There are relatively few studies on headache associated with brain tumors despite the long-recognized association. The prevalence of headache reported in modern studies containing unselected types of brain tumors ranges from 48 to 72 % (Table 116-1). In one study (39) of patients with accidentally detected brain tumors, Kamiguchi et al. found that of 113 patients, 53 had complained of headache not typical of increased intracranial pressure.

Headaches are reported with all types of tumors but some seem more likely to cause headache (Table 116-2).

Isolated headache as the only symptom of a brain tumor is uncommon and usually short-lived. In a prospective study, only 8% of 183 adult patients presented with isolated headache and no patient had headache as their only symptom for longer than 10 weeks (88). Rossi and Vassella (68) studied 67 children with headache in whom a brain tumor was diagnosed and found that progressive neurologic signs and symptoms occurred in 70% within 1 month of the onset of headache and in 94% within 4 months. Less than 1% of 3291 children with brain tumors had headache as their only symptom and less than 3% had no neurologic abnormality on examination (12). Intracranial metastases were found in 32.4% of adult cancer patients presenting with new or changed headache (13). In contrast, only 12% of headaches in children with systemic cancer were caused by brain metastases (3).

The presence of headache is age dependent. In a cohort study of 714 patients, headache was the presenting symptom in 44% of patients 18 to 24 years of age and only in 8% of patients older than 75 years (49). Sixty-two percent of all children with brain tumors had chronic or frequent headaches, but at least 72% of children aged 4 to 20 had headache while only 8% of those under 1 year had headache (12).

The etiology of headache in brain tumors is multifactorial. In a classic series of experiments on patients undergoing craniotomy, Ray and Wolff mapped the pain-sensitive structures of the head (66). The venous sinuses, dural arteries, some of the dura at the base of the brain, and the cerebral arteries at the base of the brain were pain sensitive. The brain parenchyma, lining of the ventricles, and most of the dura and pia-arachnoid were insensitive to pain.

These observations (66) led the authors to postulate six mechanisms of headache pain: (1) traction on the veins draining into the large venous sinuses with resulting displacement, (2) traction on the middle meningeal artery, (3) traction on the major arteries at the base of the brain, (4) direct pressure on cranial nerves with pain
afferent fibers from the head, (5) distension and dilation of the intracranial and extracranial arteries, and (6) inflammation in or around the pain-sensitive structures of the head. Further extension of this work in 67 patients with brain tumors led to the conclusion that local and distant traction on pain-sensitive structures, mass effect, and hydrocephalus account for most of the headaches (46).

The role of increased intracranial pressure in the generation of headache is debated. In a study of migraineurs (76) it was found that headache induced by intravenous histamine could be abolished by increasing the cerebrospinal fluid (CSF) pressure. Other studies have found that increasing intracranial pressure using saline infusions caused headache (81). Northfield proposed that sudden variation in intracranial pressure caused headache rather than the absolute value of the pressure (56). For example, plateau waves are acute elevations in intracranial pressure due to a rapid increase in intracerebral blood volume in a patient with reduced intracranial compliance (e.g., from a tumor) and impaired autoregulation of cerebral perfusion pressure. Symptoms include headache associated with dizziness and alterations in consciousness and motor control that are triggered by standing (92).

Pituitary adenomas have been postulated to cause headache by raising intrasellar pressure (4) and by the expression of somatostatin receptors coupled to the pain pathways in growth hormone-secreting tumors (47).

The IHS 2004 (36) clinical features of headache attributed to intracranial neoplasms, code 7.4, are listed as follows, with this chapter having emphasis on the first three subcodes:

Diagnostic criteria are as follows:

A. Diffuse nonpulsating headache with at least one of the following characteristics and fulfilling criteria C and D:

1. Associated with nausea and/or vomiting

2. Worsened by physical activity and/or maneuvers known to increase intracranial pressure (such as Valsalva maneuver, coughing, or sneezing)

3. Occurring in attacklike episodes (see Notes, #1)

B. Space-occupying intracranial tumor demonstrated by computed tomography (CT) or magnetic resonance imaging (MRI) and causing hydrocephalus (see Notes, #2).

