Pro-inflammatory and vasoconstricting prostanoid PGF 2α causes no headache in man

Abstract

Background: During two decades of migraine provocation studies with naturally occurring signalling molecules, vasodilators such as prostaglandin E₂, prostaglandin I₂ (prostacyclin) and prostaglandin D₂ were shown to be able to induce headache in man. To elucidate the role of inflammation and vasodilatation in the generation of headache, we investigated whether the pro-inflammatory and vasoconstricting prostanoid prostaglandin F_2α (PGF_2α) would cause headache in a human model of headache.

Methods: Twelve healthy volunteers were randomly allocated to receive 3.5 µg/kg/min PGF_2α or placebo over 20 min in a two-way crossover study. We recorded headache intensity on a verbal rating scale, middle cerebral artery blood flow velocity (V_MCA) and the diameters of the superficial temporal artery (STA) and radial artery (RA).

Results: We found no difference in the area under the curve (AUC) for immediate headache (0–90 min) between PGF_2α and placebo (p = 0.144). The McNemar's test showed no difference in the incidence of immediate and delayed headache between verum and placebo (p = 0.500 and p = 1.000, respectively). There was no difference in V_MCA (p = 0.776) and in the diameter of the STA (p = 0.460) or RA (p = 0.780) between PGF_2α and placebo.

Conclusion: The present study shows that PGF_2α, unlike vasodilating prostaglandins, does not provoke headache. We suggest that the vasodilating abilities of prostaglandins are important for the induction of experimental headache in healthy volunteers.

Keywords

headache prostaglandins vasoconstriction vasodilatation healthy volunteers

Introduction

Two highly discussed topics in migraine research today are the role of inflammation and vasodilatation. Non-steroidal anti-inflammatory substances are effective in the treatment of acute migraine (1), as well as in the prophylaxis of migraine (2), and a whole avenue of research has focused on neurogenic inflammation in migraine (3). Vasodilatation has a traditional role in migraine, but in recent years increasing evidence has suggested that it may not be as important as initially thought. Strongest among such evidence is the fact that calcitonin gene-related peptide (CGRP) receptor antagonists (4) and 5-hydroxytryptamine receptor 1F (5HT_1F) agonists (5) are effective in the acute treatment of migraine but have no vasoconstrictive action. We still favour that vasodilatation has some role in migraine pain because during two decades of migraine provocation studies with naturally occurring signalling molecules only vasodilators were able to induce migraine (6–8). In addition, the intravenous administration of norepinephrine (noradrenaline), well known for its vasoconstricting effect, did not cause headache or vascular responses in healthy volunteers (9).

Over the past 5 years we have systematically investigated the role of prostaglandins (PGs) in the generation of headache in healthy volunteers and migraine sufferers (8,10–12). The intravenous administration of vasodilative PGs, such as PGI₂ (prostacyclin), PGE₂ and PGD₂ induced headache in healthy volunteers and PGI₂ triggered migraine-like attacks in migraine sufferers (8,10–12). To what extent the dilatation of cerebral or extra-cerebral arteries contributes to PG-induced headache or migraine is still unknown. One might argue that PGs are inflammatory mediators (13) capable of activating sensory afferents and that PG-induced vasodilatation does not contribute to provoked headache. To elucidate the role of inflammation and vasodilatation we examined the effects of an infusion of PGF_2α, a product of arachidonic acid, which is involved in tissue inflammation (13) and pain processing (14). In contrast to the other PGs, PGF_2α acts as a vasoconstrictor (15). Elevated levels of urinary PGF_2α metabolite in children with migraine (16) and, more recently, increased ictal levels in the saliva of migraine sufferers have been reported (17). We tested the hypothesis that the pro-inflammatory and vasoconstricting prostanoid PGF_2α could cause headache.

Methods

Participants and study design

The study was designed as a randomised, double-blind, placebo-controlled, two-way intraindividual crossover study. Trial subjects were recruited through the www.forsogsperson.dk website (18), through the MOK.dk magazine for medical students of Copenhagen University, and by advertising on the bulletin boards of hospitals and higher education institutions. Twelve healthy volunteers (five male and seven female), mean age 24.4 years (range: 19–33 years) and mean weight 66 kg (range: 55–90 kg) completed the study. Participants had no history of migraine or any other type of headache (except for episodic tension-type headache less than once a month) and no previous serious somatic, psychiatric or infectious diseases. Physical and neurological examinations, and electrocardiography (ECG), were carried out on the day of enrolment.

