-
7 Clin Pathol 1993;46:790-794
ACP Broadsheet No 141 September 1993
Role of endocrine biochemistry laboratoriesin the investigation
of infertility
G H Beastall
IntroductionInfertility is a descriptive term that embracesmany
specific diagnoses. For practical pur-poses, infertility is defined
as the failure of acouple to achieve a pregnancy despite oneyear of
regular unprotected sexual inter-course. Infertility is common:
estimates sug-gest that as many as one in six couples willneed
specialist help because of infertility atsome time.
Infertility may result from abnormalities ineither or both
partners or from their incom-patibility. Of the many specific
causes ofinfertility, some have a genetic or anatomicalbasis, some
result from a previous disease ortreatment, and some are endocrine
in origin.'Regrettably, it is not always possible to reacha
specific diagnosis. The challenge to theinvestigating physician is
great, and it isheightened still further by the pressure toreach a
specific diagnosis and start treatmentquickly despite a failure of
central funding tomatch the clinical demand.
It is now generally accepted that agreedstrategies are required
to maximise the effi-ciency of investigation of infertility.23
Detailsof such strategies will vary from centre to cen-tre, but in
all cases the services and staff ofthe pathology laboratory will be
vital. Thisarticle seeks to identify the role of theendocrine
biochemistry service in the investi-gation of infertility and to
recommend waysin which that service may best be provided.
This Broadsheet has beenprepared by the author at theinvitation
of the Association ofClinical Pathologists whoreserve the
copyright. Furthercopies of this Broadsheet maybe obtainedfrom
thePublishing Manager, Journalof Clinical Pathology, BMAHouse,
Tavistock Square,London WClH 9JR
Institute ofBiochemistry, RoyalInfirmary, GlasgowG4 OSFG H
BeastallAccepted for publication21 October 1992
Strategy of investigation of infertilityThe perfect strategy for
the investigation ofthe infertile couple would achieve the idealsof
reaching a specific diagnosis in the shortesttime, with minimum
inconvenience to thepatient and at least cost to the NHS.
Theperfect strategy that is applicable to all infer-tile couples
has yet to be designed, but withexperience, teamwork, and new
technologythe best centres are achieving an ever betterbalance
between these ideals. These centreshave each adopted a strategy of
investigationthat can be considered in a number of stages.
STAGE 1 THE INITIAL CLINICAL INVESTIGATIONIt is highly desirable
that both partners from
the infertile couple attend the initial investiga-tion, which
comprises a detailed clinical his-tory and physical
examination.
As part of the clinical history, bothpartners should be asked
about congenitalabnormalities, previous pregnancies,
seriousillnesses and infections, past chemotherapy orradiotherapy,
and current habits in relation tosmoking and the use of alcohol or
recre-ational drugs. Sensitive areas such as psycho-logical
disorders, venereal disease, and thefrequency and satisfaction of
intercourseshould be addressed. The male partnershould be
questioned about the descent of thetestes at puberty, penile
erections and ejacu-lation, and specifically asked about
mumpsorchitis, correction of varicocele, or exposureto radiation,
toxins or heat. The female part-ner will be required to give a full
menstrualhistory, details of all contraceptive prepara-tions used,
and of any changes in bodyweight. Specifically, she should be
questionedabout abortions, pelvic inflammatory disease,or pelvic
surgery.
Physical examination of both partnersshould include a careful
documentation ofsecondary sex characteristics and any featuresthat
could be consistent with hypothalamic orpituitary disease, thyroid
disease, orCushing's syndrome. The male genitaliashould be
examined, noting the size and con-sistency of the testes,
epidydimis, vas deferensand evidence of varicocele. In the
femalepartner it is important to examine the breastsfor
galactorrhoea, and the face and trunk forhirsutism. A pelvic
examination is mandatoryin a woman with primary amenorrhoea andmay
help to identify tenderness behind theuterus that could be
explained byendometriosis.