C. Headache develops and/or deteriorates in close temporal relation to the hydrocephalus.

D. Headache improves within 7 days after surgical removal or volume-reduction of tumor.

Notes:

1. Onset of headache can be sudden (thunderclap headache) and, in such cases, associated with loss of consciousness.

2. For example, colloid cyst of the III ventricle.

Diagnostic criteria are follows:

A. Headache with at least one of the following characteristics and fulfilling criteria C and D:

1. Progressive

2. Localized

3. Worse in the morning

4. Aggravated by coughing or bending forward

B. Intracranial neoplasm shown by imaging.

C. Headache develops in temporal (and usually spatial) relation to the neoplasm.

D. Headache resolves within 7 days after surgical removal or volume-reduction of neoplasm or treatment with corticosteroids.

Diagnostic criteria are as follows:

A. Diffuse or localized headache fulfilling criterion C.

B. Carcinomatous meningitis proven by (repeated) CSF examination and/or dural enhancement on MRI.

C. Headache develops and/or deteriorates with advancing disease.

Comment: Headache may improve temporarily with intrathecal chemotherapy or prednisone (prednisolone).

Diagnostic criteria are as follows:

A. Bilateral, frontotemporal, and/or retro-orbital headache fulfilling criteria C and D.

B. At least one of the following:

1. Prolactin, growth hormone (GH), and adrenocorticotropic hormone (ACTH) hypersecretion associated with microadenomas <10 mm in diameter

2. Disorder of temperature regulation, abnormal emotional state, altered thirst and appetite, and change in level of consciousness associated with hypothalamic tumor

C. Headache develops during endocrine abnormality.

D. Headache resolves within 3 months after surgical resection or specific and effective medical therapy.

The classical description of tumor headache is severe, progressive, morning headache associated with nausea and/or vomiting. While this headache pattern is always worrisome and deserves immediate investigation, it is relatively infrequent (22). Indeed, studies have shown tension-type-like dull ache in 58% or pressure headache in 77% of patients (22,69). Nausea and vomiting were present in 46% (69) and 48% (22) of patients. Another study demonstrated that only 16% of patients had tension-type headache and in the other patients, the pain was throbbing (63%) or shooting (38%) and usually associated with nausea and vomiting (60%) (61). A minority of patients (7 to 9%) have migraine-type headaches and usually there are atypical features (22,61). The pain is typically intermittent, not occurring daily in 62 to 88% of patients (22,46,61,69), moderate to severe in intensity, and progressive (61). Nocturnal and morning headaches are reported in 25 to 36% of patients and exacerbation of headache by the Valsalva maneuvers occurs in 23% (22,61,69).

In children, the most common features (79% of patients) in one study were nocturnal headache or headache present on arising associated with nausea and vomiting (68). Nausea and vomiting were associated with headache in 72% of children with supratentorial tumors and 86% with infratentorial tumors (12).

The site of headache may not reflect the tumor location. In one series, headache lateralization and tumor coincided in only one third of patients (61). Only 41% of patients with hemispheric tumors had isolated ipsilateral headache and 12% had only contralateral headaches. These results contrast with another study where all patients with unilateral headache had an ipsilateral tumor (22). Correct lateralization of headache in 80% of supratentorial tumors and 62% of infratentorial tumors (84) was present in another study, and it was noted that in patients without papilledema or other signs of increased intracranial pressure, correct lateralization was found in 100% of supratentorial tumors (84). Frontal headaches were poorly localizing and the frontal area was the most common site of headache (49 to 68%) (22,61). Bilateral headaches were present in 18 to 72% of patients (21,58). Infratentorial tumors usually present with supratentorial pain (73%), with nuchal and occipital pain in only 27% of patients (61). Headaches are more common with infratentorial (70 to 84%) than supratentorial tumors (55 to 60%) (12,61,84). Intraventricular and midline tumors are associated with headache in 92 to 95% of patients (61,84).

How often is headache the only symptom of a brain tumor? As previously noted in one series (84) of 183 adult patients with brain tumors, most patients had at least one other neurologic symptom besides headache at initial review, and only 15 patients (8%) exhibited headache as the only symptom. Age, sex, tumor localization, or tumor pathology did not correlate with headache presence; however, there was some suggestion that posterior fossa location of tumor and hydrocephalus were more common in patients who presented with headache. At diagnosis, 59 (31%) of the patients had headache, which rarely was an isolated symptom for longer than 10 weeks. The longest time any patient had headache as their only symptom was 77 days. Collectively, these observations indicate that a brain tumor does not cause chronic headache without any other major new symptoms.