The subjects were randomly assigned, by means of the www.randomization.com website (19), to receive 3.5 µg/kg/min PGF_2α (Prostin® F2 alpha; Pfizer Ltd, Tadworth, UK) or placebo (saline) over 20 min and crossed over to receive the other alternative treatment after a period of at least 1 week. The randomisation and preparation of the study drugs was carried out by medical staff not involved in the study. The randomisation code remained in Glostrup Hospital during the study and was not made available to the investigators until the study was completed. Before the main experiment we conducted a pilot study. Four different doses of PGF_2α (2 µg/kg/min, 2.5 µg/kg/min, 3 µg/kg/min and 3.5 µg/kg/min) were tested to determine the highest well-tolerated dose. The average maximum mean arterial pressure (MAP) increase during the 20-min injection of 2 µg/kg/min PGF_2α was 80 (60–89 safety limits, defined in the protocol). For the 2.5 µg/kg/min PGF_2α dose it was 82 (62–94), for the 3 µg/kg/min PGF_2α dose it was 88 (60–90) and for the 3.5 µg/kg/min PGF_2α dose it was 89 (61–91). We found that the maximal dose of 3.5 µg/kg/min PGF_2α was well tolerated and, therefore. it was used in the main experiment.

The study protocol was formally approved by the Ethics Committee of the City of Copenhagen (ref VEK H-B-2008-025) and by the Danish Data Protection Agency and was performed in accordance with the Declaration of Helsinki, as revised in 2008. The study was registered on the ClinicalTrials.gov website (www.clinicaltrials.gov) (20). All subjects gave written informed consent to participate in the study. The trial was conducted according to the protocol and good clinical practice.

Headache intensity and adverse events

To record headache intensity, a 10-point verbal rating scale (VRS) was used, where 0 indicated no headache; 1 indicated a different sensation, pounding or throbbing, but not necessarily painful; 5 indicated moderate headache; and 10 indicated the worst imaginable headache (21). Participants were encouraged to self-report any changes in their well-being during the study. Subjects were questioned about the presence of adverse events (AEs), headache and accompanying symptoms, according to the 2nd edition of the International Headache Classification (ICHD-2) (22), at T_-10, T₀ and then every 10 min until T_90. During the out-of-hospital period, defined as the period after discharge and until bedtime, all subjects were carefully instructed to make hourly recordings of headache and accompanying symptoms according to the ICHD-2 (22) and any other AEs. The AEs were classified as related to or not related to the study drug by the investigator. Subjects were allowed to take an analgesic medication of their own choice after consulting the study physician.

Transcranial Doppler ultrasonography and C-series ultrasound scan

Middle cerebral artery blood flow velocity (V_MCA) was recorded using transcranial Doppler ultrasonography (2 MHz) by means of handheld probes (Multi-Dop X digital; Compumedics DWL, Sipplingen, Germany) (8,10,11). The recordings were performed bilaterally and simultaneously and measurements of the partial pressure of end-tidal carbon dioxide (PetCO₂) were obtained with an open mask without any respiratory resistance (Propaq Encore®; Welch Allyn Protocol, Beaverton, Oregon, USA), as previously described (23). A fixed point was used with the best possible signal along the middle cerebral artery (MCA), as close as possible to the bifurcation of the anterior cerebral artery and MCA. The fix point was marked and noted and was reused in each participant for all recordings. All measurements were obtained by the same skilled laboratory technician.

A high-resolution, C-series ultrasound scanner (20 MHz, bandwidth 15 MHz; DermaScan C, Cortex Technology ApS, Hadsund, Denmark) was used to measure the diameter of the frontal branch of the left superficial temporal artery (STA) and the left radial artery (RA). All C-series scans were performed on the same location, as delimited by markings drawn on the skin. The coordinates of the marks were kept for reuse in the following trial days. All measurements within the same study subject were performed by the same skilled study personnel.

Experimental procedure

Participants had to completely avoid alcoholic beverages, caffeine, cocoa, chocolate and smoking for at least 8 h before the infusion was administered. The use of pharmacological agents apart from oral contraceptives was not permitted. Subjects fasted overnight and reported to the laboratory at 08:00 and were confirmed to be headache-free. Subjects rested in the supine position throughout the study period from time -10 min (T_-10) (the start of the in-hospital period) to T₉₀ (the end of the study period, 70 min post-infusion). The procedures were performed in a quiet room at room temperature between 23.3 and 24.1°C. Two intravenous Venflon™ catheters (Becton Dickinson AB, Stockholm, Sweden) were inserted into the antecubital vein for the infusion. The subjects rested for at least 30 min before the T_-10 values of V_MCA, the diameters of the STA and RA, MAP, heart rate (HR), PetCO₂, transcutaneous arterial oxygen saturation (SatO₂), ECG, headache score and AEs were recorded.

