Sage Journals: Discover world-class research

Abstract

Prostacyclin [prostaglandin I₂ (PGI₂)] activates and sensitizes meningeal sensory afferents. In healthy subjects PGI₂ triggers headache in healthy subjects. However, the migraine-eliciting effect of PGI₂ has not been systematically studied in patients with migraine. We hypothesized that intravenous infusion of the stable prostacyclin analogue epoprostenol would trigger migraine-like attacks in migraineurs. We infused 10 ng kg⁻¹ min⁻¹ PGI₂ or placebo over 25 min in 12 migraineurs without aura in a controlled, double-blind, cross-over study and recorded headache intensity and associated symptons, velocity in the middle cerebral artery (V_MCA) and diameter in the superficial temporal artery. In the period 0–14 h, 12 subjects reported headache on PGI₂ day compared with three subjects on placebo day (P = 0.004), and six subjects fulfilled the criteria for an experimentally induced migraine-like attack compared with two subjects on placebo (P = 0.219). During infusion and post-infusion phases the AUC under the headache curve on PGI₂ was significantly larger than on placebo (P < 0.05). There was a significant V_MCA decrease (P = 0.015) and superficial temporal artery diameter increase (P < 0.001) on PGI₂ compared with placebo. In conclusion, PGI₂ may trigger a migraine-like attack in migraine sufferers. We suggest sensitization of perivascular nociceptors and arterial dilation as the mode of action of PGI₂-induced headache and migraine-like attacks.

Keywords

Migraine prostacyclin sensitization nociception

Introduction

The headache- or migraine-provoking properties of naturally occurring signalling molecules have provided new insights into the pathophysiology of migraine (1). Human models of migraine (2,3) have predicted efficacy of nitric oxide synthase inhibition (4) and calcitonin gene-related peptide receptor blockade (5) and subsequently there has been proof of efficacy in Phase II trials of these compounds (6).

Although non-steroidal anti-inflammatory drugs are widely used in the treatment of migraine (7,8), the role of prostanoids in the pathogenesis of migraine and their possible migraine-eliciting effect are not fully clarified. Prostanoids are important inflammatory mediators involved in pain transmission (9). Prostaglandin I₂ (PGI₂ or prostacyclin) is a member of the prostaglandin family of lipid mediators produced by endothelial cells (10), mast cells (11) and astrocytes (12–15). PGI₂ is a potent vasodilator and an important mediator of oedema and inflammatory pain (10). Of particular relevance for headache is the release of PGI₂ upon mast cell degranulation (11) causing activation and sensitization of meningeal sensory afferents (16,17). Recently, we examined the headache-eliciting effect of PGI₂ in healthy subjects (18). PGI₂ induced a mild to moderate headache associated with dilation of cranial arteries (18). Given that headache-inducing substances in healthy volunteers have the potential to induce migraine in migraine sufferers (19), we hypothesized that PGI₂ may trigger migraine attacks in migraineurs. PGI₂ is a chemically unstable molecule, but a stable analogue is available as a freeze-dried preparation (epoprostenol) for intravenous administration in man (20). Therefore, we studied the migraine-eliciting effect of epoprostenol in patients with migraine without aura in a double-blind crossover study. In addition, we aimed to study the vascular effect of epoprostenol in relation to possible migraine induction.

Design and methods

Main experiment

We recruited 12 patients with migraine without aura (MoA) (four male and eight female) but otherwise healthy, mean age 35.3 years (range 20–53 years). Exclusion criteria were: any other type of headache (except episodic tension-type headache less than three times a month); any daily medication apart from oral contraceptives; serious somatic or psychiatric diseases. The study was approved by the Ethics Committee of the County of Copenhagen (KA-20060086), Danish Medicines Agency and the Danish Data Protection Agency, and was undertaken in accordance with the Helsinki Declaration of 1964, as revised in Edinburgh in 2000. The study was registered on www.clinicaltrials.gov and monitored by the Good Clinical Practice unit at Copenhagen University Hospital. All subjects gave informed consent to participate.

Experimental design

In a double blind, placebo-controlled, crossover design, the subjects were randomly allocated to receive PGI₂ 10 ng kg⁻¹ min⁻¹ or placebo (isotonic saline) over 25 min on 2 days, separated by at least a week. The central pharmacy performed the randomization and prepared the study drug. The randomization code remained in the hospital during the study and was not available to the investigators until the study was complete.

All subjects reported to the laboratory at 08.30 h headache free. The subjects had to be without migraine or tension-type headache 5 days before start of each study day to avoid any sensitization of neurons. Coffee, tea, cocoa, alcohol or other methylxanthine-containing foods or beverages were not allowed for at least 8 h before start of the study. All procedures were performed in a quiet room, and room temperature was noted at the beginning and at the end of each experiment. The subjects were placed in the supine position and a venous catheter (Venflon®) was inserted into the right antecubital vein for infusion. The subjects then rested for 30 min before baseline recordings. After baseline measurements, infusion started using a time and volume controlled infusion pump (Braun Perfusor, Melsungen, Germany). Headache intensity, accompanying symptoms, velocity in the middle cerebral artery (V_MCA), diameter of the superficial temporal artery and radial artery, end-tidal partial pressure of pCO₂ (P_etCO₂), adverse events and vital signs were recorded at baseline and then every 10 min until 90 min after start of infusion.

After 90 min the subject was offered bread and lemonade/water to prevent headache caused by hunger or dehydration. The subjects were carefully instructed to complete a headache diary with accompanying symptoms according to the International Headache Society (21), including questions concerning premonitory symptoms (tired, yawning, stiff neck, blurred vision, thirst, intolerant/irritable, emotional, difficulty with concentration, any other symptoms) (22) and allodynia (combing, shaving, shower, cold, earrings, eyeglasses, heat, contact lenses, tight clothes, pillow, ponytail, necklace) (23) and any rescue medication every hour until 12 h after discharge from the hospital. In addition, the subjects were asked if the reported headache mimicked and was located as during a spontaneous migraine attack. Subjects were allowed to take rescue medication of their own choice at any time.