STAGE 2 SELECTION OF INITIAL LABORATORYTESTSThe selection of
appropriate initial laboratorytests will be determined by the
outcome ofthe initial clinical investigation (table
1).Over-requesting of laboratory tests should bestrongly
discouraged. No endocrine investiga-tions are indicated in a
eugonadal male.Measurement of serum progesterone is theonly valid
request in a woman with normal
790
on July 3, 2021 by guest. Protected by copyright.
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.46.9.790 on 1 S
eptember 1993. D
ownloaded from
http://jcp.bmj.com/
-
Role of endocrine biochemistry laboratories in the investigation
of infertility
Table I Selection of initial laboratory tests for the
investigation of infertility
Initial clinicalfindings Appropriate hormone tests Other
laboratory tests
Male-eugonadal Nil Semen analysisMale-hypogonadal FSH, LH,
prolactin Semen analysis
testosterone karyotypeFemale-regular periods Luteal progesterone
NilFemale-primary amenorrhoea Pregnancy test Karyotype
FSH, LH, prolactin, oestradiolFemale-secondary amenorrhoea
Pregnancy test Nil
FSH, LH, prolactin, oestradiolFemale-hirsutism Add testosterone,
Nil
sex hormone binding globulin
periods; paradoxically, assessment of proges-terone is unlikely
to be helpful in a womanwith amenorrhoea.
STAGE 3 INTERPRETATION OF INITIALLABORATORY TESTSTable 2
summarises the relevant referenceintervals used by the laboratory
at GlasgowRoyal Infirmary; other laboratories may quoteslightly
different figures, and it is importantthat physicians are
acquainted with the refer-ence intervals appropriate to the
laboratoriesthat they use.
In both men and women clinically impor-tant hyperprolactinaemia
is unlikely unlessthe serum prolactin concentration is morethan 700
mU/l on two consecutive occasions.If the prolactin is more than
4000 mU/l thepatient is either already pregnant or has apituitary
prolactin-secreting tumour. Resultsbetween 700-4000 mU/l may be due
to amicroprolactinoma or the result of a hypo-thalamic disorder,
polycystic ovarian disease,primary hypothyroidism or drug
treatment.4The interpretation of initial gonadal hor-
mone results in men is straightforward. Anincreased serum
follicle stimulating hormoneconcentration (FSH), accompanied by a
lowor low-normal testosterone value is diagnosticof primary
testicular disease. A subnormal
Table 2 Reference intervals for hormones used in the
investigation of infertility
Hormone analyte Normal subjects Reference interval
Adrenocorticotrophic hormone Adults 0700-0900 h
-
792
mU/l). Ultimately, the differential diagnosis isbetween
microprolactinoma (normal radio-logical findings) and idiopathic
hyperprolacti-naemia (deficient supply of dopamine);dynamic tests
of prolactin secretion may helpto differentiate.9
Cushing's syndromeThis is a rare cause of infertility in either
sex.Clinical features are usually present and thehypercortisolaemia
may be confirmed by fol-lowing a standard protocol based on
cortisoland adrenocorticotrophic hormone (ACTH)measurement.'0
Primary testicularfailurePrimary testicular failure is readily
identifiedby the initial endocrine tests. Damage to boththe
interstitial and tubular elements results inincreases in both FSH
and luteinising hor-mone; when tubular function is
selectivelyimpaired only FSH is increased.
Testosteroneconcentrations are low in primary testicularfailure, in
anorchia, and usually in Klinefelter'ssyndrome. Repeat analysis is
required to con-firm the pattern. Thereafter, the
endocrinelaboratory is restricted to checking the ade-quacy of
androgen replacement treatment, forsuch patients are almost
invariably sterile.
