- Email: [email protected]
Effect of inhibition of nitric oxide synthase on chronic tension-type headache: a randomised crossover trial
EARLY REPORTS
Effect of inhibition of nitric oxide synthase on chronic tension-type headache: a randomised crossover trial M Ashina, L H Lassen, L Bendtsen, R Jensen, J Olesen
Summary Background Studies in animals have shown that nitric oxide plays an important part in central sensitisation and that inhibitors of nitric oxide synthase (NOS) decrease sensitisation in models of persistent pain. The efficacy of inhibitors of NOS has not been tested in patients with tension-type chronic headache. We aimed to show whether NG-monomethyl-L-arginine hydrochloride (L-NMMA), an inhibitor of NOS, is effective in relieving pain in such patients. Methods We undertook a randomised double-blind, crossover trial of 16 patients with chronic-tension-type headache. Patients were assigned intravenous infusion of 6 mg/kg L-NMMA or placebo on 2 days separated by at least 1 week in a randomised order. Headache intensity was measured on a 100 mm visual analogue scale, and on a verbal rating scale at baseline and at 30 min, 60 min, and 120 min after start of treatment. The primary endpoint was reduction of pain intensity on the visual analogue scale by the active treatment compared with placebo. Findings L-NMMA reduced pain intensity on the visual analogue scale significantly more than placebo: 120 min after start of treatment, the mean pain score was decreased from 49 to 33 with L-NMMA and from 44 to 40 with placebo (p=0·01). Pain intensity on the verbal rating scale was also significantly lower for treatment with L-NMMA than for treatment with placebo (p=0·02). Interpretation Inhibition of NOS had an analgesic effect in chronic tension-type headache. Further tests are required before clinical application.
Lancet 1999; 353: 287–89
Introduction Nitric oxide is a messenger molecule involved in various biological functions including neurotransmission.1–3 Experiments in animals have shown that nitric oxide is an important transmitter in pain pathways of the spinal cord and that central sensitisation of these pathways (increased excitability of neurons in the central nervous system) may be caused by or associated with activation of neuronal nitric oxide synthase (NOS) and the generation of nitric oxide.4,5 Inhibitors of NOS decrease central sensitisation in animal models of persistent pain.6–10 Inhibitors of NOS could, therefore, be a treatment for human chronic pain. We used chronic tension-type headache as a model for chronic pain, since this type of headache is one of the most common types of chronic pain,11 and since central sensitisation due to prolonged nociceptive input from pericranial myofascial tissues may play an important part in Department of Neurology, Glostrup Hospital, University of Copenhagen, DK-2600 Glostrup, Copenhagen, Denmark ( M Ashina MD, L H Lassen MD, L Bendtsen MD, R Jensen MD, J Olesen MD) Correspondence to: Dr M Ashina (e-mail: [email protected])
THE LANCET • Vol 353 • January 23, 1999
the pathophysiology of this disorder.12 We aimed to find out whether NG-monomethyl-L-arginine hydrochloride (L-NMMA), which is the only inhibitor of NOS tested thoroughly in human beings, is effective in the treatment of chronic tension-type headache.
