Headache and pituitary disease: a systematic review

Authors

  • I. Kreitschmann-Andermahr,

    1. Department of Neurosurgery, University of Essen, Essen, Germany
    2. Department of Neurosurgery, University of Erlangen, Erlangen, Germany
    Search for more papers by this author
  • S. Siegel,

    1. Department of Neurosurgery, University of Erlangen, Erlangen, Germany
    Search for more papers by this author
  • R. Weber Carneiro,

    1. Department of Neurosurgery, University of Erlangen, Erlangen, Germany
    Search for more papers by this author
  • J. M. Maubach,

    1. Experimental and Clinical Endocrinology, Med Clinic I, University of Lübeck, Lübeck, Germany
    Search for more papers by this author
  • B. Harbeck,

    1. Experimental and Clinical Endocrinology, Med Clinic I, University of Lübeck, Lübeck, Germany
    Search for more papers by this author
  • G. Brabant

    Corresponding author
    1. Experimental and Clinical Endocrinology, Med Clinic I, University of Lübeck, Lübeck, Germany
    • Correspondence: Georg Brabant, Experimental and Clinical Endocrinology, Med Clinic I - University of Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany and Endocrinology and Diabetes Group, Faculty of Medical and Human Sciences, University of Manchester, Manchester, UK. Tel.: 0049 451 500-6131; Fax: 0049 451 500-4807; E-mail: georg.brabant@uksh.de

    Search for more papers by this author

Summary

Headache is very common in pituitary disease and is reported to be present in more than a third of all patients with pituitary adenomas. Tumour size, cavernous sinus invasion, traction or displacement of intracranial pain-sensitive structures such as blood vessels, cranial nerves and dura mater, and hormonal hypersecretion are implicated causes. The present review attempts to systematically review the literature for any combination of headache and pituitary or hormone overproduction or deficiency. Most data available are retrospective and/or not based on the International Headache Society (IHS) classification. Whereas in pituitary apoplexy a mechanical component explains the almost universal association of the condition with headaches, this correlation is less clear in other forms of pituitary disease and a positive impact of surgery on headaches is not guaranteed. Similarly, invasion into the cavernous sinus or local inflammatory changes have been linked to headaches without convincing evidence. Some studies suggest that oversecretion of GH and prolactin may be important for the development of headaches, and treatment, particularly with somatostatin analogues, has been shown to improve symptoms in these patients. Otherwise, treatment rests on general treatment options for headaches based on an accurate clinical history and a precise classification which includes assessment of the patient's psychosocial risk factors.

Introduction

Headache is a very common disorder. Worldwide, the percentages of the adult population with an active headache disorder are 46% for headache in general, 11% for migraine (MIG), 42% for tension-type headache (TTH) and 3% for chronic daily headache.[1] Headaches are frequently linked to pituitary disease, and it is generally accepted that headache is a common presenting feature of pituitary adenomas and other sellar pathologies. However, any predisposition of a patient for primary headache will also contribute significantly to the development of headache in conjunction with pituitary disease. In the following systematic review, we address the literature on headache and pituitary disorders with regard to prevalence, aetiology, pathophysiological mechanisms and treatment. To this end, we searched PubMed for any combination of the terms ‘pituitary’ or ‘hypophysitis’, or ‘hypopituitarism’ or ‘Rathke's cleft cyst’ or ‘cavernous sinus’ or ‘craniopharyngioma’ or ‘thyroid hormones’ or ‘cortisol’ or ‘testosterone’ or ‘oestrogen’ or ‘growth hormone’ or ‘prolactin’ and headache’ or ‘migraine’ or ‘trigeminal pain’. All articles reporting original data and review articles on headache and pituitary disease in peer reviewed journals were screened for inclusion into the present review. We also searched the reference lists of articles identified by this search strategy. Finally, we screened for published case reports on pituitary adenoma, pituitary apoplexy, Rathke's cleft cyst, craniopharyngioma and hypophysitis. During the interval between 2007 and 2012, we identified 48 case reports which, however, lead to extremely heterogeneous results which could not sensibly be included into the meta-analytic approach. We thus omitted single cases from the final analysis. We largely selected articles from 2000 to 2013 but did not exclude commonly referenced and highly regarded older publications.

We apologize for not being able to cite all references but have used a personal selection to provide readers with more details and more references than this review can accommodate. We also provide a brief overview of the general classification of headaches.

