Headache and prolonged dilatation of the middle meningeal artery by PACAP38 in healthy volunteers

Participants

We recruited 15 healthy volunteers (8 female and 7 male; mean age 25 years (range 20–28) and mean weight 73 kg (range 54–95 kg)). Exclusion criteria were: a history of serious somatic disease; migraine or any other type of headache (except episodic tension-type headache less than once a month); daily intake of any medication except contraceptives; and contraindications for MRI scan. All female participants used safe contraceptive methods. All participants gave informed consent to participate. The Ethical Committee of Copenhagen approved the study (H-1-2008-108), which was conducted in accordance with the Helsinki II Declaration of 1964, as revised in Edinburgh in 2000.

Design and randomization

15 participants were randomly allocated in a 2:1 order to receive 10 pmol/kg/min PACAP38 (n = 10) (Calbiochem®, Darmstadt, Germany) or placebo (n = 5) (isotonic saline) for 20 min. The chosen PACAP38 dose has previously been shown to induce headache and haemodynamic changes in the brain in healthy volunteers and migraine without aura patients (6,7). The reason for selecting an unequal allocation was that we expected no placebo response on the vessels. Furthermore, a small placebo group would be sufficient to blind the researcher.

Participants who reported a delayed headache of intensity ≥ 2 on a verbal rating scale (VRS) from 0 to 10 received a subcutaneous injection of sumatriptan 6 mg (Imigran® injection, Glaxo Wellcome Operations, Barnard Castle, UK). The data analysis was performed blindly with respect to both the experimental drug and the time sequence of the MRA images.

Experimental procedures

All participants reported to the laboratory headache free. Coffee, tea, cocoa or other methylxanthine-containing foods or beverages, and tobacco were not allowed for at least 12 h before start of the study. Participants were placed in the supine position in the MR scanner room and a venous catheter (Venflon, Becton Dickinson Infusion Therapy AB, Sweden) was inserted into the left antecubital vein for infusion. We collected blood samples to determine the baseline haematocrit. The participants were monitored with ECG (by ECG monitoring function attached to the MR scanner), end-tidal partial pressure of CO₂ (P_etCO₂) (capnograph, Datex, Finland), mean arterial blood pressure (MAP) and heart rate (HR) (Veris monitor, Medrad, USA).

MR imaging was performed on a 3.0 Tesla Philips Achieva Scanner (Philips Medical Systems, Best, The Netherlands) using an eight-element phased-array receiver head coil. We obtained repeated MRA measurements covering the MMA and MCA before and after PACAP38/placebo infusion (Figure 1). All other variables were recorded at fixed time points (baseline, every 10 min in the period 0–40 min, at 70, 90, 110, 140 min, and every 30 min in the period 3–5 h) throughout the study (Figure 1). Based on previous data (6) we defined immediate (0–119 min) and delayed (2 to 5 h) phases. Headache intensity was recorded on a verbal rating scale (VRS) from 0 to 10 (0, no headache; 1, a feeling of pressure; 10, worst imaginable headache). We expected that PACAP38 would induce minimum of 1 on the VRS in all participants during the immediate and delayed phases (2 to 5 h). Participants who reported 2 or higher on the VRS were treated with the migraine-specific abortive drug sumatriptan (a 5-HT_1B/D agonist) and they underwent additional scans to investigate the effect of sumatriptan on arteries (Figure 1). OPEN IN VIEWER

If the participants fulfilled IHS criteria C, D and E for migraine without aura (10), they were characterized as having migraine-like attacks.

Data acquisition and imaging protocols

For vessel imaging, a 3D inflow gradient echo sequence was used. First, a scout MR-angiography (MRA) was performed using field-of-view (FOV) 180 × 180 × 120 mm³; acquired matrix size (MxP) of 120 × 120, acquired voxel resolution 1.5 × 1.5  × 2.4 mm³, reconstructed resolution 0.70 × 0.70 × 1.20 mm³, repetition time (TR) 23 ms; echo time (TE) 5.4 ms; flip angle 20°; SENSE factor 2; two chunks; total scan duration of 73 s. The scout MRA was used to plan the subsequent MCA scans. The MRA of MCA used: FOV 200 × 200 × 74 mm³; MxP 800 × 406, acquired voxel resolution 0.25 × 0.49 × 1.00 mm³; reconstructed resolution 0.20 × 0.20 × 0.50 mm³; TR 25 ms; TE 3.5 ms; flip angle 20°; SENSE factor 2; four chunks; total scan time of 9 min 03 s. The first acquired angiography (baseline) had a large vertical FOV to ensure that the MCA was imaged, and was used to place the subsequent angiography of MMA. MMA scans had FOV 200 × 200 × 16 mm³; MxP 800 × 571; acquired voxel resolution of 0.25 × 0.35 × 0.70 mm³; reconstructed resolution 0.20 × 0.20 × 0.35 mm³; TR 25 ms; TE 3.5 ms; flip angle 20°; SENSE factor 2; two chunks; total scan duration of 5 min 29 s. When the participant was not removed from the MR scanner between two angiographies (between baseline (T₀) and 20 min (T₂₀), and between before and after sumatriptan), we used a MCA scan with shorter FOV 200 × 200 × 37 with total scan duration time 4 min 31 s.

