A 47-Year-Old Woman With Tension-type Headaches

DR BURNS: Ms T is a 47-year-old divorced librarian with a long history of headaches. She primarily receives treatment from Dr D, an anesthesiologist and neurologist at Beth Israel Deaconess Medical Center. She has managed care insurance.

Ms T had her first severe headache at 4 years of age, and she has had headaches regularly (1-4 times per year) since the age of 15 years. Over the years, her headaches became more frequent except for the 2 years when she was pregnant and nursing her child. Ms T saw multiple neurologists and received, at varying times, metoclopramide, prednisone, methysergide, amitriptyline, low-dose verapamil, isometheptene/dichloralphenazone/acetaminophen (Midrin), propranolol, ergotamine/belladonna/phenobarbital (Bellergal-S), diazepam, sumatriptan, aspirin/butalbital/caffeine (Fiorinal), pseudoephedrine, and dihydroergotamine. Her past medical history was otherwise unremarkable except for sinus problems and depression prominent particularly during her menses. She had normal findings on a magnetic resonance imaging (MRI) scan of the head. Family history is notable for a brother who died of a brain tumor.

Initially, Dr D thought Ms T suffered from migraine headaches with hormonal effects. She saw her gynecologist to address the hormonal component of her headaches. By March 1999, she had obtained good control of her migraine headaches using leuprolide along with fluoxetine, acetaminophen/butalbital/caffeine (Fiorocet), and rizatriptan.

More recently, Ms T developed recurrent headaches and complained of feeling fatigued and tired at times. She attributed this to stress, as work had become more demanding over the years. Her headaches were often present on awakening, but could occur at any time during the day. Usually located on the left side of the head, her pain also could be felt as a band over the occipital region and upper neck. The headache had a steady pressing quality that could be severe and disabling. Her neurologic examination was nonfocal. Dr D thought these were tension-type headaches. The plan was for her to continue her current migraine regimen. In addition, she was to begin physical therapy and to participate in a "mind-body" program to reduce and manage her symptoms.

Despite actively participating in these programs, Ms T continued to have almost daily headaches. She tried various other medications including gabapentin and rofecoxib without improvement. With Ms T's agreement, Dr D decided to proceed with injections of botulinum toxin (Botox), which Ms T received between December 1999 and August 2000. She noted marked improvement following the Botox injections, including fewer headaches, less disability, and a substantial decrease in her use of other medications. She was able to discontinue taking leuprolide. Her current medications are fluoxetine (20 mg twice daily), aspirin/butalbital/caffeine (twice daily as needed to a maximum of 6 tablets per week), rizatriptan (twice daily as needed to a maximum of 2 tablets per week), diazepam, hydrocodone (5 mg twice daily to a maximum of 5 tablets per month), and prochlorperazine (25-mg suppository) as needed.

I have a combination of tension and migraine headaches. Sometimes it's hard for me to tell the difference. I know when it's definitely a migraine, but I can't always tell when it's definitely a tension headache.

My tension headaches are not that predictable. They start when I'm under stress, or I wake up with them in the morning. Sometimes I think that I sleep wrong, and when I wake up, my neck feels very uncomfortable. Sometimes the headaches start all over, a band of pain, and every once in a while, they start in the back and work their way up. Almost all the headaches settle on the left temporal area.

Right now, all I'm doing is the drug treatment. Alternative medicines haven't helped reduce the number of headaches. They sometimes help in bringing down the severity. Nothing has actually taken the headaches away.

I am trying to take as few medications as possible. I try not to treat the headaches unless I have to. When I do treat the headache, then I try to treat it aggressively and hope that it will go away entirely; then I'm good for a few days.

For the last several years, Ms T has been under my care for what I believe is a combination of menstrual migraine and tension-type headaches, which have been very hard to control. The difficulty is that some of her tension-type headaches remain unresponsive to her current treatments and evolve into a migraine attack. But many patients with intractable tension-type headaches have that combination. Very few patients with pure tension-type headache actually need care by headache specialists. Many patients with intractable headaches, who make up more than half of my headache practice, do not respond to single medications. I would like to know Dr Welch's view on what to do with those patients.

