Abstract
We aimed to investigate whether transient increases in blood pressure (BP) are associated with headache, by provoking hypertension in 50 non-hypertensive cephalalgia sufferers (Ce), 50 hypertensive cephalalgia sufferers (HyCe) and 50 healthy subjects (C) with a treadmill stress test. Occurrence of headache among groups did not differ (P = 0.81); however, women developed headache more frequently than men (P < 0.0001). The 35 individuals who developed headache during the test compared with those who did not (regardless of their original group) had higher absolute diastolic BP and higher absolute and relative diastolic BP increase during the test (P < 0.0001). HyCe patients that developed headache had higher diastolic BP than those that did not (P < 0.002). For each group, the absolute/relative mean increases of systolic/diastolic BP did not differ between patients that developed headache and those that did not excepting relative diastolic BP increase in Ce (28± vs. 15±, P = 0.006). The appearance of headache during dynamic exercise is not associated with a history of headache or hypertension. In non-hypertensive cephalalgia sufferers, relative diastolic BP increases of >27± are associated with headache attacks whereas in HyCe sufferers absolute diastolic BP values may be more important.
Introduction
The relationship between headache and increased blood pressure (BP) has been controversial. Some investigators have suggested that headache is a symptom of hypertension and may have typical features (1, 2). Others have claimed that the coexistence of headache and hypertension is coincidental (3, 4). Headache is a common complaint and by chance alone should be present in a large proportion of hypertensive patients. Moreover, the perception of patients of changes in their BP and the accompanying anxiety may manifest as headache (3). The International Headache Society (IHS) classification states that mild to moderate hypertension does not cause headache but recognizes headache as a symptom in patients who develop a paroxysmal increase in systolic (>160 mmHg) and/or diastolic (>120 mmHg) BP with or without hypertensive encephalopathy; in patients with pheochromocytoma, preeclampsia or eclampsia; and finally, in patients with an acute pressor response to an exogenous agent (e.g. cocaine) (5).
The treadmill stress test (TST), a clinically useful procedure for the investigation of cardiac ischaemia, is associated with transient increases in BP and offers a good setting for the investigation of the possible association between BP rise and the development of cephalalgia. Although cardiac cephalalgia (5, 6)—a rare type of exertional headache associated with acute myocardial ischaemia—may develop during the TST in patients suffering from coronary insufficiency, hitherto, the relationship between transient BP changes during dynamic exercise and headache has not been investigated in subjects without coronary heart disease.
Methods
Subjects
The study included 50 non-hypertensive cephalalgia sufferers (Ce); 50 hypertensive cephalalgia sufferers (HyCe); and 50 age- and sex-matched healthy controls (C). Patients were selected from the pool of patients referred to the headache out-patient clinic of our hospital between January 2004 and March 2005. Patients with a history of primary exertional headache, chest pain on exertion, cardiac failure, coronary heart disease, pulmonary, hepatic, renal disease, and mental or physical impairment leading to inability to exercise were excluded from the study. Patients taking antihypertensive or other treatment that could possibly influence the appearance of headache (e.g. β-blockers, calcium antagonists, antiepileptics, selective serotonin reuptake inhibitors, tricyclics, oral contraceptives) were not included. Analgesics were not permitted for 72 h prior to the study. No patient was on prophylactic treatment for migraine during the study and no woman was pregnant.
Controls were selected as follows: 162 employees of our institution consented to complete a detailed questionnaire focusing on the occurrence and characteristics of cephalalgia during the past 10 years. Based on their responses, 97 individuals were immediately excluded, 65 underwent a medical interview and 50 individuals were ultimately judged suitable for inclusion. These individuals had experienced ≤15 episodes of cephalalgia of any kind during the past 10 years and ≤2 episodes per year for the past 2 years. All patients were interviewed by a physician using a standardized questionnaire, which included information about the presence and characteristics of headaches during the previous 12 months. Subsequently, headaches were classified according to the IHS criteria (5).
We measured BP according to established guidelines (7). High BP was defined as systolic BP of ≥140 mmHg and/or diastolic BP of ≥90 mmHg confirmed on three to five measurements during 1 week, while normal blood pressure was defined as systolic BP of <120 mmHg and/or diastolic BP of <80 mmHg (7). The study was approved by the ethics committee of our hospital and all individuals gave informed consent prior to their inclusion in the study.