Headache intensity and migraine definition

Headache intensity was repeatedly recorded on a verbal rating scale (VRS) from 0 to 10 [0, no headache; 1, a very mild headache (including a feeling of pressing or throbbing—pre-pain); 10, worst imaginable headache] (24). Headache characteristics and associated symptoms were also recorded to determine the quality, localization (uni- or bilateral) and type of headache.

Headache induced experimentally by infusion of a neurotransmitter can not fulfil strict criteria for MoA (21). First, the migraine-like attacks reported are induced by pharmacological substances and can therefore not be spontaneous, although they often phenotypically mimic spontaneous migraine attacks in the majority of patients (2,3). Second, most spontaneous migraine attacks develop in a matter of hours and often go through a phase where they phenomenologically fulfil the criteria for tension-type headache before the headache gets worse, becomes unilateral and has the associated symptoms required for migraine. For this reason, attacks aborted by migraine-specific treatment were accepted in the new criteria for chronic migraine (25). Patients in experimental provocation studies cannot be denied treatment of the induced attacks and often treat before all migraine criteria are fulfilled. Based on these circumstances, we have used the following two criteria for a migraine-like attack induced 0–14 h after infusion of an experimental drug:

Headache attacks fulfilling either 1 or 2: headache attack fulfilling criteria C (moderate to severe pain intensity is considered ≥ 4 on VRS scale) and D for MoA (21).

Headache attack described as mimicking usual migraine attack and treated with a triptan.

Middle cerebral artery blood flow velocity

V_MCA was recorded bilaterally by transcranial Doppler (TCD) with hand-held 2-MHz probes (Multidop X; DWL, Sipplingen, Germany), as previously described (26,27). All recordings were done by the same skilled examiner (T.W.). P_etCO₂ (end-tidal CO₂) was recorded simultaneously to the TCD measurements using an open mask that caused no respiratory resistance (ProPac Encore^®; Welch Allyn Protocol, Beaverton, OR, USA). V_MCA was corrected for P_etCO₂ changes by e^0.034 per mmHg change in P_etCO₂ (28).

Diameter of the superficial temporal artery and radial artery

Diameter of the frontal branch of the left superficial temporal artery (STA) and left radial artery was measured by a high-resolution ultrasonography unit (20 MHz, bandwidth 15 MHz; Dermascan C; Cortex Technology, Hadsund, Denmark) as previously described (29,30).

Vital signs

Heart rate (HR) and blood pressure were measured by an auto-inflatable cuff (ProPac Encore^®; Welch Allyn Protocol). ECG (Cardiofax V; Nihon-Kohden, Tokyo, Japan) was monitored on an LCD screen and recorded on paper.

Data analysis and statistics

Headache scores are presented as median and quartiles. Vascular baseline values are presented as mean ± S.D. and vascular peak responses as mean including 95% confidence interval, unless otherwise stated. Baseline was beforehand defined as an average of T-10 and T0 before the start of infusion.

We defined an in-hospital phase, consisting of an infusion phase as a period from 0 to 30 min (0–30 min) because of a short half-life (2–4 min) of PGI₂ in plasma (31–33), and a postinfusion phase as a period from 30 to 90 min (30–90 min). Furthermore, a post-hospital phase was defined as the period of 1.5–14 h after start of the infusion. Primary end-points were incidence of any headache or reported migraine-like headache (confer criteria for an experimentally induced migraine attack above), and differences in response calculated as area under the curve (AUC) according to the trapezium rule (34) for headache score (AUC_headache), V_MCA (AUC_VMCA), diameter of STA (AUC_STA) between PGI₂ and placebo. Secondary end-points were differences in response in AUC_PetCO2, AUC_RA, mean arterial blood pressure (MAP) (AUC_MAP), systolic blood pressure (SBP) (AUC_SBP), diastolic blood pressure (DBP) (AUC_DBP), and HR (AUC_HR) between PGI₂ and placebo. Baseline was subtracted before calculating the AUC to reduce variation between sessions within subject.

To test the differences between incidences we used McNemar's test. To test differences between variables the Wilcoxon signed rank test for headache score and a paired, two-way t-test for vascular data were used. We tested for period and carry-over effects for all baseline variables with Mann–Whitney test and independent t-test. All analyses were performed with spss for Mac 11.0 (SPSS Inc., Chicago, IL, USA). Five per cent (P < 0.05) was accepted as the level of significance.

Results

All subjects (n = 12) (four men and eight women), mean age 35.3 years (range 20–53 years), completed the study (Table 1).

Table 1.

Mean baseline values (± s.d.) of velocity of blood flow in the middle cerebral artery (V_MCA), end-tidal CO₂ (PetCO₂), diameter of superficial temporal artery (STA) and radial artery (RA), mean arterial blood pressure (MAP), systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), in 12 migraineurs without aura on prostaglandin I₂ (PGI₂) and placebo days

	PGI₂	Placebo	P-value
V_MCA	78.51 ± 13.64	78.15 ± 14.71	0.901
PetCO₂ (mmHg)	34.70 ± 2.32	31.96 ± 6.71	0.137
STA (mm)	1.11 ± 0.18	1.07 ± 0.21	0.267
RA (mm)	2.42 ± 0.46	2.35 ± 0.30	0.448
MAP (mmHg)	80.36 ± 8.96	80.05 ± 6.93	0.828
SBP (mmHg)	112.05 ± 10.60	112.21 ± 7.63	0.948
DBP (mmHg)	64.51 ± 9.12	63.97 ± 7.64	0.687
HR	61.82 ± 10.76	59.46 ± 8.54	0.238

P-value: paired t-test.

There were no serious adverse events and patients cooperated well. Due to technical problems P_etCO₂ was missing on subject 1 on the PGI₂ day. There was no difference in the V_MCA between the left and right sides at baseline on PGI₂ (P = 0.522) or placebo (P = 0.268). There was no difference between pain side and non-pain side on PGI₂ (P = 0.764) or placebo (P = 0.630) at baseline in subjects reporting unilateral headache. We therefore used an average of the left and the right side. There was no carry-over or period effect for baseline values of headache, V_MCA, STA, RA, MAP, SBP, DBP or HR (P > 0.05).