Methods We recruited patients with a diagnosis of chronic tension-type headache according to the criteria of the International Headache Society13 (headache frequency Ä15 days per month for Ä6 months) from the outpatient headache clinic at Glostrup University Hospital. Five of the patients had coexisting infrequent migraine (<4 days/year). All patients completed a diagnostic headache diary14 during a 4-week introductory period to ensure that inclusion criteria were met. Patients were included whether or not they had myofascial tenderness. At screening, we undertook a full physical and neurological examination, 12-lead electrocardiography, routine haematological and biochemical testing, and urine-analysis for each patient. The patients were not allowed to take analgesics, including vasoactive drugs for migraine, for 12 h before treatment. Exclusion criteria were: use of any kind of daily medication, including prophylactic headache therapy but not oral contraceptives; pregnancy or breastfeeding; excessive use of analgesics (>2 aspirin or equivalent daily); excessive alcohol use; serious somatic or psychiatric disorders including depression (Hamilton depression score Ä17);15 ischaemic heart disease; and supine diastolic blood pressure of more than 90 mm Hg or heart rate of less than 50 beats per min. All patients gave written consent. The study was approved by the Danish Board of Health and the local ethics committee, and was undertaken in accordance with the Declaration of Helsinki. The trial had a double-blind, placebo-controlled crossover design. The primary endpoint was reduction of pain intensity on a visual analogue scale by the active treatment compared with placebo. The secondary endpoint was reduction of pain intensity on a verbal rating scale by the active treatment compared with placebo. Patients were randomly assigned 6 mg/kg L-NMMA (Clinalfa, Switzerland) or placebo (isotonic glucose) on 2 days separated by at least 1 week. The patients were treated during a typical day of tension-type headache—patients had pain characteristic of tension-type headache with no more than one associated symptom.13 The patients were randomly assigned by computer (MedStat, version 2.12) into two groups: one group received L-NMMA then placebo at least a week later; the other group received placebo first, then L-NMMA. The investigators were not aware that the non-restricted nature of the randomisation programme might lead to the treatment groups being of unequal size. There was no stratification for coexisting migraine. Randomisation and preparation of study drugs was done by medical staff not involved in the study. The randomisation code remained in the hospital during the study and was not available to the investigators until the study was completed. For each patient, L-NMMA or placebo was infused over 15 min into an antecubital vein, as allocated by the randomisation code. The solutions for infusion looked the same. Headache intensity was measured at baseline and 30 min, 60 min, 90 min, and 120 min after start of infusion on a 100 mm visual analogue scale (0=no pain, 100=worst pain imaginable) and on a verbal rating scale (0=no headache, 5=moderate headache, 10=worst headache imaginable). Blood pressure and pulse rate were measured 5 min before the start of treatment and every 5 min up to 30 min, then every 30 min to 120 min after the start of treatment. Electrocardiography was done continuously, and any adverse events were recorded. Patients with unrelieved headache
287
EARLY REPORTS
Time
16 patients randomised
6 assigned L-NMMA then placebo
10 assigned placebo then L-NMMA
6 completed trial
10 completed trial
Figure 1: Trial profile at 120 min were allowed to take rescue medication. All patients were asked to record details of headache intensity on the visual analogue scale, any medication taken, and adverse events on a diary card every 4 h up to 24 h after the start of treatment. The patients returned to the clinic 4–7 days after each treatment, when the diary cards were collected and any adverse events noted. Calculation of sample size was based on detection of a difference between treatments in reduction of pain intensity on the visual analogue scale at 5% significance (one-sided) with 80% power. We assumed that each patient would show 20% variation on the visual analogue scale. A 15% reduction of pain intensity was taken to be clinically significant. We estimated that 14 patients should be included, but we increased the sample size to 16 patients since this is statistically more desirable. The visual analogue scale was chosen as the primary endpoint because this method for pain measurement is used widely in both clinical and experimental studies. We calculated the sum differences between the pretreatment measurements on each scale (visual and verbal) and the measurement at 30 min, 60 min, 90 min, and 120 min after treatment to obtain a summary pain score for each treatment.16 To test for period and carry-over effects we used independentsamples t tests on the differences and the sums, respectively, of the summary scores for the first and second treatments. The scores on the visual analogue scale were assumed to be normally distributed and summary scores for active and placebo treatment were therefore compared by independent-samples t tests. However, for the ordinal scale of the verbal rating scale we used Wilcoxon signed-ranks tests on the changes in scores from baseline to 24 h after treatment. To assess decreases in blood pressure and pulse rate over the first 30 min after active treatment or placebo, we used paired samples t tests. All these analyses were done with SPSS (version 7.5.1).
