Eur J Clin Invest 2011; 41 (3): 334–342
Background Prolactinomas are the most common hormone-secreting pituitary tumours and are amenable to medical therapy with dopamine agonists. Indication for treatment will most commonly result from hypogonadism, infertility or symptoms related to tumour size. Thus, both diagnosis and treatment will essentially depend on the patients’ stage of life, namely prepubertal, reproductive or postreproductive stage. This review will focus on a lifespan-dependent diagnosis and treatment for prolactinoma and hyperprolactinaemia.
Methods PubMed, the Cochrane Library, the Web of Science and EMBASE were searched electronically. No restriction was made with respect to language. Relevant current articles will be included in this review.
Results Prevalence of prolactinomas and clinical symptoms are age group-specific, and treatment of first choice is dopamine agonists over the whole lifespan. Open questions in the treatment for hyperprolactinaemia include optimal choice and duration of pharmacological treatment. In addition, concerns have been raised on the safety of dopamine agonists since a reported association of valvular heart disease with dopaminergic treatment in patients with Parkinson’s disease.
Conclusions Clinical presentation and consequences of hyperprolactinaemia and prolactinoma will differ in the specific stages of reproductive life and require an adequate lifetime-dependent diagnostic and therapeutic approach.
Prolactinomas are the most common neuroendocrine tumour representing about 40% of all pituitary tumours. By definition, prolactinomas express and secret prolactin, are mostly benign and are classified as microadenomas with a size of less than 10 mm in diameter or macroadenomas if greater than 10 mm in diameter . The prevalence of prolactinomas has been estimated to be approximately 100 per million . More recent data suggest that the incidence of pituitary tumours may be much higher than previously reported, reaching 10% in the general population with more than 40% of pituitary adenomas being prolactinomas . The frequency of prolactinomas varies with age and sex, and they are most frequently observed in women between 20 and 50 years of age. The female-to-male ratio is about 10 : 1 in the first five decades of life and similar thereafter [4,5]. In children, prolactinomas are rare, but it appears that tumours are more aggressive because they present more frequently as macroadenoma and with accompanying neurological symptoms . Most prolactinomas are sporadic but may occur as part of genetic syndromes such as the multiple endocrine neoplasia syndrome 1 . Although risk factors for the development of prolactinomas have not been identified, oestrogens as stimulators of prolactin secretion have been discussed . However, oestrogen therapy is not associated with an increased risk of prolactinomas . Prolactin secretion is inhibited by hypothalamic dopamine. Situations that alter dopamine production or transport to the pituitary such as drugs or tumours will result in hyperprolactinaemia without prolactinoma (Tables 1 and 2). In addition, hypothyroidism can cause hyperprolactinaemia owing to the increased synthesis  of thyrotropin-releasing hormone. Physiological causes of hyperprolactinaemia include pregnancy, lactation and breast stimulation and stress. Irrespective of their cause, elevated prolactin levels will lead to reactive elevation of hypothalamic dopamine secretion and subsequent suppression of GnRH secretion . Thus, the leading symptom of hyperprolactinaemia is primary hypogonadism in the prepubertal age and secondary hypogonadism in the reproductive age. This review will focus on age- and sex-specific effects of hyperprolactinaemia and prolactinoma and an adequate lifetime-dependent diagnostic and therapeutic approach.
|Physiological causes||Pathological causes|
|Breastfeeding||Decreased dopaminergic inhibition of prolactin secretion|
(trauma, tumour, pituitary stalk compression, inflammatory diseases)
|Nipple stimulation||Drugs (see Table 2)|
|Sexual intercourse||Renal failure|
|Psychological stress||Hepatic failure|
|Macroprolactin||Chest wall injury|
|Typical antipsychotic agents||Phenothiazine drugs, haloperidol, pimozide|
|Atypical antipsychotic agents||Risperidone, olanzapine|
|Antidepressant drugs||Clomipramine, desipramine, imipramine, amitriptyline|
|Antiemetic drugs||Metoclopramid, domperidone|
|Antihistaminic drugs||Cimetidine, ranitidine|
|Hormones||Oestrogens (high dose)|
Cyproteronacetate (high dose)
|Antihypertensive drugs||Verapamil, reserpine, alpha-methyldopa|
|Opiates||Codeine, morphine and derivatives|
Preparing this review, PubMed, the Cochrane Library, the Web of Science and EMBASE were searched electronically. No search restriction was made with respect to language. Keywords used were prolactinoma, hyperprolactinaemia, dopamine agonist, paediatric hyperprolactinaemia, postmenopausal hyperprolactinaemia. Searches were carried out in July 2010. Relevant articles in English language are cited in this review and reflect a personal selection of the author.
