Abstract
The pathophysiology of primary exertional headache (EH) is unknown. Physical exertion is associated with Valsalva-like manoeuvres (VM). VM leads to increased intrathoracic pressure and reduces cerebral venous drainage. Internal jugular vein valve incompetence (IJVVI) leads to retrograde venous flow during VM with transient increase of intracranial pressure. We analysed the prevalence of IJVVI in EH patients using duplex ultrasound. Bilateral measurements were performed at rest and during VM in 20 patients and 40 controls. Incompetence was concluded if retrograde venous flow could be seen in the jugular Doppler spectrum during repeated VM. Seventy percent of EH patients and 20% of controls demonstrated IJVVI, yielding a significant difference (P = 0.0004). IJVVI was always observed on the dominant venous drainage side. Our study suggests that intracranial venous congestion caused by retrograde jugular venous flow might play a role in the pathophysiology of EH with IJVVI as a risk factor.
Introduction
Primary exertional headache (EH) is an uncommon primary headache, often causing bilateral throbbing pain of moderate or severe intensity provoked by physical exertion. EH is of abrupt onset, lasting from 5 min to 24 h. It is self-limiting and not associated with structural brain lesions (1, 2). The lifetime prevalence of EH in population-based studies varies from 1% to 12% and seems to be more frequent in younger adulthood (2–4). Data regarding a gender preponderance are contradictory (2, 3). Comorbidity of EH with headache associated with sexual activity (SH) has been described (2, 3, 5, 6).
The pathophysiology of EH is unknown. There is some evidence in the literature that Valsalva-like manoeuvres (VM) might be a provoking factor of EH (2). A VM leads to an elevation of intrathoracic pressure, a reduction of cerebral venous drainage subsequently causing transient cerebral venous congestion. The internal jugular valves are the only valvular barriers between the brain and the heart, normally preventing retrograde venous blood flow (7). Internal jugular vein valve incompetence (IJVVI) might lead to retrograde venous blood flow during VM, which increases cerebral venous congestion and raises intracranial pressure.
IJVVI can be assessed by air contrast-aided duplex ultrasound (8) or by analysing valve morphology and the functional parameters of venous blood flow during VM (9). The reliability of these methods has been confirmed by different studies analysing the prevalence of IJVVI in healthy subjects and patients with transient global amnesia (9–13). Uni- or bilateral IJVVI prevalence ranges from 29% in the young adult population to 33–45% in healthy elderly subjects (9–13).
Our study is based on the hypothesis that transient cerebral venous congestion might play a pathogenic role in EH and that IJVVI could be an underlying risk factor. The purpose of the study was to assess the prevalence of IJVVI in patients with EH compared with age-matched controls using duplex ultrasound.
Methods
Study group
Twenty patients from our emergency department (16 men, four women, mean age ± SD: 43 ± 13 years), who fulfilled the International Headache Society (IHS) criteria of EH (1), were prospectively enrolled. In addition to routine cerebral computed tomography (CT), six out of 20 patients received additional magnetic resonance or CT angiography, which yielded no pathological arterial findings. Venous analysis was not performed. Age-matched controls (18 women, 22 men, 45 ± 4 years) were recruited from the hospital staff (n = 27) and routine patients (n = 13) sent to our laboratory if they were free of any pathology of the brain-supplying arteries. Patients enrolled presented with non-specific symptoms (vertigo, tinnitus, neck pain, diplopia, paraesthesia) or routinely before cardiac surgery to exclude any pathology of the brain-supplying arteries.
The clinical exploration of patients with EH and the assessment of duplex ultrasound were performed by different physicians. Duplex ultrasound was carried out after establishment of the diagnosis approximately 24 h after presentation to our hospital, when symptoms had already subsided. Exclusion criteria for patients and controls were a history of transient global amnesia, heart failure, chronic obstructive pulmonary disease or central venous IJV cannulation. All participants gave informed consent.
Ultrasound protocol
Venous duplex ultrasound in all subjects was performed after a short period of rest in a supine body position with the head in a 0–10° neutral position. Ultrasound was performed using a 7–11-MHz linear transducer (Powervision 6000, SSA-370A; Toshiba, Tokyo, Japan).
Blood volume flow (BVF) of both IJVs was analysed as described previously (14). Briefly, the IJV vessel area was measured by cross-sectional B-mode. Care was taken to avoid any transducer-induced compression of the vessel. For BVF calculation the area was multiplied with the time-averaged blood flow velocity over at least 4 s. In case of marked respiratory variations of venous vessel area or flow velocities, ultrasound measurements were performed during short apnoea after a normal exhalation. A difference of BVF of at least 150 ml/min between both IJVs was defined as an asymmetric drainage.