Results 16 patients (12 women, four men, mean age 38·5 years, range 23–52) were randomly assigned 6 mg/kg L-NMMA or placebo on 2 days separated by at least 1 week (figure 1). Six patients received L-NMMA then placebo, and ten patients received placebo then L-NMMA. The skewed randomisation gave an efficiency loss of roughly 6·66%, which did not greatly affect the power calculation. Three patients with coexisting migraine received L-NMMA then placebo, and two received placebo then L-NMMA. The mean time between courses of treatment was 13 days for both groups. None of the patients left the study between treatment courses. Seven patients reported side-effects of L-NMMA treatment: two tiredness, three dryness of the mouth, and one each drowsiness, exhaustion, nausea, and a tingling in one arm. Four patients reported sideeffects of placebo treatment: two each tingling in the arms and shoulders, and one each dryness of the mouth, warm sensation in the body, and drowsiness. No patient withdrew from the study because of side-effects. Simple analgesics were used as rescue medication by three patients treated with L-NMMA and by seven patients treated with placebo. There was a significant difference in reduction of pain intensity on the visual analogue scale between treatment 288
Mean (SD) pain score
Baseline 30 min 60 min 90 min 120 min
L-NMMA
Placebo
49 38 35 34 33
44 41 40 42 40
(16) (18) (18) (21) (21)
(14) (17) (17) (16) (17)
Pain scores on a 100 mm visual analogue scale
with L-NMMA and placebo (table, figure 2, p=0·01). There was no significant period effect (p=0·32) or carryover effect (p=0·15). The pain intensity on the verbal rating scale was significantly lower during treatment with L-NMMA than during treatment with placebo (p=0·02), but this difference had disappeared by 24 h after the start of treatment. The changes in mean arterial blood pressure and pulse rate during treatment with L-NMMA or placebo differed significantly (blood pressure p=0·0001, pulse rate p=0·001, figure 3). The peak increase in blood pressure was 12% (SD 2) and occurred 15 min after treatment with L-NMMA. Mean pulse rate decreased by 16% (2) 10 min after the start of the L-NMMA infusion. The increase in mean arterial blood pressure and decrease in pulse rate during L-NMMA treatment are consistent with known pharmacological properties of L-NMMA.17 Patients were clinically unaffected by these changes.
Discussion Little progress has been made in the treatment of chronic pain, including low back pain, arthritis, and chronic tension-type headache. Opioids are not effective in most patients and are contraindicated owing to the risk of addiction. Non-opioids are not effective and may even aggravate chronic tension-type headache.18 Drugs are needed with new mechanisms of action based on improved knowledge of the neurobiology of chronic pain. Central sensitisation plays an important part in the pathophysiology of chronic pain.19 There is evidence to show that persistent activity in peripheral nociceptors may lead to sensitisation of neurons in the spinal dorsal horn, partly via activation of N-methyl-D-aspartate (NMDA) receptors.20 Many of the effects of NMDA-receptor activation are mediated via production of nitric oxide, and immunoreactivity related to NOS has been identified in the spinal dorsal horn.21–23 Nitric oxide may, therefore, play an important part in hyperalgesia in the spinal cord.5 Animal models of persistent pain have shown that inhibitors of NOS decrease sensitisation of the spinal dorsal horn induced by Change in pain-intensity on visual scale (%)
16 patients eligible
L-NMMA Placebo
10 0 –10 –20 –30 –40 –50 0
15
30
60 Time (min)
90
120
Figure 2: Mean percentage change from baseline in pain intensity on 100 mm visual analogue scale Bars=SE.
THE LANCET • Vol 353 • January 23, 1999
EARLY REPORTS
Change from baseline (%)
20
Mean arterial blood pressure on L-NMMA Mean arterial blood pressure on placebo Mean pulse rate on placebo Mean pulse rate on L-NMMA
10
and the Danish Hospital Foundation for Medical Research, Region of Copenhagen, the Faroe Islands, and Greenland.
References 1
0 2
–10 3 4
–20 0
10 20 30 40 50 60 70 80 90 100 110 120 Time (min) Figure 3: Mean percentage change from baseline in mean arterial blood pressure and pulse rate during and after treatment with L-NMMA and placebo
5 6
continuous painful input from the periphery.6,10,24 However, the efficacy of inhibitors of NOS has not been assessed in patients with chronic pain. We suggest that inhibition of NOS has an analgesic effect in human chronic tension-type headache, a model for chronic pain. This effect may be due to reduction of central sensitisation at the spinal dorsal horn/trigeminal nucleus. L-NMMA inhibits all three types of NOS (endothelial NOS, neuronal NOS, inducible NOS),25 and nitric oxide is involved in the transmission and modulation of nociceptive information centrally and in the periphery.26,27 Study of selective inhibitors of NOS25 is needed to show which type of NOS is involved in human chronic pain and its exact site of action. Inhibition of NOS is effective in the treatment of acute migraine attacks, probably because of interaction with specific migraine mechanisms.28 The analgesic effect of inhibition of NOS in our study might be influenced by the presence of coexisting infrequent migraine in five patients. However, we found that patients without coexisting migraine tended to have a greater reduction of pain intensity (35%) than patients with coexisting migraine (23%). This finding shows that our results were not influenced by the inclusion of patients with coexisting infrequent migraine and that the mechanisms behind the analgesic effect of NOS inhibition differ in tension-type headache and migraine. Although the effect of NOS inhibition was significant, pain was reduced by only 30%. The clinical significance of the reduction is unclear, but our patients had had chronic tension-type headache for many years and had tried many other treatments without effect. Furthermore, acute dosing with a drug that has a short half-life29 is not ideal for a chronic disorder. The purpose of our study was not to develop L-NMMA for clinical use but to test whether inhibition of NOS had any effect or not. Since the result was positive, this approach should be tested in other (nonheadache) types of chronic pain. Chronic dosing should be tested when a safe inhibitor of NOS with a longer half-life becomes available. Longer doses should also be tested, but because of the increase in blood pressure as a result of inhibition of endothelial NOS, such tests will have to await the development of selective inhibitors of neuronal NOS.
