Cerebrospinal Fluid Pressure and Venous Pressure in "Dynamite Headache" and Cluster Headache Attacks
Jan Hannerz* M.D. and Dan Greitz** M.D.
Departments of Neurology* and Neuroradiology**, Karolinska Hospital, Stockholm, Sweden. Reprint requests to: Dr. Jan Hannerz, Department of Neurology, Karolinska Hospital, S-104 01 Stockholm, Sweden. Accepted for publication: July 3, 1992 SYNOPSIS
Six patients with episodic cluster headache were investigated as to blood pressure, heart rate, cerebrospinal fluid pressure (Pcsf) and frontal vein pressure (Pvf) during five nitroglycerin (NG) provoked attacks and one spontaneous attack. In a seventh studied patient the NG failed to provoke an attack. The earlier reported decrease of systolic blood pressure and increase of diastolic blood pressure and heart rate after NG administration were also found in these patients. The "dynamite headache" was related to the start and duration of an increase of the cerebrospinal fluid pressure. There was no relationship between the start or the maximum pain of the cluster headache attack and changes in Pcsf or Pvf. On breathing oxygen during a cluster headache attack, there was a decrease of Pcsf but in some patients a temporary increase of Pvf was observed, which possibly indicates that oxygen simultaneously attains constriction of arteries and veins. Key words: Cluster headache, "dynamite headache", cerebrospinal fluid pressure. Abbreviations: Pcsf cerebrospinal fluid pressure, Pvf frontal vein pressure (Headache 1992; 32: 436-438) INTRODUCTION
Nitroglycerin (NG) is known to provoke headache in normal subjects. The duration of this "dynamite headache" has been found to be related to the duration of the NG induced increase of blood pressure and heart rate.1 Transcranial Doppler studies have shown that there is a decrease of blood flow velocity in cerebral arteries, suggesting dilatation of intracranial vessels after NG administration.1,2 These findings and the results of other studies3 have been considered to support such an effect of NG on intracranial vessels. "Dynamite headache" may possibly be due to increased load on pain sensitive vessel walls, caused by the combination of raised blood pressure, heart rate and vessel dilatation. The finding that NG induces an increase in intracranial pressure4 may indicate that there is a discrepancy in afferent and efferent blood flow to the head after NG administration, and which may be related to the appearance of the headache. NG is also known to induce attacks in patients with cluster headache who are in an active period. This headache usually appears after a latency of as long as 70 minutes5 after the administration of NG, in comparison with the "dynamite headache", which occurs only some minutes after the NG has been taken. However, there is no difference in the NG effects on blood flow velocity in intracranial arteries in cluster headache patients and controls,1,2 which can explain the mechanisms behind the cluster headache attack. Oxygen is believed to constrict intracranial vessels in man.6,7 Breathing 100% oxygen is known to often stop an attack of cluster headache.8,9 In this investigation, blood pressure, heart rate, cerebrospinal fluid pressure (Pcsf), and pressure in the frontal vein (Pvf) were studied during NG induced "dynamite headache" and cluster headache attacks and oxygen induced relief from pain, in order to better understand the mechanisms behind the two different types of pain. PATIENTS AND METHODS
Seven patients, one woman and six men, aged 29-66 years, with episodic cluster headache, participated in the study. Attacks were induced by the administration of 1 mg NG sublingually. The patients had not had any cluster headache attacks for the last 8 hours and had not taken any drugs in the 15 hours before the study. Seven liters/min of 100% oxygen was administered at maximum pain of the cluster headache attack until pain relief. During the studies all patient were lying on the left side. Blood pressure was recorded using a mercury manometer and heart rate was calculated from at least 15 successive heart cycles. Intrathecal pressure was measured by a manometer connected to a lumbar puncture needle. The superficial facial veins were compressed by a rubber headband and rubber cubes over the angular veins in order to increase the probability of the recorded vein pressure reflecting the intracranial venous pressure. The pressure in the frontal vein was measured by a thin teflon cannula with an outer diameter of 0.8 mm connected to a manometer. Heparin was infused every 15 minutes to avoid clotting. All values were recorded every other minute during the experiments.