Headache

Details on the PGI₂-induced headache or migraine during in- and post-hospital phases are given in Table 2 and Figure 1.

Figure 1.

Median (□) and individual active (a) and placebo (x) (b) headache scores on a verbal rating scale (VRS) in migraineurs without aura during and after 25 min infusion of PGI₂ (epoprostenol) 10 ng kg⁻¹ min⁻¹ or placebo. There was a significantly higher headache response in the infusion (AUC_{headache 0–30 min}, P = 0.003) and post-infusion (AUC_{headache 30–90 min}, P = 0.006) phases on PGI₂ day compared with placebo day. There was no difference in delayed headache (AUC_{Delayed headache}, P = 0.866).

Table 2.

Clinical characteristics of prostaglandin I₂-induced headache (localization, intensity/verbal rating scale, quality, aggravation by physical activity) and associated symptoms (nausea, photophobia, phonophobia) in 12 migraineurs without aura

Subject	Peak headache (onset and end)	Headache	Ass. symptoms	Mimics usual migraine	Migraine	Treatment (time)/ efficacy (pain free)
1	a		Right/6/throb./yes	+/+/−
1	b	20 (20–30) (min)	Bilat./1/throb./no	−/−/−	Yes (t_20 min)	No
	c	5 (3–7) (h)	Right/2/pres./no	+/−/−	Yes (t_3h)	Yes (t_6h)	Eletriptan 40 mg (t_6h)/yes
2	a		Left/5/throb./yes	+/+/+
2	b	20 (10–70) (min)	Left/3/throb./yes	−/+/−	Yes (t_20 min)	No
	c	5 (4–7) (h)	Left/3/throb./yes	−/+/+	Yes (t_4h)	Yes (t_5h)	Zolmatripan 2.5 mg (t_6h)/yes
3	a		Left/3/throb./yes	+/+/−
3	b	20 (20–30) (min)	Left/1/throb./no	−/−/−	Yes (t_30 min)	No
	c	9 (9–10) (h)	NS/1/NS/no	−/−/−	No	No	Sleep (t_9h)/yes
4	a		Bilat./7/throb./yes	+/+/+
4	b	30 (10–50) (min)	Bilat./4/throb./no	−/+/+	Yes (t_10 min)	Yes (t_30 min)
	c	−	−/−/−/−	−/−/−	−	−
5	a		Right/6/const.+pres./no	+/+/+
5	b	10 (10–50) (min)	Right/1/pres./no	−/−/−	Yes (t_10 min)	No
	c	−	−/−/−/−	−/−/−	−	−
6	a		Alternating unilateral/ 7/throb./yes	+/+/+
6	b	20 (10–30) (min)	Right/3/throb./no	−/+/−	Yes (t_20 min)	No
	c	7 (7–8) (h)	Left/4/pres./yes	+/+/+	Yes (t_7h)	Yes (t_7h)	Eletriptan 40 mg (t_8h)/yes
7	a		Alternating unilateral/6/ const.+pres./yes	+/+/+
7	b	−	−/−/−/−	−/−/−	−	−
	c	11 (11–13) (h)	Bilat./1/const./no	−/−/−	Yes (t_11h)	No	Sleep (t_13h)/yes
8	a		Left/6/pres./yes	+/−/−
8	b	30 (20–80) (min)	Left/4/pres./no	+/+/−	Yes (t_20 min)	Yes (t_30 min)
	c	13 (11–13) (h)	Left/2/pres./yes	−/−/−	Yes (t_11h)	No	Sleep (t_13h)/yes
9	a		Bilat./7/throb./yes	+/+/−
9	b	20 (20–90) (min)	Bilat./1/throb./no	−/−/−	Yes (t_20 min)	No
	c	1.5 (1.5–7) (h)	Bilat./1/throb./no	−/−/−	Yes (t_1.5h)	No	Sleep (t_2h)/yes
10	a		Left/7/throb./yes	+/+/−
10	b	20 (20–40) (min)	Left/1/throb.+pres./no	−/−/−	No	No
	c	−	−/−/−/−	−/−/−	−	−
11	a		Right/7/throb.+pres./no	+/+/−
11	b	10 (10–70) (min)	Right/1/throb.+pres./no	−/−/−	Yes (t_30 min)	No
	c	−	−/−/−/−	−/−/−	−	−
12	a		Alternating unilateral/ 9/throb.+pres./yes	+/+/+
12	b	30 (10–60) (min)	Left/3/throb.+pres./no	+/+/−	No	Yes (t_30 min)
	c	−	−/−/−/−	+/−/−	−	−

Peak headache including onset and cessation of headache, and associated symptoms during in-hospital (min) and out-hospital (hours) phases are shown. Time of onset for headache mimicking a usual migraine attack and time for fulfilling the criteria for migraine are given. In row b and c recorded as migraine if fulfilling the following criteria for a migraine attack; fulfilling criteria C and D for migraine without aura or headache described as mimicking a usual migraine attack and treated with a triptan (see text). The participants were allowed to take medication of their own choice at any time in the out-hospital phase. NS, not stated.

a. Spontaneous migraine attack without aura according to the International Headache Society criteria (21).

b. In-hospital phase (0–90 min).

c. Post-hospital phase (1.5–14 h).

In the period 0–14 h, 12 (100%) subjects reported headache on PGI₂ day compared with three (25%) subjects on placebo day (P = 0.004, McNemar), and six (50%) subjects fulfilled the criteria for an experimentally induced migraine-like attack compared with two (17%) subjects on placebo (P = 0.219). During the in-hospital phase (0–90 min), 11 (92%) subjects reported headache on the PGI₂ compared with two (17%) on placebo (P = 0.004, McNemar), and three subjects (25%) reported migraine-like attacks compared with no subjects on placebo (P = 0.250). During the post-hospital phase (1.5–14 h), seven (58%) subjects reported delayed headache on PGI₂ day compared with three (25%) subjects on placebo day (P = 0.125, McNemar), and three (25%) subjects reported migraine-like attacks compared with two (17%) subjects on placebo (P = 1.000). Mean time and range for migraine-like attacks were 30 min in-hospital and 6 h (5–7 h) post-hospital phase. One subject had a headache that started during the in-hospital phase and lasted throughout the post-hospital phase.