Clinical symptoms result from endocrine effects of hyperprolactinaemia and from mass effects owing to tumour expansion. Prolactin induces galactorrhoea and hypogonadotropic hypogonadism owing to an inhibitory effect on hypothalamic GnRH release. Thus, in children, primary hypogonadism and delayed puberty will be observed [6,11], and in premenopausal women, secondary oligo- or amenorrhoea. In children and adolescents, macroadenomas are more frequent in men, and tumours are diagnosed later with a higher incidence of headache and vision impairment, growth arrest and other pituitary disturbances when compared to women. In the reproductive age, hyperprolactinaemia in men causes impotence, infertility and loss of libido. Whereas the typical complex of galactorrhoea and amenorrhoea usually will result in prompt medical consultation by female patients, the more subtle complex of decreased libido and erectile dysfunction may delay diagnostic procedures in adult men . In female patients, most prolactinomas are microadenomas with a diameter <10 mm, whereas male patients present later in life and with larger tumours . In postmenopausal women, the classical features are not present and symptoms are related to mass effects, such as visual loss and headache, but oestrogen replacement may allow galactorrhoea also in postmenopausal patients. Mass effects of larger tumours include compression of pituitary cells or pituitary stalk resulting in additional defects of pituitary functions and/or neurological symptoms (Table 3). Long-term hypogonadism may lead to reduced bone mass [14,15].
|Prepubertal||Delayed puberty||Delayed puberty|
|Growth delay||Headache, impaired vision|
|Chronic anovulation||Erectile dysfunction|
|Postreproductive||Reduced bone mass||Reduced bone mass|
|Neurological symptoms||Neurological symptoms|
The diagnosis of hyperprolactinaemia and prolactinoma requires careful clinical evaluation, laboratory testing and pituitary imaging techniques. At first, physiological causes of prolactin elevation need to be ruled out by clinical history and examination (Table 1). If the patient is taking a drug, known to increase prolactin levels (Table 2), the candidate drug should be discontinued if possible or replaced by an alternative drug, which does not increase prolactin secretion. Prolactin levels should be remeasured after 72 h . The initial laboratory testing should include a pregnancy test, routine biochemical parameters to exclude renal or hepatic insufficiency and TSH determination. Excessive stress should be avoided, and the patient should be awake and fasted for at least 1 h . Normal prolactin levels in women are below 20–25 μg L−1 and below 15–20 μg L−1 in men with the commonly used assays (1 μg L−1 is equivalent to 21·2 mIU L−1). Moderate elevations of prolactin up to 100 μg L−1 can be caused by microprolactinomas, pituitary macroadenomas other than prolactinomas as well as antidopaminergic drugs, oestrogen or functional causes. Macroprolactinomas are typically associated with prolactin levels higher than 200 μg L−1, and there is an correlation between tumour size and prolactin levels [1,9]. Misleading factors exist in the presence of prolactin autoantibodies and the high-dose hook effect. In up to 16% of patients, autoantibodies against prolactin may form prolactin–IGG complexes called macroprolactin. Macroprolactin has no or reduced bioactivity, and macroprolactinaemia therefore has no pathophysiological relevance [18,19]. The presence of macroprolactin can be detected by polyethylene precipitation. Macroprolactin should be measured in patients with moderately elevated prolactin levels and lacking or atypical symptoms . The high-dose hook effect may be present in very large macroadenomas when very high prolactin levels lead to saturation of the antibody in the immunoassay and false negative or low results. If suspected – for example, in a case of a macroadenoma with normally or slightly elevated prolactin levels – a serial dilution should be performed to rule out the hook effect [21,22].