Hypogonadotrophic hypogonadism in the malepartnerThis condition
is characterised biochemicallyby subnormal testosterone
concentrations inthe presence of low or normal gonado-trophins. The
testes are capable of increasingtestosterone concentration by more
than100% in response to five days of stimulationwith human
chorionic gonadotrophin.' Thebiochemical pattern may result from
eitherhypothalamic or pituitary disorder, andalthough the
gonadotrophin releasing hor-mone (GnRH) test may discriminate, it
is notinfallible.3 Clomiphene (200 mg) can beeffective in
hypothalamic disease, and in aformal test serum luteinising
hormone, FSH,and testosterone will rise by at least 70%,45%, and
40%, respectively." Treatmentwith either pulsatile GnRH or
exogenousgonadotrophin demand laboratory support,but such treatment
is lengthy and expensive,and artificial insemination by donor is
aneffective alternative that requires no labora-tory backup for the
donor.
Women with regular periodsOnce ovulation has been confirmed,
there isno requirement for further biochemical moni-toring of the
female partner; other causes ofinfertility should be sought. Women
withanovular cycles are likely to be givenClomiphene, and endocrine
follow up shouldbe restricted to one or two progesterone
mea-surements in the luteal phase of each cycle.
Primary ovarian failureThe biochemical pattern of raised
gonado-trophins with subnormal oestradiol valuesrequires
confirmation. In all but a very few ofthese women fertility is no
longer an issue,
but hormone replacement therapy is requiredto assist libido and
prevent osteoporosis. Noconsensus exists as to the extent of
endocrinemonitoring of patients receiving oestrogen.
HirsutismThe detailed investigation of hirsutism iscomplex, and
readers are referred elsewherefor a comprehensive treatise.'2
Polycysticovarian disease is a common cause of infertil-ity that is
associated with mild to moderatehirsutism and a variable menstrual
pattern.The serum luteinising hormone is usuallyincreased relative
to the FSH. The totaltestosterone concentration may be
raised(rarely more than 10 nmol/l), but it may bewithin normal
limits. The sex hormone bind-ing globulin (SHBG) binding capacity
is usu-ally reduced, such that the free androgenindex is increased.
The biochemical patternshould be confirmed by repeat sampling,
andthe finding of modestly increased androstene-dione with or
without dehydroepiandros-terone sulphate results substantiates
thediagnosis which can be confirmed by ovarianultrasound
examinations. Patients complain-ing of infertility are likely to be
treated withClomiphene or Dexamethasone rather thanaggressive
antiandrogen treatment, and hor-mone measurement should be targeted
atmonitoring for ovulation.
Severe hirsuitism of short duration stronglysuggests an androgen
secreting tumour, usu-ally adrenal or ovarian. Serum
androgens,especially testosterone, are often grosslyincreased. Such
tumours can vary consider-ably in the pattern of steroids that
theysecrete, and the laboratory should ensure thatall possible
androgens are measured in serumand that a urinary steroid profile
is per-formed." Corticosteroid secretion may alsobe deranged and
ACTH secretion sup-pressed. Treatment may involve
surgery,chemotherapy, or radiotherapy; the labora-tory should
discuss which of the steroid mea-surements is the most appropriate
tumourmarker.
So-called late onset congenital adrenalhyperplasia is
increasingly being recognised asa cause of hirsutism and
infertility. It isimportant to discriminate this partial steroid2
1-hydroxylase deficiency from polycysticovarian disease, because
corticosteroidreplacement is very effective in the former.The
measurement of serum 17-hydroxyprog-esterone is the key analysis,
an exaggeratedresponse 1 hour after synthetic ACTH is thedefinitive
test.'4
Hypogonadotrophic hypogonadism in the femalepartnerA few women,
often with primary amenor-rhoea, will have subnormal FSH or
luteinis-ing hormone and oestradiol results. Theyhave
hypopituitarism resulting either from atumour in the hypothalamus
or pituitary, orfrom failure of the hypothalamic-pituitary axisto
mature. Detailed investigation of the ante-rior pituitary,
including hormone testing, isindicated.
Beastall
on July 3, 2021 by guest. Protected by copyright.
