IGF-1 potentiatesthe action of LH to stimulate the thecal and interstitial components ofthe ovary10 and this synergistic effect may lead to an increase in theactivity of cytochrome P450c-17alpha hydroxylase, which is animportant intermediary in the production of androgens. This LHregulated cytochrome P450 steroidogenic enzyme catalyses the finalrate limiting step in testosterone and androstenedione biosynthesis,which is then used as a substrate by the ovarian granulosa cells toconvert androgenic precursors to oestrogen. Sharp et alu. demonstratedthat in vivo, the major effect of insulin on androgen secretion wasmediated by changes in sex hormone binding globulin (SHBG). Insulinlevels in a study by Hamilton-Fairley et al. were not significantly raisedin the PCOS group as a whole, but were elevated in anovulatorycompared with ovulatory women12 , despite having similar androgenlevels; this may indicate that it is insulin resistance in the ovary thatcontributes to the mechanism of anovulation, and that insulin is notdirectly involved in the increased androgen production in these women.Insler's group found that serum IGFBP-1 concentration was inverselycorrelated to insulin concentration and was significantly lower in obesePCOS than in non obese13 . They also showed SHBG levels to besignificantly lower in obese women, compared to non obese patients. Itmay thus be postulated that in obese women, hyperinsulinaemia is asecondary disorder, resulting in a decrease in SHBG and IGFBP-1 levels,creating an increase in free androgens, which are in turn converted to

British Medical BuU,tm 1997;53 (No. 2) 3 4 3



Downloadedfrom


4/18

Obesity

GeneticFactors Insulin resistance

SecondaryFactors

Dysregulanon of Cyt p45017 alphaIOver production of androgens

Increased follicle recruitment

PCO

Obesity

Insulin reisistance

No secondary factors

OvarianConsequences

raised circulatinginsulinIncreased responsiveness

toLH Normal responsivenesstoLHJ IArTested

foll iculo gen es is

IANOVULATORY PCONormalfolliculogenesis

OVULATORY PCO

F i g . 1 Diagramillustrating a hypothesisfor the pathogenesis of

anovulation in PCOS.

oestrone due to the excessive adipose tissue. Figure 1 illustrates ahypothesis for the underlying biochemical abnormality.Two important factors account for the prevalence of menstrualabnormalities in obese women with PCOS. Firstly, obese womenproduce more oestrogen by extraglandular conversion of androgensthan do non obese subjects and, therefore, have a greater excess ofcirculating oestrone over oestradiol concentrations than lean women; apositive correlation of the oestradiol/oestrone ratio with BMI has beenconfirmed by Kiddy et alu. Secondly, as concentrations of SHBG aremuch lower in obese women, there is an increased free biologicallyactive fraction of oestradiol in this group of women. Dysfunctional344 British Madical Bullttin 19973 (No. 2)

byg

uestonMarch7,2011


Downloa

dedfrom


5/18


uterine bleeding will often improve after weight loss and up to 80%improvement in menstrual function has been reported following weightloss15.Kiddy and colleagues observed a marked improvement in the clinicaland endocrinological profile of obese women with PCOS following along term, low calorie diet (less than 1000 kcal per day over a period of6-7 months)16 . This improvement occurred in those women who lostmore than 5% of their initial body weight, with an improvement inreproductive function (demonstrated by a 70% spontaneous conceptionrate) in previously subfertile women who lost weight. Menstrualregularity was also improved and hirsuitism was reduced by 40% inthe women in this group. There was no significant change ingonadotrophin concentrations, or total serum testosterone levels, but amarked increase in concentrations of SHBG with a reciprocal change infree testosterone levels. These changes in the weight loss group wereaccompanied by reduced fasting serum insulin levels, and a decreasedresponse to an oral loading dose of glucose. There is now evidence tosuggest that insulin inhibits hepatic secretion of SHBG both in vivou andin vitron. There were no significant changes in either clinical orendocrine indices in the group who lost less than 5% of theirpretreatment weight. The hypothesis suggested is that a reduction ininsulin concentrations either directly or indirectly affects ovarianfunction. These results emphasize the endocrine impact of obesity andbody fat distribution. Balen et al. also confirmed a positive correlationbetween body mass index (BMI) with an increased rate of hirsuitism,cycle disturbance, infertility and raised serum testosterone concentra-tions19 . Again, therapeutic weight loss was encouraged, and was foundto improve the symptoms and endocrine profile of women with PCOS.Although routine measurement of serum insulin concentration wasnot recommended, Conway et al. reported an 11% prevalence ofdiabetes in obese PCO patients, which underlines the importance ofassessing glucose tolerance in these women20 . Women who areoverweight should be encouraged to lose weight, and can then expectto experience an improvement in the symptoms of menstrual dis-turbance, infertility and hyperandrogenism. The effects of BMI,however, are part of a spectrum, and the amount of weight loss shouldnot be based on an all or nothing effect with a fixed BMI as an end p oin t,since improvement may occur after a relatively small weight loss.Hamilton-Fairley et al. studied the impact of body weight uponanovulatory women who were undergoing ovulation induction with lowdose gonadotrophin therapy, to assess the influence of those women whowere moderately overweight upon fecundity and miscarriage rates.Women with marked overweight (BMI more than 28) were excludedfrom this trial. The conception rate was found to be similar in both