Treadmill stress test
All individuals underwent a TST beginning at 09.00 h in a temperature-controlled room (24 ± 2°C) after an overnight fast. The subjects were brought to the exercise laboratory where the heart rate and BP were recorded at rest. Sticky electrodes were attached to the chest, shoulders and hips and connected to the ECG portion of the stress test machine. A 12-lead ECG was recorded on paper. A cuff was fitted to arm diameter to avoid overestimation of measured BP. The treadmill was then started at a relatively slow ‘warm-up’ speed. The treadmill speed and its slope or inclination were increased every 3 min according to a preprogrammed protocol. Each 3-min interval is known as a stage. Thus, an individual after the completion of stage 3 had exercised for 9 min. The test was stopped prior to achievement of 85% of the maximal heart rate predicted for one's age, or if BP rose or fell beyond acceptable limits. BP was regularly recorded during the second minute of each stage. However, it was recorded more frequently if the readings were too high or if the individual presented headache. If the examined individual presented headache during the TST, its time of onset and characteristics were meticulously recorded. Relative BP changes were calculated using the formula 100 × (BP during headache − BP prior to headache)/BP prior to headache.
Statistics
The χ2 test was used for the comparison of dichotomous variables among the three groups of subjects. One-way
Results
The type of cephalalgia either before or during the TST did not differ between hypertensive and non-hypertensive headache sufferers; more than two-thirds of headaches that were reported during the TST were classified as tension-type headache (Table 1). The frequency of cephalalgia appearance during the TST did not differ among the three groups of individuals (P = 0.96); however, women developed cephalalgia much more often than men (P < 0.0001) (Fig. 1). Mean values of systolic and diastolic BP either at baseline or during the TST did not differ between C and Ce (P > 0.6). HyCe had higher mean systolic and diastolic BP values at baseline and during the TST compared with the other two groups (P < 0.0001). The relative increase of diastolic BP during the TST compared with baseline values did not differ among the three studied groups (P = 0.8). Conversely, the relative increase of systolic BP in C and Ce was higher compared with HyCe (P < 0.0001) (Fig. 2).
Frequency of cephalalgia (%) during the treadmill stress test was comparable among the three studied groups. Headache was much more frequent in women than in men [only 1/25 in controls (C) and hypertensive cephalalgia sufferers (HyCe) and 2/25 in non-hypertensive cephalalgia sufferers (Ce) compared with 10/25 women in C and Ce and 11/25 in HyCe].
Relative blood pressure (BP) increase during the treadmill stress test compared with baseline values in the three groups. The relative increase of diastolic BP (D) was identical among all groups. This was not the case for systolic BP (S). Arrows indicate statistically significant differences. C, Controls; Ce, non-hypertensive cephalalgia sufferers; HyCe, hypertensive cephalalgia sufferers.
Group characteristics
TST, Treadmill stress test; BP, blood pressure.
When the 35 individuals who developed headache during the TST were compared with the 135 individuals who did not (regardless of the group to which they belonged), the former had higher absolute diastolic BP values and higher absolute and relative diastolic BP increase during the TST compared with the latter (Table 2). None of the above associations held true for systolic BP.
Differences in BP between individuals with and without headache during the TST (regardless of the original group)
TST, Treadmill stress test; BP, blood pressure.
For each of the three groups (C, HyCe, Ce) absolute systolic BP values of individuals that developed headache during the TST did not differ compared with those of individuals that did not develop headache (P > 0.8) (Fig. 3). Conversely, absolute diastolic BP values of HyCe patients that developed headache were higher than those of HyCe patients that did not (P < 0.002). In the Ce group, the differences in diastolic BP between patients who presented headache and those that did not were not significant (P = 0.055). For each one of the three groups (C, HyCe, Ce), absolute systolic and diastolic BP changes (increases) during the TST were comparable between individuals that presented headache and those that did not (Fig. 4). Accordingly, for all groups relative BP increases during the TST were identical between individuals that presented headache and those that did not, excepting the relative diastolic BP increase in the Ce group (27.9 ± 9% vs. 15.3 ± 8.1%, P = 0.006) (Fig. 5).
Absolute blood pressure (BP) values in the three groups stratified by the appearance of headache during the treadmill stress test (TST). Only statistically significant differences are displayed. S, Systolic BP; D, diastolic BP; C, controls; Ce, non-hypertensive cephalalgia sufferers; HyCe, hypertensive cephalalgia sufferers. Hatched, Cephalalgia during TST; □, no cephalalgia during TST.