During the infusion phase, the AUC_0–30 min on PGI₂, 25 (15–47.5), was significantly larger than on placebo, 0 (0–0) (P = 0.003) (Fig. 1). During the post-infusion phase, the AUC_{30–90 min} on the PGI₂, 35 (6.25–57.5), was significantly larger than on the placebo, 0 (0–0) (P = 0.006) (Figs 1 and 2). The median peak headache, 1 (1–2.75), occurred 20 min after start of PGI₂ infusion (Fig. 1a). During the post-hospital phase, there was no difference in AUC_1.5–14h between the PGI₂, 2.25 (0–6.25), and the placebo days, 0 (0–1.88) (P = 0.866).

We asked patients whether headache reported during and after experiments mimicked their spontaneous migraine attacks. In the period 0–14 h, 10 (83%) subjects reported the headache to mimic a spontaneous attack on the PGI₂ day and three (25%) on the placebo day. Furthermore, 11 (92%) subjects reported that headache was located as during a spontaneous attack after PGI₂ and two (17%) subjects after placebo.

Figure 2.

Individual and mean flow velocities (cm/s) in the middle cerebral arteries (V_MCA) during and after 25 min infusion of PGI₂ (epoprostenol) 10 ng kg⁻¹ min⁻¹ or placebo in 12 migraineurs without aura. There was a significant difference in the infusion (AUC_{VMCA 0–30 min}, P = 0.015) phase and no difference in the post-infusion (AUC_{VMCA 30–90 min}, P = 0.291) phase after correction for changes in P_etCO₂.

Middle cerebral artery velocity and diameter

We found a significant decrease in the AUC_VMCA on PGI₂ compared with placebo during the infusion phase (P < 0.001) and no difference in the post-infusion phase (P = 0.275) (Fig. 3).

Figure 3.

Individual and mean diameter (mm) of the superficial temporal artery (STA) during and after 25 min infusion of PGI₂ (epoprostenol) 10 ng kg⁻¹ min⁻¹ or placebo in 12 migraineurs without aura. There was a significant difference in the infusion phase (AUC_{STA 0–30 min}, P < 0.001) and post-infusion (AUC_{STA 30–90 min}, P = 0.005) phases.

There was a significant difference in the AUC_PetCO2 recordings during TCD scans between PGI₂ and placebo days during the immediate phase (P = 0.001), but no difference in the post-infusion phase (P = 0.055). anova showed a significant change over time on PGI₂ day (P < 0.001) but no difference on placebo day (P = 0.508). Post hoc analysis revealed a significant difference at time 20, 30, 40, 50, 70 and 90 min (P < 0.05).

After correction for P_etCO₂ changes, there was a significant decrease in the AUC_VMCA on PGI₂ day compared with placebo day during the infusion phase (P = 0.015) and no difference in the post-infusion phase (P = 0.291).

Superficial temporal artery and radial artery

During the infusion phase, the AUC_{STA 0–30 min} on the PGI₂ day was significantly larger than on the placebo day (P < 0.001) (Fig. 4). In the post-infusion phase, the AUC_{STA 30–90 min} on the PGI₂ day was significantly larger than on the placebo day (P = 0.005). There was no difference in AUC_{RA 0–30 min} (P = 0.052) or AUC_{RA 30–90 min} (P = 0.783) between the PGI₂ day and the placebo day. See Table 3 for peak responses.

Figure 4.

Median headache and mean per cent changes from baseline in the middle cerebral artery blood flow velocity after correction for changes in P_etCO₂ (V_MCA) (□), diameter of the superficial temporal artery (STA) (▴) and radial artery (RA) (○) from baseline during and after 25 min infusion of PGI₂ (epoprostenol) 10 ng kg⁻¹ min⁻¹ in 12 migraineurs without aura. Compared with baseline, peak responses occurred at 20 min for headache [1 (1–2.75) (median and quartiles)], 10 min for V_MCA [−10.5% (−14.9, 6.0, 95 CI)], 20 min for STA [32.9% (26.4, 39.4, 95% CI)], and 20 min for RA (9.0% (3.7, 14.3, 95% CI)] after infusion start.

Table 3.

Maximal responses to PGI₂ compared with baseline during the in-hospital phase (0–90 min) for headache (median and quartiles) and vascular responses (mean per cent and 95% CI)

Variable (peak time)	Per cent changes from baseline
Headache (20 min)	1 (1–2.75)
V_MCA (10 min) (uncorrected for changes in P_etCO₂)	−13.5% (−17.5, −9.5, 95% CI)
V_MCA (10 min) (corrected for changes in P_etCO₂)	−10.5% (−14.9, −6.0, 95% CI)
STA (20 min)	32.9% (26.4, 39.4, 95% CI)
RA (20 min)	9.0% (3.7, 14.3, 95% CI)
HR (20 min)	49.7% (38.7, 60.6, 95% CI)
MAP (10 min)	−4.5% (−7.3, −1.6, 95% CI)
SBP (80 min)	5.9% (1.9, 10.0, 95% CI)
DBP (10 min)	−9.4% (−13.2, −5.6, 95% CI)

Mean blood flow in the middle cerebral artery (V_MCA), diameter of the superficial temporal artery (STA), radial artery (RA), mean arterial blood pressure (MAP), systolic blood pressure (SBP), and diastolic blood pressure (DBP) are given.

Premonitory symptoms and allodynia

During spontaneous migraine attacks 11 subjects usually experienced premonitory symptoms. Five subjects (42%) reported some of their usual premonitory symptoms in the in-hospital phase on the PGI₂ day with onset 10–40 min after start of the infusion compared with one subject (8%) reporting usual premonitory symptoms with onset 50 min after start of the infusion on the placebo day (P = 0.125, McNemar). Ten subjects (83%) reported some of their premonitory symptoms in the post-hospital phase on the PGI₂ day compared with five subjects (42%) on the placebo day (P = 0.125, McNemar).