What characteristics define tension headache? What is the epidemiology and natural history? How is the diagnosis established? How can physicians distinguish it from other types of headache? What are the best pharmacologic approaches? What is the role of nonpharmacologic therapies? What new treatments are on the horizon? What do you recommend for Ms T?

DR WELCH: I agree with Dr D that Ms T has 2 primary headaches that meet criteria suggested by the International Headache Society classification for chronic tension-type headache and for migraine without aura.¹ Table 1 describes the clinical features of episodic tension-type headache, chronic tension-type headache, and migraine. Although tension-type headache is the most prevalent of the primary headaches, it is also the least distinct, the diagnostic classification being based mostly on the absence of particular features (Table 2). Ms T began to experience episodic tension-type headaches in her late 20s. Chronic tension-type headaches, experienced nearly daily, now constitute her main disability because of their frequency and severity. Often, the pain is present on awakening, but it may start at any time of day and last throughout the day. Pain is usually located on the left side of the head, but it occasionally can be felt in a hatband location or over the occipital region and upper neck. Typically, the headache has a steady pressing quality that can become severe and disabling. A small proportion of patients with typical tension-type headache also experience occasional pulsating pain, unilateral pain, pain worsening on movement, anorexia, nausea, or photophobia—symptoms that may characterize migraine as well.² For example, when pain becomes severe, Ms T may experience mild nausea. Also, her headaches are consistently experienced on the left side, from where the pain of a severe migraine attack often emanates, and her headaches occasionally are made worse by movement. She sometimes is unable to distinguish the pain of the tension headache and that of migraine in the early stages when both are less severe, causing her confusion in choosing the appropriate self-medication regimen. Differentiating headache types is difficult for patients who experience more than 1 primary headache and may often pose a dilemma for the most experienced headache specialists.

Tension-type headache is one of the most common medical conditions, with prevalence ranging from 30% to 80% in North America, depending on definition and the epidemiological methods used to obtain the data.³ Most comprise mild and infrequent episodic attacks, while the prevalence of tension-type headache that occurs more than once a month is around 20% to 30%. Chronic tension-type headache affects 2% to 3% of the population.^{4 - 5} In subjects with coexisting migraine, the severity and frequency of tension-type headache attacks increase further. Prevalence of tension-type headache peaks around 30 to 39 years of age in both sexes and then declines. The disorder is only slightly more prevalent in women than men, about 1.3 to 1, increasing to twice the male prevalence in chronic tension-type headache.^{5 - 6} Menstruation is more frequently a precipitating factor in tension-type headache than migraine, although the latter is influenced to a greater degree by menarche, pregnancy, menopause, and oral contraceptives.⁶ African Americans have a lower prevalence compared with whites.⁵ Episodic tension-type headache is more prevalent with increasing education, but chronic tension-type headache is the opposite.⁵ Possibly, the latter is due to poor diet, stress, and inadequate medical care, or else disabling headache results in low income by disrupting education and occupation. Psychosocial stress adversely affects the prognosis of tension-type headache.⁷ Chronic tension-type headache is associated with reduction in quality of life, to a degree that is predicted by the emotional component of the headache.^{8 - 9} Nevertheless, few patients with episodic tension-type headache consult a general practitioner, and even fewer consult a specialist.⁹

The mechanisms of tension-type headache remain to be established. The only genetic study of chronic tension-type headache found a 3 to 1 increased risk in first-degree relatives, indicating genetic predisposition but a complex inheritance pattern.¹⁰ Because prevalence of the disorder is high, it appears that most individuals possess the potential to develop tension-type headache when exposed to certain environmental factors, and a single mechanism is unlikely to account for the disorder. In fact, the International Headache Society classification of tension-type headache categorizes the disorder by multiple causes (Table 3).¹ Thus, tension-type headache may be associated with oromandibular dysfunction, psychosocial stress, anxiety, depression, muscular stress, and drug overuse, but these should be viewed as associations rather than scientifically proven mechanisms.