British Mtdical Bulletin 1997;53 (No. 2) 3 4 5

by

guestonMarch7,2011


Downloadedfrom


6/18

Obesity

groups, but the risk of miscarriage was found to rise with increasingBMI21. Overweight women were found to be more likely to receivegonadotrophin therapy, as they were less likely to respond to clomipheneoyulation induction than their lean counterparts22. This study further foundthat overweight women undergoing this treatment required significantlyhigher doses of gonadotrophins than lean women in order to achievesatisfactory ovarian stimulation. It was apparent that overweight womenwere relatively resistant to the action of endogenous gonadotrophin, as thelevels of mid follicular FSH were appropriate for the higher dose received bythe overweight group. Although the conception rate did not appear to beaffected between the lean and overweight groups, there was a significantincrease in the rate of miscarriage, and appeared to be independent of raisedLH concentrations as serum LH levels were similar in both groups. Theprevalence of early pregnancy loss in a population of women with PCOS wassignificantly greater in the overweight group (47%) compared with 27% inthe lean group.The distribution of fat is also important. Pasquali published a reportupon the effect of fat distribution on the clinical, hormonal andmetabolic features of women with PCOS 23 . After adjustment for BMIand age, the waist to hip ratio (WHR) was found to be positivelycorrelated with an increased LH, androstenedione and oestroneconcentration s, as well as higher levels of fasting and glucose stimulatedinsulin, higher levels of triglycerides, elevated serum very low densitylipoprotein (VLDL) and apolipoprotein B, and lower levels of highdensity lipoprotein (HDL) cholesterol. These results are perhaps notsurprising, as it has been previously reported that this androiddistribution of fat is associated with atherogenesis and poor metabolicparameters. Android obesity is the result of fat deposited in theabdominal wall and visceral mesenteric locations. This fat is moresensitive to catecholamines and less sensitive to insulin, which renders itmetabolically more active, with a resulting association of hyperinsulin-aemia and cardiovascular risk factors.It would appear that obesity is both a result of, and a contributingfactor to, the endocrine imbalance in PCOS and these patients will becaught in a vicious circle of increasing hyperandrogenism. Improvingmatters using dietary restriction requires low carbohydrate diets, andthus provides an endocrine basis for a dietary approach to treatment.The role of insulin in the pathogenesis of PCOS may suggest a futurepharmacological treatment in PCOS. As there is no change in ovarianmorphology following weight loss, this would suggest, however, thatobesity contributes to the clinical and biochemical abnormalities inPCOS, but is not the underlying cause of the disorder. The persistentstate of hyperinsulinaemia associated with obesity causes hyperandro-genism, which may also be associated with menstrual irregularities and

3 4 6 British Mtdkal Bulletin 1997^3 (No. 2)



Downloa

dedfrom


7/18


anovulatory infertility. It is obvious that, in the absence of ovulation,women will fail to become pregnant. The importance of weight loss forthose obese women with PCOS must be emphasised, as this may restoreovulation and improve fertility.

Fertility and obesityObesity is common among multiparous women but this would suggestthat weight gain is the result of many pregnancies, rather than it being afavourable factor for conception. The impact of obesity upon theendocrinological profile has already been discussed, particularly inreference to the relationship of obesity and polycystic ovaries.

There are no data available for fertility rates in obese women who donot have PCOS, but there is evidence for an increased risk ofmiscarriage. Hamilton-Fairley et al.lx studied the risk of miscarriageand associated BMI and miscarriage rate in 13,128 primiparous womenin the North West Thames region (1992). These data included womenwith normal ovarian morphology and women with PCOS. Of 9239women with a BMI of 19-24.9, 11% miscarried. In the moderatelyoverweight group (BMI of 25-27.9), 14% went on to miscarry, whichwhen adjusted for maternal age, represents an odds ratio of 1.26. In theoverweight and obese group (BMI greater than 28), 15% of thesewomen miscarried, (the adjusted odds ratio for this group being 1.37).These data confirm that increasing obesity significantly reduces thechances of a successful pregnancy.The distribution of fat seems to be important in the generalpopulation. To assess the effects of this variable upon female fecundity,a prospective cohort study was performed using women attending afertility clinic requiring donor insemination. The WHR, BMI, age andconception rates were recorded. It was found that the percentage ofwomen who became pregnant fell from 63% to 32% with increasingW HR . Cumulative pregnancy rates for women w ith WH Rs less than 0.8(gynaecoid obesity) were significantly higher than for w omen with ratiosmore than 0.8 (android or central obesity). Both underweight andoverweight women had a lower chance of becoming pregnant thanwomen with normal weight, although this effect was greater for obesewomen than lean w omen24 . One hypothesis suggested that an increase inthe androgenic microenvironment of the follicle could decrease oocytequality. Weight loss has been shown to restore cycle regularity, and sexhormone concentrations, which does indicate that it is the obesity whichexacerbates reproductive problems.

