Hyperprolactinaemia can manifest clinically in many different ways, varying significantly depending on the patient's characteristics, as well as the cause and magnitude of prolactin (PRL) excess. It is normally unmasked during the evaluation of presenting complaints, such as amenorrhoea, galactorrhoea, erectile dysfunction, or headaches.
Once the diagnosis of hyperprolactinaemia has been confirmed through detection of serum PRL levels above the normal reference range (for the particular laboratory), its underlying cause is evaluated in a step-wise manner. A single measurement of serum PRL is recommended. Dynamic tests for PRL secretion are not needed. In patients with symptomatic non-physiological hyperprolactinaemia, recommended first steps are excluding medication use, renal failure, hyperthyroidism, and parasellar tumours as possible causes.
On detection of hyperprolactinaemia:
A detailed history should be obtained and a thorough physical examination performed (to aid diagnosis of the underlying cause).
History and examination should try to exclude physiological causes.
Medicines known to cause hyperprolactinaemia should be withdrawn (may require consultation with a specialist physician, such as a psychiatrist or cardiologist, depending on the offending drug).
Laboratory tests to help identify the underlying cause:
Liver and renal function tests
Thyroid function tests
Macroprolactin levels (measured in asymptomatic or atypical presentations).
Magnetic resonance imaging (MRI) scan of the pituitary gland is performed once physiological, drug-induced, and secondary hyperprolactinaemias have been excluded.
If a macroadenoma is present, the patient should be investigated for hypopituitarism.
If no lesion is detected on MRI, idiopathic hyperprolactinaemia is diagnosed.
When there is a discrepancy between a large pituitary tumour on imaging and mildly elevated PRL, serial dilution of serum samples can be used to eliminate an artifact that can occur with some immunoradiometric assays, which can lead to a falsely low PRL reading.
Clinical history and examination
During the evaluation of a patient with hyperprolactinaemia, a detailed past medical, drug, and family history should be obtained, and a thorough physical examination performed, focusing on signs and symptoms indicative of PRL excess.
The predominant clinical consequence of hyperprolactinaemia, in both men and women, is hypogonadotrophic hypogonadism, which is due to an interruption in the pulsatile secretion of gonadotrophin-releasing hormone by the elevated PRL. This leads to decreased levels of follicle-stimulating hormone (FSH) and luteinising hormone (LH) and a concomitant reduction in oestrogen and testosterone concentrations that manifest with characteristic clinical features.
Hyperprolactinaemia presents with a variety of signs and symptoms depending on its severity and underlying cause:
In both men and women, prolactinomas, particularly macroprolactinomas (>10 mm in diameter), produce signs and symptoms of a 'mass effect'. These include headaches, visual loss or visual field defects, cranial nerve neuropathies, and seizures. Hypopituitarism may also occur as a result of compression of the hypothalamic-pituitary stalk, or other pituitary cells. Hypopituitarism may involve any or all of the endocrine axes regulated by the pituitary. In order of frequency, it leads to growth hormone (GH) deficiency, secondary hypogonadism, secondary hypothyroidism, and secondary adrenal failure. It may also cause diabetes insipidus.
In women, moderate PRL excess (i.e., 2217-3261 picomol/L [51-75 micrograms/L]) is usually associated with oligomenorrhoea, whereas mild PRL elevation (i.e., 1348-2174 picomol/L [31-50 micrograms/L]) results in a short luteal phase and decreased fertility and libido.
Hyperprolactinaemia also presents with different signs and symptoms depending on the age and sex of the patient:
Pre-menopausal women usually have PRL levels >4348 picomol/L (100 micrograms/L) and present with galactorrhoea, oligo/amenorrhoea, infertility, menstrual disturbances, decreased libido, sexual dysfunction, hirsutism, vaginal dryness with dyspareunia, habitual abortions, and the possibility of pregnancy.
Post-menopausal women usually present with signs and symptoms of a mass effect, and galactorrhoea occurs only in the presence of oestrogen replacement therapy.
Other clinical manifestations in post-menopausal women include osteopenia and osteoporosis (loss of height and back pain, but without an increased risk of pathological fractures) secondary to hypo-oestrogenaemia resulting from elevated PRL levels, as well as hirsutism and acne, which may be due to increased secretion of dehydroepiandrosterone sulphate (DHEAS) from the adrenal glands, also known as hyperandrogenaemia.
Men present with infertility (due to reduced sperm production), impotency (due to erectile dysfunction), decreased libido, galactorrhoea, gynaecomastia, reduced bone density, and signs and symptoms of a mass effect.
Children present with delayed puberty, and signs and symptoms of a mass effect in both sexes, and primary amenorrhoea and galactorrhoea in girls.
The patient should also be evaluated for signs and symptoms related to possible underlying causes of hyperprolactinaemia such as hypothyroidism (fatigue; sensitivity to cold; constipation; menstrual abnormalities; muscle cramps; weight gain; thyroid goitre; cool, rough, and dry skin; puffy face and hands; husky voice; slow reflexes; thin and brittle nails; carotenaemic skin colour) and other systemic diseases, including chronic renal failure (signs of uraemia: hypertension, yellow skin, uraemic fetor, decreased mental status) and cirrhosis (hepatosplenomegaly, weight loss, ascites, jaundice, peripheral oedema, encephalopathy, pyrexia, dilatation of superficial veins). Hyperprolactinaemia has also been associated with preclinical atherosclerosis (assessed by carotid intima media thickness), metabolic abnormalities (such as increased insulin resistance), and systemic inflammation (assessed by increased levels of high-sensitivity C-reactive protein).