RESULTS
Three to five minutes after NG administration and simultaneously with the occurrence of the "dynamite headache", the Pcsf increased in all patients from 3.0-8.4 cm water, (average 5.8 cm water), and returned to the initial pressure simultaneously with the resolution of the "dynamite headache". At the appearance of cluster headache and at maximum pain of the attack, there were no consistent changes in Pcsf. In two patients there was a slight increase in Pcsf and in three a slight decrease compared to the initial pressure and also to the pressure which prevailed before the pain started. These changes were not different from those found in the patient in whom NG failed to provoke an attack, and in whom there was a slight decrease in Pcsf during the 70 minutes the patient was studied after NG was given. In five trials with 100% oxygen breathing at maximum pain until disappearance of the pain in a cluster headache attack, the Pcsf decreased 0.5-5.5 cm water in four patients and increased 6.8 cm water in the fifth patient. The frontal vein pressure usually parallelled the changes of the Pcsf, (cf Fig. 1), but not during the 3-5 minutes after NG administration when Pvf increased less than Pcsf or even decreased compared to initial values. However, during oxygen breathing, there was an increase in vein pressure for about five minutes in comparison to the Pcsf in three out of five tests, before the initial relationship between the two pressures was resumed. In one patient a spontaneous attack was studied. There were no changes in either Pcsf or Pvf with the development of the cluster headache or at maximum pain compared to the initial pressure.
DISCUSSION
NG is considered to dilate intracranial arteries.1-3 As to the effect of NG on intracranial veins, six orbital phlebograms were reviewed from an earlier study.10 The phlebograms were performed before and after NG administration in episodic cluster headache patients who were in an active phase. It was found that NG usually dilated intraorbital veins in these patients. The "dynamite headache" has earlier been shown to be related to duration of the NG induced increase of blood pressure and heart rate.1 The start and the duration of the "dynamite headache" is now found to be related to an increase of the pressure of the cerebrospinal fluid (Pcsf), which can be assumed to be equal to the intracranial pressure. The findings support the hypothesis that the "dynamite headache" is due to a usually short-lasting discrepancy between afferent and efferent cerebral blood flow, which temporarily results in an increased intracranial pressure. There were no findings indicating a relationship between changes in Pcsf and the start or the development of the pain in the cluster headache attack either in NG induced attacks or the spontaneous attack. Oxygen is believed to constrict intracranial vessels in man.6,7 0xygen has also been used in cluster headache attacks with beneficial effects on the pain.8,9 The findings of temporary increased pressure in the frontal vein and simultaneously a decrease of Pcsf with pain relief during oxygen administration in three of five cluster headache attacks, indicate that oxygen may have effects on both arteries and veins during these circumstances. Cluster headache has been suggested to be due to venous vasculitis causing a unilateral sympathicoplegia in the middle fossa,1 i.e., a local dysfunction in the intracranial blood circulation, which should not interfere significantly with the Pcsf. The findings of this study do not invalidate such a notion. Footnote: Consent obtained from all patients and investigations were approved by the ethics committee of Karolinska Hospital, Stockholm, Sweden. REFERENCES
1.
Hannerz J, Hellstrom G, Klum T, Wahlgren NG. Cluster headache and "dynamite headache": Blood flow velocities in the middle cerebral artery. Cephalalgia 1990; 10:31-38.
2.
Dahl A, Russel D, Nyberg-Hansen R, Rootvelt K. Cluster headache: Transcranial Doppler ultrasound and rCBF studies. Cephalalgia 1990; 10:87-94.
3.
Iversen HK, Holm S, Friberg L. Nitroglycerin-induced headache and intracranial hemodynamics. Migraine and other headaches: The vascular mechanisms. Ed:J Olesen. Raven press 1991:327-330.
4.
Dohi S, Matsumoto M, Takahashi T. The effects of nitroglycerin on cerebrospinal fluid pressure in awake and anesthesized humans. Anesthesiology 1981; 54:511-514.
5.
Ekbom K. Nitroglycerin as a provocative agent in cluster headache. Arch Neurol 1968; 19:487-493.
6.
Cohen PJ, Alexander SC, Smith TC, Reivich M, Wollman H. Effects of hypoxia and normocarbia on cerebral blood flow and metabolism in concious man. J Appl Physiol 1967; 23:183-189.
7.
Sakai F, Meyer JS. Regional cerebral hemodynamics during migraine and cluster headaches measured by the 133Xenon inhalation method. Headache 1988; 18:122-132.
8.
Horton BT. Histamine cephalgia. Lancet 1952; 72:92-98.
9.
Kudrow L. Response of cluster headache attacks to oxygen inhalation. Headache 1981; 21:1-4.
10.
Hannerz J, Ericson K, Bergstrand G. Orbital phlebography in patients with cluster headache. Cephalalgia 1987; 7:207-211.