British Medical Bulletin 1997;53 (No. 2) 3 4 7



Downloadedfrom


8/18

Obesity

To assess the effect of morbid obesity on fertility, Bilenka et al. studieda group of massively obese women who were undergoing bariatricsurgery using vertical banded gastroplasty25. The report studied 9women and their pregnancies before and after treatment and theirsubsequent weight loss. Of 18 pregnancies prior to the operation, 6ended in miscarriage. Of the 13 pregnancies following weight loss, onlyone miscarried. Of this group, the conceptions were all spontaneous,whereas previously, 5 out of 6 attempting to conceive required ovulationinduction. Although the numbers were small, this paper demonstratesthe importance of excessive weight upon conception rates and the rate ofmiscarriage.The relationship between obesity and fertility, and the impact of bodyfat distribution on miscarriage rates appears to warrant pre-conceptualadvice concerning w eight loss. Although weight loss is not recommendedwhile pregnant, the presence of obesity, and the related hyperinsulinae-mia is thought to be responsible for early pregnancy loss.

Weight gain and pregnancyObesity is a common nutritional problem, and many recent studies haveassessed the pregnancy course and outcome in association with maternalobesity. However, there remains a lack of consensus on appropriateweight gain in obese pregnant women. This was evident in the 1990Institute of Medicine report

26, which recommended a minimum weightgain of at least 7 kg (15 lb) for obese women, but due to lack of cleardata, was unable to specify an upper limit of weight gain.Weight gain in pregnancy reflects both fat deposition, a physiologicaladap tation, and water, due to expansion of the plasma volume and totalbody water, as well as increases in the size of the uterus and breasts. Ifweight gain is excessive, the maternal fat stores may not be expendedduring the pregnancy, and will thus be retained in the postpartumperiod.As there is no accepted defining value for obesity in pregnancy, theprevalence is unreported. However, obesity has been reported asincreasing in women of reproductive age. A true assessment of themagnitude of this problem is important, and its subsequent effect uponweight gain in the fetus, and maternal and neonatal morbidity andmortality. The BMI is the generally accepted quantifying term forobesity and an important initial assessment of any pregnant womanshould involve a calculation of the BMI. Routine measurement ofmaternal weight gain has been an integral part of prenatal care for manyyears, but the amount of gain considered desirable has varied over the

3 4 8 British Medical Bulletin 199 7^3 |No. 2)

byg

uestonMarch7,2011


Downloa

dedfrom


9/18


years. Prospective trials evaluating BMI and correlating this withpregnancy outcome have yet to be performed. The relationship betweenmaternal weight gain and fetal growth is not clear and the range ofweight gain which is acceptable to reduce the risk of delivering eithermacrosomic or small for gestational age babies is still controversial.High and low weight gain in women of ideal weight have beenassociated with increased frequency of perinatal death27 and pretermdelivery has been associated with low weight gains, particularly in thethird trimester. Publications in 1989 concerning weight gain inpregnancy noted that low birth weight was 2.3 times more likely andfetal death was 1.5 times more likely in women who gained less tha n 20lb (9 kg) during pregnancy, when compared with those who gained morethan 20 lb. It was assumed that there was a direct correlation betweenthe amount of weight gained in pregnancy and the pregnancy outcome interms of low birth weight babies and infant mortality. Following thisreport, recommendations of increased weight gain during pregnancywere instituted.The distribution and range of maternal gain is more variable in obesewomen, and some studies have reported that women who are obesewhen pregnancy begins, gain less weight on average than those who arenot obese. The relationship concerning obese pregnant women, weightgain and birthweight is uncertain, although it would appear that obesewomen are at higher risk of both maternal and fetal complications.Current recommendations from the Institute of Medicine (IOM)suggest a minimum gain of 15 lb (6.8 kg) in obese women, commencingjust prior to pregnancy, until delivery at term, with no upper limitspecified. However, the pattern of weight gain remains unquantified.Based on the limited literature available, the IOM committee recom-mended a weekly maternal weight gain of approximately 1 lb (0.45 kg)per week during the second and third trimesters for women of normalpre-pregnancy BMI. Just over 1 lb per week w eight gain is recommendedfor underweight women, and just under for obese gravidas. Because pre-pregnancy weight and body size has been shown to modify therelationship between gestational weight gain and birth weight, separaterecommendations are made for under weight, normal weight, over-weight and obese women28 . Many studies have suggested that the effectof a given weight gain on fetal growth is greatest in thin women and leastin overweight women29-30.Low maternal weight gain is undesirable due to its association withvery low birth weight babies; conversely, high maternal weight gain maybe com plicated by the difficulties in delivering large infants secondary toclinically overt or subclinical gestational diabetes. Complications arisingin pregnancy are often exaggerated in obese mothers. How ever, it is alsotrue that obese mothers are generally older than their lean counterparts,