Prevalence and clinical characteristics of headache in pituitary disease

Pituitary adenoma

A PubMed search for ‘headache’ and ‘pituitary’ provided more than 1400 hits, but the number of systematic, prospective studies elucidating the prevalence of headache in patients with pituitary disease is very limited, and there is surprisingly only one prospective study using the criteria of the International Headache Society (IHS), the International Classification of Headache Disorders-II (ICHD-II; for details see below).[2] In this study, performed by Levy et al.,[3] 84 patients with pituitary tumour and troublesome headache were analysed prospectively. Chronic (46%) MIG and episodic (30%) MIG were the predominant complaints followed by primary stabbing headache (27%) and, to a much lesser extent, by short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) (5%), cluster headache (4%) and hemicrania continua (1%). Six headache presentations (7%) could not be classified according to the IHS criteria. Abe et al. reported on 51 patients with pituitary adenomas for headache not using standard criteria. A total of 19 (37%) mostly female patients reported headache, characterized predominantly as generalized and bifrontal headache or ‘head heaviness’.[4] In another prospective study by Levy and colleagues,[5] 63 patients initially seen with pituitary tumours were evaluated for headache which was present in 70%, whereas Pereira-Neto et al. found moderate to severe headache, graded by means of the Headache Impact Test Scale (HIT-6) in 68% (17/25 patients) consecutive patients undergoing surgery for pituitary adenomas.[6] Gondim and colleagues assessed the presence of headache and headache characteristics (by means of a self-made questionnaire) in 64 prospective patients with a primary pituitary mass.[7] Headache was a major complaint in 44 (69%) patients in the study group and was localized mostly in the frontal, followed by the orbital and retro-orbital regions. Its quality was described as throbbing, followed by dull pressure, sharp or burning.[7]

The frequency of headaches obtained partially from large institutions specializing in the treatment of headache differs substantially from retrospective data from endocrinological institutions. A study of 517 patients primarily investigated for initial symptoms leading to the diagnosis of a pituitary adenoma suggests a frequency of headaches between 11 and 15%.[8] When intrasellar tumours not originating from the pituitary itself are included,[9] 34% of these patients reported headache as an initial symptom, which resolved in 63% following surgery. Again, none of these studies used a systematic evaluation of headaches according to the IHS criteria. A recent report, published in 2013[10] found tumour-attributed headache in 24 of 58 patients admitted for surgery for a pituitary adenoma. Seven patients had migrainelike headache, 11 tension-type headache and three had both. Cluster headache-like headache was found once, and two headaches could not be classified.

Secondary SUNCT or short-lasting unilateral neuralgiform headache attacks with cranial autonomic features (SUNA) associated with pituitary micro- or macroadenoma are rare types of headache and are usually described in case reports or small case series. These headaches belong to the group of trigeminal autonomic cephalalgias (TAC) because of the combination of head pain and activation of cranial autonomic efferent fibres (for an overview see[11]). Chitsantikul et al. identified six patients with such forms of headache (5 SUNCT, 1 SUNA).[12] Interestingly, all five SUNCT patients had pituitary tumours ipsilateral to the side of the headache, but tumour removal provided major pain relief only in three patients.

Thus, a large proportion of patients with pituitary adenomas (37·5–70%) suffer from headache, primarily presenting as MIG or TTH. The much less frequently reported TAC, may, however, constitute a true secondary headache type associated with pituitary adenomas.

Pituitary apoplexy

Headache associated with pituitary apoplexy, a rare sudden haemorrhagic infarction of a pre-existing pituitary adenoma, is described as quite specific. It is acknowledged by a separate headache category within the IHS. Several retrospective series[13-15] and numerous case reports have shown that pituitary apoplexy almost always presents with sudden-onset headache (92–100%). It is frequently associated with meningism and mimics the clinical presentation of subarachnoid haemorrhage or meningitis.[16, 17] Other frequent symptoms and signs of pituitary apoplexy include nausea and vomiting, visual field disturbances, oculomotor paresis, an altered state of consciousness and meningism.[13-15] A similar clinical headache presentation has been described in patients with haemorrhage into a pre-existing Rathke's cleft cyst.[18]