Dynamic tests of prolactin secretion have been suggested to be helpful in the differential diagnosis of hyperprolactinaemia including TRH stimulation, L-dopa, domperidone and insulin-induced hypoglycaemia . However, in general, it is appropriate to make a diagnosis of prolactinoma on the basis of basal prolactin levels, imaging of the pituitary after exclusion of differential causes of hyperprolactinaemia as described earlier [23,24]. Testing of pituitary function is usually not necessary in patients with microadenoma but should be performed in patients with macroadenoma because additional defects are present in approximately 45% of patients . Visual field testing is advisable in all patients with macroadenoma as suprasellar tumour expansion and compression of the optic chiasm require immediate treatment.
After exclusion of secondary causes of hyperprolactinaemia, radiologic imaging further ascertains the diagnosis of prolactinoma. Gadolinium-enhanced MRI is preferred to contrast-enhanced computed tomography . The high prevalence of hormonally inactive microadenomas (incidentalomas) in up to 10% of the normal population has to be considered . Normal MRI findings, on the other hand, do not absolutely exclude a microprolactinoma , which may be explained by the detection limit of the MRI apparatus. The presence of an adenoma on MRI is consistent with a prolactinoma but does not prove it, because any pituitary tumour may cause hyperprolactinaemia because of pituitary stalk compression. Proof of the diagnosis would require histological analysis, but this will be rarely available as only a minority of prolactinomas requires surgical treatment. Empirical confirmation of the diagnosis can be deduced from a tumour response with decreasing prolactin levels and significant reduction in tumour size following dopaminergic therapy.
Treatment for hyperprolactinaemia and prolactinoma will aim at reducing tumour size, normalizing prolactin levels and controlling biological consequences of prolactin elevation. Indications for treatment are gonadal and sexual dysfunction such as delayed puberty, infertility, testosterone deficiency, oligo- and amenorrhoea, acne and hirsutism, intolerable galactorrhoea, an enlarging microadenoma and all macroadenomas. No medical treatment and watchful waiting may be justified in asymptomatic patients with hyperprolactinaemia or a microadenoma. Studies on untreated microprolactinomas indeed showed little or insignificant tumour growth over a time up to 8 years [25–27]. In case of no treatment, PRL levels and tumour size (by MRI) should be checked on a regular basis once yearly in a stable clinical situation . In the reproductive age, in women, oral contraceptives may be safely used with an expected increase in PRL levels of 20–30% but no increase in tumour size . Spontaneous resolution of hyperprolactinaemia may occur and is especially likely in eumenorrheic [30,31] or postmenopausal  women. Therefore, in most postmenopausal women, microadenomas require no treatment unless there is bothersome galactorrhoea together with hormone replacement therapy.
Treatment of first choice is dopamine receptor agonists. First-generation dopamine agonists are bromocriptine and the less frequently used lisuride and pergolide, and second-generation drugs are cabergoline and quinagolide. Bromocriptine, cabergoline and pergolide are ergot derivatives, and quinagolide is a non-ergot derivative. The most commonly used compounds are bromocriptine and cabergoline, followed by quinagolide, which is not available in the United States (Table 4).
|Medication||Dose||Specific aspects||Side effects|
|Bromocriptine||–10 (max. 30) mg|
2–3 times day−1
|More side effects|
Available data for children and pregnancy
Nausea, headache, hypotension, constipation, nasal congestion
Available data for children and pregnancy
Fatigue, anxiety, depression
|Quinagolide||75–150 (max. 600) μg|
Limited data for children and pregnancy
Limited data concerning duration of therapy
Typical dose regimen is as follows:
Bromocriptine: starting dose 0·625–1.25 mg daily with an 1·25-mg increase per week, medium dose 2–3 times 2·5 mg daily up to 3 × 10 mg day−1.
Cabergoline: starting dose 0·25–0·5 mg once weekly with a monthly increase by 0·5 mg, medium dose 0·5 mg 2 times per week up to 3·5 mg week−1.
Quinagolide: starting dose 25 μg day−1, with a 25-μg increase every 3 days up to 75 μg, medium dose 75–150 μg daily up to 600 μg.
