Nitric oxide-induced headache in patients with chronic tension-type headache

Transcription

1 Brain (2000), 123, Nitric oxide-induced headache in patients with chronic tension-type headache M. Ashina, L. Bendtsen, R. Jensen and J. Olesen Department of Neurology, Glostrup Hospital, University of Copenhagen, DK-2600 Glostrup, Copenhagen, Denmark Correspondence to: Messoud Ashina, Department of Neurology, Glostrup Hospital, University of Copenhagen, DK-2600 Glostrup, Copenhagen, Denmark Summary An experimental model of headache offers unique GTN treatment. In patients, the AUC on a GTN day possibilities to study the mechanisms responsible for head [2221 ( ); median with quartiles in paren- pain. Using the glyceryl trinitrate [GTN; nitric oxide theses], was significantly greater than on a placebo day (NO) donor] model of experimental headache, we studied [730 ( ), P 0.008]. On the GTN day, the AUC the intensity, quality and time profile of headache after in patients [2221 ( )] was significantly higher infusion of GTN in 16 patients with chronic tensionpeak pain intensity occurred 8 h after infusion of GTN, than in controls [43 (0 972), P ]. In patients, type headache and in 16 healthy controls. Subjects were randomized to receive intravenous infusion of GTN whereas in controls it occurred 20 min after the start of infusion. The present study demonstrates that an NO- (0.5 µg/kg per minute for 20 min) or placebo on two induced biphasic response with an immediate and a headache-free days separated by at least 1 week. Headache delayed headache is common to chronic tension-type intensity was measured on a 10-point verbal rating scale headache and migraine. Furthermore, the NO-induced during 2 h of observation and for the next 10 h after delayed headache has the characteristics of the primary discharge from hospital. The primary endpoints were the headache disorder. This suggests that NO contributes to difference between the area under the curve (AUC the mechanisms of several types of primary headaches intensities duration) for headache recorded on the day and that NO-related central sensitization may be an of GTN treatment and on the day of placebo treatment important common denominator in the pain mechanisms in patients, and in patients and controls on the days of of primary headaches. Keywords: tension-type headache; nitric oxide; central sensitization; glyceryl trinitrate; primary headaches Abbreviations: AUC area under the curve; IHS International Headache Society; GTN glyceryl trinitrate; NO nitric oxide; NOS nitric oxide synthase Introduction Nitric oxide (NO) plays a key role in migraine (Thomsen and Olesen, 1998) and cluster headache (Fanciullacci et al., 1997). This freely diffusible molecule is involved in a variety of biological functions, including neurotransmission (Garthwaite and Boulton, 1995). NO also contributes to sensory transmission in the peripheral and central nervous system (Aley et al., 1998; Wu et al., 1998). We have observed previously that systemic administration of the NO donor glyceryl trinitrate (GTN) may induce strong immediate headache in patients with episodic tension-type headache compared with healthy subjects (Olesen et al., 1993), and we demonstrated recently that nitric oxide synthase (NOS) inhibition has an analgesic effect in patients with chronic tension-type headache (Ashina et al., 1999). It remains Oxford University Press 2000 unknown, however, whether NO induces more intense headache in patients with chronic tension-type headache than in healthy controls and, if so, what the characteristics and time profile of such headache might be. In the present study we recorded the intensity, quality and time profile of headache after infusion of the NO donor GTN in patients with chronic tension-type headache and compared GTN-induced headache in patients and healthy controls. Material and methods Subjects We recruited 16 patients with a diagnosis of chronic tensiontype headache according to the criteria of the International

2 Table 1 Clinical data on headache patients and healthy controls Nitric oxide in tension-type headache 1831 Patients Controls Number Females/males 13/3 13/3 Age (years) 39 (23 52) 37 (23 52) Frequency of tension type headache (days/4 weeks) 20 (15 25) 1 Amount of analgesics used (g/4 weeks) 6 (0 26) Hamilton depression score 2 (0 6) Data are mean (range). Headache Society (IHS) (Headache Classification Committee, measurement has been used in previous studies of GTN- 1988) (headache frequency 15 days per month for 6 induced headache in migraineurs (Thomsen et al., 1994). months) from the out-patient headache clinic at Glostrup Twelve-lead ECG was performed before examination. Any Hospital (Table 1). All patients underwent a general physical adverse events were recorded. Subjects with headache examination and a neurological examination and completed 120 min after the start