Other pituitary diseases

In a series of 116 consecutive patients with nonadenomatous sellar lesions operated upon by one neurosurgeon, headache, affecting 34%, was the second most common clinical complaint after visual field disturbances.[9] Unfortunately, the quality of headache was not reported in this series. Numerous case reports mention a close association between lymphocytic hypophysitis and headache which, according to one review article, is supposed to precede visual field disturbances or oculomotor palsies.[19] In another report, 60% of patients with lymphocytic hypophysitis suffered from headache which commonly presented before pituitary hormone changes could be characterized.[20, 21] Other inflammatory conditions, which are even more rarely diagnosed such as granulomatous hypophysitis (GRH), xanthogranulomatous hypophysitis (XGH), xanthomatous hypophysitis (XH), and necrotizing hypophysitis,[22] are reported to be similarly associated with headache but are almost exclusively investigated in case reports. Only in granulomatous hypophysitis, the most frequent of these conditions, was a series of 38 patients reported. Headache was the most common symptom present in 61% of cases. Additional clinical symptoms related to sellar compression, anterior hypopituitarism, diabetes insipidus and hyperprolactinaemia.[23, 24] Further differential diagnoses include infections due to tuberculosis, syphilis or fungal causes, systemic inflammatory conditions such as sarcoidosis, Wegener's granulomatosis, Takayasu's arteritis, Crohn's disease, histiocytosis X and foreign body reactions such as a ruptured Rathke's cyst or mucocele. All are described in case reports to be associated with headaches at an early stage or as the presenting symptom, but there are no large series to evaluate any underlying pathological mechanism.

In contrast to these rare entities, larger series exist in patients with craniopharyngioma. Headache and visual field defects seem to be the most common presenting clinical features, in 64% and 55%, respectively, of 121 childhood and adult patients.[25] This was confirmed by a large series from the Danish national cohort with 189 children and adults suffering from craniopharyngioma. Headache was the leading symptom, being present in more than 70% of children and almost 60% of adults.[26] Again, as in other nonadenomatous pituitary pathologies, the quality of the headache has not been systematically assessed.

International headache classification and its relevance for pituitary disease

The International Classification of Headache Disorders-II (ICHD-II) of the International Headache Society (IHS) is the most frequently used system for the classification of headaches.[2] It differentiates between primary headaches (such as TTH, MIG, cluster headache and other trigeminal neuralgias), secondary headaches and cranial neuralgias, facial pain and other headaches and contains numerous subgroups. The aetiology of many primary headache syndromes is not conclusively elucidated. Pathophysiological explanations include genetic, neurobiological and psychosocial mechanisms.[27-29] When headache occurs de novo in temporal relation with another disorder thought to be capable of causing it, such as traumatic brain injury, acute cerebrovascular disease or a brain tumour, it is classified as secondary, even if the headache phenomenology corresponds to MIG, THH or cluster headache.[2] An example of a secondary headache related to pituitary disease is ‘headache attributed to pituitary apoplexy’. The third large group of headaches according to the ICHD-II consists of pain syndromes mediated by afferent fibres (e.g. trigeminal, glossopharyngeal and vagus nerves and the nervus intermedius) as well as headache syndromes not fitting any of the other existing categories.[2]

A clear differentiation between primary and secondary headaches is not always possible. Thus, the ICHD-II introduced a new way of diagnosing and coding primary headaches which worsen significantly in close temporal relation to another disorder known to cause headache. Many patients with an association of pituitary disease and headache fit to this classification (for an overview see[30]). TAC, a rare subgroup of headache disorders presenting with lateralized pain accompanied by cranial autonomic features, constitute another example of headaches associated with pituitary disease (for an overview see[11, 30]) and can be classified as primary or secondary headache or both. Cluster headache, paroxysmal hemicrania, SUNCT/SUNA and hemicrania continua belong into this category. As the key feature of TACs is the lateralization of symptoms and signs, it has been suggested that an underlying pituitary region pathology is considered and that appropriate imaging diagnostics are initiated.[11] Table 1 summarizes this classification of headaches and lists the current pharmacological therapeutic approaches when an underlying disease is not classified.

Table 1. Overview of headache types and pharmacologic therapies
Headache typeClinical presentationPharmacological therapy
  1. Please note that there may be country-specific pharmacological treatment preferences.

Part I: Primary headache types
migraine (MIG)

Mostly but not always unilateral

Highest incidence between 35 and 45 y

Most frequent subgroups are migraine without aura and migraine with aura

Often positive family history of migraine

Anti-emetics and NSAIDs are first-line therapy for migraine attacks

Triptans are used as migraine-specific acute treatment

Pharmacological migraine prophylaxis is indicated in case of 3 or more attacks/month

First-line preparations include propranolol, metoprolol, flunarizine, valproate and topiramate

Tension-type headache (TTH)

Usually bilateral

Differentiation between episodic (<15 days/month) and chronic (>15 days/month for 3 consecutive months) TTH.