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.46.9.790 on 1 S
eptember 1993. D
ownloaded from
http://jcp.bmj.com/
-
Role of endocrine biochemistry laboratories in the investigation
of infertdity
A large proportion ofwomen with amenor-rhoea or oligomenorrhoea
show no obviousabnormality in their initial hormone
results.Repeated baseline hormone measurement isof no value in
these patients. A proportion ofthe women will have polycystic
ovarian dis-ease despite the absence of hirsutism-a rela-tive
increase in the luteinising hormone:FSH ratio may be a pointer, and
limitedendocrine follow up is justified as a comple-ment to ovarian
ultrasound examination.Many women in this group, however, willhave
a subtle failure of the hypothalamic-pituitary axis, most notably
of the positivefeedback mechanism responsible for the mid-cycle
surge.
Clomiphene is commonly used to induceovulation in the latter
patients; progesteronemeasurements during treatment are indi-cated.
The dose of Clomiphene will be lowerand the success rate will be
higher in patientswho have good oestrogen concentrations.Although
the initial serum oestradiol result isof some value in this
judgment, the progesto-gen challenge test to induce
withdrawalbleeding is still widely used.A proportion of patients
will fail to respond
to Clomiphene. These women are candidatesfor treatment with
exogenous gonado-trophins, perhaps following down-regulationof
endogenous pituitary FSH or luteinisinghormone secretion with a
GnRH agonist.Laboratory monitoring of the oestradiolresponse to
gonadotrophin treatment willhelp to minimise ovarian
hyperstimulation.
Laboratory support for programmes of assistedconceptionGamete
intrafallopian transfer (GIFT) and invitro fertilisation (IVF) are
being usedincreasingly to treat carefully selected
infertilecouples. Follicular growth and oocyte matu-ration are
closely monitored both by ultra-sound examination and measurement
ofeither luteinising hormone or oestradiol.Timing is critical to
the success of these pro-cedures and laboratories must provide a
rapid(2-4 hours) turnround of results. In practicalterms this
usually means that special assaysare required on a daily basis,
including week-ends. The laboratory analysis is often bestperformed
adjacent to the patient treatmentarea.
Range of endocrine biochemistryanalysesA wide range of hormone
assays is needed ifall possible causes of infertility are to
beinvestigated (table 2). Two major factors willdetermine which
assays will be offered locallyand which will be sent to a reputable
refer-ence laboratory. Firstly, the clinical demandfor each analyte
will influence the assay fre-quency and result turnround time. The
labo-ratory should seek to return results fromnon-urgent
investigations within the followingtimescales: 7 days, FSH,
luteinising hor-mone, prolactin, oestradiol,
progesterone,testosterone, SHBG; 21 days, ACTH, 17-
hydroxyprogesterone, androstenedione, dehy-droepiandrosterone
sulphate. Results forpatients entered in assisted conception
pro-grammes must be returned much morerapidly (see previous
section).
Secondly, not all laboratories have thefacilities and expertise
necessary to performand interpret the results of all the
analyseslisted in table 2. Modern immunoassay kitsare technically
simple to perform but theyoften have compromises built into
theirdesign which can create pitfalls for theunwary. Circumstances
will vary considerablyamong laboratories, but as a guideline it
issuggested that FSH, luteinising hormone,prolactin and
progesterone assays be providedlocally. Once direct immunoassay
kits foroestradiol and testosterone are shown to bevalid for the
investigation of infertility, it islikely that they too should be
providedlocally. Other analyses are probably best pro-vided by a
reputable reference centre.
Specimen requirementMost hormone measurements are made inblood.
A 10 ml specimen is collected withoutvenous stasis and with minimum
stress. Ofthe analytes listed in table 2, all but ACTHare stable in
serum kept at room temperaturefor 24 hours. Longer term storage of
serumshould be at - 20°C.The timing of blood collection is crucial
to
the successful investigation of infertility.Several of the
hormones listed in table 2 aresecreted in pulses or exhibit a
circadianrhythm, but in practical terms any specimencollected
between 0900-1700 h will suffice,with the exception of cortisol and
ACTHspecimens which should be collected at0700-0900 h and 2200-2400
h.
