Abstract
The Queckenstedt's (Q)-est can aggravate headache intensity during migraine attacks (Q-test effect). The objective of this study was to delineate the Q-test effect in patients experiencing migraine attacks. We performed a 30-s Q- and a sham test on 39 patients with acute migraine attacks in both supine and sitting positions. Headache intensities during and 30 s after the Q- or sham tests were recorded on a 0–10 verbal scale. Brushing allodynia (BA) was recorded after using a gauze-brushing test over the patient's face and forearms. The Q- but not the sham test aggravated headache intensity in both sitting and supine positions. The presence of throbbing pain and higher pain intensities was associated with the Q-test effect in the supine position. However, the presence or absence of BA was not correlated. We concluded that the Q-test effect is likely to be related to peripheral sensitization of the meninges but not central sensitization. The Q-test effect may be used as an objective marker for peripheral sensitization.
Introduction
Queckenstedt's manoeuvre (Q-test) is able to induce cephalic venous congestion, which aggravates headache intensity in patients with migraine but not tension-type headache during their headache attacks (Q-test effect) (1, 2). We hypothesized that Q-test-induced aggravation of headache during migraine reflects intracranial sensitization by increased intracranial pressure or pressure in the dural sinuses (2). Whether peripheral or central sensitization is contributing to this effect and its relationship to the headache's characteristics have not been clarified.
Meningeal primary afferent neurons become hypersensitive after stimulation (3), which explains why headaches worsen when patients engage in physical activities that increase intracranial pressure such as coughing, bending over, etc. (4, 5). Theoretically, the pressure caused by the patient's arterial pulse is mechanically transmitted across the subarachnoid space (6), so that the sensitized meningeal afferent neurons sense the pulsation, thus generating the characteristic ‘throbbing pain’ (7). This process is called peripheral sensitization because the activated trigeminovascular neurons that innervate the meninges are outside of the brain (8). Cutaneous allodynia, i.e. pain resulting from application of a non-noxious stimulus to normal skin, is common during a migraine attack (9). The origin of this type of allodynia is in the trigeminal nucleus caudalis or thalamus due to sensitization of their second- and third-order neurons (10, 11). This type of allodynia results from central sensitization and is found to be associated with activation of glutamate and substance P receptors (12–14). Four types of cutaneous allodynia have been studied: dynamic mechanical (brush), static mechanical (pressure), heat and cold (14). The gauze-brushing test can be used to detect brushing allodynia (BA) (15–17). Unlike other forms of allodynia, BA is solely dependent on A-δ but not C fibres (17–19).
Since the Q-test can cause venous congestion, thereby affecting the dural sinus, we hypothesize that the aggravation of headache intensity by the Q-test during migraine attacks is related to peripheral but not central sensitization. In this study, we used ‘throbbing pain’ to indicate the presence of peripheral sensitization and BA to represent central sensitization during migraine attacks.
Subjects and methods
We recruited patients who were currently experiencing acute migraine attacks and visiting the Neurology Out-patient Clinic at Yi-Lan General Hospital, or Yuan Shan Veterans Hospital, Yi-Lan, Taiwan. The study protocol was approved by the Institutional Review Board of National Yang-Ming University, Taipei, Taiwan. All patients provided written informed consent prior to enrolment.
We recorded the current headache characteristics of the study patients, including aura, duration, headache frequency, headache intensity (mild, moderate or severe), location of pain, throbbing quality, aggravation by physical activities, associated symptoms (nausea, vomiting, photophobia and phonophobia) and painkiller usage. The headache intensity before the Q- or sham test was reported based on a 0–10 verbal scale: 0 (no pain), 1–3 (mild pain), 4–6 (moderate pain) and 7–10 (severe pain).
Diagnosis of acute headache attack
The diagnosis of the migraine attack was based on the diagnostic criteria in the International Classification of Headache Disorders, 2nd edn (ICHD-II), 2004 (20) (codes 1.1 and 1.2). Patients who reported a headache frequency ≥ 15 days/month and those patients who had taken abortive treatment within the past 12 h were excluded.
Brushing allodynia
We applied a 100 × 100-mm gauze that was folded diagonally to brush the patients' chins, foreheads and forearms bilaterally at a frequency of about 1–2 Hz 10 times, and asked them if they felt any discomfort as a result of the ‘gauze brushing test’ other than the ‘brushing sensation’. If the answer was ‘yes,’ the response to BA was recorded as ‘positive’ (BA+). If the answer was ‘no’ we recorded a ‘negative’ (BA−) response.