of infusion were allowed to take a diagnostic headache diary (Russell et al., 1992) during a rescue medication. All subjects were asked to record details 4-week run-in period. Exclusion criteria were: history of of the following on a diary card 4, 8, 12, 16, 20 and 24 h migraine or any other type of primary headaches; any kind after start of infusion: headache location; headache intensity of daily medication (including prophylactic headache therapy (verbal rating scale); headache quality and aggravation by but not oral contraceptives); excessive use of analgesics physical activity; associated symptoms; any medication taken; (corresponding to 2 g aspirin/day); serious somatic or and adverse events. The diary cards were mailed to the clinic psychiatric diseases, including depression (Hamilton after each treatment. depression score 17) (Hamilton, 1960). Sixteen healthy volunteers served as controls (Table 1). The controls never had migraine and had fewer than 12 Data analysis and statistics headache days per year. Results are presented as median with quartiles in parentheses. All subjects gave written consent to participation in the The primary endpoint was the difference between the area study, which was approved by the local ethics committee under the headache curve (AUC intensity duration) and conducted in accordance with the Declaration of Helsinki. recorded over 12 h of observation on an active day and on a placebo day in patients. The secondary endpoint was the difference between the AUC recorded on the active day in Procedures patients and controls. The AUC was calculated according to In a double-blind, placebo-controlled crossover design, the the trapezium rule (Matthews et al., 1990). The difference patients and controls were allocated randomly to receive between the AUCs recorded on an active day and on a GTN at 0.5 µg/kg per minute or placebo (isotonic saline) on placebo day within patients and controls was compared by 2 days separated by at least 1 week. Medical staff not the use of the Wilcoxon signed ranks test. The difference involved in the study performed the randomization and between the AUCs recorded on an active day in patients and prepared the study drug. The randomization code remained controls was compared by the use of the Mann Whitney test. in the hospital during the study and was not available to the To assess changes in blood pressure and pulse rate in patients investigators until the study was complete. All subjects, who over the first 60 min after active treatment or placebo, we had been completely headache-free for at least 12 h before used the paired samples t-test. Five per cent was accepted as the examination, were examined on days without headache the level of significance. All data were analysed with SPSS at hours. The subjects were not allowed to take any version software (SPSS Inc., Chicago, Ill., USA). kind of analgesic for 24 h before the examination. GTN or placebo was infused over 20 min into an antecubital vein, as allocated by the randomization code. The study solutions looked the same. The following measurements were recorded Results at baseline and every 5 min until 60 min after the start of GTN and placebo infusions in patients infusion and then every 15 min until 120 min after the start The AUC on a GTN day [2221 ( )], was of infusion: headache intensity on a verbal rating scale significantly higher than on a placebo day [730 ( )] (0 10: 0, no headache; 5, moderate headache; 10, worst (P 0.008) (Fig. 1). During GTN infusion, 15 out of 16 imaginable headache); blood pressure and pulse rate. The patients developed immediate headache, with peak intensity verbal rating scale was chosen because this method of pain 20 min after the start of infusion (Table 2 and Fig. 1). One

3 1832 M. Ashina et al. Fig. 1 Median headache intensity during (20 min) and after infusion of GTN and placebo in 16 patients with chronic tension-type headache. Headache was scored on a 10-point verbal rating scale (VRS). The area under the headache curve (intensity duration) on a GTN day (filled squares) was significantly higher than on a placebo day (open squares) (P 0.008). The headache intensity reached its peak value 8 h after the start of the GTN infusion and was higher than after placebo. patient (patient 16) developed a migraine-like headache GTN infusion in patients versus controls lasting ~15 min (Table 2). During placebo infusion, three out The AUC in patients [2221 ( )] was significantly of 16 patients developed a bilateral, mild (range 2 3 on the higher than that in controls [43 (0 972)] (P ). verbal rating scale), pressing headache without associated symptoms. After GTN infusion, 15 out of 16 patients developed a delayed headache (Table 2) with peak intensity GTN infusion in patients with pressing 8 h after the start of infusion (Fig. 1). In 13 patients the headache versus controls headache