25
Contributors
26
Messoud Ashina designed the study, collected and processed data, and was responsible for writing the paper. Lisbeth Hjorth Lassen, Lars Bendtsen, Rigmor Jensen, and Jes Olesen helped to design the study and undertook data interpretation. All investigators contributed to writing the paper.
7
8
9
10
11
12 13
14 15 16 17
18 19
20
21
22
23 24
27 28
Acknowledgments We thank Hanne Andresen for technical assistance, and Peter Dalgaard from the Department of Biostatistics at the University of Copenhagen for statistical advice. The study was supported by the Danish Headache Society
THE LANCET • Vol 353 • January 23, 1999
29
Ignarro LJ, Lippton H, Edwards JC, et al. Mechanisms of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprusside and nitric oxide: evidence for the involvement of S-nitrosothiols as active intermediates. J Pharmacol Exp Ther 1981; 218: 739–49. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol Rev 1991; 43: 109–41. Garthwaite J. Nitric oxide signalling in the nervous system. Neurosciences 1993; 5: 171–80. Haley JE, Dickenson AH, Schachter M. Electrophysiological evidence for a role of nitric oxide in prolonged chemical nociception in the rat. Neuropharmacology 1992; 31: 251–58. Meller ST, Gebhart GF. Nitric oxide (NO) and nociceptive processing in the spinal cord. Pain 1993; 52: 127–36. Meller ST, Cummings CP, Traub RJ, Gebhard GF. The role of nitric oxide in the development and maintenance of the hyperalgesia produced by intraplantar injection of carrageenan in the rat. Neuroscience 1994; 60: 367–74. Minami T, Nishihara I, Ito et al. Nitric oxide mediates allodynia induced by intrathecal administration of prostaglandin E2 or prostaglandin F2 alpha in conscious mice. Pain 1995; 61: 285–90. Coderre TJ, Yashpal K. Intracellular messengers contributing to persistent nociception and hyperalgesia induced by L-glutamate and substance P in the rat formalin pain model. Eur J Neuroscience 1994; 1: 1328–34. Hao J-X, Xu X-J. Treatment of chronic allodynia-like response in spinally injured rats: effects of systemically administered nitric oxide synthase inhibitors. Pain 1996; 6: 313–19. Mao J, Price DD, Zhu J, Lu J, Mayer DJ. The inhibition of nitric oxideactivated poly (ADP-ribose) synthase attenuates transsynaptic alteration of spinal cord dorsal horn neurons and neuropathic pain in the rat. Pain 1997; 72: 355–66. Rasmussen BK, Jensen R, Schroll M, Olesen J. Epidemiology of headache in a general population: a prevalence study. J Clin Epidemiol 1991; 44: 1147–57. Bendtsen L, Jensen R, Olesen J. Qualitatively altered nociception in chronic myofascial pain. Pain 1996; 65: 259–64. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988; 8 (suppl 7): 1–96. Russell MB, Rasmussen BK, Brennum J, et al. Presentation of a new instrument: the diagnostic headache diary. Cephalalgia 1992; 12: 369–74. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 56–62. Matthews JN, Altman DF, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990; 300: 230–35. Haynes WG, Noon JP, Walker BR, Webb DJ. Inhibition of nitric oxide synthesis increases blood pressure in healthy humans. J Hypertens 1993; 11: 1375–80. Olesen J. Analgesic headache. BMJ 1995; 310: 479–80. Woolf CJ, Doubell TP. The pathophysiology of chronic pain-increased sensitivity to low threshold A beta-fibre inputs. Curr Opin Neurobiol 1994; 4: 525–34. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993; 52: 259–85. Mizukawa K, Vincent SR, McGeer PL, McGeer EG. Distribution of reduced-nicotinamide-adenine-dinucleotide-phosphate diaphorasepositive cells and fibres in the cat central nervous system. J Comp Neurol 1989; 279: 281–311. Anderson CR. NADPH diaphorase-positive neurons in the rat spinal cord include a subpopulation of autonomic preganglionic neurons. Neurosci Lett 1992; 139: 280–84. Morris R, Southam E, Gittins SR, et al. The NO-cGMP pathway in neonatal rat dorsal horn. Eur J Neuroscience 1994; 1: 876-79. Roche AK, Cook M, Wilcox GL, Kajander KC. A nitric oxide synthesis inhibitor (L-NAME) reduces licking behaviour and Fos-labelling in the spinal cord of rats during formalin-induced inflammation. Pain 1996; 6: 331–41. Southan GJ, Szabo C. Selective pharmacological inhibition of distinct nitric oxide synthase isoforms. Biochem Pharmacol 1996; 4: 383–94. Kindgen-Milles D, Arndt JO. Nitric oxide as a chemical link in the generation of pain from veins in humans. Pain 1996; 64: 139–42. Holthusen H, Arndt JO. Nitric oxide evokes pain at nociceptors of the paravascular tissue and veins in humans. J Physiol 1995; 15: 253–58. Lassen LH, Ashina M, Christiansen I, Ulrich V, Olesen J. Nitric oxide synthase inhibition in migraine. Lancet 1997; 349: 401–02. Blitzer ML, Loh E, Roddy MA, et al. Endothelium-drived nitric oxide regulates systemic and pulmonary vascular resistance during acute hypoxia in humans. J Am Coll Cardiol 1996; 28: 591–96.
289
Effect of inhibition of nitric oxide synthase on chronic tension-type headache: a randomised crossover trial M Ashina, L H Lassen, L Bendtsen, R Jensen, J Olesen
Summary Background Studies in animals have shown that nitric oxide plays an important part in central sensitisation and that inhibitors of nitric oxide synthase (NOS) decrease sensitisation in models of persistent pain. The efficacy of inhibitors of NOS has not been tested in patients with tension-type chronic headache. We aimed to show whether NG-monomethyl-L-arginine hydrochloride (L-NMMA), an inhibitor of NOS, is effective in relieving pain in such patients. Methods We undertook a randomised double-blind, crossover trial of 16 patients with chronic-tension-type headache. Patients were assigned intravenous infusion of 6 mg/kg L-NMMA or placebo on 2 days separated by at least 1 week in a randomised order. Headache intensity was measured on a 100 mm visual analogue scale, and on a verbal rating scale at baseline and at 30 min, 60 min, and 120 min after start of treatment. The primary endpoint was reduction of pain intensity on the visual analogue scale by the active treatment compared with placebo. Findings L-NMMA reduced pain intensity on the visual analogue scale significantly more than placebo: 120 min after start of treatment, the mean pain score was decreased from 49 to 33 with L-NMMA and from 44 to 40 with placebo (p=0·01). Pain intensity on the verbal rating scale was also significantly lower for treatment with L-NMMA than for treatment with placebo (p=0·02). Interpretation Inhibition of NOS had an analgesic effect in chronic tension-type headache. Further tests are required before clinical application.
Lancet 1999; 353: 287–89
Introduction Nitric oxide is a messenger molecule involved in various biological functions including neurotransmission.1–3 Experiments in animals have shown that nitric oxide is an important transmitter in pain pathways of the spinal cord and that central sensitisation of these pathways (increased excitability of neurons in the central nervous system) may be caused by or associated with activation of neuronal nitric oxide synthase (NOS) and the generation of nitric oxide.4,5 Inhibitors of NOS decrease central sensitisation in animal models of persistent pain.6–10 Inhibitors of NOS could, therefore, be a treatment for human chronic pain. We used chronic tension-type headache as a model for chronic pain, since this type of headache is one of the most common types of chronic pain,11 and since central sensitisation due to prolonged nociceptive input from pericranial myofascial tissues may play an important part in Department of Neurology, Glostrup Hospital, University of Copenhagen, DK-2600 Glostrup, Copenhagen, Denmark ( M Ashina MD, L H Lassen MD, L Bendtsen MD, R Jensen MD, J Olesen MD) Correspondence to: Dr M Ashina (e-mail: [email protected])
THE LANCET • Vol 353 • January 23, 1999
the pathophysiology of this disorder.12 We aimed to find out whether NG-monomethyl-L-arginine hydrochloride (L-NMMA), which is the only inhibitor of NOS tested thoroughly in human beings, is effective in the treatment of chronic tension-type headache.