For IJVVI assessment, the Doppler sample volume was set between 0.5 and 1 cm and was placed in the centre of the IJV lumen approximately 2 cm above the internal jugular valves. During continuous monitoring using the triplex mode of the ultrasound system a maximal VM over 5 s was performed at least twice for each side. Combined B-mode and Doppler spectra images before and during VM were stored on magneto-optical disk. The sufficiency of VM was confirmed by a clearly visible increase of jugular vein diameter. Incompetence was concluded if a retrograde venous flow lasting ≥ 1 s was observed in the jugular Doppler spectrum analysis during repeated VM (Fig. 1). Decrease of venous flow or zero flow during VM was considered as competent jugular valve. Retrograde flow of <1 s duration was considered to be a short reflux during closing of competent valves (9)
Blood flow in the internal jugular vein (IJV) at rest (A) and during Valsalva manoeuvre (VM) in a patient with exertional headache. The decrease of blood flow during VM in the left IJV confirms a competent valve (B), whereas the transient retrograde blood flow during VM (>1 s) in the right IJV indicates the incompetence of the valve (C).
Statistical analysis
All presented values are given as means ± standard deviation. Two-sided Fisher's exact test was performed for comparison of controls and patients. P < 0.05 was considered to be significant.
Results
Measurements were successful in all subjects. Fourteen of 20 EH patients (70%) and eight of 40 controls (20%) demonstrated IJVVI, yielding a significant difference (P = 0.0004, two-sided Fisher's exact test). IJVVI in EH occurred in eight patients on the right side, in two patients on the left side and bilaterally in four patients. Eleven patients showed an asymmetric jugular drainage. A right-sided dominance was seen in nine, left-sided dominance in two and symmetrical drainage was observed in the remaining nine patients. IJVVI in patients with EH was always observed at the dominant drainage side or bilaterally.
In 29 controls (73%) an asymmetric venous drainage was found with a right- and left-sided dominance in 21 (53%) and eight (20%) subjects, respectively. The IJVVI was unilateral in six and bilateral in two subjects. Unilateral IJVVI occurred four times on the right, twice on the left side and in three subjects was on the dominant side.
Discussion
To our knowledge, this is the first study to analyse the IJVVI prevalence in EH. We found a significantly greater number of IJVVI in EH patients compared with age-matched controls. The prevalence of IJVVI in our controls (20%) was found to be in good agreement with control data previously published by other authors (range 20–29%) (8, 9), so that selection bias by our inclusion criteria seems unlikely.
Twelve (60%) of our included patients suffered from pure EH, eight (40%) had an additional history of SH. Comorbidity of EH and SH with similar pain characteristics occurs in 31–40%, and similar underlying pathophysiological mechanisms have been assumed by different authors (3, 5, 15). However, other authors have stressed the differences in clinical symptoms and preceding factors (16). Furthermore, the two conditions do not share the same treatment response (5). In our series, five of eight EH patients with additional SH demonstrated IJVVI.
The aetiology of EH remains unknown. A vascular origin of this headache is hypothesized by most authors, whereas different, mostly unproven pathophysiological mechanisms have been proposed in the past. One is the hypothesis of a transient cerebral arterial vasospasm, which has been described in one patient with EH and SH (17) but could not be confirmed by further studies. A second postulated mechanism is impaired cerebrovascular autoregulation caused by metabolic and myogenic factors reported in EH and SH patients (15, 18). Furthermore, comorbidity of EH and SH with migraine has been reported by several authors (3, 5) and a pathophysiological relationship between these disorders assumed (19). However, in our study only three EH patients (two with unilateral IJVVI) additionally fulfilled the IHS criteria for migraine, which is quite similar to the prevalence in a normal population. The third hypothesis is that a marked increase of cerebral arterial pressure might cause dilation of the pain-sensitive venous sinuses (20). Analyses in heavy weight-lifters have demonstrated extreme blood pressure values exceeding systolic values of 400 mmHg and diastolic values of 300 mmHg (21). However, the extent of blood pressure rise during ‘normal’ physical exertion probably does not lead to comparable blood pressure alterations. A second mechanism leading to an increase of intracranial pressure is venous congestion, which is facilitated by IJVVI and VM (22). The significantly higher prevalence of IJVVI in patients with EH compared with controls in our study seems to support this assumption and suggests that IJVVI might be a risk factor for EH. Furthermore, our study has confirmed the right-sided dominance of IJV drainage well known from previous studies (16, 23, 24). IJVVI in all of our patients was found on the dominant drainage side, i.e. more frequently on the right side, which has also been observed by other authors (8). As a potential underlying mechanism, higher retrograde intrathoracic pressure during VM caused by the closer spatial relation of the right jugular valve to the intrathoracic cavity has been assumed (8).
Although a suggestive finding, IJVVI can not explain all aspects of EH. EH often resolves without treatment, whereas IJVVI remains unchanged. Not every VM in a EH patient leads to headache and, finally, many individuals with IJVVI never develop EH. Therefore, a more complex mechanism, which might, however, include a transient rise of intracranial pressure as a potential trigger, has to be considered.
In summary, our study has shown a significantly higher prevalence of IJVVI in patients with EH compared with normal controls. This finding supports the hypothesis of a relationship between EH and cerebral venous congestion, facilitated by a VM, and indicates that IJVVI might be a risk factor for EH.