Seven out of 12 subjects reported allodynia during their spontaneous migraine attacks. During the in-hospital phase two subjects (17%) reported allodynia on the PGI₂ day compared with none on the placebo day (P = 0.500, McNemar). In the post-hospital phase two subjects (17%) reported allodynia on the PGI₂ day compared with two subjects (17%) on the placebo day (P = 1.000, McNemar).

Arterial blood pressure and heart rate

We found no difference in MAP between PGI₂ and placebo days during either the infusion (AUC_{MAP 0–30 min}, P = 0.351) or the post-infusion phase (AUC_{MAP 30–90}, P = 0.344). There was no difference in SBP in the infusion phase (AUC_{SBP 0–30 min}, P = 0.080), but a significant difference in the post-infusion phase (AUC_{SBP 30–90 min}, P = 0.003). There was a significant difference in DBP in the infusion phase (AUC_{DBP 0–30 min}, P = 0.046) and no difference in the post-infusion phase (AUC_{DBP 30–90 min}, P = 0.701). We found that AUC_{HR 0–30 min} was significantly larger on PGI₂ than on placebo during the infusion phase (P < 0.001), but there was no difference in the post-infusion phase (AUC_{HR 30–90 min}, P = 0.980).

Adverse events

All subjects reported adverse events on the PGI₂ and two subjects reported adverse events on placebo (Table 4).

Table 4.

Adverse events were recorded every 10 min during the in-hospital phase (0–90 min) during and after 25 min infusion of PGI₂ (epoprostenol) 10 ng kg^- ¹ min^- ¹ or placebo in 12 migraineurs without aura

Time period	Symptoms	PGI₂ (n = 12)	Placebo (n = 12)	P-values
0–90 min	Headache	11	1	0.004
	Nausea	2	0	0.500
	Photophobia	5	0	0.063
	Phonophobia	1	0	1.000
	Flushing	12	0	< 0.001
	Heat sensation	12	0	< 0.001
	Palpitation	9	0	0.004
	Abdominal pain	1	0	1.000
	Abdominal pain mimicking menstrual pain (n = 8)	1	0	1.000
	Bearing down sensation	1	0	1.000
	Tension in the ears	2	0	0.500
	Discomfort to turn head	1	0	1.000
	Yawning	1	1	1.000
	An intracranial tone of high frequency	1	0	1.000
	Increased saliva production	1	0	1.000
	A whistling intracranial sound	1	0	1.000
	Sensitive facial skin ipsilateral to headache	2	0	0.500

Symptoms in italics are usual premonitory symptoms and symptoms assumed to represent allodynia during spontaneous migraine attacks in subjects reporting this adverse event during the in-hospital phase.

McNemar.

Discussion

The major finding of the present study is that PGI₂ triggered more headaches than placebo in migraine sufferers and 50% of the patients reported experimentally induced migraine attacks.

PGI₂-induced head pain or migraine

In 1981, Peatfield (35) reported that 8 ng kg⁻¹ min⁻¹ of PGI₂ induced a migraine-like headache in seven out of eight migraineurs. There was no detailed reporting of headache features in that study, but some of the migraineurs reported a short-lasting headache that stopped after the infusion and mimicked a spontaneous migraine attack. Interestingly, one of the subjects had treated a mild migraine attack before the infusion and reported aggravation of migraine pain and associated symptoms during infusion that returned to pre-infusion intensity after the infusion.

In the present study with a larger dose and detailed reporting, we observed that PGI₂ infusion induced a migraine-like headache in half of patients (50%). Interestingly, 75% (n = 9) of the subjects reported headache to mimic a spontaneous attack during the in-hospital phase. Furthermore, 83% of patients reported the headache to be located as during a spontaneous attack and 42% reported their usual premonitory symptoms in the in-hospital phase with onset during or shortly after PGI₂. Three of these subjects fulfilled the criteria for an experimentally induced migraine attack without aura. However, we found no statistical difference in pre- or postmonitory symptoms between two experimental days. Therefore, we can not exclude that reported symptoms were unspecific or related to adverse events after infusion. It is possible that a full-blown migraine attack might have occurred if a higher dose or a longer infusion time had been used. In support, we have previously shown a clear dose–response relationship between headache and PGI₂ in healthy subjects (18), and we did not observe a plateau in headache intensity within the 25-min infusion. It would be tempting to increase dose or prolong the PGI₂ infusion. However, considering the marked side-effects of the present regime, neither a higher dose nor a longer infusion time would be ethically acceptable. Adverse events of PGI₂ may to some extent have compromised blindness of the present study, but adverse events were caused by the physiological response to PGI₂ and could not have been avoided. The present double-blind approach was the best possible way, although not ideal, of coping with methodological error.

During the post-hospital phase, three patients (25%) treated the migraine-like attacks with a triptan, and all subjects reported pain-free response (Table 2). Four other subjects (33%) reported that sleep successfully terminated headache in the post-hospital phase. Interestingly, three of these subjects reported that headache preceded sleep and mimicked a spontaneous migraine attack. Given that sleep is often a natural abortive of migraine attacks, it could be speculated that these patients aborted the development of a migraine attack by sleeping. Tiredness and sleep may, however, be unspecific in this context, but it is noteworthy that tiredness is the most common non-headache feature during the premonitory, headache and postmonitory phases of a migraine attack (22). Thus, sleeping might affect incidence of reported migraine attacks in the present study.