fulfilled IHS criteria for tension-type headache, When the five patients whose usual headache quality was while two patients (patients 14 and 15) developed a headache throbbing were excluded, the remaining 11 patients, whose which fulfilled the diagnostic criteria for migraine without headache was of a pressing quality, still had a significantly aura. Eight patients took simple analgesics as a rescue higher response to GTN [2003 ( )] than 16 controls medication on the GTN day. After placebo infusion, 11 [43 (0 972)] (P ). In these patients, the AUC on patients reported a delayed headache. The headache intensity the GTN day [2003 ( )] was also significantly of these patients reached its peak 10 h after the start of higher than on a placebo day [660 (0 1348)] (P 0.02). infusion (Fig. 1). None of the patients developed a migraine None of these patients developed a migraine headache. headache. Two patients took simple analgesics as a rescue medication on the placebo day. Immediate and delayed headache in patients GTN and placebo infusions in controls The AUC recorded over the initial 120 min (immediate The AUC on the GTN day [43 (0 972)] was significantly headache) observation on the GTN day [171 (31 426)] was higher than on the placebo day [0 (0 0)] (P 0.008) (Fig. 2). significantly higher than on the placebo day [0 (0 91)] (P During the GTN infusion, nine out of 16 controls developed 0.02). The AUC recorded from 120 min to 12 h (delayed a headache without associated symptoms. The headache headache) after infusion on the GTN day [2108 ( )] intensity reached its peak 20 min after the start of infusion. was significantly higher than on the placebo day [720 (60 After discharge from hospital, six out of 16 controls developed 1410)] (P 0.008). a delayed headache (Table 3). One subject (subject 3) developed a headache, which fulfilled the IHS diagnostic criteria for migraine without aura. On the placebo day, none Haemodynamics in patients of the subjects developed a headache during the infusion or The mean arterial blood pressure (MAP) decreased in the 12 h following infusion. significantly during treatment with GTN compared with

4 Nitric oxide in tension-type headache 1833 Table 2 Clinical characteristics of patients Patient Time* (h) Headache Associated symptoms Migraine Rescue (location/intensity/quality/ (nausea/photophobia/ medication aggravation ) phonophobia) (h) 1 a Right/mild/pres./no No/no/no No b Bilateral/1/pres./no No/no/no No No c 12 Bilateral/8/pres./no No/no/no No Yes (12) 2 a Bilateral/mild/pres./no No/no/no No b Right/1/throb./no No/no/no No No c 8 Bilateral/6/pres./no No/no/no No No 3 a Bilateral/mod./pres./no No/no/no No b Bilateral/5/pres./no No/no/no No No c 4 Bilat./6/pres./yes No/no/no No Yes (4) 4 a Left/mild/pres./no No/no/no No b Bilateral/4/pres./no No/no/no No No c 4 Left/2/pres./no No/no/no No No 5 a Bilateral/mild/pres./no No/no/no No b Bilateral/2/pres./no No/no/no No No c 8 Bilateral/5/throb./yes No/no/no No No 6 a Bilateral/sev./pres.throb./no No/no/no No b No headache No/no/no No No c 10 Unilateral/6/throb./yes No/no/no No No 7 a Bilateral/mod./pres.throb./no No/no/no No b Bilateral/8/pres.throb./no No/no/no No No c 4 Bilateral/7/throb./no No/no/no No Yes (5; 8) 8 a Bilateral/mild/pres./no No/no/no No b Bilateral/1/pres./no No/no/no No No c No headache No/no/no No No 9 a Bilateral/mod./pres./no No/no/no No b No headache No/no/no No No c 8 Bilateral/5/pres./no No/no/no No Yes (9) 10 a Bilateral/mod./pres./no No/no/no No b Bilateral/2/pres./no No/no/no No No c 10 Bilateral/3/pres./yes No/yes/no No No 11 a Bilateral/mod./pres./no No/no/no No b Bilateral/10/throb./no No/no/no No No c 4 Bilateral/10/throb./yes No/no/no No Yes (4; 9) 12 a Bilateral/mild/pres./no No/no/no No b Bilateral/1/pres./no No/no/no No No c 10 Bilateral/5/pres./no No/no/no No No 13 a Bilateral/mild/pres./yes No/no/no No b Bilateral/6/pres./no No/no/no No No c 4 Bilateral/3/pres./yes No/no/no No No 14 a Bilateral/mod./pres.throb./no No/no/no No b Bilateral/2/throb./no No/no/no No No c 12 Bilateral/9/throb./no Yes/no/no Yes Yes (8) 15 a Bilateral/mod./pres.throb./no No/no/no No b Bilateral/7/pres.throb./no No/no/no No No c 6 Bilateral/10/throb./yes Yes/no/no Yes Yes (7) 16 a Bilateral/mod./pres.throb./no No/no/no No b Bilateral/3/throb./yes Yes/no/no Yes No c 4 Bilateral/8/throb./yes No/yes/no No Yes (6) a usual headache; b headache during GNT infusion (20 min); c peak headache after glyceryl trinitrate infusion; sev. severe; mod. moderate; pres. pressing; throb. throbbing; not applicable. *Time to postinfusion peak; headache aggravation by physical activity; fulfilling IHS criteria for migraine without aura; intake and time of intake of rescue medication. placebo (P 0.01) (Fig. 3). Patients were clinically unaffected by these changes. There was no difference in pulse rate during treatment with GTN and placebo (P 0.50) (Fig. 3).