British Medico/ Bulletin 1997S3 (No. 2) 349



Downloadedfrom


10/18

Obejity

and, therefore, maternal age may be a confounding factor for thedevelopment of medical disorders. Most studies have demonstratedthat maternal obesity is an independent factor for the development ofdiabetes and hypertension. Diabetes is associated with a hyperinsuli-naemic state, hyperglycaemia and insulin resistance, which are allconditions which are common in both advancing gestation, andobesity. Gestational diabetes is seen more frequently among obesepregnant women31 and can also be attributed to insulin resistance.Although gestational diabetes usually resolves shortly followingdelivery, there remains an increased risk of the mother developingdiabetes in later life.Chronic hypertension is another medical condition which may arise inobese gravidas. Gross et al. reported 7% in obese women (compared to1.5% of lean women) having chronic hypertension32. A reported 17 % ofobese gravidas compared with lean women were noted to develop pre-eclampsia, although there was no significant increase in eclampsia.Several studies have assessed the intrapartum risk involved for obesewomen and have shown increased risks associated with high pre-pregnancy BMI and large weight gain during pregnancy. In 1992,Johnson et al.33 performed a longitudinal retrospective study observing7407 term pregnancies delivered over a 2 year period. 3191 cases weresuitable for analysis after elimination for fetal abnormalities, selectedmedical or surgical complications, and those with no antenatal care orincomplete records. Pregnancy outcome was evaluated in regard to BMIcategories recommended by the National Academy of Sciences, that islow (29).Increased maternal pre-pregnancy BMI and increased maternal gesta-tional weight gain were both associated with an increased risk ofmacrosomia (P


11/18


and these data indicate that as the pre-pregnancy weight is increased, therecommended gestational weight gain should be decreased.Serious medical obstetric complications more likely to arise in obesewomen include hypertension, cardiomyopathy, acute fatty liver ofpregnancy (AFLP) and cholestasis in pregnancy, and should beconsidered during antenatal care. Women who are lean should betreated as a different population, requiring very different nutritionaladvice to those women who are obese.Prenatal diagnosis is made more difficult in those women who aregrossly obese. Wolfe et al. studied the effect of increasing maternal sizeon ultrasonographic visualisation of fetal anatomy, and found thatvisualisation of fetal organs was significantly decreased in the obesegroup. Thus important fetal abnormalities may fail to be detectedantenatally.The level of obesity at which visualisation was affectedsubstantially34-35, was marked, namely a BMI of 36.2.The obese patient is also at risk for morbidity arising from operativedelivery. There is an increased risk of wound infection, urinary tractinfection, blood loss, and thromboembolic complications. Shouldoperative delivery be necessary, regional anaesthetic blocks are theanaesthetic of choice due to compromised respiratory function in theobese gravid; this may, however, be technically difficult in an obesepatient. If an emergency delivery is necessary, then rapid sequenceinduction is a technique which is common in obstetric anaesthesia andhelps prevent the aspiration of gastric contents.Postpartum weight retention is another important issue which is

related to weight gain in pregnancy. O'Scholl found that women whohad a normal pre-pregnancy BMI, and who gained weight at anexcessive and essentially unlimited rate, did not greatly enhance fetalgrowth, but were more likely to develop postpartum obesity36 .Advancing age and lifestyle changes may be confounding factors, butbeginning a subsequent pregnancy with extra weight leads to an increaserisk of gestational diabetes, hypertension and the complications inlabour which have already been discussed. There is little informationavailable concerning postpartum loss of weight, but the strongest factorcontributing to postpartum weight changes is prenatal weight gain.Clinical recommendations from the IOM report suggest that womenshould generally lose 0.45-0.9 kg per month during the first 6 monthspostpartum, although there is much variability. Women who are breastfeeding need to be encouraged to consume at least 1800 kcal per day,and any intake below this is not recomm ended. Obese wom en, however,can aim to lose up to 2 kg per month without jeopardising their milkquality, although losing weight within the first few weeks postpartum isgenerally not recommended, as this is the point at which milk supply isbeing established36 .Brihih Mtdical Bulletin 199 723 (No. 2) 35 1



Downloadedfrom


12/18

Obesity

Obesity is a severe health hazard, and adequate nutritional adviceshould be given to all pregnant women. Obesity is associated with anincreased risk of miscarriage, but those fortunate enough to have anongoing pregnancy need particular attention, and regular antenatalscreening for glucose intolerance, hypertension, macrosomia andexcessive weight gain should be routine. The mother can then becounselled appropriately throughout her pregnancy. Nutritional adviceshould be altered according to the individual depending upon the pre-pregnancy weight, and weight gain recommendations in obese womenshould vary from that advice given to lean pregnant women. Weight losspostpartum must also be encouraged, particularly in those womenretaining weight, and they need to be assisted in attaining their idealweight37.