Tenderness of the pericranial muscles, escalating with increasing frequency and severity of headaches, is the most common abnormality in patients with tension-type headache,¹¹ enough that it is used to subclassify the disorder (Table 3).¹ In chronic tension-type headache, pericranial muscle tenderness is associated with abnormally increased activity of the surface electromyogram over the frontal and temporal muscles and increased muscle stiffness.^{12 - 13} Increased muscle tenderness can initiate headache, possibly mediated by low threshold mechanosensitive A afferent neurons projecting to sensitized dorsal horn neurons of the trigeminal system.¹¹ Persistent and prolonged sensory input, including from C fibers, may cause central sensitization to pain stimuli and be part of the process that converts episodic tension-type headache to chronic headache.¹⁴ The hypothesis that descending control of these pain pathways is altered, contributing to sensitization, is supported by results of functional MRI.¹⁵

The diagnosis of chronic tension-type headache is essentially clinical, based on the features that have been described. Neurologic examination results are normal. However, as a simple way of assessing the involvement of pericranial muscles and subclassifying tension-type headache, the clinician should perform a cranial examination focused on evaluating pericranial muscle and tissue tenderness and trigger points of pain.¹⁶ The temporal, pterygoid, masseter, sternocleidomastoid, and trapezius muscles and the temporomandibular joint, mastoid process, occipital canal, and muscle insertion sites should be firmly palpated using the second and third fingers in a rotational movement and comparing side to side. If pain is elicited, a simple scoring system can be useful to quantify pain, eg, 0 for no pain, 1 for local pain, and 2 if withdrawal occurs to painful palpation. Recording a total tenderness score can be a useful means of focusing therapy on pericranial muscle tenderness, of measuring response to treatment, and of demonstrating to patients like MsT that there is an organic basis to their problem.

There are no established diagnostic tests for tension-type headache. If tension-type headaches become severe, increasingly frequent, or unresponsive to medication, it is important to reassess the diagnosis, particularly questioning for new features in onset of headache and related symptoms and examination findings that might suggest secondary causes of headache. Although routine diagnostic evaluation of tension-type headache does not require imaging of the brain,¹⁷ under these circumstances it is prudent to do so. When in the past Ms T had normal findings on brain imaging, reassurance of no life-threatening disease also reduced her stress level, particularly since her brother died of a brain tumor.

A retrospective study of headache clinic patients with almost daily headache revealed that up to 30% of subjects could not be categorized using International Headache Society criteria for primary tension-type, migraine, or other headaches.¹⁸ Although controversial, some headache specialists consider that this 30% predominately comprises a chronic headache syndrome, popularly termed "chronic daily headache" or "chronic migraine," deserving separate and unique classification.¹⁹ Although its cause and mechanisms remain to be established, chronic daily headache may be induced by repeated attacks of migraine and tension-type headache, especially under circumstances that include overuse of drugs such as codeine, hydrocodone, butalbital (as Fiorinal or Fiorocet), ergotamine preparations, and triptans.²⁰ However, this must be distinguished from chronic headaches associated with either drug ingestion or withdrawal, known as "rebound headache."²¹ In general, the head pain of chronic daily headache resembles that of mild to moderate migraine attacks. The time intervals for chronic daily headache to develop are approximately 5 years of exposure to medication and a prior history of primary headache for 10 years.²⁰ Acute drug withdrawal worsens headache, but clinical experience shows that treatment of the primary headache complaint fails if the drugs are not terminated.²² Up to 75% of patients improve when drug overuse is discontinued and minor analgesics are used with selected preventive medications. Headache usually remains, becoming less severe and less frequent with time, although unpredictable in duration.²⁰ Transformed migraine essentially is identical to chronic daily headache, but without a history of substantial drug overuse.²³ An argument might be made that Ms T's headaches should be categorized as chronic daily headache. Making this diagnosis is always problematic, however, because studies of the condition are retrospective and uncontrolled, the clinical features lack uniformity, and clear diagnostic criteria have not been determined.