Experimental mechanical dilation of the internal carotid artery, MCA or STA induces focal head pain in non-migraineurs (36,37), but there has been contradicting report of vasodilation during both spontaneous (30,38) and experimentally induced (2,3,39–41) migraine attacks. Vasodilating substances often induce headache or migraine-like attacks (2,41) and we therefore measure vasoactive responses even though vasodilation in a modern context is considered as an epiphenomenon (17,39,42,43). We have previously examined headache and vascular responses after PGI₂ in healthy subjects (18). The incidence of headache (n = 11), median maximal headache intensity [1 on VRS (1–2.75, quartiles)] and STA dilation (38.8%) in healthy subjects are similar to the present data. This may suggest that the physiological effect of PGI₂ in humans is not restricted to migraineurs. It appears, however, that patients and healthy subjects may differ in the number of reported unilateral (67% vs. 9%) and throbbing (75% vs. 58%) headaches and decrease in V_MCA (−10.5% vs. −4.6%) and incidence of headache in the post-hospital phase (seven vs. four). We have previously seen enhanced arterial responses in migraineurs compared with healthy controls (19,44). One could speculate that hypersensitivity to PGI₂ is explained by a small difference in vasodilation or decreased interictal pain thresholds of perivascular nociceptors in migraine sufferers. The post-hoc exploration showed no difference in vasodilation between painful and non-painful sides (n = 8) (unpublished observation). This may suggest that vasodilation cannot explain lateralization of pain in reported migraine-like attacks. However, these data need cautious interpretation because of the post hoc nature of our analysis.

Possible mechanisms of a PGI₂-induced migraine-like attacks

In the last 12 years there has been increased focus on the role of meningeal sensory afferents and mast cells in the generation of migraine pain. The question is, where and how can intravenously administrated PGI₂ activate and sensitize sensory afferents relevant for migraine pain? PGI₂ acts via a subtype of prostanoid receptors (IP) predominantly coupled to a Gs-type G-protein causing an increase in cAMP (45–47). The cAMP–PKA pathway has been implicated in the generation of headache and migraine-like headache (3,48) and mechanical sensitization of dural nociceptors (49). The trigeminovascular system is considered to be crucial in all aspects of cranial pain and haemodynamics (50) and IP-receptors are expressed in both trigeminal neurons (51) and cerebral arteries (52). It has been shown that pain can be triggered in humans by stimulating the dura mater close to meningeal arteries (36). Mediators like PGI₂ released upon mast cell degranulation cause sensitization of dural afferents in animals (16). Moreover, infusion of cyclooxygenase (COX)-1/COX-2 inhibitors blocks sensitization in both meningeal nociceptors (53) and trigeminal ganglion cells (54). PGI₂ dose-dependently sensitizes dural afferents directly in both Aδ and C units for at least 30 min after topical application (16). Interestingly, the timing corresponds to the reported headache (median range 10–50 min) in the present study. It should be noticed that PGI₂ also dilates cranial arteries via IP-receptors in vascular smooth muscle cells via potassium channels and cAMP–PKA pathway (55–58). It could therefore be speculated that cephalic dilation or changes in vascular tone may contribute to PGI₂-induced head pain. We observed a throbbing quality of head pain in most of the migraineurs. Sensitization of dural afferents in the present study may contribute to intracranial mechanical hypersensitivity and contribute to the throbbing quality of pain, as suggested during spontaneous migraine attacks (17). Allodynia was reported by two subjects during the in-hospital phase. This could be due to unspecific side-effects or activation of skin nociceptors, and not, as in spontaneous migraine, to a central sensitization phenomenon, because this would require 2–4 h to develop (59–62). We did not use Quantitative Sensory Testing and the presence of allodynia can not be verified. Interestingly, intravenous infusion of PGI₂ in both healthy subjects (18) and migraineurs does not induce general or focal pain in other parts of the body, except a few cases of abdominal pain. It is therefore plausible to suggest that intravenous PGI₂ in the dose given is more selective in activating trigeminal perivascular nociceptors and thereby head pain and migraine.

Collectively, these data suggest that PGI₂ may cause sensitization of perivascular dural nociceptors via IP-receptors located in peripheral terminals of trigeminal ganglia (51,63). This, together with arterial dilation, may cause migraine-like attacks. Furthermore, it appears that incidence of migraine-like attacks after PGI₂ (50%) is quite similar to other models of migraine such as glyceryl trinitrate (50–80%) (2,64–66), calcitonin gene-related peptide (∼33%) (3), histamine (∼70%) (67) and sildenafil (∼80%) (39). Future human and animal research should consider PGI₂ as a biomarker to study neurovascular mechanisms relevant to migraine.

Footnotes

Acknowledgements

The authors thank laboratory technicians Lene Elkjær and Winnie Grønning for their dedicated and excellent assistance. The study was supported by grants from the Lundbeck Centre of Neurovascular Signaling (LUCENS).

References

Thomsen

Olesen

. Nitric oxide in primary headaches. Curr Opin Neurol 2001; 14: 315–21.

Thomsen

Kruuse

Iversen

Olesen

. A nitric oxide donor (nitroglycerin) triggers genuine migraine attacks. Eur J Neurol 1994; 1: 73–80.

Lassen

Haderslev

Jacobsen

Iversen

Sperling

Olesen

. CGRP may play a causative role in migraine. Cephalalgia 2002; 22: 54–61.

Lassen

Ashina

Christiansen

Ulrich

Olesen

. Nitric oxide synthase inhibition in migraine. Lancet 1997; 349: 401–2.

Petersen

Lassen

Birk

Lesko

Olesen

. BIBN4096BS antagonizes human alpha-calcitonin gene related peptide-induced headache and extracerebral artery dilatation. Clin Pharmacol Ther 2005; 77: 202–13.

Olesen

Diener

Husstedt

Goadsby

Hall

Meier

. Calcitonin gene-related peptide receptor antagonist BIBN 4096 BS for the acute treatment of migraine. N Engl J Med 2004; 350: 1104–10.

Tfelt-Hansen

Olesen

JGP

Ramadan

Tfelt-Hansen

Welch

KMA

Nonsteroidal anti-inflammatory drugs in the acute treatment of migraines

The headaches 2006 3rd edn

Philadelphia, PA

Lippincott Williams & Wilkins

449–57

Diener

. Efficacy and safety of intravenous acetylsalicylic acid lysinate compared to subcutaneous sumatriptan and parenteral placebo in the acute treatment of migraine. A double-blind, double-dummy, randomized, multicenter, parallel group study. The ASASUMAMIG Study Group. Cephalalgia 1999; 19: 581–8, discussion 42

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.