Horm one replacement the rap y (HRT) and obesityThe menopause is defined as one year following the last menstrualperiod; the median age of menopause in most developed societies isaround 50 years. The essential feature of the menopause is the failure ofovarian follicular development. A finite number of oocytes are laid downin fetal life (about 7 million) with approximately 450,000 oocytes beingavailable by puberty. It is thought that once this number has reachedcritically low levels, then irreversible ovarian failure occurs. Inpremenopausal women, the follicle is responsible for more than 90%of oestradiol produced, and when folliculogenesis ceases serum levels ofoestradiol decline. Associated with this is an increase in FSH and LH,thus giving the characteristic biochemical picture of hypergonadotrophichypogonadism. The postmenopausal ovary does secrete some oestradiol,but most is produced by peripheral conversion from androgens.Androstenedione is the most important androgen used for the produc-tion of oestrogens. It is a C19 steroid which exhibits little androgenicactivity, and is converted in the liver and adipose tissue to oestrone. Theproduction of adrenal androgens is related to body weight and it is,therefore, likely that the higher amount of oestrogen produced in obesewomen is due to a higher production of androgen precursor, as well asits higher rate of aromatisation in adipose tissue. Menopause has beenassociated with increases in upper body (android) fat distribution.Oestrogen is thought to promote a gynaecoid fat distribution, whichcarries less risk for diabetes, and cardiovascular risk factors than andro idfat deposition.

A low oestrogen status causes several symptoms to a woman who isperi or postmenopausal. The vasomotor flush is the hallmark feature.3 5 2 British Medico/Bulletin 1997^3 (No. 2)

byg

uestonMarch7,2011


Downloa

dedfrom


13/18


Oestrogen deficiency causes atrophic changes in the urethra and vaginawhich may present as dyspareunia, pruritis, vaginitis or urinaryfrequency and dysuria. Oestrogen deficiency also changes trabecularbone and is thus important in the genesis of osteoporosis. The risk offracture from osteoporosis is know n to be dependant upon bone mass atthe time of menopause, and the rate of bone loss following themenopause. In general, bone mass is higher in obese women and in theAfro-Carribean population, and lower in thin, white women who leadsedentary lives. Oestrogen therapy will act prophylactically to preventosteoporosis or stabilize the process. Epidemiological evidence suggeststhat oestrogen therapy decreases hip and arm fractures by 50-60% 38 .This reduction is seen primarily in patients who have taken oestrogen formore than 5 years. The an nual m ortality from hip fractures in wom en ishigh and oestrogen replacement will considerably reduce the associatedmorbidity and mortality rate. A number of studies show that obesity isprotective against osteoporosis. Pocock and colleagues demonstratedthat the BMI was an important predictor of bone mass in the lumbarspine, distal forearm and femur39. This effect is thought to be due bothfrom the mechanical stress which excess body weight places upon themusculoskeletal system, as well as increased postmenopausal oestrogenconcentrations. This theory was supported by Krolner, who observedthat in patients who had undergone gastric bypass operations, the degreeof loss of mineral co ntent in the lumbar spine was closely associated withthe degree of weight loss, and hence mechanical stress 40 and was notrelated to calcium and vitamin D absorption, which would remainunaffected in these pa tients. The increased synthesis of oestrogen via theconversion of androstenedione into oestrogen in adipocytes is thought tooffer an explanation why skeletal regions which are not exposed to anyweight related stress also benefit from the protective effect associatedwith an increased BMI. In a review of lifestyle, obesity, and osteoporosis,Notelovitz reported that untreated menopause, particularly if naturallypremature or surgically induced, frequently overrides the protectiveeffect of obesity and following bone density evaluation in their patients,they found that most women required hormone additive therapy41 .Coronary heart disease (CHD) is now the leading cause of death inwomen and the risk of this disease markedly rises following the loss ofovarian function. During the reproductive years, women are protectedfrom coronary heart disease, presumably by the raised serum levels ofhigh density Upoprotein (HDL). Oestrogen replacement therapy increasethe HDLs and reduces LDL (low density lipoprotein) and cholesterollevels42. HRT has been found to reduce the incidence of CHD inpostmenopausal women by up to 50%. Obesity is a risk factor for thedevelopment of CHD and, therefore, an additional risk would be aprolonged low oestrogen state. HRT causes changes in lipids and