A comprehensive medical history is required to reveal the presence of drugs known to raise PRL levels. Such drugs should be withdrawn.
Physiological causes of hyperprolactinaemia, such as pregnancy, lactation, hypoglycaemia, myocardial infarction, surgery, psychological stress, exercise, food ingestion, sexual intercourse, and sleep, should also be excluded.
The physical examination should include evaluation of the chest wall, abdomen, and breast searching for signs of cirrhosis, as well as trauma (e.g., burns, bruising) and surgical scars. Palpation of the thyroid gland (to identify a goitre), ophthalmological assessment (with particular attention to the visual fields) and assessment for the presence of hypogonadism (e.g., testicular size and texture, pattern of body hair growth) should also be carried out.
The initial diagnostic focus should be on identification of a plausible non-tumour aetiology, as this will avoid the associated risks and expense of unnecessary computed tomography (CT) and MRI scans.
When PRL values are not diagnostic, levels should be re-evaluated on another day, at least 1 hour after the patient has awakened or eaten. To avoid pulsatile secretion, 2 or 3 samples separated by at least 15 to 20 minutes, are required (although some trials have shown no significant difference between such values). Stress is a confounding parameter that can cause pseudohyperprolactinaemia in 27% of cases. In these cases, PRL measurement drawn from an indwelling antecubital cannulation, either as a repeated sample or after 120-minute bed rest, can quite reliably solve these problems. A resting sample may be considered if random PRL values are below 4087 picomol/L (94 micrograms/L). Drugs known to elevate PRL levels should be withdrawn, with appropriate advice from a specialist physician, such as a psychiatrist or cardiologist, before a second sample is taken. The duration of withdrawal depends on the half-life of the drug but is usually between 48 and 72 hours. There is no need to suspend oral contraceptives or oestrogen-progesterone hormone therapy.
Renal, liver, and thyroid function should be tested with measurement of electrolytes, urea, and urinalysis; liver function tests and albumin; thyroid function tests (thyroid-stimulating hormone [TSH], free thyroxine [T4]) and prothrombin time, respectively.
In asymptomatic or atypical presentations, or when menstrual disorders can be attributed to other causes, such as polycystic ovary syndrome, macroprolactinaemia should be measured via polyethylene glycol (PEG) precipitation and excluded. Harmonisation of PEG precipitation process and reporting by laboratories may improve diagnostic accuracy. Gel filtration chromatography is another method, but it is time consuming, expensive, and not used routinely in clinical laboratories. Assessment of PRL concentrations after separation of PRL macroforms and the use of a more appropriate reference range for PRL can help to differentiate true hyperprolactinaemia and pseudohyperprolactinaemia.
PRL stimulation tests including thyrotropin-releasing hormone and anti-dopamine (metoclopramide, domperidone) tests have been proposed, but their sensitivity and diagnostic value in determining the need for subsequent imaging techniques are low.
After exclusion of physiological, pharmacological, and other secondary causes of hyperprolactinaemia, MRI of the head with selective pituitary cuts is required to evaluate patients for a mass lesion. MRI is preferred to CT, as CT may miss small lesions while exposing the patient to large doses of radiation.
A CT scan is indicated in cases of calcified sellar or suprasellar tumours such as craniopharyngiomas, or when bony structures are involved.
In the absence of symptoms, the clinical utility of finding a pituitary adenoma with MRI is questionable, as up to 10% of imaging studies on asymptomatic patients yield abnormal pituitary findings suggestive of adenoma known as an incidentaloma.
Some investigators suggest evaluation with MRI only if PRL levels are >4348 picomol/L (100 micrograms/L), whereas others recommend it for all patients with persistently elevated PRL levels in the absence of an identifiable secondary cause for the hyperprolactinaemia.
The optimal PRL cut-off value that distinguishes between prolactinoma and non-functioning pituitary adenoma (NFPA) has not been established. However, 98.7% of patients with NFPA with suprasellar extension have PRL levels below 4087 picomol/L (94 micrograms/L or 2000 milliunits/L), suggesting that serum PRL levels of up to 4348 picomol/L (100 micrograms/L) may be a useful cut-off (based upon 1087 picomol/L [25 micrograms/L] PRL as the upper limit of normal for women). Similarly, a retrospective review found that only 4.2% of patients with NFPA had PRL levels >3913 picomol/L (90 micrograms/L).
If a macroadenoma is detected, the patient should be further screened for hypopituitarism through evaluation of the levels of the other pituitary and end-organ hormones (adrenocorticotrophic hormone, LH, FSH, GH, TSH, free T4, glucose tolerance test, cortisol, testosterone, estradiol, DHEAS, insulin-like growth factor 1). Some clinicians perform initial basal determinations of these hormones to rule out associated hypersecretion and to establish baseline reference values for comparison following future interventions.
If a macroadenoma is detected in the presence of mild hyperprolactinaemia, the most likely diagnosis is a non-PRL-producing pituitary adenoma or other sellar mass causing the 'stalk effect'.
Some cases of giant prolactinomas (large macroadenomas >2 cm) present with factitiously normal or even low PRL levels, a phenomenon known as the 'hook effect'. This effect is due to saturation of anti-PRL antibodies in an assay (normally immuno-radiometric) tube causing artificially low results. Further measurement of PRL following a 1/100 dilution results in diagnosing this situation.
If physiological, drug-induced, and secondary hyperprolactinaemias have been excluded, and no lesion is detected on MRI, the diagnosis of idiopathic hyperprolactinaemia is made.
Use of this content is subject to our disclaimer