Q- or sham tests
The Q-test was done by placing equal and constant manual pressure on the patients' bilateral internal jugular veins at the level of the limbs of the thyroid cartilage. The sham test was performed by placing equal and constant manual pressure at the same level of the Q-test on both of the sternocleidomastoid muscles (about 2 cm lateral to previous positions) (2, 21) and avoiding the internal jugular veins. In order to keep the procedures consistent and maintain a uniform and constant manual pressure for each patient, one examiner, C-H. Chou, performed all of the procedures.
Study procedure
This was a single-blind study, i.e. patients did not know the difference between the Q- and the sham tests. Patients were asked to sit down for at least 5 min for a rest. We performed the gauze brushing test first. Then, the Q- or the sham test was performed for 30 s and then relieved for 30 s in the sitting position. Afterwards, the sham test or the Q-test followed. The order of the performance of the Q- or the sham test was randomly decided. Each patient's headache intensity was recorded using a 0–10 verbal scale at 10-s intervals during the Q- and the sham tests. The same procedures were then repeated with the patient in the supine position.
Statistical methods
In this study, headache intensity changes were recorded during the Q- or the sham tests for 10, 20 and 30 s and relief for 10, 20, 30 s, and the results classified in comparison with the demographics and different headache characteristics, such as BA, aura, current headache duration, headache frequency, headache intensity, laterality, throbbing pain, aggravation by physical activities, nausea, vomiting, photophobia and phonophobia. The independent t-test and paired t-test were used when appropriate. We used general linear models to evaluate the independent effect of the relevant factors for the Q-test effect, which were obtained from univariate analyses above. A P-value < 0.05 was considered to be statistically significant.
Results
Thirty-nine patients (10 male, 29 female, mean age 37.0 ± 9.7 years, range 19–53 years) with ICHD-II migraine attacks participated in this study. Their demographic data and current headache profiles are shown in Table 1. Thirty patients (76.9%) had throbbing headache and 10 (25.6%) had BA. The duration of headache did not differ statistically between the throbbing pain group (34.6 ± 18.4 h) and the non-throbbing pain group (31.6 ± 22.7 h) (P = 0.68), nor between those with BA (26.1 ± 20.9 h) and those without (36.6 ± 18.2 h) (P = 0.14).
The demographic and headache profiles of the studied migraine patients
The effect of supine and sitting positions
Headache intensity increased during the Q-test in both the supine and sitting positions, but the degree of increment was higher in the supine position (Fig. 1).
The headache intensity changes (by a 0–10 verbal scale) during and after the Q- or the sham tests in migraine patients (39 cases). L, supine position; S, sitting position; Q, Q-test; Sh, sham test. #P < 0.05 compared between the Q- and the sham tests by paired t-test.
Univariate analysis
The effect of brushing allodynia
Both patients with BA (10 cases) and without BA (29 cases) showed higher headache intensities during the Q-test vs. the sham test in supine positions. However, the increment did not differ between patients with BA and without BA (headache intensity changes, BA+ vs. BA− at 10 s, 0.80 ± 0.92 vs. 0.79 ± 0.81, P = 0.98; 20 s, 1.1 ± 1.1 vs. 1.2 ± 1.1, P = 0.73; 30 s, 1.2 ± 1.1 vs. 1.5 ± 1.3, P = 0.50) (Fig. 2).
The headache intensity changes (by a 0–10 verbal scale) in relation to the presence or absence of brushing allodynia during and after the Q- or the sham tests in migraine patients in supine position. BA+, with brushing allodynia (10 cases); BA−, without brushing allodynia (29 cases); Q, Q-test; Sh, sham test.
The effect of the presence of throbbing pain
The throbbing pain group had a higher increment of headache intensity during the Q-test than the non-throbbing group in the supine position [P = 0.015 (10 s), < 0.001 (20 s) and 0.003 (30 s)] (Fig. 3), but in the sitting position the two groups were different only at 20 s (P = 0.011) after the Q-test, but not at 10 s (P = 0.17) and 30 s (P = 0.13).
Headache intensity changes (by a 0–10 verbal scale) in relation to the presence or absence of throbbing pain during and after the Q- or the sham tests in migraine patients in supine position. T, with throbbing pain (30 cases); nT, without throbbing pain (nine cases); Q, Q-test; Sh, sham test. ∗P < 0.05 compared between the T and the nT groups by independent t-test.