Following the baseline measurements, subjects were randomised to either PGF_2α or placebo. At T₀, the infusion was initiated by a time- and volume-controlled infusion pump (B. Braun Perfusor, Melsungen, Germany). All measurements were recorded at T_-10, T₀ and then every 10 min until T₉₀. MAP and HR were measured by an auto-inflatable cuff (Propaq Encore®; Welch Allyn Protocol, Beaverton, Oregon, USA). The ECG was recorded continually on paper using a Cardiofax V electrocardiograph (Nihon-Cohden Corporation, Tokyo, Japan).

Statistics

Vascular variables are presented as mean ± SD and as mean percentage from baseline. Headache scores are presented as median and quartiles. Baseline was defined as the average of T_-10 and T₀ before the start of the infusion. Immediate headache was defined as any headache during the in-hospital period (0–90 min) and delayed headache was defined as any headache occurring during the out-of-hospital period.

Sample size was calculated for a 90% power and a two-sided 5% significance level. Based on our previous studies with prostaglandins (8,11) we assumed the incidence of headache after PGF_2α as being 100% and after placebo as being 25%. The sample size was calculated as n = 21p(1−p)/(pA−pB)², where p = (pA + pB)/2, n = 21 × 0.625(1−0.625)/(0.75)² = 8.75 (24). As we also wanted to find a difference in the MCA of at least 10% between two study days we conducted an explorative sample size calculation: n = 2σ²/(μ₂ − μ₁)²× f (α, 1 − β) = 2 × 0.1²/(0, 1²) × 10.5 = 21 (σ = 10%, μ₂−μ₁ = 10%) (25). Therefore, 12 subjects were considered to be sufficient for a crossover study.

The data were baseline-corrected and the V_MCA data were also PetCO₂ corrected (26). This was followed by calculating the area under the curve (AUC) for the time period T₀–T₉₀ for V_MCA, and for the headache score, MAP, HR and PetCO₂, according to the trapezium rule (27).

The primary end points were the difference in incidence of headache and the AUC for headache score (AUC headache score) between PGF_2α and placebo. The secondary end points were the differences in the AUC for the VMCA (AUC _VMCA), STA (AUC_STA), RA (AUC _RA), PetCO₂ (AUC _PetCO2), MAP (AUC _MAP) and HR (AUC _HR) between the placebo day and the active day. To test the statistical difference between the variables we applied a paired, two-way t-test for the vascular data; the Wilcoxon signed-rank test was used for the AUC _{headache score} and the McNemar's test was used for the incidence of headache and AEs. To explore any possible changes over time for vascular variables we conducted a post hoc analysis by means of a repeated measures one-way ANOVA (which included the Dunnett's test). Five per cent (p < 0.05) was accepted as the level of significance. All analyses were performed with the PASW Statistics 18 for Windows software (SPSS Inc., Chicago, Illinois, USA).

Results

We recruited a total of 16 healthy volunteers. During the study, four participants withdrew. Therefore, 12 subjects were retained, as planned in the study protocol. Three of the four participants who withdrew wished to abort the infusion on either the first or second study day due to severe nausea or increased intestinal motility and discomfort. The infusion was stopped at approximately T₁₀ on the active day. By this time none of the participants had experienced headache. One participant reported headache on the placebo day. One participant was withdrawn from the study due to an increase in MAP of 36%, which was 16% over the safety limit (20%), as defined in the protocol.

There were no differences in baseline recordings for any variables between placebo and active day (Table 1). There were no differences in V_MCA between the left and the right side on the two study days (p = 0.148; p = 0.544).

Table 1.

Mean baseline values (± SD) of measured variables

Measured variables	Placebo	PGF_2α	P value^a
V_MCA (cm/s)	70.8 ± 11	71.6 ± 11.8	0.719
C-scan: temporal (mm)	1.1 ± 0.16	1.1 ± 0.19	0.922
C-scan: radial (mm)	2.5 ± 0.35	2.5 ± 0.32	0.647
Mean arterial blood pressure (mmHg)	72.2 ± 4.7	72.2 ± 4.5	0.998
Heart rate (bpm)	58 ± 6	58 ± 8	0.972
PetCO₂	35.2 ± 2.8	34.1 ± 2.5	0.119

PGF_2α: prostaglandin F2α, V_MCA: middle cerebral artery blood flow velocity (average of four cycles); PetCO₂: end-tidal partial pressure of carbon dioxide.