Data analysis and statistics

Effect of PACAP38 on MMA and MCA during immediate phase

ANOVA showed significant changes in MMA circumference after infusion of PACAP38 over time (p < 0.00004) but not after placebo (p = 0.27). Post hoc analysis revealed significant dilatation of the MMA at 20 min: 5.97 mm± 0.14 (p = 0.0003) and at 110 min: 5.76 mm± 0.10 (p = 0.001), after start of PACAP38 infusion compared with baseline: 4.85 mm± 0.11. We found no difference in MMA circumference at 20 min: 4.77 mm± 0.38 (p = 0.805) or at 110 min: 4.99 mm± 0.31 (p = 0.130) after placebo compared with baseline: 4.73 ± 0.40. Mean percent dilatation and CI are shown in Figure 2. The AUC_{0–110 min} for MMA was significantly larger after PACAP38: 102.3 ± 13.8, compared with placebo: 13.8 ± 11.2 (p = 0.001) (Figure 2). OPEN IN VIEWER

ANOVA showed no statistical changes in MCA circumference after infusion of PACAP38 over time (p = 0.39) or after placebo (p = 0.60). Post hoc analysis showed no changes of the MCA at 20 min: 9.27 mm± 0.43 (p = 6.02) or at 110 min: 9.13 mm± 0.41 (p = 0.458) after infusion start compared with baseline: 9.34 mm± 0.43. Placebo infusion showed no MCA changes either at 20 min: 9.93 mm± 0.47 (p = 0.650) or at 110 min: 9.69 mm± 0.35 (p = 0.641) compared with baseline: 9.82 mm± 0.54. Mean percentage changes and CI are shown in Figure 3. The AUC_{0–110 min} for the MCA showed no difference between PACAP38: –17.4 ± 16.2, and placebo: 0.33 ± 23.9, p = 0.536 (Figure 3). OPEN IN VIEWER

MMA and MCA circumference before and after sumatriptan

Five participants (56%) reported headache intensity ≥ 2 (range 2 to 6 on VRS) during the delayed phase. Four participants were scanned before and after injection of sumatriptan 6 mg subcutaneously, as one participant declined the administration of sumatriptan.

MMA circumference was significantly reduced 20 min after sumatriptan: 5.21 mm± 0.08 compared with the pre-treatment scan: 5.79 mm± 0.23 (p = 0.045). MCA remained unchanged 20 min after sumatriptan: 9.31 mm± 0.63 compared with the pre-treatment recording: 9.19 mm± 0.50 (p = 0.78) (Figure 4). All participants reported pain free at 50 min (mean time) after sumatriptan. OPEN IN VIEWER

OPEN IN VIEWER

Figure 5.

Individual and median (thick lines) headache intensity after PACAP38 (A) and after placebo (B) infusion. Participants who reported VRS ≥ 2 and received sumatriptan (Group A on Figure 1) are not shown in this figure.

Effect of PACAP38 on heart rate, MAP and end-tidal CO₂

During the immediate phase, the AUC_HR was significantly larger after PACAP38, compared with placebo (p = 0.003). There was no difference between PACAP38 and placebo with regard to AUC_MAP and AUC_PetCO2 (p > 0.05) during the immediate phase (0–119 min) (Table 3).

OPEN IN VIEWER

Table 3.

Adverse events after PACAP38 and placebo infusion

	PACAP38 (n = 9)	Placebo (n = 5)
Immediate phase (0–119 min)
Heat sensation	9 (100%)	3 (60%)
Palpitation	9 (100%)	0 (0%)
Flushing	9 (100%)	0 (0%)
Delayed phase (2–5 h)
Heat sensation	8 (89%)	2 (40%)
Palpitation	1 (11%)	1 (20%)
Flushing	8 (89%)	1 (20%)

There was no difference in AUC_MAP, AUC_HR and AUC_PetCO2 between PACAP38 and placebo (p > 0.05) during the delayed phase (2–5 h) (Table 3).