Absolute blood pressure (BP) changes (increases) in the three groups stratified by the appearance of headache during the treadmill stress test (TST). None of the differences between patients who presented headache during the test and those that did not reached statistical significance, although there was a tendency for higher BP increases in the hypertensive cephalalgia sufferer (HyCe) and non-hypertensive cephalalgia sufferer (Ce) groups (arrows). S, Systolic BP; D, diastolic BP; C, controls. Hatched, Cephalalgia during TST; □, no cephalalgia during TST.
Relative blood pressure (BP) increases in the three groups stratified by the appearance of headache during the treadmill stress test (TST). No comparisons were statistically significant except for diastolic BP in non-hypertensive cephalalgia sufferers (Ce). S, Systolic BP; D, diastolic BP; C, controls; HyCe, hypertensive cephalalgia sufferers. Hatched, Cephalalgia during TST; □, no cephalalgia during TST.
Discussion
The association between hypertension and cephalalgia was put forward in the medical literature almost a century ago (1) and remains a controversial issue. In the present study, the frequency of cephalalgia during the TST was identical in non-hyperetensive and hypertensive cephalalgia sufferers, and almost identical to that of controls. This is in accordance with a large prospective study (8) which elegantly demonstrated the absence of any positive association between initial high BP and subsequent risk of headache. Furthermore, a cross-sectional study (2), which found an overall positive relationship of headache to hypertension, has demonstrated that this relationship was inversely related to age and was not valid for individuals older than 45 years. In our study, all groups were well balanced in terms of age and the mean age of the individuals in each group approximated to this 45-year ‘threshold’. A retrospective study, conducted exclusively in patients with diagnosed hypertension lacking a control group of non-hypertensive individuals, has demonstrated a weak correlation between the prevalence of headache and diastolic BP but no correlation with systolic BP (9). To the best of our knowledge, this is the first study to investigate the association between acute rise of BP and headache appearance in non-hypertensive individuals suffering (Ce) or not (C) from cephalalgia, as well as in hypertensive headache sufferers (HyCe).
We observed that TST-induced headache was much more common in women than in men. This may simply reflect the preponderance of headache in women in the general population, where the female:male ratio has been assessed at 2–3 : 1 (10, 11). Instead, we found a 10-fold frequency of TST-induced headache in women. Whether this suggests a different response to exercise-induced BP increases between women and men or merely represents a statistical error in view of the small sample size is unknown.
Interestingly, when the three groups (Ce, HyCe, C) were pooled together and all the individuals that presented headache during the TST were compared with those that did not, we found that the former had significantly higher diastolic BP values as well as higher absolute and relative diastolic BP increases during exercise. In contrast, systolic BP had no association with the presence of headache during the TST. A study using ambulatory BP monitoring, lacking a control group of normotensive individuals, failed to document any association between BP and headache (12). Other authors applying the same method found no difference in BP between women suffering from chronic daily headache and women without headache (13). Nevertheless, this study did not include any men and chronic daily headache is a different entity from the episodic headache that is usually attributed to transient BP elevations.
When the individuals in each group were stratified by the appearance of headache during the TST, non-hypertensive cephalalgia sufferers that presented headache had higher relative diastolic BP increases than those that did not, whereas hypertensive cephalalgia sufferers that presented headache had clearly higher diastolic BP values during the TST among all participants in the study (115 ± 6.7 mmHg). Despite this fact, and although HyCe patients as a whole had evidently higher diastolic BP values during exercise compared with the other two groups, the frequency and type of headache during the TST were not different. Of the patients that presented headache during exercise in our study, 80%, regardless of their group, described the characteristics of tension-type headache. At least in some patients, acute hypertensive headache unaccompanied by neurological dysfunction is presumed to result from an abrupt increase in BP which remains within the range of cerebral autoregulation (14). The resultant constriction of the resistance vessels of the brain raises transmural pressure in the larger pain-sensitive arteries at the base of the brain, which are mainly located in the posterior cranial fossa, are innervated by the cervical roots and refer pain to the back of the head and neck, mimicking a tension-type headache (14). However, this study focused on the occurrence and not on the type of cephalalgia during the TST, and no evidence in favour of the aforementioned hypothesis can be provided.
In conclusion, although a history of hypertension or cephalalgia per se is not associated with increased frequency of cephalalgia during dynamic exercise, diastolic BP increase plays a role in the induction of headache during dynamic exercise.