5 1834 M. Ashina et al. Fig. 2 Median headache intensity during (20 min) and after infusion of GTN and placebo in 16 healthy controls. The area under the curve on a GTN day (circles) was significantly higher than on a placebo day (sunbursts) (P 0.008). Table 3 Clinical characteristics of peak headache in six out of 16 controls who developed a delayed headache after glyceryl trinitrate infusion Subjects Time* (h) Headache Associated symptoms Migraine Rescue medication (localization/intensity/ (nausea/photophobia/ (h) quality/aggravation ) phonophobia) 1 6 Bilateral/5/pres./no No/no/no No No 2 6 Bilateral/3/pres./no No/yes/no No Yes (5.5) 3 4 Bilateral/6/throb./yes No/yes/yes Yes Yes (4) 4 4 Bilateral/1/pres.throb./yes No/no/no No No 5 6 Bilateral/7/pres.throb./yes No/no/no No Yes (6) 6 4 Bilateral/2/pres./no No/no/no No No pres. pressing; throb. throbbing. *Time to post-infusion peak; headache aggravation by physical activity; fulfilling IHS criteria for migraine without aura; intake and time of intake of rescue medication. Discussion and duration intermediate between those of migraineurs and Previous studies of experimental headache in healthy controls. tension-type headache In 1980, Krabbe and Olesen infused histamine intravenously into patients with chronic tension-type headache (Krabbe and Present results in relation to findings in Olesen, 1980). The patients developed pressure-like, constant migraine headache with an intensity intermediate between those in It was important to compare the GTN response with the migraine patients and healthy controls. It was not reported placebo response in patients with chronic tension-type whether the patients developed delayed headache. Histamine headache, because a high proportion of these patients were presumably causes production of NO via an endothelial H 1 expected to develop their usual headache in the experimental receptor (Lassen et al., 1996). In another study, by Olesen period. In the present study, 15 out of 16 patients developed and colleagues, the exogenous NO donor GTN was given a delayed headache with mean peak intensity 8 h after the intravenously to 17 migraineurs, nine patients with episodic start of GTN infusion. Eleven patients developed a delayed tension-type headache and 17 healthy control subjects (Olesen headache of low intensity with a maximum 10 h after placebo et al., 1993). Seven out of nine patients developed a infusion. This was likely to represent their usual headache. nitroglycerin-induced headache (immediate) with intensity Both the immediate (0 2 h observation) and the delayed (2

6 Nitric oxide in tension-type headache 1835 Fig. 3 Changes in mean arterial blood pressure (MAP) and pulse rate during and after infusion of GTN and placebo in patients with chronic tension-type headache. The MAP decreased significantly during treatment with GTN (filled squares) compared with placebo (open squares) (P 0.01). The peak decrease in MAP was 11 2% (mean SEM) and occurred 15 min after the start of GTN infusion. There was little difference in pulse rate during treatment with GTN (circles) and placebo (sunbursts) (P 0.50). The maximum increase in pulse rate was 20 6% (mean SEM) and occurred 10 min after the start of GTN infusion. The plots show mean scores. 12 h observation) headache were significantly stronger on Our results indicate that patients with chronic tension-type the GTN day than on the placebo day. In controls, GTN headache similar to that of patients with migraine are induced a headache during infusion, which rapidly supersensitive to NO, and that the majority of patients in disappeared after the infusion had stopped. Interestingly, the both groups develop their usual headache several hours after time profile of the GTN-induced headache in patients with the infusion of GTN. chronic tension-type headache was strikingly similar to the time profile of GTN-induced headache in patients with migraine without aura (Thomsen et al., 1994). Thus, patients with migraine without aura developed an immediate headache Possible mechanisms of immediate and delayed during GTN infusion and a delayed headache fulfilling the headache IHS criteria for migraine several hours after the infusion was The most important finding in the present study was that stopped. Peak intensity occurred 