P value: difference in baseline variables between two experimental days (paired t-test).

Prostaglandin F2α-induced headache and vascular responses

The incidence of immediate and delayed headache is shown in Table 2. In all cases of headache it was reported as pressing and localised to the forehead. We found no difference in the AUC for immediate headache (0–90 min) between PGF_2α and placebo (p = 0.144) (Figure 1). There was no difference in the AUC_VMCA between PGF_2α and placebo (p = 0.776) (Figure 2A). Due to a difference in the AUC_PetCO2 between the placebo and PGF_2α day (p = 0.039), the MCA data were corrected with e^0,034 for each mmHg decrease in PetCO₂ (23). Exploratory ANOVA analysis revealed no changes over time in V_MCA on either study day (p > 0.05). We found no difference in the diameter of the STA or RA between the PGF_2α and the placebo day [AUC_STA difference (p = 0.460) and AUC_RA difference (p = 0.780)] (Figure 2B and C).

Figure 1.

Individual (□) and median (▪) headache scores on a verbal rating scale after intravenous infusion of prostaglandin F_2α (PGF_2α) and placebo in 12 healthy volunteers. There was no difference in the area under the curve (AUC) for headache (0–90 min) between PGF_2α and placebo (p = 0.144).

Figure 2.

Individual (□) and mean (▪) (A) middle cerebral arteries blood flow velocities (cm/s) (V_MCAs); (B) diameter (mm) of the superficial temporal artery (STA); (C) diameter (mm) of the radial artery (RA) after intravenous infusion of prostaglandin F_2α (PGF_2α) and placebo in 12 healthy volunteers. There was no difference in AUC _VMCA (p = 0.776), AUC_STA (p = 0.460) and AUC_RA (p = 0.780) between PGF_2α and placebo. AUC: area under the curve.

Table 2.

Incidence of prostaglandin F_2α-induced immediate and delayed headache in 12 healthy subjects

	Placebo	Prostaglandin F_2α	P value
Incidence of immediate headache^a	4	2	0.500
Incidence of delayed headache^a	2	3	1.000

The McNemar's test showed no difference in the incidence of immediate and delayed headache between the placebo and active day.

Vital signs and adverse events

The AUC_MAP and the AUC_HR were significantly larger on the PGF_2α day than on the placebo day (p < 0.0001) (Figure 3). The maximum mean increase in MAP from baseline by 18.9% [confidence interval (CI) 14.4–23.3] (Figure 4) was observed at T₂₀ and the maximum mean increase in HR from baseline by 16% (CI 10.7–21.3) was also observed at T₂₀ on the PGF_2α day.

Figure 3.

Mean MAP (▪) and HR (♦) after intravenous infusion of PGF_2α and placebo in 12 healthy volunteers. The AUCs for MAP and HR were significantly larger on the PGF_2α day compared with the placebo day (p < 0.0001). MAP: mean arterial pressure, HR: heart rate, PGF_2α: prostaglandin F_2α, AUC: area under the curve.

Figure 4.

Median headache (□) and mean percentage changes from baseline for the mean arterial blood pressure (MAP) (▪), middle cerebral artery blood flow velocity (V_MCA) (⋄), diameter of the superficial temporal artery (STA) (▴) and diameter of the radial artery (RA) (♦). PGF_2α: prostaglandin F_2α, VRS: verbal rating scale.

During the in-hospital period participants reported increased incidence of heat sensation on the PGF_2α day compared with the placebo day (p = 0.004) (Table 3). During the out-of-hospital period participants reported the following: tiredness (33%), yawning (8%), thirst (25%) and nausea (8%) on the PGF_2α day. One subject had diarrhoea after discharge from the hospital lasting until bedtime. On the placebo day, the following were reported: tiredness (25%), yawning (25%) and thirst (8%). No difference in the incidence of the above mentioned symptoms was found between the placebo and the active day (p > 0.05, McNemar’s test).

Table 3.

Adverse events reported and recorded during the in-hospital period (0–90 min)

	Placebo	PGF_2α
Immediate headache^b	4	2
Nausea^b	0	4
Photophobia^b	1	2
Phonophobia^b	0	1
Flushing^b	0	2
Heat sensation^a	2	11
Palpitation^b	0	0
Tightness in chest^b	0	5
Muscle pain^b	0	1
Chills^b	0	5
Increased mucus production in throat^b	0	4
Cough^b	0	2
Local irritation around site of injection^b	0	4
Low abdominal pain which mimics menstruation^b	0	4
Increased intestinal motility^b	0	1
Abdominal pain^b	0	6

PGF_2α: prostaglandin F_2α.