All dopamine agonists are effective in normalizing prolactin levels, restore reproductive function in men and women and decrease tumour size [33,34] in the majority of patients. Normalization of prolactin levels is also associated with increased bone density [35,36]. Cabergoline appears to be the most effective compound with relatively fewer side effects compared to bromocriptine . Failure to normalize prolactin levels or reduce tumour size by 50% has been suggested as criteria defining dopamine agonist resistance . The mechanisms involved in dopamine agonist resistance are not completely understood. Alterations in dopamine receptors on tumour cells concerning receptor number and ligand affinity have been demonstrated and could explain an escape phenomenon under therapy. Primary resistance may be explained by dopamine receptor variants with differential affinity to dopamine agonists . In patients taking bromocriptine, resistance can be observed in 25–50% of cases, and in patients taking cabergoline, 5–15%. Cabergoline has been shown to be effective in patients who are resistant to and did not respond to other dopamine agonists before . Thus, in patients who are resistant to one medical regimen, a change to another dopamine agonist may lead to improvement in prolactin levels , followed by increasing the dose of the dopamine agonist to the maximum dosage tolerated. In rare cases, surgery and irradiation remain as therapeutic options. Experimental agents such as interferon-alpha or somatostatin have not been clearly demonstrated to be efficient. In the rare case of a malignant prolactinoma, temozolomide appears to be the most effective chemotherapeutic agent [reviewed in 41].
Apart from normalizing gonadal function in macroadenomas, reduction in tumour size is an important goal. Reduction in tumour size can frequently be observed within weeks after starting therapy, which results in visual field improvement in patients with compromised vision. Dopamine agonists are therefore used as primary therapy even in patients with large macroadenomas and visual field defects before or instead of neurosurgery [33,42]. Impaired function of the anterior pituitary may also improve with tumour size reduction . If levels of gonadal steroids remain low after maximum treatment with dopamine agonists, replacement therapy in men and premenopausal women may be indicated. In some cases of large macroadenomas, gonadotroph function will remain defective even after prolactin normalization because of cell destruction. Most of the aforementioned data have been obtained with bromocriptine and cabergoline. Comparable data exist for quinagolide, which appears as effective as cabergoline (Table 4). Relatively sparse data exist for pergolide, lisuride and terguride, as well as metergoline, which are therefore not currently used for the treatment for hyperprolactinaemia .
Side effects and safety
Side effects of dopamine agonist treatment are common and consist of frequent orthostatic, gastrointestinal and neurological symptoms. Orthostatic symptoms are dizziness and syncopes, and the most common gastrointestinal side effects are nausea and vomiting in up to 30% of patients. Some patients may experience cold-induced vasospasms, especially when taking high doses of bromocriptine . Neurological symptoms include headache and drowsiness. Psychotic reactions have been observed but are rare and reversible on withdrawal of the dopamine agonist . To minimize side effects, therapy should be initiated at low doses and increased stepwise in weekly or monthly intervals. Taking the medication with some food and before bedtime is helpful in improving tolerability.
Rare side effects include CSF rhinorrhea or intratumoural haemorrhage following rapid decrease in tumour size. Pleural effusions and thickening, pulmonary infiltrations and fibrosis and retroperitoneal fibrosis have been described . Major concerns about the safety of dopamine agonists have been raised by studies in patients with Parkinson’s disease, which showed an increased frequency of cardiac-valve abnormalities associated with pergolide and cabergoline [46,47]. As responsible pathomechanism, the stimulation of 5-hydroxytryptamine receptor 2B by ergot-derived dopamine agonists and subsequent fibromyoblast proliferation may be involved . Two studies in patients with prolactinoma reported an increased prevalence of mild regurgitation of the tricuspid valve, which was not clinically relevant [49,50]. A recent meta-analysis analysed the data from six studies with a total of 393 patients including both female and male patients, pre- and postmenopausal patients but no children . This meta-analysis showed a significant increased risk of mild plus moderate tricuspid valve regurgitation. Limiting factors in these studies include small sample size, duration of studies and study design, and a larger prospective study was postulated to answer the question of clinical relevance of these findings . Up to date, there are no studies showing an association of the non-ergot derivative quinagolide with cardiac abnormalities, and therefore, quinagolide may be considered as an alternative option to ergot derivatives especially in this respect.
Taken together, even if the risk of cardiac valvulopathy appears low, it seems advisable to discuss these concerns with a patient on an individual basis as there is no consensus statement based on the available data. Treatment should be carried out with the lowest dose and shortest duration possible. An echocardiograph can be offered to patients taking ergot-derived dopamine agonists and should be performed in patients with pre-existing cardiovascular disease. Follow-up echocardiographs are recommended on an individual basis and will be necessary especially in patients with macroprolactinomas taking cabergoline in high doses for an extended period of time.