British Medical BuHehn 19 97 ^3 (No. 2) 3 5 3

y

g

,

j

g


14/18

Obesity

lipoprotein profiles42. The effects of gonadal steroid hormones haveimportant influences upon the cardiovascular system, and these non lipideffects may contribute to the pathogenesis of CHD. These include directarterial effects as well as the metabolic consequences of insulin resistanceand hyperinsulinaemia43 . Oestradiol reverses the effects of menopauseon glucose and insulin metabolism, causing insulin resistance to bedecreased. Androgenic progestagens may oppose this potentiallybeneficial effect on insulin resistance. HRT also reverses the increasedcentral fat distribution, which is associated with atherogenic biochem-ical parameters.Breast cancer and endometrial cancer are increased by long termoestrogen replacement therapy. The incidence of endometrial carcinomais increased in women who are obese, nulliparous and infertile. Obesewomen already have a substantially increased risk of developingendometrial cancer, varying from 3 times normal when 9.5-22.5 kgoverweight, to 10 times when the excess of weight is greater than22.5 kg44. A close correlation has also been shown between daily dietaryfat consumption and the development of endometrial cancer45. Theincreased peripheral conversion of androgens to oestrogens, and theresultant effect of the unopposed oestrogen upon the endometrium isthought to be responsible. Women who use unopposed oestrogens havea dose dependent risk of developing endometrial cancer, but currentknowledge assumes that mortality can be reduced by an increase inclinical surveillance. Women who take progestogens with the oestrogenare protected from any increased risk of developing endometrial cancer.Risks for postmenopausal breast cancer have been identified as obesity,late age at menopause, a family history of breast disease and an early ageat menarche. The association between HRT and breast cancer is weakand controversial. Thrombosis is a further reported risk factor in HRTuse, but current evidence would suggest that the arterial risks associatedwith obesity are not significantly worsened by the administration ofHRT. In 1988, Henderson et al.46 reviewed literature on mortality fromosteoporotic hip fractures, endometrial cancer, breast cancer andischaemic heart disease in women aged 65-74 years and calculated thehypothetical risk from postmenopausal oestrogen use in each of thesediseases. The total mortality from coronary heart disease andosteoporotic hip fractures is many times greater than from breast cancerand, therefore, on balance, the beneficial effects of therapy appear tooutweigh the small increased risk of breast cancer. The addition ofprogestagens considerably decreases the risk of endometrial cancer.The relationship between obesity and weight gain with postmeno-pausal hormone replacement therapy remains uncertain. While weightgain is common at the time of the menopause, it has also been suggestedthat oestrogen replacement is directly associated with an increased

3 5 4 British Midkal Bullmlm 1997;53 (No. 2)

byg

uestonMarch7,2011


Downloa

dedfrom


15/18


prevalence of obesity. There have been several studies upon the effect ofHRT on fat distribution, which have yielded contradictory results.Several of these studies have found HRT use associated with lowerWHRs, although other centres have reported that, after correcting forage and degree of obesity, this correlation disappears. In a randomisedprospective trial using a cross sectional cohort over a 15 year period byKritz-Silverstein et al., it appeared that women using intermittent orcontinuous hormone replacement therapy, had significantly lower meanBMIs at baseline than women who never used HRT. After adjustmentfor potentially confounding covariates, there were no significantdifferences between oestrogen users, and those not on oestrogenreplacement in BMI at follow up, change in weight or BMI betweenbaseline and follow up, or WHR or fat mass at follow up, therebysuggesting no association between H RT and weight gain and the cen tralobesity distribution of fat that is common in postmenopausal women47 .Weight gain seen around the time of the menopause is multifactorial andvaries substantially within ethnic and socioeconomic groups and may bemore related to age and lifestyle changes, rather than to the effect ofpostmenopausal oestrogen therapy.Obesity at the time of the menopause is a common occurrence andjustifies dietary advice and therapeutic action. Central obesity increasesthe risk from CHD, and the postmenopausal state is associated withmore atherogenic lipid profiles. Treatment with oral exogenousoestrogens induces favourable effects on lipids and lipoproteins, andepidemiological evidence has established that the risk of cardiovascularmortality is reduced by 40-60% in women receiving HRT. Althoughobesity does confer a protective effect upon bone loss in menopausalwomen, evidence would again suggest that mortality from osteoporoticfractures is considerably reduced in women using HRT. The contro-versial link between exogenous oestrogen use in peri and postmenopau-sal women and breast cancer has not been established, and withincreased surveillance for endometrial carcinoma, the relative benefitsconferred upon both obese and non obese postmenopausal womenwould appear to outweigh the detrimental effects of HRT.

ConclusionsThe implications and consequences of obesity are far reaching and obesewomen trying to become pregnant, or those who have alreadyconceived, will require advice on the effects of their body mass uponconception rates, and the risks and complications associated withpregnancy and childbirth. Those women with PCOS must be strongly

British Mtdical Bulletin 1997^3 (No. 2) 35 5

yg

,

j

g


16/18

Obesity

encouraged to lose weight in order to improve their menstrual functionand androgenic profile. Hirsuitism and subfertility may be effectivelytreated with a simple weight loss regimen, and may result in spontaneousovulation, thereby avoiding the need for ovulation induction usingpharmacological intervention. Should a pregnancy be achieved, bothwomen with PCOS and those with normal ovarian morphology who areobese are at a significantly greater risk of miscarriage, and nutritionaladvice and counselling should be offered at the earliest opportunity.Regular antenatal screening should be offered to monitor bloodpressure, maternal weight gain, glucose tolerance, and a sonographicassessment of fetal growth to exclude macrosomia. These women shouldbe advised of potential complications that may arise during the deliveryand while weight loss is not recommended in pregnancy, weight gainshould be restricted and monitored carefully. HRT is generally agreed tobe beneficial for postmenopausal women and, although body mass andobesity confer a degree of protection against osteoporosis, studies havedemonstrated that exogenous hormone replacement is still important toprevent CHD and osteoporotic changes, and the benefits of this therapywould appear to outweigh the disadvantages.