Methods We recruited patients with a diagnosis of chronic tension-type headache according to the criteria of the International Headache Society13 (headache frequency Ä15 days per month for Ä6 months) from the outpatient headache clinic at Glostrup University Hospital. Five of the patients had coexisting infrequent migraine (<4 days/year). All patients completed a diagnostic headache diary14 during a 4-week introductory period to ensure that inclusion criteria were met. Patients were included whether or not they had myofascial tenderness. At screening, we undertook a full physical and neurological examination, 12-lead electrocardiography, routine haematological and biochemical testing, and urine-analysis for each patient. The patients were not allowed to take analgesics, including vasoactive drugs for migraine, for 12 h before treatment. Exclusion criteria were: use of any kind of daily medication, including prophylactic headache therapy but not oral contraceptives; pregnancy or breastfeeding; excessive use of analgesics (>2 aspirin or equivalent daily); excessive alcohol use; serious somatic or psychiatric disorders including depression (Hamilton depression score Ä17);15 ischaemic heart disease; and supine diastolic blood pressure of more than 90 mm Hg or heart rate of less than 50 beats per min. All patients gave written consent. The study was approved by the Danish Board of Health and the local ethics committee, and was undertaken in accordance with the Declaration of Helsinki. The trial had a double-blind, placebo-controlled crossover design. The primary endpoint was reduction of pain intensity on a visual analogue scale by the active treatment compared with placebo. The secondary endpoint was reduction of pain intensity on a verbal rating scale by the active treatment compared with placebo. Patients were randomly assigned 6 mg/kg L-NMMA (Clinalfa, Switzerland) or placebo (isotonic glucose) on 2 days separated by at least 1 week. The patients were treated during a typical day of tension-type headache—patients had pain characteristic of tension-type headache with no more than one associated symptom.13 The patients were randomly assigned by computer (MedStat, version 2.12) into two groups: one group received L-NMMA then placebo at least a week later; the other group received placebo first, then L-NMMA. The investigators were not aware that the non-restricted nature of the randomisation programme might lead to the treatment groups being of unequal size. There was no stratification for coexisting migraine. Randomisation and preparation of study drugs was done by medical staff not involved in the study. The randomisation code remained in the hospital during the study and was not available to the investigators until the study was completed. For each patient, L-NMMA or placebo was infused over 15 min into an antecubital vein, as allocated by the randomisation code. The solutions for infusion looked the same. Headache intensity was measured at baseline and 30 min, 60 min, 90 min, and 120 min after start of infusion on a 100 mm visual analogue scale (0=no pain, 100=worst pain imaginable) and on a verbal rating scale (0=no headache, 5=moderate headache, 10=worst headache imaginable). Blood pressure and pulse rate were measured 5 min before the start of treatment and every 5 min up to 30 min, then every 30 min to 120 min after the start of treatment. Electrocardiography was done continuously, and any adverse events were recorded. Patients with unrelieved headache
287
EARLY REPORTS
Time
16 patients randomised
6 assigned L-NMMA then placebo
10 assigned placebo then L-NMMA
6 completed trial
10 completed trial
Figure 1: Trial profile at 120 min were allowed to take rescue medication. All patients were asked to record details of headache intensity on the visual analogue scale, any medication taken, and adverse events on a diary card every 4 h up to 24 h after the start of treatment. The patients returned to the clinic 4–7 days after each treatment, when the diary cards were collected and any adverse events noted. Calculation of sample size was based on detection of a difference between treatments in reduction of pain intensity on the visual analogue scale at 5% significance (one-sided) with 80% power. We assumed that each patient would show 20% variation on the visual analogue scale. A 15% reduction of pain intensity was taken to be clinically significant. We estimated that 14 patients should be included, but we increased the sample size to 16 patients since this is statistically more desirable. The visual analogue scale was chosen as the primary endpoint because this method for pain measurement is used widely in both clinical and experimental studies. We calculated the sum differences between the pretreatment measurements on each scale (visual and verbal) and the measurement at 30 min, 60 min, 90 min, and 120 min after treatment to obtain a summary pain score for each treatment.16 To test for period and carry-over effects we used independentsamples t tests on the differences and the sums, respectively, of the summary scores for the first and second treatments. The scores on the visual analogue scale were assumed to be normally distributed and summary scores for active and placebo treatment were therefore compared by independent-samples t tests. However, for the ordinal scale of the verbal rating scale we used Wilcoxon signed-ranks tests on the changes in scores from baseline to 24 h after treatment. To assess decreases in blood pressure and pulse rate over the first 30 min after active treatment or placebo, we used paired samples t tests. All these analyses were done with SPSS (version 7.5.1).