There was a difference in occurrence of heart sensation between placebo and PGF_2α (p = 0.004, McNemar’s test).

There was no difference between the two experimental days regarding other adverse events (p > 0.05, McNemar's test).

Mean percentage changes of vascular parameters and median headache on the PGF_2α and the placebo day are shown in Figure 4.

Discussion

The major outcome of the present study is that pro-inflammatory and vasoconstrictive PGF_2α, as opposed to vasodilating prostaglandins, failed to induce headache.

PGF2α and nociception

PGF_2α is one of the most abundant prostaglandins in the brain (28). Immunohistochemical studies reported the distribution of PGF synthase (PGFS) and PGF receptor (FP) in the central nervous system (29,30). PGFS immunoreactivity was found in the grey matter at all segmental levels in rat spinal cord and immunoelectron microscopy revealed increased immunoreactivity in neuronal dendrites (29). Yoshikawa et al. reported increased PGFS immunoreactivity in the white matter of the brain and spinal cord of mice (30). Furthermore, functional FP receptors were found in the spinal cord of mice (31) and mRNA for the FP receptor was expressed in cultured rat astrocytes and oligodendrocytes (32).

Intrathecal administration of PGF_2α caused touch-evoked allodynia in mice (33–35) and dose-dependent mechanical hyperalgesia in rats (36). PGF_2α-induced mechanical allodynia was attenuated in mice lacking the FP receptor (FP^-/-) (37). Electrophysiological experiments showed that PGF_2α-induced mechanical hyperalgesia corresponded well with a dose-dependent increases in mechanically evoked discharges of spinal nociceptive-specific neurons (36). Based on these data and on the ability of PGs to cross the blood–brain barrier (38), we expected that PGF_2α, as a pro-inflammatory mediator, could induce headache in healthy volunteers. However, the data collected in our study showed that PGF_2α did not cause headache. In the next section we discuss why PGF_2α, unlike other pro-inflammatory PGs, failed to induce headache.

Why did PGF2α fail to induce headache?

The role of PGE₂ and PGI₂ in nociception is well established (14). Thus, PGE₂ contributes to the inflammation-induced sensitisation of sensory neurons (39,40) and to the mediation of allodynia (41). Degranulation of mast cells causes PGI₂-induced activation and the sensitisation of meningeal sensory afferents (42,43). In contrast, PGD₂ is not known to activate or sensitise dural afferents (43). However, it dilates cerebral arteries and induces a mild headache in healthy volunteers (12). To show the differences between PGs in headache and vascular responses we have listed these responses in Table 4. During PGE_2, PGI₂ and PGD₂ infusion the maximum velocity drop in the MCA, compared to baseline, was 14.3% (range:−4.6–28.3%) at T₁₀ and T₂₀ (8,10–12). The STA dilated by 37.7% (range: 23.5–55.7%) and the RA dilated by 9.15% (range: 4.4–16.3%) (8,10–12). PGI₂ caused significant dilatation of extra-cranial arteries (10), whereas PGE₂ and PGD₂ caused significant dilatation of both extra- and intra-cranial arteries (11,12). In the current study we observed no vascular responses after the infusion of PGF_2α in either intra- or extra-cranial arteries. Taken together, previous and present data suggest that the pronociceptive abilities of PGs alone are not sufficient to induce headache and that PG-induced vasodilatation may play a crucial role in the generation of head pain.

Table 4.

Incidence of immediate headache during infusion of PGE_2, PGI_2, PGD₂ and PGF_2α and maximum per cent changes in V_MCA and the diameter of the STA

	Immediate headache	V_MCA	STA
PGE₂ (11)	11 (11)^a	−13.9% (−20.6–7.3)^a	23.5% (14–37.8)^a
PGI₂ (8)	11 (12)^a	−10.5% (−14.9–6.0)^a	32.9% (26.4–39.4)^a
PGD₂ (12)	10 (12)^a	−28.3% (−33.6–22.9)^a	55.7% (46.7–64.6)^a
PGF_2α	2 (12)^b	−2.7% (−8.1–2.6)^b	10.9% (5.9–15.8)^b

PGE₂: prostaglandin E₂, PGI₂: prostaglandin I₂ (prostacyclin), PGD₂: prostaglandin D₂, PGF_2α: prostaglandin F_2α, V_MCA: middle cerebral artery blood flow velocity, STA: superficial temporal artery.

Significant from the placebo day. ^bNot significant from the placebo day.