5.5 h after the infusion was systemic administration of the NO donor GTN in patients with stopped. The characteristics of the delayed headache in chronic tension-type headache, as in migraineurs, resulted in chronic tension-type headache were, however, different from a biphasic nociceptive response with a headache peak during those in patients with migraine. Thus, 80% of migraine infusion (immediate headache), a reduction in headache patients developed migraine without aura after infusion of intensity for ~1.5 h (intermediate phase), and a second and GTN (Thomsen et al., 1994), whereas only 13% of patients more pronounced headache peak several hours later (delayed with chronic tension-type headache did so in the present headache). We suggest that a direct effect of NO on study. The other 87% developed a tension-type headache. perivascular sensory afferents and/or NO-induced arterial None of our patients had previously experienced migraine dilatation are responsible for the immediate headache and fulfilling the IHS criteria. To exclude the possibility that that enhanced central sensitization at the spinal/trigeminal previous migraine had been overlooked, patients who had level is responsible for the delayed headache. previously experienced throbbing headaches were analysed NO evokes pain in humans when injected paravascularly separately. Two patients who developed migraine headache or perfused through a vascularly isolated segment of a hand were in this group, but the three others did not develop vein (Holthusen and Arndt, 1995). These findings suggest migraine. Furthermore, patients with exclusively pressing that NO may directly activate or sensitize nociceptors around pain in the past also developed a significantly stronger blood vessels. Intravenous infusion of GTN (0.5 µg/kg per headache during and after GTN infusion than controls. minute) over 20 min induces dilatation of the middle cerebral

7 1836 M. Ashina et al. artery in healthy subjects (Iversen et al., 1989) and in in the CNS, thereby increasing the sensitization of nociceptive migraineurs and patients with episodic tension-type headache pathways in the CNS. (Thomsen et al., 1993). In these studies the dilatation lasted Sustained NO-induced vascular nociception may lead to at least until 1 h after GTN infusion had stopped (Iversen central sensitization and the subsequent convergence of et al., 1989; Thomsen et al., 1993) and 3 h in another study nociceptive input from blood vessels and myofascial tissue. (Lassen et al., 1996). Furthermore, vasodilatation of the As mentioned earlier, NO may activate or sensitize middle meningeal artery in rats increases the firing rate in nociceptors around blood vessels directly or by dilatation. the trigeminal nucleus caudalis (Cumberbatch et al., 1999). Dilatation of the meningeal blood vessels in rats causes Collectively, these studies suggest that immediate headache sensitization of central trigeminal neurones and facilitation after infusion of GTN in patients with chronic tension- of convergent sensory responses (Cumberbatch et al., 1999). type headache may originate in NO-induced activation or Interestingly, trigeminal neurones became maximally sensitization of sensory nerves around the cephalic/ sensitized after the peak increase in vessel diameter and the extracephalic arteries or from NO-induced arterial dilatation, increased excitability exceeded the period of vasodilatation or both. (Cumberbatch et al., 1999). It is, therefore, possible that Basic pain research has demonstrated that central excessive vascular nociception caused by GTN may gradually sensitization, i.e. increased excitability of neurones in the augment the sensitizing effect of pre-existing myofascial input CNS, generated by prolonged nociceptive input from the in chronic tension-type headache sufferers (Olesen, 1991). periphery, plays an important role in the pathophysiology of The questions of why patients with chronic tension- chronic pain (Woolf, 1983; Hu et al., 1992; Woolf and type headache and patients with migraine develop delayed Doubell, 1994). NO is released in the spinal cord during headache of greater intensity with different characteristics central sensitization, and prolonged elevation of NO within and why most healthy subjects develop no delayed