10.

Hata

Breyer

. Pharmacology and signaling of prostaglandin receptors: multiple roles in inflammation and immune modulation. Pharmacol Ther 2004; 103: 147–66.

11.

Roberts

2nd Lewis

Oates

Austen

. Prostaglandin thromboxane, and 12-hydroxy-5,8,10,14-eicosatetraenoic acid production by ionophore-stimulated rat serosal mast cells. Biochim Biophys Acta 1979; 575: 185–92.

12.

Zonta

Sebelin

Gobbo

Fellin

Pozzan

Carmignoto

. Glutamate-mediated cytosolic calcium oscillations regulate a pulsatile prostaglandin release from cultured rat astrocytes. J Physiol 2003; 553: 407–14.

13.

Hull

Muksch

Akundi

Waschbisch

Hoozemans

Veerhuis

. Amyloid beta peptide (25-35) activates protein kinase C leading to cyclooxygenase-2 induction and prostaglandin E2 release in primary midbrain astrocytes. Neurochem Int 2006; 48: 663–72.

14.

Liao

Chen

Chiang

Chen

. Induction of cyclooxygenase-2 expression by manganese in cultured astrocytes. Neurochem Int 2007; 50: 905–15.

15.

Hirst

Young

Newton

Allport

Marriott

Wilkin

. Expression of COX-2 by normal and reactive astrocytes in the adult rat central nervous system. Mol Cell Neurosci 1999; 13: 57–68.

16.

Zhang

Strassman

Burstein

Levy

. Sensitization and activation of intracranial meningeal nociceptors by mast cell mediators. J Pharmacol Exp Ther 2007; 322: 806–12.

17.

Strassman

Raymond

Burstein

. Sensitization of meningeal sensory neurons and the origin of headaches. Nature 1996; 384: 560–4.

18.

Wienecke

Olesen

Oturai

Ashina

. Prostacyclin (epoprostenol) induces headache in healthy subjects. Pain 2008; 139: 106–16.

19.

Olesen

Iversen

Thomsen

. Nitric oxide supersensitivity: a possible molecular mechanism of migraine pain. Neuroreport 1993; 4: 1027–30.

20.

Vane

Corin

. Prostacyclin: a vascular mediator. Eur J Vasc Endovasc Surg 2003; 26: 571–8.

21.

IHS The International Classification of Headache Disorders, 2nd edn. Cephalalgia 2004; 24(Suppl. 1): 9–160.

22.

Giffin

Ruggiero

Lipton

Silberstein

Tvedskov

Olesen

. Premonitory symptoms in migraine: an electronic diary study. Neurology 2003; 60: 935–40.

23.

Jakubowski

Silberstein

Ashkenazi

Burstein

. Can allodynic migraine patients be identified interictally using a questionnaire?. Neurology 2005; 65: 1419–22.

24.

Iversen

Olesen

Tfelt-Hansen

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

25.

Olesen

Bousser

Diener

Dodick

First

Goadsby

. New appendix criteria open for a broader concept of chronic migraine. Cephalalgia 2006; 26: 742–6.

26.

Thomsen

Iversen

. Experimental and biological variation of three-dimensional transcranial Doppler measurements. J Appl Physiol 1993; 75: 2805–10.

27.

Hansen

Sitarz

Birk

Rahmann

Oturai

Fahrenkrug

. Vasoactive intestinal polypeptide evokes only a minimal headache in healthy volunteers. Cephalalgia 2006; 26: 992–1003.

28.

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: 368–72.

29.

Nielsen

Iversen

Tfelt-Hansen

Olesen

. Small arteries can be accurately studied in vivo, using high frequency ultrasound. Ultrasound Med Biol 1993; 19: 717–25.

30.

Iversen

Nielsen

Olesen

Tfelt-Hansen

. Arterial responses during migraine headache. Lancet 1990; 336: 837–9.

31.

Longenecker

McNeil

Longenecker

. Plasma protein binding and biologic half-life of prostacyclin in steady state sickle cell disease. Thromb Res 1991; 64: 751–6.

32.

de Leval

Hanson

David

Masereel

Pirotte

Dogne

. New developments on thromboxane and prostacyclin modulators part II: prostacyclin modulators. Curr Med Chem 2004; 11: 1243–52.

33.

Hassan

Pickles

. Epoprostenol (prostacyclin, PGI2) increases apparent liver blood flow in man. Prostaglandins Leukot Med 1983; 10: 449–54.

34.

Matthews

Altman

Campbell

Royston

. Analysis of serial measurements in medical research. BMJ 1990; 300: 230–5.

35.

Peatfield

Gawel

Rose

. The effect of infused prostacyclin in migraine and cluster headache. Headache 1981; 21: 190–5.

36.

Ray

Wolff

. Experimental studies on headache: pain sensitive structures of the head and their significance in headache. Arch Surg 1940; 1: 813–56.

37.

Nichols

3rd Mawad

Mohr

Stein

Hilal

Michelsen

. Focal headache during balloon inflation in the internal carotid and middle cerebral arteries. Stroke 1990; 21: 555–9.

38.

Friberg

Olesen

Iversen

Sperling

. Migraine pain associated with middle cerebral artery dilatation: reversal by sumatriptan. Lancet 1991; 338: 13–17.

39.

Kruuse

Thomsen

Birk

Olesen

. Migraine can be induced by sildenafil without changes in middle cerebral artery diameter. Brain 2003; 126: 241–7.

40.

Schoonman

van der Grond

Kortmann

van der Geest

Terwindt

Ferrari

. Migraine headache is not associated with cerebral or meningeal vasodilatation—a 3T magnetic resonance angiography study. Brain 2008 1

41.

Lassen

Jacobsen

Haderslev

Sperling

Iversen

Olesen

. Involvement of calcitonin gene-related peptide in migraine: regional cerebral blood flow and blood flow velocity in migraine patients. J Headache Pain 2008; 9: 151–7.

42.