PGF2α-induced vascular responses in humans

PGF_2α constricted isolated canine basilar, middle cerebral (44) and posterior communicating arteries (45), as well as the basilar artery in the rhesus monkey (46) and bovine MCA (47). Several human in vitro studies also reported PGF_2α-induced contraction of the middle cerebral and basilar arteries of infants (gestational age: 30–40 weeks) (48) and PGF_2α-induced contraction of adult pial (49), temporal (50) and radial arteries (51,52). The majority of studies which used in vivo cranial window models in rats and cats showed constriction of the basilar (53) and pial arteries (54,55) after extraluminal application of PGF_2α. Interestingly, one open cranial window study in cats demonstrated dilatation of small and large feline pial arterioles during topical application of PGF_2α (56). Another study which used a closed cranial window model demonstrated no response from the same vessels (57). In contrast to preclinical studies, the present study of intravenous PGF_2α in human volunteers did not show constriction of the investigated brain arteries. Since we observed a significant increase in blood pressure during PGF_2α infusion, a specific immediate constriction of arterioles must have been induced by PGF_2α. This is in agreement with a previous study in FP (^−/−) mice which showed abrogation of an immediate, dose-dependent increase in blood pressure during PGF_2α infusion (58). Although there are some in vitro human studies showing constriction of both cerebral and extra-cerebral arteries in the presence of PGF_2α (48–52), little is known about FP receptor distribution in human intra- and extra-cranial arteries. Our finding of a lack of arterial constriction after PGF_2α administration might be explained by species-specific responses to PGF_2α exposure in vivo or possibly to a simultaneous activation of other dilating prostanoid receptors. Thus, a crosstalk between FP and PGE₂ (EP₂) receptors could result in an increase of cyclic AMP (cAMP) (59) and an attenuation or decrease of vasoconstriction caused by the FP receptor.

Conclusion

The present study demonstrates that PGF_2α, as opposed to PGE_2, PGI₂ and PGD₂, does not induce headache. The apparent lack of a dilating effect of PGF_2α on cerebral arteries could explain the absence of headache. Based on these data we suggest that the vasodilating abilities of PGs are important in generating pain in healthy volunteers.

Footnotes

Acknowledgements

We thank Lene Elkjær and Winnie Grønning for their excellent technical support.

References

Tfelt-Hansen

. Triptans vs other drugs for acute migraine. Are there differences in efficacy? A comment. Headache 2008; 48: 601–605.

Evers

. Treatment of migraine with prophylactic drugs. Expert Opin Pharmacother 2008; 9: 2565–2573.

Goadsby

. Pathophysiology of migraine. Neurol Clin 2009; 27: 335–360.

Edvinsson

Goadsby

. CGRP and its receptors provide new insights into migraine pathophysiology. Nat Rev Neurol 2010; 6: 573–582.

Ramadan

Skljarevski

Phebus

Johnson

. 5-HT1F receptor agonists in acute migraine treatment: a hypothesis. Cephalalgia 2003; 23: 776–785.

Asghar

Hansen

Amin

. Evidence for a vascular factor in migraine. Ann Neurol 2011; 69: 635–645.

Schytz

Birk

Wienecke

Kruuse

Olesen

Ashina

. PACAP38 induces migraine-like attacks in patients with migraine without aura. Brain 2009; 132: 16–25.

Wienecke

Olesen

Ashina

. Prostaglandin I2 (epoprostenol) triggers migraine-like attacks in migraineurs. Cephalalgia 2010; 30: 179–190.

Lindholt

Petersen

Tvedskov

Iversen

Olesen

Tfelt-Hansen

. Lack of effect of norepinephrine on cranial haemodynamics and headache in healthy volunteers. Cephalalgia 2009; 29: 384–387.

10.

Wienecke

Olesen

Oturai

Ashina

. Prostacyclin (epoprostenol) induces headache in healthy subjects. Pain 2008; 139: 106–116.

11.

Wienecke

Olesen

Oturai

Ashina

. Prostaglandin E2(PGE2) induces headache in healthy subjects. Cephalalgia 2009; 29: 509–519.

12.

Wienecke

Olesen

Ashina

. Discrepancy between strong cephalic arterial dilatation and mild headache caused by prostaglandin D₂ (PGD₂). Cephalalgia 2011; 31: 65–76.

13.

Hata

Breyer

. Pharmacology and signaling of prostaglandin receptors: multiple roles in inflammation and immune modulation. Pharmacol Ther 2004; 103: 147–166.

14.