Follow-up and duration of treatment
Short-term follow-up in patients with hyperprolactinaemia without detectable tumour and microadenomas includes repeated measurement of prolactin levels to adjust the necessary dose of the dopamine agonist. Initially, monthly intervals are adequate; once prolactin levels are normalized, we measure prolactin in 6–12 monthly intervals. The dose of the dopamine agonist should be stepwise decreased as long as prolactin levels stay within the normal range. Tumour size should be re-evaluated regularly (e.g., after 1 year) or if prolactin levels rise significantly . The same scheme can be followed in patients with uncomplicated macroadenomas, but it has been suggested to aim at suppressing prolactin levels because this may increase tumour reduction . In patients with large macroadenomas and neurological or visual defects, the dopamine agonist dose needs to be increased faster. Prolactin levels should therefore be measured more frequently in weekly intervals, visual field defects re-evaluated in intervals of 2–4 weeks and an MRI repeated after 2–3 months or as clinically indicated, as data on the optimal scheme of follow-up are not available .
Optimal duration of therapy and the question if and when therapy can be discontinued are under intensive discussion . Responsible pathomechanisms for persisting remission may be tumour necrosis and fibrosis . Persisting normalization of prolactin has been described for bromocriptine . In this retrospective study, 21% of patients showed sustained normal prolactin levels after a median duration of therapy of 47 months. Colao et al.  described persisting normoprolactinaemia after treatment with cabergoline in 70% of patients with microprolactinoma and 64% of patients with macroprolactinoma, but later a reduced success rate of approximately 50% was reported in a larger and extended study . The following criteria for withdrawal of treatment were applied: normal prolactin level, no visible tumour or tumour reduction of >50%, distance of >5 mm to the optic chiasm, no intrasellar invasion. In a study, Kharlip and coworkers modified criteria with tumour reduction to less than 10 mm for macroadenomas or significant size reduction for microadenomas and duration of therapy for 2 years. In this study, an overall recurrence rate of 54% was reported . For quinagolide, there are no studies showing successful withdrawal of therapy. A meta-analysis of 19 studies including 743 patients (mostly women in the reproductive age) reported persisting normoprolactinaemia in 21% of patients. Patients with a minimum duration of therapy of 2 years and no visible tumour had the highest likelihood of persisting normoprolactinaemia . Withdrawal of treatment thus appears justified in a subgroup of patients if inclusion and exclusion are met (Table 5); however, overall recurrence rates are relatively high, and regular follow-up is necessary.
|Inclusion criteria||Exclusion criteria|
|Duration of therapy >2 years||Duration of therapy <2 years|
|Normalized prolactin||Persisting elevation of prolactin|
|Significant decrease in tumour size:|
Disappearance of tumour or
>50% reduction or
Decrease in macroadenoma to <10 mm
|Increase in tumour size|
|Follow-up warranted||Follow-up not warranted|
Given the high efficacy of dopamine agonists, trans-sphenoidal surgery is rarely the treatment of first choice in hyperprolactinaemia. Absolute indications are rapidly increasing tumour size with tumour mass related–neurological symptoms despite medical treatment, pituitary infarction with haemorrhage and loss of consciousness , sudden loss or impairment of vision with headache, cystic prolactinoma and cerebrospinal fluid leak under dopamine agonist treatment . Relative indications include intolerance of or contraindications against dopamine agonists, persisting chiasmal compression despite medical treatment, increasing tumour size despite medical treatment. Surgical cure rates are highly dependent on the experience of the neurosurgeon and reach 75–90% in microadenomas  but are much variable for macroadenomas ranging from 18 to 80% [44,64]. The main complication to be considered is the induction of disturbance of other pituitary functions. The overall mortality is very low with less than 0.5% .
Radiation therapy has a limited role in the treatment for hyperprolactinaemia and will be considered after unsuccessful medical and surgical treatment. Side effects are frequent including hypopituitarism, neurological dysfunction and less frequently optic, cranial nerve damage and secondary malignancies, and normalization of prolactin levels is infrequent with a prolonged onset of prolactin suppression . It is unclear whether stereotactic radiation is superior to conventional radiotherapy concerning reduction in side effects and faster effects on tumour growth and prolactin secretion.
