Headaches and Hypertension: Primary or Secondary?

A 48‐year‐old woman was referred to the hypertension clinic because of refractory stage II hypertension and headaches. She was taking lisinopril, atenolol, and extended‐release nifedipine with clinic blood pressures consistently 160–180/100 mm Hg and constant daily and unremitting headaches of more than 2 months' duration. There had been no recent change in her medication regimen. Whenever she called a nurse to report severe headache with a self‐obtained blood pressure of 180 mm Hg systolic, she was referred to the nearest emergency department. Within the prior 2 weeks she had been to three different emergency departments and had two negative computed tomography brain scans.

History revealed significant job‐related stress with daily, constant, nonthrobbing frontal and posterior headaches for 1 month. Ibuprofen and sulindac b.i.d. were not effective, but cyclobenzaprine had been helpful to a moderate degree. A diagnosis of tension‐type headache was made, and it was explained to the patient that her headaches were not secondary to her hypertension or indicative of an impending stroke. Cyclobenzaprine 10 mg b.i.d. was initiated with once‐ to twice‐weekly clinic visits for reassurance. This led to a cessation of emergency department visits. She was referred to the headache clinic where the antispasticity agent tizanidine (at 8 mg q.d.) was substituted for cyclobenzaprine and she attended a biofeedback class. The headaches subsided with follow‐up blood pressures of 140–145/85 mm Hg on the same three antihypertensive agents.

CASE TWO

A 67‐year‐old woman was referred to the hypertension clinic because of variable blood pressures with associated headaches. She started taking hydrochlorothiazide for hypertension 5 years prior and over the past 6 months had recorded particularly highly variable blood pressures with a self‐reported automated apparatus as well as a home sphygmomanometer. She had been taking her blood pressure several times a day. Blood pressures as high as 200/118 mm Hg would be reduced to 98/76 mm Hg following clonidine. She denied recent chest pain, dizziness, or syncope. Blood pressures were more likely to be elevated in the morning, and she was able to predict these changes with symptoms of headache, ringing in the ears, tenseness, and facial flushing. In the past 2 years she had seen a cardiologist and undergone a negative radionucleotide imaging examination for atypical chest pain and had consulted a pulmonologist for vague breathing complaints, also with a negative workup. The patient was compliant with an exercise prescription, and followed through with a stress‐reduction program. She was advised to reduce her home blood pressure determinations to no more than three times weekly. Within 6 weeks, consistent clinic and home blood pressures of 120–135/70–85 mm Hg were achieved on hydrochlorothiazide alone.

DISCUSSION

It is not uncommon in hypertension practices to hear a patient say, "I can tell my pressure is high because I get a headache." However, not very many undiagnosed hypertensives approach their physician asking, "I have a headache, will you check my blood pressure?" The Headache Classification Committee of the International Headache Society states that mild‐to‐moderate hypertension does not cause headaches, ¹ but what about more severe levels?

In Stewart's ² 1953 observations of headache and hypertension, 88 patients with headache and 112 without headache had mean diastolic blood pressures of 140 mm Hg and 133 mm Hg, respectively, with considerable overlap, not significantly different. Interestingly, 71 (81%) of those with headache were knowledgeable about their elevated pressures, and only 25 (22%) of those without headache knew of their hypertension. Individuals who knew they had hypertension were more likely to report headaches than comparable patients without knowledge of their blood pressure elevation. Likewise, the National Health and Nutrition Examination Survey (NHANES) study of headache related to degrees of hypertension found no association even with systolic pressures >160 mm Hg. ³ Cantillon et al. ⁴ looked at 51 hypertensive patients in a general practice who claimed ability to predict their blood pressure and compared the predictors' blood pressure estimates to those of an identical number of nonpredictor hypertensives. Predictors were more anxious and complained of headache more frequently than nonpredictors (37% vs. 15%), but were unable to forecast their blood pressures reliably.

Physician and patient perception of a headache association with hypertension dates from the 1913 Janeway ⁵ description that details a typical hypertensive headache as a nonmigrainous headache on awakening that resolves later in the morning, a description promulgated by other authors. ⁶ , ⁷ Spierings ⁸ anecdotally described four headache syndromes associated with acute and chronic hypertension, malignant hypertension, and hypertensive encephalopathy with postulated mechanisms, but he lacked supportive data. In fact, two ambulatory blood pressure trials looking for an association between headache and blood pressure fluctuations found nothing. ⁹ , ¹⁰

It has been postulated that the morning headaches experienced by some hypertensive patients are actually due to obstructive sleep apnea. One study compared the prevalence of morning headaches in a group of 148 snorers, 24% of whom had hypertension with an apnea/hypopnea index <5, to a group of 164 sleep apnea patients with an apnea/hypopnea index ≥15, 44% of whom had hypertension defined as a sustained blood pressure >160/95 mm Hg. ¹¹ There was no statistical difference between the two groups regarding morning headaches, which occurred in 44% of snorers and 58% of sleep apnea patients. Furthermore, morning headaches were not correlated with sleep architecture, nocturnal respiratory parameters, or hypertension and instead were strongly related to a measure of depression. Another study did come up with an association between severity of sleep apnea and the frequency of morning headaches. This study noted that successful treatment of the sleep apnea with nasal continuous positive airway pressure (CPAP) improved the headaches. ¹² It appears that morning headaches believed to be associated with hypertension are more likely related to mood disorders and sleep apnea. Some of these headaches will respond to nasal sleep apnea with nasal CPAP.