Often positive family history of headache in chronic TTH

TTH responds to nonopioid analgesics such as NSAIDs, paracetamol and metamizol

Caffeine may enhance the effect.

Too frequent use (> 9 days/month) can lead to medication-overuse headache.

Peppermint oil applied topically to the temples may relieve TTH.

Antidepressants (i.e. amitryptiline) and anticonvulsants (i.e. valproate, topiramate) are effective for pharmacologic prophylaxis of chronic TTH

Cluster headache and other trigeminal autonomic cephalalgias (TAC)

  Cluster headache

  Paroxysmal hemicrania

  Short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT)

  Probable trigeminal autonomic cephalalgia

Male preponderance

Strong association with smoking

Always unilateral

Begin usually around the eye or temples

Quick onset of pain, exacerbates within minutes

Especially in SUNCT tearing and redness of the eye on same side as headache, although these and other autonomic symptoms (ptosis, miosis, nasal congestion, ipsilateral sweating of the brow or face) can also occur in any form of TACs

Alcohol may precipitate attacks

Inhalation of oxygen relieves pain in acute cluster headache attacks

Some triptans (sumatriptan) may also be effective

Intranasal application of lidocaine with reclined and rotated head may reach the sphenopalatine ganglion and thus relieve pain and autonomic symptoms

Choice of pharmacoprophylaxis depends on frequency and course of attacks and includes verapamil, steroids and lithium

Other primary headaches

This group includes primary stabbing headache, primary cough headache and other entities

For a complete overview, please refer to the IHS classification.

 
Part II: Secondary headache types
For a complete overview of secondary headache types, see the IHS classification. Listed below are the secondary headache types in association with pituitary disease.

Secondary headaches are de novo headaches occurring in close temporal conjunction with another disorder (i.e. brain tumour, head and/or neck trauma, infection) that is per se capable of causing headache

Phenomenologically they can mimic any type of primary headache

Pre-existing headache worsened by a disorder known to cause headache by itself receive two diagnoses: the primary and the secondary headache type diagnosis

Pharmacological treatment depends on aetiology and clinical presentation of headache

Headache attributed to cranial or cervical vascular disorder

  headache attributed to pituitary apoplexy

Severe acute headache

Developing simultaneously with acute haemorrhagic pituitary infarction

May be accompanied by meningism

Treatment is that of underlying disorder

Headache attributed to nonvascular intracranial disorder

Headache attributed to intracranial neoplasm

Headache attributed to increased intracranial pressure or hydrocephalus caused by neoplasm Headache attributed directly to neoplasm

Headache attributed to carcinomatous meningitis

Headache attributed to hypothalamic or pituitary hyper- or hyposecretion

Please refer to IHS classification for complete criteria of subgroups

Note that ‘headache attributed to hypothalamic or pituitary hyper- or hyposecretion’ is defined as associated with microadenomas only

Treatment is that of underlying disorder plus symptomatic analgesic treatment
Part III: Cranial neuralgias central and primary facial pain and other headaches

For a complete overview of secondary headache types, see the IHS classification.

Of relevance in conjunction with pituitary disease may be trigeminal neuralgia

Includes pain syndrome mediated by afferent fibres in the trigeminal nerve, nervus intermedius, glossopharyngeal and vagus nerves as well as the upper cervical roots

It also includes a subclassification of headaches that do not fit into any of the other chapters

 

Pathophysiology of headache in pituitary disease

The pathophysiology of headache in patients with pituitary adenomas remains obscure. Putative explanations of headache evolution include mechanical, biochemical, vascular and biopsychosocial mechanisms. These are described below in more detail. Additionally, Table 2 provides an overview of the systematic studies on headache and pituitary disease and lists the major findings.

Table 2. Overview of systematic studies on headache and pituitary disease
StudyNPrevalenceTumour sizeCavernous sinus invasionHormonal secretionaIntrasellar pressureOther factorsRemark
  1. a

    Hormonal secretion was measured in various ways throughout the different studies; variables used to measure hormonal secretion are specified in the remark section.

  2. +, Positive relation; 0, no relation; -, negative relation; NA, not assessed; HS, hormonal secretion.