In the male partner blood may be collectedon any day for basal
investigation. In thefemale partner the day of collection in
rela-tion to the last menstrual period is crucial;where periods do
occur the date of the lastmenstrual period should appear on
therequest form. Women with regular periodsgenerally only require
progesterone measure-ment to monitor ovulation-sampling shouldbe
between 19-24 days after the testmenstrual period for a 28 day
cycle, and cor-respondingly later for a longer cycle.
Amenorrhoeic women may be investigatedon any day before
treatment, but specimensshould be timed to coincide with
treatment.
Fluids other than blood may be used toinvestigate infertility.
Pregnancy tests are usu-ally performed on urine, and there are
nowseveral commercial systems for urine testingat home, both for
pregnancy and ovulation.There is likely to be a resurgence of
urineanalysis in the laboratory with the refinementof non-isotopic
assays for the simultaneousmeasurement of oestrogen and
progestogenglucuronides. Measurement of salivary prog-esterone has
been advocated as a simple andeffective non-invasive method of
monitoringovulation'5; it is perhaps surprising that theapplication
has not been more widely used.
793
on July 3, 2021 by guest. Protected by copyright.
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.46.9.790 on 1 S
eptember 1993. D
ownloaded from
http://jcp.bmj.com/
-
Beastall
Potential pitfalls ofhormonemeasurementModem isotopic and
non-isotopic immuno-metric assays are the methods of choice forthe
measurement of FSH, luteinising hor-mone, and prolactin. Individual
assays showexcellent precision and may be automated.There is method
dependent bias arising froma combination of impure calibrants and
anti-body pairs, however, that recognise differentantigenic
determinants.16 Some normal indi-viduals also secrete an isoform of
luteinisinghormone that may not be recognised by alldual monoclonal
antibody assays.'7
Problems persist with direct (non-extrac-tion) immunoassays for
serum steroids. Validassays are available for cortisol and
proges-terone (which bind to cortisol binding globu-lin) but most,
if not all, direct assays areinvalid for the measurement of
testosterone infemale sera and oestradiol at concentrationsbelow
300 pmolIl. The problem is caused bythe endogenous SHBG which
compromisesthe steroid-antibody binding reaction. Untilthis problem
can be resolved prior solventextraction is recommended.
Traditionally, the measurement of ACTH,SHBG,
17-hydroxyprogesterone, androstene-dione and dehydroepiandrosterone
sulphatehas been confined to a few reference centres.Although
methodology is now improving forthese analytes, the small workload
and lack ofexperience of the average laboratory suggestthat there
will be no major expansion in thenumber of centres offering these
tests.
The role of clinical biochemistsIf effective strategies for the
investigation ofinfertility are to be formulated and audited itis
essential for senior staff from the endocrinebiochemistry
laboratory to interact directlyand regularly with the physicians in
charge.The role of the clinical biochemist (chemicalpathologist)
may be summarised as follows:
(a) Ensure direct supply or access to a fullrange of hormone
assays of appropriateturnround time and be able to vouchfor the
validity of the results.
(b) Agree with the physicians a detailedstrategy for the
investigation of infertil-ity.
(c) Monitor the progress of patients beinginvestigated according
to the strategyand, where appropriate, offer specificinterpretative
comment.
(d) Attend regular audit meetings withphysicians to discuss the
success of thestrategy, workload trends and the out-come for
individual patients.
(e) Wherever possible, agree with physi-cians joint research
projects that aim toimprove the effectiveness of the investi-gation
or management of infertility.
SummThe staff and services of the endocrine bio-chemistry
laboratory are essential to the effi-cient investigation of
infertility. Each centreshould adopt a detailed strategy for the
inves-tigation of the infertile couple which specifiesthe hormone
analyses required at each stage.Appropriate first-line hormone
tests shouldbe selected after a thorough clinical historyand
physical examination of both partners.Second-line hormone testing
should be deter-mined from the results of the initial
investiga-tion and should be restricted to requests thateither
confirm or clarify an endocrine basis toinfertility or monitor the
response to treat-ment. The clinical biochemist should adviseon
specimen timing and collection, haveresponsibility for guaranteeing
time and validhormone results, and be part of the team thataudits
the overall strategy and the outcomefor individual patients.I
gratefully acknowledge the expert secretarial assistance ofMiss
Myra Ogilvie.