Effect of PACAP38 on extra and intracerebral arteries in humans

PACAP38 dilates human cerebral arteries (6,7) and acts as a neuromodulator (3). Birk and colleagues systematically examined cerebral haemodynamics in response to PACAP38 infusion in healthy volunteers (6). The authors reported headache and a dilatation of the MCA immediately after PACAP38 infusion. This important finding was followed by an investigation of PACAP38 in migraine patients (7). Schytz et al. found a short-lasting dilatation (8.5%) of the MCA, and a marked (37%) and prolonged (90 min) dilatation of the superficial temporal artery (STA) (7). In addition, PACAP38 caused a delayed headache and migraine-like attacks in patients and controls (6,7). However, a possible relationship between head pain and vasodilatation after PACAP38 remains unclear. The above mentioned studies used transcranial Doppler (TCD) combined with single photon emission computed tomography (SPECT) that have a few major limitations. First, the measurement of the MCA diameter was calculated from blood velocity in the MCA and cerebral blood flow and hence indirect. Secondly, the SPECT method is influenced by extracranial flow due to scattered radiation (Compton scatter). Finally, TCD cannot be used to explore a possible role of meningeal blood vessels (e.g. the MMA). The latter is very important as MMA is the principal artery supplying the dura mater, which is a pain-sensitive intracranial structure, suggested as a primary site of nociception in migraine pain (13–15). However, MRA is a direct way to measure diameter of arteries and superior to above mentioned methods.

PACAP38 induced vasodilatation and headache

In the present study we applied MRA to examine the MMA and the MCA directly with regard to the arterial circumference. We found PACAP38 to induce a marked and sustained dilatation of the MMA up to 23% and co-occurring with headache, whereas the MCA diameter remained unchanged. However, the most interesting observation was that dilatation of the MMA lasted more than 5 h, was associated with delayed headache and was reversed after administration of sumatriptan.

PACAP38 has to cross the blood brain barrier (BBB) to reach its receptors in the smooth muscle cells in the MCA (4,16,26). The transport of PACAP38 across the BBB is dependent on a saturable carrier mediated transport system (16). Only a very small percentage (0.053%) PACAP38 does, however, cross the BBB in mice in the 5 min period after intravenous administration (16). Birk et al. (6) reported no major direct effect on rCBF after PACAP38 infusion in healthy volunteers after CBF was corrected for hyperventilation. This suggests that a lack of vasodilatation of the MCA after PACAP38 could be explained partly by its inability to cross the BBB in humans.

PACAP38 exerts its effect via the VPAC1, VPAC2 and PAC1 receptors (17,18). VPAC1 is considered to be the main receptor responsible for the vasodilatation (19,20), whereas VPAC2 is primarily responsible for mast cell degranulation as VPAC1 is not expressed on human mast cells (21). Interestingly, PACAP38 and vasoactive intestinal polypeptide (VIP) belong to the same secretin/glucagon family and both share their effect on VPAC1 and VPAC2 receptors (1,2). In a study of cutaneous pain in humans, PACAP38 and VIP showed similar responses with regard to short-lasting vasodilatation and neurogenic inflammation in skin (22). However, provocation studies using intravenous infusion revealed differences between PACAP38 and VIP. Although both peptides have a short half life (PACAP38: 3.5 min (7); VIP: 1.0 min (23)) and plasma concentration of both returned to baseline 2 h after start of infusion we observed differences in vascular and headache responses. Thus, dilatation of the superficial temporal artery was relatively short-lasting after VIP (24) and sustained after PACAP38 throughout the observation period of 90 min (7). Moreover, VIP failed to induce migraine attacks (25), whereas PACAP38-induced delayed migraine-like attacks (7). These data suggest that long-lasting vasodilatation and delayed migraine after PACAP38 are unlikely to be due to an effect on VPAC1 and VPAC2 receptors. However, PACAP38, in contrast to VIP, also acts on the PAC1 receptor. Therefore, it is possible that the present data of marked and sustained effect of PACAP38 on the MMA and delayed headaches is due its selective effect on the PAC1 receptor. The present findings could also indicate that the delayed headache and migraine-like attacks might be due to a sustained dilatation of extracerebral arteries.

Another interesting observation in the present study is that anti-migraine drug sumatriptan reversed PACAP38-induced dilatation of the MMA and headache. We did not observe any effect of sumatriptan on the MCA. The same pattern was reported after CGRP administration in healthy volunteers (8). Taken together these findings suggest that head pain may arise from the extracerebral arteries and that contraction of the MMA, but not the MCA, has a role in headache relief after sumatriptan.