Results 16 patients (12 women, four men, mean age 38·5 years, range 23–52) were randomly assigned 6 mg/kg L-NMMA or placebo on 2 days separated by at least 1 week (figure 1). Six patients received L-NMMA then placebo, and ten patients received placebo then L-NMMA. The skewed randomisation gave an efficiency loss of roughly 6·66%, which did not greatly affect the power calculation. Three patients with coexisting migraine received L-NMMA then placebo, and two received placebo then L-NMMA. The mean time between courses of treatment was 13 days for both groups. None of the patients left the study between treatment courses. Seven patients reported side-effects of L-NMMA treatment: two tiredness, three dryness of the mouth, and one each drowsiness, exhaustion, nausea, and a tingling in one arm. Four patients reported sideeffects of placebo treatment: two each tingling in the arms and shoulders, and one each dryness of the mouth, warm sensation in the body, and drowsiness. No patient withdrew from the study because of side-effects. Simple analgesics were used as rescue medication by three patients treated with L-NMMA and by seven patients treated with placebo. There was a significant difference in reduction of pain intensity on the visual analogue scale between treatment 288
Mean (SD) pain score
Baseline 30 min 60 min 90 min 120 min
L-NMMA
Placebo
49 38 35 34 33
44 41 40 42 40
(16) (18) (18) (21) (21)
(14) (17) (17) (16) (17)
Pain scores on a 100 mm visual analogue scale
with L-NMMA and placebo (table, figure 2, p=0·01). There was no significant period effect (p=0·32) or carryover effect (p=0·15). The pain intensity on the verbal rating scale was significantly lower during treatment with L-NMMA than during treatment with placebo (p=0·02), but this difference had disappeared by 24 h after the start of treatment. The changes in mean arterial blood pressure and pulse rate during treatment with L-NMMA or placebo differed significantly (blood pressure p=0·0001, pulse rate p=0·001, figure 3). The peak increase in blood pressure was 12% (SD 2) and occurred 15 min after treatment with L-NMMA. Mean pulse rate decreased by 16% (2) 10 min after the start of the L-NMMA infusion. The increase in mean arterial blood pressure and decrease in pulse rate during L-NMMA treatment are consistent with known pharmacological properties of L-NMMA.17 Patients were clinically unaffected by these changes.
Discussion Little progress has been made in the treatment of chronic pain, including low back pain, arthritis, and chronic tension-type headache. Opioids are not effective in most patients and are contraindicated owing to the risk of addiction. Non-opioids are not effective and may even aggravate chronic tension-type headache.18 Drugs are needed with new mechanisms of action based on improved knowledge of the neurobiology of chronic pain. Central sensitisation plays an important part in the pathophysiology of chronic pain.19 There is evidence to show that persistent activity in peripheral nociceptors may lead to sensitisation of neurons in the spinal dorsal horn, partly via activation of N-methyl-D-aspartate (NMDA) receptors.20 Many of the effects of NMDA-receptor activation are mediated via production of nitric oxide, and immunoreactivity related to NOS has been identified in the spinal dorsal horn.21–23 Nitric oxide may, therefore, play an important part in hyperalgesia in the spinal cord.5 Animal models of persistent pain have shown that inhibitors of NOS decrease sensitisation of the spinal dorsal horn induced by Change in pain-intensity on visual scale (%)
16 patients eligible
L-NMMA Placebo
10 0 –10 –20 –30 –40 –50 0
15
30
60 Time (min)
90
120
Figure 2: Mean percentage change from baseline in pain intensity on 100 mm visual analogue scale Bars=SE.