Ito

Okuda-Ashitaka

Minami

. Central and peripheral roles of prostaglandins in pain and their interactions with novel neuropeptides nociceptin and nocistatin. Neurosci Res 2001; 41: 299–332.

15.

Vysniauskiene

Allemann

Flammer

Haefliger

. Vasoactive responses of U46619, PGF2alpha, latanoprost, and travoprost in isolated porcine ciliary arteries. Invest Ophthalmol Vis Sci 2006; 47: 295–298.

16.

Uchiyama

Sakai

. Increased main urinary metabolite of prostaglandin F2 alpha excretion in childhood migraine. Arch Dis Child 1988; 63: 342–342.

17.

Durham

Vause

Derosier

McDonald

Cady

Martin

. Changes in salivary prostaglandin levels during menstrual migraine with associated dysmenorrhea. Headache 2010; 50: 844–851.

18.

Forsogsperson.dk, Trial volunteers, http://www.forsogsperson.dk (accessed August 2011).

19.

Randomization.com, Randomization plan generators, http://www.randomization.com (accessed August 2011).

20.

US National Institutes of Health, ClinicalTrials.gov, http://www.clinicaltrials.gov (accessed August 2011).

21.

Iversen

Olesen

Tfelt-Hansen

. Intravenous nitroglycerin as an experimental model of vascular headache. Basic characteristics. Pain 1989; 38: 17–24.

22.

Olesen

Steiner

. The International classification of headache disorders, 2nd edn (ICDH-II). J Neurol Neurosurg Psychiatry 2004; 75: 808–811.

23.

Thomsen

Iversen

. Experimental and biological variation of three-dimensional transcranial Doppler measurements. J Appl Physiol 1993; 75: 2805–2810.

24.

Campbell

Julious

Altman

. Estimating sample sizes for binary, ordered categorical, and continuous outcomes in two group comparisons. BMJ 1995; 311: 1145–1148.

25.

Pocock

. Clinical trials: a practical approach. Chichester and New York: John Wiley and Sons, 1983.

26.

Markwalder

Grolimund

Seiler

Roth

Aaslid

. Dependency of blood flow velocity in the middle cerebral artery on end-tidal carbon dioxide partial pressure–a transcranial ultrasound Doppler study. J Cereb Blood Flow Metab 1984; 4: 368–372.

27.

Matthews

Altman

Campbell

Royston

. Analysis of serial measurements in medical research. BMJ 1990; 300: 230–235.

28.

Yoshikawa

Kita

Kishimoto

Shimizu

. Profiling of eicosanoid production in the rat hippocampus during kainic acid-induced seizure: dual phase regulation and differential involvement of COX-1 and COX-2. J Biol Chem 2006; 281: 14663–14669.

29.

Suzuki-Yamamoto

Toida

Watanabe

Ishimura

. Immunocytochemical localization of prostaglandin F synthase II in the rat spinal cord. Brain Res 2003; 969: 27–35.

30.

Yoshikawa

Takei

Hasegawa-Ishii

. Preferential localization of prostamide/prostaglandin F synthase in myelin sheaths of the central nervous system. Brain Res 2011; 1367: 22–32.

31.

Muratani

Nishizawa

Matsumura

. Functional characterization of prostaglandin F2alpha receptor in the spinal cord for tactile pain (allodynia). J Neurochem 2003; 86: 374–382.

32.

Kitanaka

Hashimoto

Gotoh

. Expression pattern of messenger RNAs for prostanoid receptors in glial cell cultures. Brain Res 1996; 707: 282–287.

33.

Minami

Uda

Horiguchi

Ito

Hyodo

Hayaishi

. Allodynia evoked by intrathecal administration of prostaglandin F2 alpha to conscious mice. Pain 1992; 50: 223–229.

34.

Minami

Nishihara

Ito

Sakamoto

Hyodo

Hayaishi

. Nitric oxide mediates allodynia induced by intrathecal administration of prostaglandin E2 or prostaglandin F2 alpha in conscious mice. Pain 1995; 61: 285–290.

35.

Minami

Matsumura

Okuda-Ashitaka

. Characterization of the glutamatergic system for induction and maintenance of allodynia. Brain Res 2001; 895: 178–185.

36.

Turnbach

Spraggins

Randich

. Spinal administration of prostaglandin E(2) or prostaglandin F(2alpha) primarily produces mechanical hyperalgesia that is mediated by nociceptive specific spinal dorsal horn neurons. Pain 2002; 97: 33–45.

37.

Kunori

Matsumura

Mabuchi

. Involvement of prostaglandin F 2 alpha receptor in ATP-induced mechanical allodynia. Neuroscience 2009; 163: 362–371.