References

1011

12

Stein IF, Leventhal ML. Amenorrhoea associated with bilateral polycystic ovaries. Ann ]Obstet Gynecol 1935; 29: 181-91Poison DW , Adams J, Wadsworth J , Franks S. Polycystic ovariesa common finding in normalwomen. Lancet 1988; i: 87 0-2Franks S. Polycystic ovary syndromea changing perspective. Clin Endocrinol 1989; 31 : 8 7 -120Adams J, Franks S, Poison DW et al. Multifollicular ovaries: clinical and endocrine features andresponse to pulsatile gonadotrophin releasing ho rmone. Lancet 1985; ii: 1375-8Robinson S, Kiddy D, Gelding S et al. Decreased insulin sensitivity in women with polycysticovaries is related to menstrual disturbance. Clin Endocrinol 199 3: 39 ; 36871Singh KB, Mahajan DK, Wortsman J. Effect of obesity on the hormonal characteristics of thepolycystic ovary syndrome. / Reprod Med 1990: 39: 805-8Burghen GA, Givens JR, Kitabci AE. Correlation of hyperandrogenism with hyperinsulinaemiain polycystic ovarian disease. / Clin Endocrinol Metab 1980; 50: 113-5Nestler JE, Clore JN, Blackard WG. The central role of obesity (hyperinsulinaemia) in thepathogenesis of the polycystic ovary syndrome. Am ] Obstet Gynecol 1989; 161 : 1095-7Adashi EY, Resnick CE, Hernandez ER et al. Insulin-like growth factor-1 as an amplifier offollicle stimulating hormone action: studies on mechanisms and sites of action in cultured ratgranulosa cells. Endocrinology 1988; 122; 1583-91Cara JF , Fan J, A zzraello J, Rosenfield RL. Insulin like growth factor 1 enhances luteinizinghormone binding to rat ovarian theca-interstsitial cells. J Clin Invest 1990; 86: 560-5Sharp PS, Kiddy DS, Reed MJ, Anyaoku V, Johnstone DG, Franks S. Correlation of plasmainsulin and insulin like growth factor-1 with indices of androgen transport and metabolism inwomen w ith polycystic ovary syndrome. Clin Endocrinol 1991; 35: 25 3-7Hamilton-Fairley D, Kiddy D, Anyaoku V, Koistinen R, Seppala M, Franks S. Response of sexhormone binding globulin and insulin-like growth factor binding protein-1 to an oral glucose

35 6 British Mmdical Bull.tin 1997;53 (No. 2)