The effect of headache intensity
Twenty-two (56.4%) patients reported severe intensity of their current headache attacks, and 17 patients mild to moderate intensity. The severe intensity group had higher increment of headache intensity during the Q-test than the mild to moderate intensity group in the supine position during the Q-test at all time points [P = 0.048 (10 s), 0.017 (20 s) and 0.014 (30 s)] (Fig. 4), but in the sitting position the two groups did not differ at any of the time points [P = 0.48 (10 s), 0.27 (20 s), 0.46 (30 s)].
Headache intensity changes (by a 0–10 verbal scale) between patients with severe and mild to moderate headache during and after the Q- or the sham tests in migraine patients in the supine position. HA, severe headache (22 cases); Ha, mild to moderate headache (17 cases); Q, Q-test; Sh, sham test. ∗P < 0.05 compared between the HA and the Ha groups by independent t-test.
Other demographics or headache characteristics were not associated with the increment of headache intensity during the Q-test.
Multivariate analysis: general linear models
We calculated the independent effect of the time sequences (during Q-test at time points 0, 10, 20 and 30 s), throbbing pain and headache intensity on the headache intensity changes during the Q-test in migraine patients in the supine position (Table 2). All three factors remained significant in the multivariate analyses.
The results of general linear models for analyses of throbbing pain, headache intensity, and time sequences in relation to ‘headache intensity changes’ by the Q-test in the supine position
Throbbing, present or absent; headache intensity, by a 0–10 verbal scale; time sequence (Q-test time points), 0, 10, 20 and 30 s.
Discussion
Our study has demonstrated that the Q-test aggravated the headache intensity in migraine patients with throbbing pain, which supports our peripheral sensitization hypothesis. Studies have demonstrated that regardless of the type of stimulation (mechanical, electrical or temperature) on one's intracranial structures, the intracranial meninges can only sense pain (22, 23). Since the Q-test can induce congestion of the dural sinuses, mechanical stimulation is elicited on the intracranial meninges. Presence of throbbing pain indicates sensitization of peripheral trigeminovascular neurons innervating the meninges (9). Therefore, the Q-test, by increasing mechanical stimulation, aggravates the sensitized meninges and thus increases pain intensity. In this study, the Q-test effect was demonstrated in both sitting and supine positions, with a higher response for the latter position. This discrepancy is due to a difference in cerebral venous drainage in these two positions, with greater venous congestion in the supine position (2, 24).
Time sequence results were also related to the degree of the Q-test effect in this study. This may not be completely explained by venous congestion, and thus increased intracranial pressure by the Q-test must be considered. Yamazaki and Hirayama (25) in a human study have shown time-dependent increase of intracranial pressure within 30 s after the Q-test in the lateral recumbent position. Headache intensity was also related to the headache increment after the Q-test. The exact reason is unknown. One possible reason is that patients with severe headache intensity are likely to have a higher degree of sensitization and therefore may experience a higher headache increment during the Q-test.
In contrast, our study did not find any association between the development of BA and the Q-test effect in migraine patients. This result implies that central sensitization may not play a major role in producing the Q-test effect. However, our study only utilized dynamic allodynia (BA), not other types of allodynia, such as static, heat or cold allodynia, which might in part explain the low incidence of cutaneous allodynia in our patients even though the mean duration of migraine attacks was > 30 h. Without using thermal modalities, it is estimated that about 36% of patients with cutaneous allodynia could have been missed if the study adopted only mechanical allodynia (11). Therefore, the lack of association between the Q-test effect and the presence of cutaneous allodynia should be interpreted cautiously.
Our study has limitations. First, the average duration of migraine attacks (33.9 ± 19.2 h) in our participants was longer than the average appearance of cutaneous allodynia (1 h) (10). Therefore, we do not know if our findings still hold in what would be the early hours of a migraine attack. Second, throbbing pain is an index of peripheral sensation; however, it is subjective and cannot be reliably graded. Nevertheless, there is no other objective surrogate for peripheral sensitization during migraine attacks.
In summary, our study found the Q-test effect was related to peripheral but not central sensitization mechanisms during migraine attack. The Q-test effect may be used as an objective marker for peripheral sensitization in future clinical research or practice.
Footnotes
Competing interests
None to decalre.
Acknowledgements
This study was supported in part by grants from the Taiwan Headache Society and the Veterans Affairs Commission, Executive Yuan, Taiwan. We thank all patients who participated in this study, and Miss Ping-Hsuang Lin for her great help in data recording.