Endocrine Function Tests
Clinical setting
In general, patients with endocrine disorders present to clinicians
because they are thought to have either hormone excess or hormone deficiency
and each of these states has a variety of underlying causes. The accurate diagnosis
of clinical endocrine disease depends upon knowledge of the principles of
feedback control described in Chapter 7, on an understanding of basic endocrine
biochemistry and on the availability of high quality assay systems provided by
chemical pathology laboratories (Fig. 8a).
Many hormones are secreted in a pulsatile fashion, often subject to
diurnal or ultradian rhythms, such that a single, untimed blood sample may be
of little or no diagnostic value. There are important exceptions to this rule,
in particular thyroid disease in which basal measurements of TSH and free
thyroid hormone concentrations are diagnostic in the vast majority of cases,
hyperprolactinaemia in which a single result in an unstressed patient is
reliable and calcium and PTH measurements which are stable. Characteristically,
endocrine disorders disrupt normal feedback mechanisms and this feature is
exploited in the interpretation of a number of endocrine function tests.
Furthermore, certain hormones rise in response to stressful stimuli and this
too can be utilized for diagnostic purposes. Because of these special
considerations, collection of anything other than basal and straightforward
blood samples must be undertaken by experienced staff who are aware of the
appropriate local protocols. In hospital endocrine referral centres, specialist
nurses are skilled in correct patient preparation for the tests, the delivery
of drugs required for hormone stimulation or suppression, careful clinical
monitoring of patients, observation for side effects during the tests and
correct management of samples. The latter forms a vital part of dynamic testing
protocols and more complex investigations should always be performed in
conjunction with chemical pathology staff to ensure that samples are collected
into the correct pre- servatives at the necessary temperature and maintained in
ideal conditions prior to transfer to the laboratory.
Careful recording of the timing of the test, any symptoms experienced by
the patient and the results are essential.
There are many dynamic function tests employed in clinical endocrinology
and clinicians must refer to local protocols and normal ranges. Examples of
some commonly used dynamic endocrine function tests are shown in Fig. 8b. Two
examples follow which illustrate some important principles of endocrine
testing.
Insulin tolerance test
The insulin tolerance test (ITT) is used to assess the anterior pituitary
reserve of growth hormone (GH) and adrenocorticotrophin (ACTH), both of which
are stress hormones and rise in response to illness and hypoglycaemia. The ITT
tests the response to a hypoglycaemic stimulus which acts at the level of the
hypothalamus to stimulate the production of these pituitary counter-regulatory
hormones. The ITT is contraindicated in
patients with significant ischaemic heart disease, epilepsy, glycogen storage
diseases and severe hypoadrenalism (0900 h cortisol <100 nmol/l). Thus a
0900 h cortisol measurement and ECG must be performed prior to the test and 25%
dextrose and hydrocortisone available for intravenous injection during the test
if required. An experienced nurse and a doctor must be present throughout and
resuscitation equipment be available. If in doubt, a glucagon test should
replace the ITT, although it produces less reliable results.
Patients fast from 2200 h the night before the test, which should start
at 0900 h. After weighing, soluble insulin (Actrapid) is given as a bolus dose
of 0.15 U/kg. Blood samples are taken for glucose, cortisol and GH at regular
intervals – the blood glucose must fall below 2.2 mmol/l to provide an adequate
hypoglycaemic stimulus (further insulin may be given if this is not achieved).
At the end of the test the patient must be given food to eat and 100 mg iv
hydrocortisone if the hypoglycaemia was severe. Fig. 8c shows a typical ITT
recording chart. Results may vary from laboratory to laboratory and advice
about normal responses should be checked locally. In general, severe GH
deficiency is indicated by a peak GH of 3 µg/L or less and a normal peak
cortisol should exceed 550 nmol/L.
Water deprivation test
A water deprivation test (WDT) is performed when there is a clinical
suspicion of either central or nephrogenic diabetes insipidus (see Chapter 36)
or to investigate thirst and polyuria. Like the ITT, extreme care is needed to
perform a WDT and constant supervision to prevent the patient from drinking, to
monitor body weight and to handle plasma and urine samples appropriately.
Extreme caution should be taken in patients with severe DI and the diagnosis
may be made on overnight basal samples alone where the plasma osmolality is >295
mosmol/kg and the urine
osmolality/plasma osmolality ratio (U/P) is <2.0.
Before starting the WDT the patient should be allowed to drink freely to
0800 h but should avoid tea, coffee or smoking. The patient should be weighed
and 97% of this weight recorded. From 0800 h all fluid intake is discontinued
for 8 hours. The patient is weighed and urine and plasma samples taken hourly
for measurement of urine volume and urine and plasma osmolalities. If the
weight loss exceeds 3% of body weight the test is discontinued, plasma
osmolality measured urgently and desmopressin given if osmolality >305
mosmol/kg. Assuming the test continues, at 1600 h desmopressin 2 µg is given
intra- muscularly and urine and plasma samples continued for a further 4 h.
In cranial diabetes insipidus the plasma osmolality rises with
inappropriately high urine volumes and no evidence of concentration. After
desmopressin, urine volumes fall with normal concentration. In nephrogenic
diabetes insipidus the urine fails to concentrate after desmopressin injection.
Results in primary polydipsia are variable and urine concentration may not maximize
due to previous high urine volumes causing a decrease in the osmotic gradient in the loop of Henle.