Age-dependent therapeutic aspects
Children and adolescents. Prolactinomas are the most frequent pituitary tumour in children and adolescents. As in adults, dopamine agonists are treatment of choice . Doses are comparable to adults as well as side effects (Table 4). For bromocriptine, the effectiveness appears to be slightly lower in children, possibly related to the relatively higher frequency of macroadenomas . Cabergoline has shown to be effective in paediatric patients . Data for quinagolide are limited  owing to low patient numbers. As outlined earlier, men more frequently present with macroadenomas and more aggressive tumour growth, making therapy more difficult . Surgery is a rarely used option with the trans-sphenoidal access except for very young patients where the sphenoidal sinus is not yet pneumatized.
Reproductive age. After starting dopamine agonist therapy, ovulation and fertility may be restored immediately in women in the reproductive age. Therefore, adequate contraception needs to be advised when beginning medication. Reappearance of menses can serve as a clinical guide towards effectiveness of therapy. In patients seeking fertility, normalization of prolactin levels is necessary, and ideally, tumour size is reduced to <10 mm before conception. As discussed earlier, trans-sphenoidal surgery is an option in patients with microadenoma and dopamine agonist intolerance or resistance . In men, treatment with dopamine agonists will lead to improvement in semen quality in addition to restoration of gonadal function and decrease in tumour size .
Pregnancy. In a patient with microprolactinoma, pregnancy usually is without problems and dopamine agonist therapy can be discontinued. Increasing oestrogen levels during pregnancy will be accompanied by increasing prolactin levels, but significant tumour enlargement is rare (in approximately 3% of patients only) . Physiologically, prolactin levels and pituitary size will increase significantly during pregnancy, and serum prolactin levels will not reliably indicate an increase in size of the prolactinoma. Routine measurements of prolactin levels are therefore not necessary according to the guidelines, and MRI is not indicated as the decision for treatment will be based on clinical findings . In patients with macroprolactinomas, a significant increase in tumour size has been reported in 30% of patients , and regular testing of visual fields is recommended every trimester or dependent on clinical instability. If required, a MRI scan should be performed, but without contrast medium. Dopamine agonists are in general not approved for treatment in pregnancy, and the respective product information will contain graded warnings considering discontinuation of the drug in pregnancy, which may differ in individual countries. Reinstitution of treatment nevertheless may be necessary if visual field defects or other neurological symptoms owing to tumour growth develop, based on an individual decision. Available data from more than 2500 pregnancies are compatible with a safe application of bromocriptine . Data on cabergoline are more limited but quite reassuring with 380 uncomplicated pregnancies . Post-partum, a dopamine agonist should not be given to allow lactation, if the patient wishes to breastfeed. Pregnancy may be associated with remission of hyperprolactinaemia post-partum, so prolonged discontinuation of prepregnancy dopamine agonist therapy and regular follow-up can be considered [30,70].
Postmenopause. Data on the treatment for hyperprolactinaemia in postmenopausal women are limited and can be obtained from subgroup analysis of interventional studies . Microadenoma usually requires no treatment, and for macroadenomas, the above-mentioned criteria apply . Menopause has been suggested to be associated with remission of hyperprolactinaemia , so treatment withdrawal is justified and regular follow-up measurements of prolactin levels should be performed.
Hyperprolactinaemia and prolactinomas are the most common neuroendocrine disorder in clinical practice. Clinical presentation and consequences of hyperprolactinaemia and prolactinoma are different in the specific stages of reproductive life and require an adequate lifetime-dependent diagnostic and therapeutic approach. Treatment of first choice is dopamine agonists over the whole lifespan. Open questions in the treatment for hyperprolactinaemia include optimal choice and duration of pharmacological treatment. In addition, concerns have been raised on the safety of dopamine agonists since a reported association of valvular heart disease with dopaminergic treatment in patients with Parkinson’s disease. Long-term prospective studies are required to address these questions. A guideline published by the Pituitary Society offers guidance for the diagnosis and treatment for hyperprolactinaemia and prolactinomas, and this review in general is concordant with this guideline.
Sources of funding
Endokrinologikum Frankfurt, Academic Teaching Unit of the Goethe University Frankfurt, Frankfurt am Main, Germany (W.A. Mann).