Because Ms T suffers from chronic, often severe, headache, Dr D must always be alert to the potential coincidental occurrence of secondary headache syndromes. These include headache associated with trauma, vascular disease, nonvascular intracranial disorders, substance use or withdrawal, noncephalic infections, metabolic disorders, disorders of facial or cranial structures, and cranial neuralgia. Several conditions may mimic episodic or chronic tension-type headache. Usually, differentiating headaches from acute eye diseases or sinusitis is straightforward. Chronic tension-type headaches are often erroneously attributed to chronic sinusitis, but clinical and radiological evidence must be obtained before making this diagnosis. A causal relationship between tension-type headache and oromandibular dysfunction is controversial, but the 2 conditions often coexist.²⁴ The chronic headache of intracranial hypotension is most often distinguished by an increase of pain on standing. If there is uncertainty, MRI with gadolinium may detect meningeal enhancement characteristic of the disorder.²⁵ Idiopathic benign intracranial hypertension, also known as pseudotumor cerebri, may sometimes cause daily headache but without papilledema on examination.²⁶ Obesity may increase suspicion of the disorder, and a lumbar puncture to measure cerebrospinal fluid pressure may be necessary to distinguish the headache from chronic tension-type headache. Although tension-type headache generally is not a presenting feature of cervical spondylosis,²⁷ the persistent location of the left hemicranial pain in Ms T's case raises the consideration of structural compression of the upper cervical nerve roots, so-called cervicogenic headache.²⁸ For the same reason, atypical pain from occipital nerve entrapment should be considered. Also, a rare persistently unilateral headache of unknown mechanism called "hemicrania continua" is distinguishable by autonomic features similar to cluster headache or a response to indomethacin treatment.²⁹ Ms T does not attribute her headache to caffeine withdrawal, which can have the same features as tension-type headache,³⁰ but she feels her headaches improve when she ingests caffeine-containing drinks or caffeine-containing drugs. Because drinking coffee is a national pastime, caffeine withdrawal headache is a common differential diagnosis and can be treated effectively. Daily persistent headache is another headache syndrome of undetermined cause in which a headache with an abrupt onset never goes away.¹⁹ Patients usually remember precisely the time and day the headache began, reflecting no antecedent problem. The features are similar to tension-type headache. The cause is unknown but suggests some organic event such as a viral illness that causes persistent sensitization of peripheral or central nociceptive pathways. Finally, posttraumatic headache with minor head trauma, for which there are no confirmatory signs, is a complex problem with regard to cause and management but will be clear from the history.³¹

In general, clinical experience shows that management of headache requires a strong patient-clinician relationship and a team effort in which the patient plays an equal role. Ms T expressed disappointment that biofeedback and meditation had not cured her headache. Patients should be informed that most primary headaches have no cure, but in most cases they can be controlled. Informing patients that they have an organic, perhaps genetically determined, disorder, rather than a psychological one, is extremely important. In discussing the management plan, physicians should explain that treatment aims to relieve the symptoms of acute headaches and prevent further episodes by either behavioral or pharmacological means.

Rest and minor analgesics are the best treatments for episodic tension-type headache.³² Physicians should seek precipitating factors and provide reassurance and psychological support for these patients. Gentle massage of the head and neck and application of hot packs to these regions are physical maneuvers that help alleviate some of the pain. Given the risk of analgesic addiction, only minor analgesics should be used, and these are generally effective in subjects with infrequent, mild to moderately severe tension-type headache (Table 4). Nonsteroidal anti-inflammatory drugs (NSAIDs) and cyclooxygenase 2 (COX-2) inhibitors are useful alternative analgesics, although experience with the latter is limited. The Food and Drug Administration has approved aspirin in combination with caffeine and butalbital (Fiorinal) for treatment of tension-type headache. This preparation may have addictive potential, and it may induce rebound headache or daily headache syndromes.³³ If the patient has used scheduled analgesics consistently, the pain of severe episodic tension-type headache may not respond to anything but parenteral narcotics. Instead of pursuing this line of treatment, it is preferable to wean the patient from parenteral narcotics and treat the patient with tranquilizers that have some central action on neuromuscular relaxation, such as diazepam. Diazepam may have the added beneficial effect of causing drowsiness and sleep. Like Fiorinal, diazepam should be avoided by patients with chronic tension-type headache because of its addictive potential and tendency to exacerbate tension-type headache.