Jan Hannerz* M.D. and Dan Greitz** M.D.
Departments of Neurology* and Neuroradiology**, Karolinska Hospital, Stockholm, Sweden. Reprint requests to: Dr. Jan Hannerz, Department of Neurology, Karolinska Hospital, S-104 01 Stockholm, Sweden. Accepted for publication: July 3, 1992 SYNOPSIS
Six patients with episodic cluster headache were investigated as to blood pressure, heart rate, cerebrospinal fluid pressure (Pcsf) and frontal vein pressure (Pvf) during five nitroglycerin (NG) provoked attacks and one spontaneous attack. In a seventh studied patient the NG failed to provoke an attack. The earlier reported decrease of systolic blood pressure and increase of diastolic blood pressure and heart rate after NG administration were also found in these patients. The "dynamite headache" was related to the start and duration of an increase of the cerebrospinal fluid pressure. There was no relationship between the start or the maximum pain of the cluster headache attack and changes in Pcsf or Pvf. On breathing oxygen during a cluster headache attack, there was a decrease of Pcsf but in some patients a temporary increase of Pvf was observed, which possibly indicates that oxygen simultaneously attains constriction of arteries and veins. Key words: Cluster headache, "dynamite headache", cerebrospinal fluid pressure. Abbreviations: Pcsf cerebrospinal fluid pressure, Pvf frontal vein pressure (Headache 1992; 32: 436-438) INTRODUCTION
Nitroglycerin (NG) is known to provoke headache in normal subjects. The duration of this "dynamite headache" has been found to be related to the duration of the NG induced increase of blood pressure and heart rate.1 Transcranial Doppler studies have shown that there is a decrease of blood flow velocity in cerebral arteries, suggesting dilatation of intracranial vessels after NG administration.1,2 These findings and the results of other studies3 have been considered to support such an effect of NG on intracranial vessels. "Dynamite headache" may possibly be due to increased load on pain sensitive vessel walls, caused by the combination of raised blood pressure, heart rate and vessel dilatation. The finding that NG induces an increase in intracranial pressure4 may indicate that there is a discrepancy in afferent and efferent blood flow to the head after NG administration, and which may be related to the appearance of the headache. NG is also known to induce attacks in patients with cluster headache who are in an active period. This headache usually appears after a latency of as long as 70 minutes5 after the administration of NG, in comparison with the "dynamite headache", which occurs only some minutes after the NG has been taken. However, there is no difference in the NG effects on blood flow velocity in intracranial arteries in cluster headache patients and controls,1,2 which can explain the mechanisms behind the cluster headache attack. Oxygen is believed to constrict intracranial vessels in man.6,7 Breathing 100% oxygen is known to often stop an attack of cluster headache.8,9 In this investigation, blood pressure, heart rate, cerebrospinal fluid pressure (Pcsf), and pressure in the frontal vein (Pvf) were studied during NG induced "dynamite headache" and cluster headache attacks and oxygen induced relief from pain, in order to better understand the mechanisms behind the two different types of pain. PATIENTS AND METHODS
Seven patients, one woman and six men, aged 29-66 years, with episodic cluster headache, participated in the study. Attacks were induced by the administration of 1 mg NG sublingually. The patients had not had any cluster headache attacks for the last 8 hours and had not taken any drugs in the 15 hours before the study. Seven liters/min of 100% oxygen was administered at maximum pain of the cluster headache attack until pain relief. During the studies all patient were lying on the left side. Blood pressure was recorded using a mercury manometer and heart rate was calculated from at least 15 successive heart cycles. Intrathecal pressure was measured by a manometer connected to a lumbar puncture needle. The superficial facial veins were compressed by a rubber headband and rubber cubes over the angular veins in order to increase the probability of the recorded vein pressure reflecting the intracranial venous pressure. The pressure in the frontal vein was measured by a thin teflon cannula with an outer diameter of 0.8 mm connected to a manometer. Heparin was infused every 15 minutes to avoid clotting. All values were recorded every other minute during the experiments.