Stoffel-Wagner, [8]51711·3–15·5 %NANANANA 

Retrospective study design

Prevalence is calculated for headache as first symptom of pituitary adenomas

Abé, [4]5137·3 %

0

Descriptive

0

Descriptive

0

Descriptive

NA 

Prevalence is calculated for preoperative headache

HS: hypopituitarism

Arafah, [33]4951 %

0

P = 0·23

NA

0

P = 0·713

+

P = 0·000

 

Prevalence is calculated for > 2 headache episodes/week requiring analgesics for relief

HS: hypopituitarism

Levy, [5]6370 %

0

P = 0·100

0

P > 0·050

+

Descriptive

NA

Family history,

+,

P = 0·004

Prevalence is calculated for preoperative headache

HS: GH- and prolactin-producing tumours

Mercado, [54]16066 %

0

P = 0·117

NANANA Only patients with acromegaly included
Gondim, [7]6468·8 %

+

P = 0·026

+

P = 0·014

0

P = 0·106

NA

Optic chiasm compression, +, P = 0·004

Solid vs. cystic lesion, 0, P = 0·700.

Sellar destruction, +, 0·280

Family history 0, P = 0·384

HS: nonfunctioning vs. secreting tumours

Effect of sella destruction is stated to be significant with P = 0·28, probably typing error

Pereira-Neto, [6]2568 %NANANA

0

P = 0·880

  
Schankin, [10]5841 %

0

P = 0·880

0

P = 0·620

0

P = 0·630

NA

History of headache, + , P = 0·030

Nicotine abuse, + , P < 0·010

Acute headache treatment, + , P = 0·040

Ki67-index > 3%, + , P = 0·020

Prevalence calculated for preoperative, pituitary related headache with postoperative amelioration

HS: secreting vs. nonsecreting tumours

Mechanical aspects

Historically, headaches have been claimed to be provoked by the size of the tumour and invasion into the cavernous sinus where they may lead to traction and/or displacement of intracranial pain-sensitive structures in the blood vessels and involve cranial nerves and the dura mater. Such mechanically driven conceptions of headaches are supported by early studies correlating tumour volume and headache in brain tumours.[31, 32] Detailed investigations focusing on pituitary tumours only showed heterogeneous results. Gondim et al. suggested that larger tumours were significantly more frequently associated with headaches than smaller microadenomas,[7] whereas no association to tumour size or the involvement of cavernous sinuses was found by Abe and coworkers.[4] In a particularly detailed study, Levy and coworkers evaluated the location of the tumour in 63 subjects with respect to headaches and again found no significant association to tumour volume or site.[5] All these data, however, are hampered by the relatively small number of patients making it difficult to draw firm conclusions.

An increase in tumour size may raise the intrasellar pressure (ISP).[6, 7, 33-39] Even though ISP measurements are not available in healthy persons, ‘normal’ ISP was defined in patients with small microadenomas or in patients with empty sella.[38, 40] Pituitary apoplexy within a pre-existing pituitary tumour leading to rapid expansion in the size of the pituitary lesion[41] has been shown to grossly increase ISP in parallel to acute-onset, throbbing headaches.[33] In pituitary adenomas without pituitary apoplexy, data are ambiguous. Measurement of ISP in 25 patients with pituitary adenomas[6] found no correlation between ISP values and tumour area, the number of visual quadrants affected, adverse scores on the generic quality of life questionnaire total RAND SF-36 scale and the more specific HIT-6 scale for headaches. In a large study of 107 patients, 75% of macroadenomas showed significantly raised pressure with highest levels in tumours with parasellar invasion (30 ± 2 vs. 12 ± 3 in ‘normal’; P < 0·01).[38] Another recent series of 49 patients confirms that patients presenting with headaches (n = 25) had significantly higher ISP than controls without headaches (n = 24).[33] There was no significant correlation with tumour size. The authors claimed a positive relation to circulating prolactin levels.