For the prevention of episodic and chronic tension-type headaches, behavioral approaches commonly involve regular sleep and meals and avoidance of initiating or trigger factors. Straightforward stress coping, meditation, or relaxation strategies best manage work-related or family stress and emotional problems. An exercise program may be helpful. Neck manipulation may be effective,³⁴ especially if evidence points to pericranial muscle involvement. Some reports suggest an association of manipulation with extracranial arterial dissection, but the incidence is unknown.³⁵ Because of expense and uncertain effect, therapist-assisted biofeedback is best reserved for patients who do not respond to other measures. Correction of oromandibular dysfunction may be undertaken if evidence of abnormality is convincing, but it is best to avoid irreversible procedures.³⁶ Alternative medicine such as acupuncture has no scientifically proven effectiveness for headache beyond the placebo effect.³²

Psychological therapy can provide up to 60% improvement in recurrent episodic or chronic tension-type headache; the intensity of the management will likely need to be greater the more frequent and severe the complaint.^{37 - 39} If a comorbid mood or anxiety disorder is present and the headache intractable, then pharmacological therapy may be added.⁴⁰ Randomized, double-blind, placebo-controlled trials have consistently proven amitriptyline to be the most useful medication for preventing chronic tension-type headache, providing 30% to 50% reduction in headache frequency and severity^{41 - 42} (Table 5). Further, in a randomized placebo-controlled trial of tricyclic antidepressant medication (amitriptyline [up to 100 mg daily] or nortriptyline [up to 75 mg daily]) combined with stress management therapy, the combination of antidepressant and stress management therapy produced greater than 50% reduction in headache activity than either alone in 64% of patients.⁴² The beneficial effect of amitriptyline is probably independent of its antidepressant activity. Amitriptyline should be started at low doses of 10 to 25 mg/d taken at night, increasing by 25 mg weekly until effective. Maintenance doses are around 75 mg/d, but occasionally up to 200 mg/d is needed. Prominent adverse effects are weight gain, dry mouth, and drowsiness. Serious adverse effects include cardiac arrhythmia, urinary retention, and glaucoma, although these are uncommon.

Other tricyclic antidepressants may be effective but have not been as rigorously tested as amitriptyline.⁴¹ In a single double-blind, placebo-controlled trial, doxepin, 100 mg/d, benefited patients with chronic tension-type headache and migraine combined.⁴³ Controlled clinical trials suggest that the selective serotonin reuptake inhibitors have no or limited analgesic effect but may be useful in patients with intractable headache and depression.⁴¹ In a double-blind placebo-controlled trial, fluoxetine, in 20 to 40 mg/d doses, provided 50% improvement in overall headache status in 64 patients with chronic tension-type headache and migraine combined.⁴⁴ Because of possible gastrointestinal ulceration and hemorrhage, prolonged prophylaxis with NSAIDs should be avoided. Because many patients have associated pericranial muscle disorder, muscle relaxants may be tried. Tizanidine, a centrally acting muscle relaxant, produced a 30% to 50% improvement in all headache measures in a placebo-controlled, double-blind crossover trial.⁴⁵ If one group of preventive drugs is ineffective, it is reasonable to change to another, although this strategy has not been tested in clinical trials.

Because the mechanisms of primary headache are not well understood, an evidence-based therapeutic approach soon may be exhausted or prove unsatisfactory. Often at the crossroads of practicing the art and science of medicine, clinicians must call on their experience with drugs that are less well scientifically proven. Ms T's headaches represent a complex and difficult management problem because she has already failed to respond to most therapeutic regimens. Fiorinal is the mainstay of her acute treatment but does not provide complete pain relief. Because of the potential of developing chronic daily headache as she has in the past with drug overuse, she should avoid Fiorinal, diazepam, and opioids. Instead, Dr D should encourage Ms T to continue using ibuprofen alone, and if this is ineffective, to try other NSAIDs or COX-2 inhibitors to obtain pain relief. Ms T now takes fluoxetine daily for headache prevention and is in a period of less frequent headache episodes. Should she relapse, a trial of tizanidine or a repeat trial of amitriptyline might be worthwhile.