RESULTS
Three to five minutes after NG administration and simultaneously with the occurrence of the "dynamite headache", the Pcsf increased in all patients from 3.0-8.4 cm water, (average 5.8 cm water), and returned to the initial pressure simultaneously with the resolution of the "dynamite headache". At the appearance of cluster headache and at maximum pain of the attack, there were no consistent changes in Pcsf. In two patients there was a slight increase in Pcsf and in three a slight decrease compared to the initial pressure and also to the pressure which prevailed before the pain started. These changes were not different from those found in the patient in whom NG failed to provoke an attack, and in whom there was a slight decrease in Pcsf during the 70 minutes the patient was studied after NG was given. In five trials with 100% oxygen breathing at maximum pain until disappearance of the pain in a cluster headache attack, the Pcsf decreased 0.5-5.5 cm water in four patients and increased 6.8 cm water in the fifth patient. The frontal vein pressure usually parallelled the changes of the Pcsf, (cf Fig. 1), but not during the 3-5 minutes after NG administration when Pvf increased less than Pcsf or even decreased compared to initial values. However, during oxygen breathing, there was an increase in vein pressure for about five minutes in comparison to the Pcsf in three out of five tests, before the initial relationship between the two pressures was resumed. In one patient a spontaneous attack was studied. There were no changes in either Pcsf or Pvf with the development of the cluster headache or at maximum pain compared to the initial pressure.
DISCUSSION
NG is considered to dilate intracranial arteries.1-3 As to the effect of NG on intracranial veins, six orbital phlebograms were reviewed from an earlier study.10 The phlebograms were performed before and after NG administration in episodic cluster headache patients who were in an active phase. It was found that NG usually dilated intraorbital veins in these patients. The "dynamite headache" has earlier been shown to be related to duration of the NG induced increase of blood pressure and heart rate.1 The start and the duration of the "dynamite headache" is now found to be related to an increase of the pressure of the cerebrospinal fluid (Pcsf), which can be assumed to be equal to the intracranial pressure. The findings support the hypothesis that the "dynamite headache" is due to a usually short-lasting discrepancy between afferent and efferent cerebral blood flow, which temporarily results in an increased intracranial pressure. There were no findings indicating a relationship between changes in Pcsf and the start or the development of the pain in the cluster headache attack either in NG induced attacks or the spontaneous attack. Oxygen is believed to constrict intracranial vessels in man.6,7 0xygen has also been used in cluster headache attacks with beneficial effects on the pain.8,9 The findings of temporary increased pressure in the frontal vein and simultaneously a decrease of Pcsf with pain relief during oxygen administration in three of five cluster headache attacks, indicate that oxygen may have effects on both arteries and veins during these circumstances. Cluster headache has been suggested to be due to venous vasculitis causing a unilateral sympathicoplegia in the middle fossa,1 i.e., a local dysfunction in the intracranial blood circulation, which should not interfere significantly with the Pcsf. The findings of this study do not invalidate such a notion. Footnote: Consent obtained from all patients and investigations were approved by the ethics committee of Karolinska Hospital, Stockholm, Sweden. REFERENCES
1.
Hannerz J, Hellstrom G, Klum T, Wahlgren NG. Cluster headache and "dynamite headache": Blood flow velocities in the middle cerebral artery. Cephalalgia 1990; 10:31-38.
2.
Dahl A, Russel D, Nyberg-Hansen R, Rootvelt K. Cluster headache: Transcranial Doppler ultrasound and rCBF studies. Cephalalgia 1990; 10:87-94.
3.
Iversen HK, Holm S, Friberg L. Nitroglycerin-induced headache and intracranial hemodynamics. Migraine and other headaches: The vascular mechanisms. Ed:J Olesen. Raven press 1991:327-330.
4.
Dohi S, Matsumoto M, Takahashi T. The effects of nitroglycerin on cerebrospinal fluid pressure in awake and anesthesized humans. Anesthesiology 1981; 54:511-514.
5.
Ekbom K. Nitroglycerin as a provocative agent in cluster headache. Arch Neurol 1968; 19:487-493.
6.
Cohen PJ, Alexander SC, Smith TC, Reivich M, Wollman H. Effects of hypoxia and normocarbia on cerebral blood flow and metabolism in concious man. J Appl Physiol 1967; 23:183-189.
7.
Sakai F, Meyer JS. Regional cerebral hemodynamics during migraine and cluster headaches measured by the 133Xenon inhalation method. Headache 1988; 18:122-132.
8.
Horton BT. Histamine cephalgia. Lancet 1952; 72:92-98.
9.
Kudrow L. Response of cluster headache attacks to oxygen inhalation. Headache 1981; 21:1-4.
10.
Hannerz J, Ericson K, Bergstrand G. Orbital phlebography in patients with cluster headache. Cephalalgia 1987; 7:207-211.