headache the spinal dorsal horn is important in maintaining the central or only a minor one are important. The most likely explanation sensitization (Lin et al., 1999). While inhibition of NOS has is that pre-existing facilitation of distinct nociceptive central been shown to reduce central sensitization in animal models pathways in chronic tension-type headache sufferers of persistent pain (Haley et al., 1992; Hao and Xu, 1996; (myofascial pathways) and migraineurs (vascular pathways) may be enhanced by NO-induced central sensitization. This Mao et al., 1997), nociceptive responses in these models are could explain why the delayed headache fulfilled tensionenhanced by NO donors (Kitto et al., 1992; Coderre and type headache criteria in patients with chronic tension-type Yashpal, 1994). Furthermore, it was reported that inhibition headache and migraine criteria in migraineurs. This could of NOS markedly reduces Fos expression in the also explain why NO does not induce strong delayed headache trigeminocervical complex of the cat after stimulation of the in healthy subjects when no pre-existing sensitization is superior sagittal sinus (Hoskin et al., 1999). Interestingly, in present. animals, formalin-induced tissue injury results in a biphasic In conclusion, the present study provides important nociceptive response (Coderre and Yashpal, 1994) similar to information about NO mechanisms in chronic tension-type that observed in the present study. The second pain phase headache. It is suggested that NO-induced delayed headache may be caused by central sensitization, and NOS inhibition in patients with chronic tension-type headache is due to reduces pain only in this phase (Coderre and Yashpal, 1994). augmentation of pre-existing central sensitization. Moreover, Recent clinical studies suggest that central sensitization at our results indicate that NO contributes to the mechanisms the spinal/trigeminal level caused by nociceptive input from of several types of primary headaches and that NO-related pericranial myofascial tissues may play an important role in central sensitization may be an important common chronic tension-type headache (Bendtsen et al., 1996; Jensen denominator in the pain mechanisms of primary headaches, et al., 1998). Furthermore, we demonstrated recently that although their basic pathophysiological mechanisms are NOS inhibition has an analgesic effect in chronic tension-type different. headache, probably due to reduction of central sensitization (Ashina et al., 1999). Studies in rats and in humans have shown that after intravenous administration of GTN very Acknowledgements little drug remains in the blood, and that the majority We wish to thank Mrs Hanne Andresen for skilful technical of the GTN is distributed to the tissues (Sorkin et al., assistance. The study was supported financially by the 1984). In anaesthetized cats, intravenous infusion of GTN University of Copenhagen. (0.25 µg/kg per minute over 20 min) induced a prolonged (60 min) increase in NO in the parenchyma of the brain References (Read et al., 1997), and a prolonged increase in the level of Aley KO, McCarter G, Levine JD. Nitric oxide signaling in pain NO in the spinal dorsal horn was demonstrated during central and nociceptor sensitization in the rat. J Neurosci 1998; 18: sensitization (Wu et al., 1998). These data indicate that Ashina M, Lassen LH, Bendtsen L, Jensen R, Olesen J. Effect of infusion of GTN may result in the storage and subsequent inhibition of nitric oxide synthase on chronic tension-type headache: liberation of NO or may trigger endogenous NO production a randomised crossover trial. Lancet 1999; 353:

8 Nitric oxide in tension-type headache 1837 Bendtsen L, Jensen R, Olesen J. Qualitatively altered nociception Lassen LH, Thomsen LL, Kruuse C, Iversen HK, Olesen J. in chronic myofascial pain. Pain 1996; 65: Histamine-1 receptor blockade does not prevent nitroglycerin induced migraine. Support for the NO-hypothesis of migraine. Eur Coderre TJ, Yashpal K. Intracellular messengers contributing to J Clin Pharmacol 1996; 49: persistent nociception and hyperalgesia induced by L-glutamate and substance P in the rat formalin pain model. Eur J Neurosci 1994; Lin Q, Palecek J, Paleckova V, Peng YB, Wu J, Cui M, et al. Nitric 6: oxide mediates the