Levy

Burstein

Strassman

. Calcitonin gene-related peptide does not excite or sensitize meningeal nociceptors: implications for the pathophysiology of migraine. Ann Neurol 2005; 58: 698–705.

43.

Rahmann

Wienecke

Hansen

Fahrenkrug

Olesen

Ashina

. Vasoactive intestinal peptide causes marked cephalic vasodilation, but does not induce migraine. Cephalalgia 2008; 28: 226–36.

44.

Lassen

Heinig

Oestergaard

Olesen

. Histamine inhalation is a specific but insensitive laboratory test for migraine. Cephalalgia 1996; 16: 550–3.

45.

Boie

Rushmore

Darmon-Goodwin

Grygorczyk

Slipetz

Metters

. Cloning and expression of a cDNA for the human prostanoid IP receptor. J Biol Chem 1994; 269: 12173–8.

46.

Katsuyama

Sugimoto

Namba

Irie

Negishi

Narumiya

. Cloning and expression of a cDNA for the human prostacyclin receptor. FEBS Lett 1994; 344: 74–8.

47.

Nakagawa

Tanaka

Usui

Harada

Sasaki

Itoh

. Molecular cloning of human prostacyclin receptor cDNA and its gene expression in the cardiovascular system. Circulation 1994; 90: 1643–7.

48.

Birk

Kruuse

Petersen

Tfelt-Hansen

Olesen

. The headache-inducing effect of cilostazol in human volunteers. Cephalalgia 2006; 26: 1304–9.

49.

Levy

Strassman

. Distinct sensitizing effects of the cAMP-PKA second messenger cascade on rat dural mechanonociceptors. J Physiol 2002; 538: 483–93.

50.

Goadsby

PJMA

Edvinsson

ElaK

Primary neurovascular headache. Migraine and cluster headache from a physiologic perspective

Cerebral blood floow and metabolism 2002 2nd edn

Philadelphia, PA

Lippincott Williams & Wilkins

457–65

51.

Jenkins

Feniuk

Humphrey

. Characterization of the prostanoid receptor types involved in mediating calcitonin gene-related peptide release from cultured rat trigeminal neurones. Br J Pharmacol 2001; 134: 1296–302.

52.

Davis

Murdoch

Ali

Purbrick

Ravid

Baxter

. EP4 prostanoid receptor-mediated vasodilatation of human middle cerebral arteries. Br J Pharmacol 2004; 141: 580–5.

53.

Levy

Zhang

Jakubowski

Burstein

. Sensitization of meningeal nociceptors: inhibition by naproxen. Eur J Neurosci 2008; 27: 917–22.

54.

Jakubowski

Levy

Goor-Aryeh

Collins

Bajwa

Burstein

. Terminating migraine with allodynia and ongoing central sensitization using parenteral administration of COX1/COX2 inhibitors. Headache 2005; 45: 850–61.

55.

Coleman

Smith

Narumiya

. International Union of Pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol Rev 1994; 46: 205–29.

56.

Wise

Jones

. Focus on prostacyclin and its novel mimetics. Trends Pharmacol Sci 1996; 17: 17–21.

57.

Siegel

Carl

Adler

Stock

. Effect of the prostacyclin analogue iloprost on K+ permeability in the smooth muscle cells of the canine carotid artery. Eicosanoids 1989; 2: 213–22.

58.

Schubert

Serebryakov

Mewes

Hopp

. Iloprost dilates rat small arteries: role of K(ATP)- and K(Ca)-channel activation by cAMP-dependent protein kinase. Am J Physiol 1997; 272: H1147–56.

59.

Burstein

. Deconstructing migraine headache into peripheral and central sensitization. Pain 2001; 89: 107–10.

60.

Burstein

Collins

Jakubowski

. Defeating migraine pain with triptans: a race against the development of cutaneous allodynia. Ann Neurol 2004; 55: 19–26.

61.

Burstein

Yarnitsky

Goor-Aryeh

Ransil

Bajwa

. An association between migraine and cutaneous allodynia. Ann Neurol 2000; 47: 614–24.

62.

Burstein

Cutrer

Yarnitsky

. The development of cutaneous allodynia during a migraine attack clinical evidence for the sequential recruitment of spinal and supraspinal nociceptive neurons in migraine. Brain 2000; 123: 1703–9.

63.

Nakae

Hayashi

Yamamoto

Iino

Yoshikawa

. Functional role of prostacyclin receptor in rat dorsal root ganglion neurons. Neurosci Lett 2005; 388: 132–7.

64.

Afridi

Kaube

Goadsby

. Glyceryl trinitrate triggers premonitory symptoms in migraineurs. Pain 2004; 110: 675–80.

65.

Sances

Tassorelli

Pucci

Ghiotto

Sandrini

Nappi

. Reliability of the nitroglycerin provocative test in the diagnosis of neurovascular headaches. Cephalalgia 2004; 24: 110–19.

66.

Christiansen

Thomsen

Daugaard

Ulrich

Olesen

. Glyceryl trinitrate induces attacks of migraine without aura in sufferers of migraine with aura. Cephalalgia 1999; 19: 660–7, discussion 26

67.

Lassen

Thomsen

Olesen

. Histamine induces migraine via the H1-receptor. Support for the NO hypothesis of migraine. Neuroreport 1995; 6: 1475–9.

Prostaglandin I 2 (epoprostenol) triggers migraine-like attacks in migraineurs

Abstract

Keywords

Introduction

Design and methods

Main experiment

Experimental design

Headache intensity and migraine definition

Middle cerebral artery blood flow velocity

Diameter of the superficial temporal artery and radial artery

Vital signs

Data analysis and statistics

Results

Headache

Middle cerebral artery velocity and diameter

Superficial temporal artery and radial artery

Premonitory symptoms and allodynia

Arterial blood pressure and heart rate

Adverse events

Discussion

PGI2-induced head pain or migraine

Possible mechanisms of a PGI2-induced migraine-like attacks

Footnotes

Acknowledgements

References

PGI₂-induced head pain or migraine

Possible mechanisms of a PGI₂-induced migraine-like attacks