Obesity occurs frequently in the American hypertensive population; the mean body mass index of subjects in the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) was 30. ¹³ Headache related to morbid obesity is occasionally due to benign intracranial hypertension and may be reversed by gastric bypass procedures. ¹⁴ , ¹⁵ A syndrome of idiopathic intracranial hypertension associated with primary aldosteronism and headache accompanied by visual disturbances has also been described. ¹⁶ One case reversed following adrenal tumor resection, and the other required a ventriculoperitoneal shunt. In these cases, headache was attributed to intracranial hypertension, perhaps caused by a central mechanism by which chronic mineralocorticoid excess may impede the flow of spinal fluid, rather than the hypertension itself.

Headache in a hypertensive patient may be affected by medications taken to treat the hypertension. Calcium channel blockers and hydralazine are particularly linked to headache causation. Beta blockers and α‐1 receptor angiotensin receptor blockers have been used prophylactically to reduce the recurrence rate of migraines. ¹⁷

Two large trials have shown that successful treatment of hypertension reduces headache in the context of overall quality‐of‐life improvement. The Treatment Of Mild Hypertension Study (TOMHS), which included 902 men and women, found a nonsignificant trend toward reduction in the incidence of headache among patients receiving treatment compared with those receiving placebo. Overall quality of life determined by an extensive questionnaire and seven different parameters was, however, significantly improved in patients on antihypertension drug therapy. ¹⁸ A substudy of the Hypertension Optimal Treatment (HOT) study (n=610) demonstrated that intensive blood pressure lowering significantly reduced the incidence of headache (p<0.001) while improving overall quality of life. ¹⁹ Two self‐administered questionnaires, the Psychological General Well‐being Index and the Subjective Symptoms Assessment Profile indicated that the lower the achieved diastolic blood pressure, the greater the improvement in self reported well‐being.

The first patient presented (Case 1) readily reported significant job‐related stress with unrelenting headaches more than a month in duration, strongly suggesting tension‐type headache. Onset of these headaches occurred several months following initiation of extended‐release nifedipine for hypertension, and therefore appeared unrelated to drug therapy. Despite consistent stage 2 hypertension, a treatment strategy directed against tension headaches was selected before further antihypertensive drug titration leading to reduced blood pressures. This helped build the patient trust in her physicians. In this case, headache pain itself accounted for much of the patient's blood pressure elevation.

Case two was a patient who believed she could forecast her blood pressure by symptoms of headache, tinnitus, and facial flushing. These accompanied morning‐awakening blood pressure elevations. However, the wide variability in blood pressures suggested an anxiety syndrome with resultant ascents in pressure. In this case, stress reduction and an exercise program led to more even blood pressure control lacking extreme spikes. Advancing her antihypertensive drug regimen might have led to excessively low pressures. This patient also illustrates the potential difficulties of managing an anxious hypertensive patient with a home blood pressure apparatus.

When a hypertension patient presents with headache, a clinician should initially seek to diagnose a primary headache syndrome. A differential diagnosis should include tension‐type headache and cervical osteoarthritis, particularly in older individuals, and also look for anxiety and depressive disorders (Table I). Treatment of these conditions is also likely to achieve improved hypertension control. Some headaches may be medication related, and a rare patient may respond to nasal CPAP for obstructive sleep apnea. Interestingly, a recent large, prospective survey demonstrated that elevated systolic and diastolic pressures were associated with reduced risk of nonmigrainous headache ²⁰ possibly as a result of hypertension associated hypalgesia due to baroreceptor stimulation and release of endogenous opioids. ²¹ Headache, including morning headache, is unlikely to be a secondary effect of hypertension—except perhaps in patients with very severe disease and sustained elevations.

Table.

Primary Etiologies of Nonmigrainous Headache in Patients with Hypertension

Mood disorders (anxiety, depression)

Tension‐type headache

Medications

Calcium channel blockers

Hydralazine

Alpha‐1 antagonists (e.g., terazosin, doxazosin)

Antiheadache medicine overuse

Nonsteroidal anti‐inflammatory agents

Nitrates

Cervical osteoarthritis

Obstructive sleep apnea

Idiopathic intracranial hypertension

Morbid obesity

Primary hyperaldosteronism