central sensitization of primate spinothalamic Cumberbatch MJ, Williamson DJ, Mason GS, Hill RG, Hargreaves tract neurons. J Neurophysiol 1999; 81: RJ. Dural vasodilation causes a sensitization of rat caudal trigeminal Mao J, Price DD, Zhu J, Lu J, Mayer DJ. The inhibition of nitric neurones in vivo that is blocked by a 5-HT1B/1D agonist. Br J oxide-activated poly(adp-ribose) synthetase attenuates Pharmacol 1999; 126: transsynaptic alteration of spinal cord dorsal horn neurons and neuropathic pain in the rat. Pain 1997; 72: Fanciullacci M, Alessandri M, Sicuteri R, Marabini S. Responsiveness of the trigeminovascular system to nitroglycerine Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of in cluster headache patients. Brain 1997; 120: serial measurements in medical research. BMJ 1990; 300: Garthwaite J, Boulton CL. Nitric oxide signaling in the central Olesen J. Clinical and pathophysiological observations in migraine nervous system. [Review]. Annu Rev Physiol 1995; 57: and tension-type headache explained by integration of vascular, supraspinal and myofascial inputs. Pain 1991; 46: Haley JE, Dickenson AH, Schachter M. Electrophysiological Olesen J, Iversen HK, Thomsen LL. Nitric oxide supersensitivity: evidence for a role of nitric oxide in prolonged chemical nociception a possible molecular mechanism of migraine pain. [Review]. in the rat. Neuropharmacology 1992; 31: Neuroreport 1993; 4: Hamilton M. A rating scale for depression. J Neurol Neurosurg Read SJ, Smith MI, Hunter AJ, Parsons AA. Enhanced nitric oxide Psychiatry 1960; 23: release during cortical spreading depression following infusion of Hao J, Xu X. Treatment of chronic allodynia-like response in glyceryl trinitrate in the anaesthetized cat. Cephalalgia 1997; 17: spinally injured rats: effects of systemically administered nitric oxide synthase inhibitors. Pain 1996; 66: Russell MB, Rasmussen BK, Brennum J, Iversen HK, Jensen RA, Headache Classification Committee of the International Headache Olesen J. Presentation of a new instrument: the diagnostic headache Society. Classification and diagnostic criteria for headache disorders, diary. Cephalalgia 1992; 12: cranial neuralgias and facial pain. Cephalalgia 1988; 8 Suppl 7: Sorkin EM, Brogden RN, Romankiewicz JA. Intravenous glyceryl Holthusen H, Arndt JO. Nitric oxide evokes pain at nociceptors of trinitrate (nitroglycerin). A review of its pharmacological properties the paravascular tissue and veins in humans. J Physiol (Lond) 1995; and therapeutic efficacy. [Review]. Drugs 1984; 27: : Thomsen LL, Iversen HK, Brinck TA, Olesen J. Arterial Hoskin KL, Bulmer DC, Goadsby PJ. Fos expression in the supersensitivity to nitric oxide (nitroglycerin) in migraine sufferers. trigeminocervical complex of the cat after stimulation of the superior Cephalalgia 1993; 13: sagittal sinus is reduced by L-NAME. Neurosci Lett 1999; 266: Thomsen LL, Kruuse C, Iversen HK, Olesen J. A nitric oxide donor (nitroglycerin) triggers genuine migraine attacks. Eur J Neurol 1994; Hu JW, Sessle BJ, Raboisson P, Dallel R, Woda A. Stimulation of 1: craniofacial muscle afferents induces prolonged facilitatory effects Thomsen LL, Olesen J. Nitric oxide theory of migraine. [Review]. in trigeminal nociceptive brain-stem neurones. Pain 1992; 48: Clin Neurosci 1998; 5: Iversen HK, Holm S, Friberg L. Intracranial hemodynamics during Woolf CJ. Evidence for a central component of post-injury pain intravenous nitroglycerin infusion [abstract]. Cephalalgia 1989; 9 hypersensitivity. Nature 1983; 306: Suppl 10: Woolf CJ, Doubell TP. (1994) The pathophysiology of chronic Jensen R, Bendtsen L, Olesen J. Muscular factors are of importance pain increased sensitivity to low threshold A beta-fibre inputs. in tension-type headache. Headache 1998; 38: [Review]. Curr Opin Neurobiol 1994; 4: Kitto KF, Haley JE, Wilcox GL. Involvement of nitric oxide in Wu J, Lin Q, McAdoo DJ, Willis WD. Nitric oxide contributes to spinally mediated hyperalgesia in the mouse. Neurosci Lett 1992; central sensitization following intradermal injection of capsaicin. 148: 1 5. Neuroreport 1998; 9: Krabbe AA, Olesen J. Headache provocation by continuous intravenous infusion of histamine. Clinical results and receptor Received February 10, Revised April 27, mechanisms. Pain 1980; 8: Accepted May 8, 2000