It has been proposed that headaches are due to cystic changes in the pituitary.[9] Supportive for such an important role of cystic changes are observations in patients with a Rathke's cyst where headache represents the leading symptom.[42-45] Successful surgery in this group of patients may substantially improve the symptoms of headache.[46, 47]

The observation that metastases to the pituitary also present with headache may argue for a mechanical mechanism to explain the symptoms in these rapidly expanding tumours. In a series of 36 patients with pituitary metastases, diabetes insipidus, anterior pituitary insufficiency and retro-orbital pain represented the leading initial symptoms.[48] This was confirmed in a recent series of 1469 patients with pituitary tumours. A total of 15 had metastatic disease and eight suffered from headache as a presenting symptom.[49] A review of 220 cases with pituitary metastases showed that 49·8% were derived from a breast cancer and 20·2% from lung cancer.[50] In an older, large series, 50·8% of metastases involved the posterior lobe alone, 33·8% both lobes and 15·4% the anterior lobe alone.[48]

Biochemical aspects

Hormonal hypersecretion has been suggested to be the underlying mechanism for some aspects of pituitary disease-related headaches. Worsening migraine in patients harbouring a functional pituitary adenoma may be related to the endocrine changes associated with this pathology, rather than the effects of a mass lesion. Especially in female patients with prolactinomas, where old series suggest a high frequency of headaches,[4] migraine could be exacerbated by menstrual disturbances related to prolactin hypersecretion.[30]

An association of headaches and acromegaly is frequently discussed. This is based on series of up to 500 patients assessed retrospectively. Up to 60% of patients reported headaches but other studies found much lower rates (13%).[4, 51-55] A causal link to GH was suggested as headaches improved dramatically in individual cases following successful surgery for acromegaly only to recur early with recurrent GH excess.[56] Further support for a direct endocrine role of GH in the pathogenesis of headache derives from data on GH replacement in GH deficiency. In a large cohort of more than 30 000 children treated with GH for GHD, headache was reported with an incidence of 793·5 of 100 000 patient years on initiation of therapy.[57] Furthermore, somatostatin analogues (SSA) can rapidly alleviate headaches in conjunction with acromegaly.[58] As this effect, shown for octreotide, is not solely dependent on biochemical normalization of acromegaly,[59] it has been speculated that SSA inhibit some, as yet undefined, pronociceptive peptide causing headache in patients with acromegaly.[30]

Inflammatory processes within the pituitary region represent another pathophysiological entity. Rapid enlargement of the pituitary due to oedema and lymphocytic infiltration is expected to lead to stretching on the meningeal structures and cause headache.[20] Bearing in mind the unconvincing results discussed for the mechanical aspects leading to headache in macro- versus microadenomas, it seems likely that cytokines or other biochemical factors play an additional or even more prominent role in the pathogenesis of headache in these conditions. However, histological expression of none of the biochemical factors investigated so far such as neuropeptide Y (NPY), calcitonin-gene-related peptide, substance P or vasoactive intestinal polypeptide (VIP) was related to headache in patients with pituitary adenomas.[60-62]

Psychosocial risk factors for headache

The biopsychosocial model is widely accepted in explaining primary headaches. In tumour-related headaches, biological factors are assumed to be pathophysiologically most important and have been investigated most frequently. Nevertheless, they seem not to be sufficient to explain the development and chronification of headaches in pituitary disease, where one study reported a high correlation between family history of headache and headache in patients with pituitary adenoma, but not with tumour size or cavernous sinus invasion.[5] There is an increasing literature on stress-related factors which may trigger acute headache attacks but more importantly represent a risk factor for chronification of headache after tumour diagnosis.[63] Sleep disturbances are another important risk factor for primary headaches,[64, 65][66] and this may be an underinvestigated reason for tumour-related headaches, as sleep quality is frequently impaired in patients with pituitary disease, especially in patients with acromegaly.

Women are more frequently affected by primary headache than men.[67] While the higher frequency of MIG in women is predominantly a combination of genetic predisposition and alterations in female endocrine status, psychological factors such as emotionally focused coping styles and catastrophizing thoughts also enhance the risk of headache development.[68, 69] These coping styles are more prevalent in women[70, 71] which might also contribute to their vulnerability for headache.