byg

uestonMarch7,2011


Downloa

dedfrom


17/18


tolerance test in obese women with polcystic ovary syndrome before and after calorierestriction. Clin Endocnnol 1993: 39: 363-713 Insler V, Shoham Z, Barash A et al. Polycystic ovaries in non obese and obese patients: possiblepathophysiological mechanisms based on new interpretation of facts and findings.Hum Reprod1993; 8: 379-8414 Kiddy DS, Sharp PS, White DM et al. Differences in clinical and endocrine features betweenobese and non obese subjects with polycystic ovary syndrome: an analysis of 263 consecutivecases. Clin Endocnnol 1990; 32: 213-2015 Hollmann M, Runnebaum B, Gerhard I. Effects of weight loss on the hormon al profile in obese,infertile women. Hum Reprod 1996: 11; 1884-9116 Kiddy DS, Ham ilton-Fairley D, Bush A et al. Improvement in endocrine and ovarian functionduring dietary treatment of obese women with polycystic ovarian syndrome. Clin Endocnnol1992; 36: 105-1117 Weaver JU, Holly JMP , Kopelman PG et al. Decreased sex hormone binding globulin (SHBG)and insulin like growth factor binding protein-1 (IGFBP-1) in extreme obesity. Clin Endocnnol1990; 33: 415-2218 Singh A, Ham ilton-Fairley DH, Koistenen R et al. Effect of insulin like growth factor type 1(IGF-1) and insulin on the secretion of sex hormone binding globulin and insulin like growthfactor binding protein-1 (IGFBP-1) by hum an hepatoma cells ./ Endocnnol 1990; 124: R1-R319 Balen AH, Conway GS, Kaltsas G et al. Polycystic ovary syndrome: the spectrum of thedisorder in 1741 patients. Hum Reprod 1995; 10: 2107-1120 Conway GS, Agrawal R, Betteridge DJ, Jacobs H S. Risk factors for coronary artery disease inlean and obese women with the polycystic ovary syndrome. Clin Endocnnol 1992; 37: 11 9-2521 Hamilton-Fairley D, Kiddy D, Watson H , Paterson C, Franks S. Association of m oderateobesity with a poor pregnancy outcome in women w ith polycystic ovary syndrome treated withlow dose gonadotrophin. Br ] Obstet Gynaecol 1992; 99: 128-3122 Poison DW, Kiddy DS, Mason HD , Franks S. Induction of ovulation with clomiphene citrate mwomen with polycystic ovary syndrome: the difference between responders and nonresponders. Fertil Stenl 1989; 51: 30-423 Pasquali R, Casimirri F, Venturo li S et al. Body fat distribution has weight independent effectson clinical and metabolic features of women with polycystic ovary syndrome. Metabolism1994; 43: 706 -1324 Zaads tra BM, Seidell JC , Van Noord PAH et al. Fat and female fecundity: prospective study ofeffect of body fat distribution on conception rates. BM] 1993; 306: 484-725 Bilenka B, Ben-Shlomo I, Cozacov C, Gold C H, Zoh ar S. Fertility, miscarriage and pregnancyafter vertical banded gastroplasty operation for morbid obesity. Acta Obstet Gynaecol Scand1995; 74: 42-426 Institute of Medicine subcommittee on nu tritional status and weight gain during pregnancy.Nutntion during Pregnancy. Washington DC: National Academy Press; 199027 Naeye RL. Weight gain and the outcome of pregnancy. Am ] Obstet Gynecol 1979; 135: 3-928 Siega-Riz AM, Adair LS, Hobel CJ. Institute of Medicine maternal weight gain recomm enda-tions and pregnancy outcome in a predominantly Hispanic population. Obstet Gynecol 1994;84: 565-7329 Abrams B, Laros RK. Prepregnancy w eight, weight gain and birth weight. Am J ObstetGynecol 1986; 154: 504-930 Wolfe H M, Z adpr IE, Gross TL, Martier SS, Sokol RJ. The clinical utility of maternal bodymass index in pregnancy. Am] Obstet Gynecol 1991; 164: 1307-1031 Edwards LE, Dickes WF, Alton IR. Pregnancy in the massively obese. Course outcome andobesity prognosis of the infant. Am J Obstet Gynecol 1978; 131: 47 9-8332 Gross TL, Sokol RJ, King KC. Obesity in pregnancy: risks and outcom e. Obstet Gynecol 1980;56: 446-5033 Johnson JWC, Longmate JA, Frentzen B. Excessive maternal weight and pregnancy ou tcome.Am] Obstet Gynecol 1992; 167: 353-7034 Wolfe H M, Sokol RJ, Martier SM, Zador IE. Maternal obesity: a potential source of error insonographic prenatal diagnosis. Obstet Gynecol 1990; 76: 3394235 Wolfe HM , Gross TL. Obesity in pregnancy. Clin Obstet G ynecol 1994: 37; 596-604

British Medical Bullmfm 1997 3 (No. 2) 3 5 7

y

g

,

j

g


18/18

Obesity

36 O'Scholl T, Hediger M L, Schall Jl, Ances IG, Smith WK. Gestational weight gain, pregnancyoutcome, and postpartum weight retention. Am ] Obstet Gynecol 1995; 86: 423737 Parker JD . Post partum weight change. Clin Obstet Gynecol 1994; 37: 528-3738 Kiel DP, Felson DT, Anderson JJ, Wilson PWF, Mokowitz M A. Hip fracture and the use ofoestrogen in postmenopausal women: the Framingham study. N Engl J Med 1987; 317: 116939 Pocock N, Eisman J, Gwinn T et al. Muscle strength, physical fitness and weight but not agepredict femoral neck bone mass. / Bone Miner Res 1989; 4: 441-840 Krolner B, Ranlov PJ, Clemmesen T, Nielson SP. Bone loss after gastrop lasty for morbidobesity: side effect or adaptive response to weight loss. Lancet 1982; i: 95 6-741 Notelovitz M . Lifestyle, exercise and osteoporosis. In: Drife JO, Studd JW, eds. HRT andOsteoporosis. London: Springer, 1990; 32 3-4 742 Bush TL, Miller VG. Effects of pharmacolog ical agents used during m enopause: impact onlipids and lipoproteins. In: Mishell DRJ, ed. Menopause: Physiology and Pharmacology.Chicago: Year Book Medical Publishers, 1987: 187-20843 Stevenson JC, Crook D, Goddsland IF, Collins P, Whitehead M I. Hormone replacementtherapy and the cardiovascular system. Non lipid effects. Drugs 1994; 47: 35-4144 Wynder EL, Escher GC, Mantel N . An epidemiological investigation of cancer of theendometrium. Cancer 1966; 19: 489-52045 Armstrong B, Doll R. Environmental factors and cancer incidence and mortality in differentcountries, with special references to dietary practices. Int ] Cancer 1975; 15: 617-3146 Hend erson BE, Ross RK, Lobo RA, Pike M C, Mack TM. Re-evaluating the role of progestogentherapy after the menopause. Fertil Steril 1988 ; 49 (Suppl): 9S-15S47 Kritz-Silverstein D , Barrett-Connor E. Long term post menopausal hormone use, obesity, andfat distribution in older women. JAMA 1996; 275: 46-9

byg

uestonMarch7,2011


Downloa

dedfrom

obesity and female reproductive function

Documents