1 Hull MGR, Glazener CMA, Kelly NJ, et al. Populationstudy of
causes, treatment and outcome of infertility. BrMedj
1985;291:1693-7.
2 Swerdloff RS, Wang C, Kandeel FR. Evaluation of theinferdle
couple. Endocrinol Metab Clin North Am 1988;17:301-8.
3 Butt WR, Blunt SM. The role of the laboratory in
theinvestigation of infertility. Ann Clin Biochem
1988;25:601-9.
4 Jeffcoate SL, Bacon RRA, Beastall GH, Diver MJ, FranksS, Seth
J. Assays for prolactin: guidelines for the provi-sion of a
clinical biochemistry service. Ann Clin Biochem1986;23:638-5 1.
5 Beastall, GH, Ferguson KM, O'Reilly DStJ, Seth J,Sheridan B.
Assays for follicle stimulating hormone andluteinising hormone:
guidelines for the provision of aclinical biochemistry service. Ann
Clin Biochem 1987;24:246-62.
6 Ratcliffe WA, Carter GD, Dowsett M, Hillier SG, MiddleJG, Reed
MJ. Oestradiol assays: applications and guide-lines for the
provision of a clinical biochemistry service.Ann Clin Biochem
1988;25:466-83.
7 Loy R, Seibel MM. Evaluation and therapy of polycysticovarian
syndrome. Endocrinol Metab Clin North Am1988;17:785-813.
8 Semple CG, Beastall GH, Teasdale GM, Thomson JA.Hypothyroidism
presenting with hyperprolactinaemia.BrMedj 1983;286:1200-1.
9 Cowden EA, Ratcliffe JG, Thomson JA, MacPherson P,Doyle D,
Teasdale GM. The role of dynamic tests ofprolactin secretion in the
diagnosis of prolactinoma.Lancet 1979;1:1 15-8.
10 Howlett TA, Rees LH. Is it possible to diagnose
pituitary-dependent Cushing's disease? Ann Clin Biochem
1985;22:550-8.
11 Davis JC, Hipkin U. Biochemical assessment of hormonalstatus.
In: Belchetz PE, ed. Management ofpituitary dis-ease. London:
Chapman & Hall Ltd, 1984; 161-85.
12 Biffigandi P, Massucchetti C, Molinatti GM. Female
hir-sutism: pathophysiological considerations and therapeu-tic
options. Endocrinol Rev 1984;5:498-513.
13 Cook B, Beastall GH. Measurement of steroid
hormoneconcentrations in blood urine and tissues. In: Green B,Leake
RE, eds. Steroid hormones: a practical approach.Oxford: IRL Press
Ltd, 1987:1-66.
14 New MI, Lorenzen F, Lerner AJ, et al. Genotyping steroid2
1-hydroxylase deficiency: hormonal reference data. JClin
EndocrinolMetab 1983;57:320-6.
15 Walker RF, Walker S, Riad-Fahmy D. Salivary proges-terone as
an index of luteal function. In: Read GF,Riad-Fahmy D, Walker RF,
Griffiths K, eds.Immunoassays of steroids in saliva. Cardiff: Alpha
OmegaPublishing Ltd, 1982:115-126.
16 Seth J, Hanning I, Bacon RRA, Hunter WM. Progressand problems
in immunoassays for serum pituitarygonadotrophins: evidence from
the UK External QualityAssessment Schemes 1980-88. Clin Chim Acta
1989;186:67-72.
17 Pettersson K, Ding Y-Q, Huhtaniemi I. Monoclonal
anti-body-based discrepancies between two-site immuno-metric tests
for lutropin. Clin Chem 1991;37:1745-8.
794
on July 3, 2021 by guest. Protected by copyright.
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.46.9.790 on 1 S
eptember 1993. D
ownloaded from
http://jcp.bmj.com/