Treatment of headache related to pituitary disease

An accurate clinical history and a precise headache diagnosis are important for correct counselling of patients with headaches and pituitary disease and to target their therapy. Chronic versus acute pain as well as symptomatic and causal therapies must be differentiated. Table 1 gives an overview of the most common pharmacological treatment options in the most prevalent types of primary headaches. For many headache types, nonpharmacological interventions such as information, reassurance, identification of trigger factors, physical therapy, relaxation training, acupuncture and psychological interventions are a mainstay of headache treatment and are widely used.[72, 73]

For the pituitary pathology itself, surgery, radiotherapy and medical options are available. Stratification of therapy is primarily targeted to the underlying primary pituitary pathology and not towards the relief of headaches. Differences in the outcome may thus be biased depending on the underlying disease. Radiotherapy may, for example, be selected much more frequently in aggressive recurrent adenomas or in cases of metastatic disease to the pituitary, whereas medical approaches are only used for certain hormone-producing tumours. A series of 84 patients investigated by Levy et al.[3] demonstrated improvement following surgery in 49% but worsening of headache in 15%. In the same study, radiotherapy was evaluated, even though in only 16 patients. Only 1 patient reported improvement of headache symptoms when the patients were followed for a median of 5 years from radiotherapy. In addition, a number of specific medical treatment options besides classical pain medication for headaches have been evaluated but the response needs to be stratified according to the histopathology of the tumour and hormone production. To date, there are no series large enough to better define the role of a given therapeutic approach.

Dopamine agonists (DA)

Dopamine agonists (DA) have been evaluated in the therapy of headaches associated with pituitary adenomas, but a beneficial effect is not convincing. The best results have been described with cabergoline where up to 80% of patients experienced a reduction in headache severity and frequency.[74, 75] These beneficial effects are considerably less frequent with bromocriptine.[3] The evaluation of results is difficult as they are not clearly related to the size of the tumour or to the substantial reduction in tumour size which can be expected with dopamine agonists, especially in prolactinomas. DA are well known to potentially exacerbate headache.[76-78] This may be due to deleterious effects on the trigeminovascular system whose activity may be altered by the ergot alkaloid properties of DA.[79, 80] The knowledge of these adverse properties of DA is important as in some cases a dramatic worsening of symptoms is observed with the use of DA including the SUNCT type of headaches.[76-78]

Somatostatin analogues (SSA)

Somatostatin and its analogues are widely used to shrink pituitary tumours, particularly GH producing adenomas. They are as well known to have analgesic actions on their own through their interference with the opioidergic system.[81, 82] In the largest study so far, by Levy, a total of 22 patients were investigated during treatment with somatostatin or various forms of SSA.[3] This treatment was, however, neither standardized in dose or injection intervals nor in the histotype and size of the tumours. A total of 12 patients improved. Whether preparation or dosing might play a role or whether certain analogues are more potent remains unclear as few patients have been reported so far (e.g. only 4 patients treated with lanreotide of whom 1 improved). On treatment with native somatostatin given in up to 12 injections a day or the long-acting form octreotide LAR, 61% of patients improved, but some of these suffered rebound headache or tachyphylaxis. This phenomenon may be explained by the observation that somatostatin receptors heterodimerize with opiate receptors, which also explains the occasional reports of octreotide dependency in patients with acromegaly and the occurrence of rebound headaches.[58, 59, 83]

Conclusions

Pituitary pathologies are linked to different types of headache, most commonly the primary headache types MIG, TTH and unclassified headaches. Currently, the number of studies which carefully follow-up patients with pituitary disease and headaches according to the IHS classification or other standardized diagnostic criteria is small.

A well-accepted pathophysiological cause of headaches in pituitary diseases is a rapid increase in tumour size such as in pituitary apoplexy or metastases to the pituitary. Data on tumour expansion to explain headaches mechanically in slowly growing pituitary tumours are much less convincing and causal relations between increased intrasellar pressure and headache have not clearly been substantiated. This fits with observations in many patients with large and invasive pituitary pathologies which are indeed pain free. A potential role of hormone hypersecretion has been discussed as an underlying mechanism, predominantly based on the positive relationship observed in some rather aggressive, rapidly growing GH producing tumours in children and adolescents. No systematic data on the effects of anterior or posterior hypopituitarism on headache have been published so far.

Currently, we lack large prospective studies on the effects of any standardized endocrinological or neurosurgical therapy in the treatment of headaches in patients with pituitary pathology. Studies on the effects of SSA or DA on headache rest on small series and were never systematically evaluated. Unfortunately, firm data on the effects of surgery on headache are also missing. Surgical reduction clearly improves symptoms only in rapidly expanding lesions such as metastases while the effects of surgery on headaches in slowly growing tumours are much less striking. Importantly, any approach on the assessment of headaches in pituitary disease needs to include an accurate clinical subtyping following the IHS or other headache guidelines to correctly stratify patients for other (frequent) causes of headaches and to help in counselling them on pharmacological and nonpharmacological treatments.

Ancillary