38.

Bito

Wallenstein

. Transport of prostaglandins across the blood–brain and blood–aqueous barriers and the physiological significance of these absorptive transport processes. Exp Eye Res 1977; 25(Suppl): 229–243.

39.

Lin

Amaya

Barrett

. Prostaglandin E2 receptor EP4 contributes to inflammatory pain hypersensitivity. J Pharmacol Exp Ther 2006; 319: 1096–1103.

40.

Southall

Vasko

. Prostaglandin receptor subtypes, EP3C and EP4, mediate the prostaglandin E2-induced cAMP production and sensitization of sensory neurons. J Biol Chem 2001; 276: 16083–16091.

41.

Okuda-Ashitaka

Minami

Matsumura

. The opioid peptide nociceptin/orphanin FQ mediates prostaglandin E2-induced allodynia, tactile pain associated with nerve injury. Eur J Neurosci 2006; 23: 995–1004.

42.

Strassman

Raymond

Burstein

. Sensitization of meningeal sensory neurons and the origin of headaches. Nature 1996; 384: 560–564.

43.

Zhang

Strassman

Burstein

Levy

. Sensitization and activation of intracranial meningeal nociceptors by mast cell mediators. J Pharmacol Exp Ther 2007; 322: 806–812.

44.

Ayajiki

Fujioka

Noda

Okamura

Toda

. Modifications by sumatriptan and acetylcholine of nitric oxide-mediated neurogenic dilatation in dog cerebral arteries. Eur J Pharmacol 2001; 420: 67–72.

45.

Toda

Onoue

. Prostaglandin F2 alpha-induced contraction in isolated dog cerebral, coronary and mesenteric arteries soaked in excess potassium. Jpn J Pharmacol 1988; 47: 204–207.

46.

Sipkema

van der Linden

Fanton

Latham

. Responses to mechanical stimuli of isolated basilar and femoral arteries of the Rhesus monkey are different. Heart Vessels 1996; 11: 18–26.

47.

Wendling

Harakal

. Effects of prostaglandin F2 alpha and thromboxane A2 analogue on bovine cerebral arterial tone and calcium fluxes. Stroke 1991; 22: 66–72.

48.

Bevan

Dodge

Nichols

. Responsiveness of human infant cerebral arteries to sympathetic nerve stimulation and vasoactive agents. Pediatr Res 1998; 44: 730–739.

49.

Wallerstedt

Reinstrup

Uski

Bodelsson

. Effects of propofol on isolated human pial arteries. Acta Anaesthesiol Scand 1999; 43: 1065–1068.

50.

White

Leseth

Möller

. Interleukin-1beta potentiates endothelin ET(B) receptor-mediated contraction in cultured segments of human temporal artery. Regul Pept 1999; 81: 89–95.

51.

Yang

. Inhibition of vasoconstriction by the thromboxane A2 antagonist GR32191B in the human radial artery. Br J Clin Pharmacol 1999; 48: 207–215.

52.

Ryszawa

Rudziński

Piatek

. Different reactivity of the proximal and distal segments of the radial artery to vasoconstrictors in patients undergoing coronary artery bypass grafting. Kardiol Pol 2007; 65: 1313–1319.

53.

Murray

Faraci

Heistad

. Signal transduction pathways in constriction of the basilar artery in vivo. Hypertension 1992; 19: 739–742.

54.

Kapp

Robertson

White

. Spasmogenic qualities of prostaglandin F2alpha in the cat. J Neurosurg 1976; 44: 173–175.

55.

Rosenblum

. Constriction of pial arterioles produced by prostaglandin F2alpha. Stroke 1975; 6: 293–297.

56.

Wahl

Schilling

Whalley

. Cerebrovascular effects of prostanoids. In-situ studies in pial arteries of the cat. Naunyn Schmiedebergs Arch Pharmacol 1989; 340: 314–320.

57.

Ellis

Wei

Cockrell

Choi

Kontos

. The effect of PGF2 alpha on in vivo cerebral arteriolar diameter in cats and rats. Prostaglandins 1983; 26: 917–923.

58.

Lucitt

Stubbe

. Prostaglandin F2alpha elevates blood pressure and promotes atherosclerosis. Proc Natl Acad Sci U S A 2009; 106: 7985–7990.

59.

Abera

Sales

Catalano

Katz

Jabbour

. EP2 receptor mediated cAMP release is augmented by PGF 2 alpha activation of the FP receptor via the calcium-calmodulin pathway. Cell Signal 2010; 22: 71–79.