From December 1999 to August 2000, due to exacerbation of her headaches, Dr D injected botulinum toxin into frontal, temporal, occipital, and other pericranial muscles on 4 occasions, averaging every 2 months. Ms T and Dr D associate her less frequent and disabling headaches with this therapy, but this opinion should be viewed with caution. Reasons for question in Ms T's case include variability in the natural history of headaches and the strong placebo effect of the injections. Other factors include the potential nociceptive effect solely of the injection procedure itself, the potential benefit of coincident use of preventive drugs, and reduction in analgesic drug overuse. Further, the rationale for using this therapy in chronic tension-type headache remains uncertain. To date, published reports are conflicting about the benefit of this treatment for chronic tension-type headache,^{46 - 47} and rigorous controlled trials should be awaited before introducing this into routine practice.

What does the future hold for Ms T? Unfortunately, chronic tension-type headache may be protracted into later life.²⁰ The problem should improve if she totally avoids using Fiorinal and narcotic analgesics. The fear remains that if she moves again into a phase of drug overuse, she may develop intractable headache that would be less responsive to medications than before. Effective pharmacological prevention of chronic tension-type headache seems distant. Unlike symptomatic medications, no drugs have been designed especially for tension-type headache prevention, in large part because even less is known of the true cause of tension-type headache than of its mechanisms. Even with the best outcome, Ms T may likely continue to have headaches to some degree, although there is some hope that the headaches will become more responsive to medications.

A PHYSICIAN: Is there any role for cortisone in this kind of patient?

DR WELCH: One possibly effective but not rigorously proven strategy on hospital admission for drug withdrawal, other than dihydroergotamine, is to give a course of intravenous hydrocortisone, 100 mg, or dexamethasone, 4 mg every 4 to 6 hours for 1 to 2 days.⁴⁸ You cannot give repeated courses of steroids because of the complications of chronic steroid exposure. The mechanism steroids act upon is likely at the level of the nociceptive stimulus, possibly against a neurogenically induced inflammatory state.

A PHYSICIAN: Do you have any insight into the mechanism of headache that follows ingestion of red wine or any suggestions about therapy?

DR WELCH: I think that alcohol, in this instance, is a trigger for migraine headache that we don't completely understand. It's obviously a cerebrovascular dilator. There is some thought that the histamine content of red wine is causing the headache. Wine may also contain components that act as false neurotransmitters.

DR D: When Ms T cannot identify her headache as a migraine, what is wrong with starting with Fiorinal, or an opioid-based analgesic, if a standard NSAID or a COX-2 inhibitor doesn't work?

DR WELCH: Well, it's important to consider the future. These types of medications are the major culprits in the longevity of the headache and the chronicity of headache. For many patients, this medication is the most common one that produces rebound headaches. If you have a predisposition to headaches, it is best to avoid medications that induce headaches. To me, it seems sensible to have the patient become accustomed to using heavy doses of nonopiate medications. Furthermore, you said that Ms T discontinued amitriptyline and it did not make any difference. However, at that time, she was still taking all the other medications. It is typical that any beneficial effect of the prophylaxis is obscured as long as you continue medications like Fiorinal, hydrocodone, and codeine.²²

DR D: Well, part of it is quality of life and her ability to continue to work. The data on rebound headaches is at best confusing.

DR WELCH: That's right. You do what you can, and if you find something that improves the quality of life for these patients, you go with it. We talk about evidence-based medicine and sticking to it, but when you deal with this kind of condition, you very rapidly run out of evidence-based medicine. You are then practicing the art of medicine. It is hard to criticize any approach when you reach the stage that Ms T is at, and she has improved under your care.