THE LANCET • Vol 353 • January 23, 1999
EARLY REPORTS
Change from baseline (%)
20
Mean arterial blood pressure on L-NMMA Mean arterial blood pressure on placebo Mean pulse rate on placebo Mean pulse rate on L-NMMA
10
and the Danish Hospital Foundation for Medical Research, Region of Copenhagen, the Faroe Islands, and Greenland.
References 1
0 2
–10 3 4
–20 0
10 20 30 40 50 60 70 80 90 100 110 120 Time (min) Figure 3: Mean percentage change from baseline in mean arterial blood pressure and pulse rate during and after treatment with L-NMMA and placebo
5 6
continuous painful input from the periphery.6,10,24 However, the efficacy of inhibitors of NOS has not been assessed in patients with chronic pain. We suggest that inhibition of NOS has an analgesic effect in human chronic tension-type headache, a model for chronic pain. This effect may be due to reduction of central sensitisation at the spinal dorsal horn/trigeminal nucleus. L-NMMA inhibits all three types of NOS (endothelial NOS, neuronal NOS, inducible NOS),25 and nitric oxide is involved in the transmission and modulation of nociceptive information centrally and in the periphery.26,27 Study of selective inhibitors of NOS25 is needed to show which type of NOS is involved in human chronic pain and its exact site of action. Inhibition of NOS is effective in the treatment of acute migraine attacks, probably because of interaction with specific migraine mechanisms.28 The analgesic effect of inhibition of NOS in our study might be influenced by the presence of coexisting infrequent migraine in five patients. However, we found that patients without coexisting migraine tended to have a greater reduction of pain intensity (35%) than patients with coexisting migraine (23%). This finding shows that our results were not influenced by the inclusion of patients with coexisting infrequent migraine and that the mechanisms behind the analgesic effect of NOS inhibition differ in tension-type headache and migraine. Although the effect of NOS inhibition was significant, pain was reduced by only 30%. The clinical significance of the reduction is unclear, but our patients had had chronic tension-type headache for many years and had tried many other treatments without effect. Furthermore, acute dosing with a drug that has a short half-life29 is not ideal for a chronic disorder. The purpose of our study was not to develop L-NMMA for clinical use but to test whether inhibition of NOS had any effect or not. Since the result was positive, this approach should be tested in other (nonheadache) types of chronic pain. Chronic dosing should be tested when a safe inhibitor of NOS with a longer half-life becomes available. Longer doses should also be tested, but because of the increase in blood pressure as a result of inhibition of endothelial NOS, such tests will have to await the development of selective inhibitors of neuronal NOS.
25
Contributors
26
Messoud Ashina designed the study, collected and processed data, and was responsible for writing the paper. Lisbeth Hjorth Lassen, Lars Bendtsen, Rigmor Jensen, and Jes Olesen helped to design the study and undertook data interpretation. All investigators contributed to writing the paper.
7
8
9
10
11
12 13
14 15 16 17
18 19
20
21
22
23 24
27 28
Acknowledgments We thank Hanne Andresen for technical assistance, and Peter Dalgaard from the Department of Biostatistics at the University of Copenhagen for statistical advice. The study was supported by the Danish Headache Society
THE LANCET • Vol 353 • January 23, 1999
29
Ignarro LJ, Lippton H, Edwards JC, et al. Mechanisms of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprusside and nitric oxide: evidence for the involvement of S-nitrosothiols as active intermediates. J Pharmacol Exp Ther 1981; 218: 739–49. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol Rev 1991; 43: 109–41. Garthwaite J. Nitric oxide signalling in the nervous system. Neurosciences 1993; 5: 171–80. Haley JE, Dickenson AH, Schachter M. Electrophysiological evidence for a role of nitric oxide in prolonged chemical nociception in the rat. Neuropharmacology 1992; 31: 251–58. Meller ST, Gebhart GF. Nitric oxide (NO) and nociceptive processing in the spinal cord. Pain 1993; 52: 127–36. Meller ST, Cummings CP, Traub RJ, Gebhard GF. The role of nitric oxide in the development and maintenance of the hyperalgesia produced by intraplantar injection of carrageenan in the rat. Neuroscience 1994; 60: 367–74. Minami T, Nishihara I, Ito et al. Nitric oxide mediates allodynia induced by intrathecal administration of prostaglandin E2 or prostaglandin F2 alpha in conscious mice. Pain 1995; 61: 285–90. Coderre TJ, Yashpal K. Intracellular messengers contributing to persistent nociception and hyperalgesia induced by L-glutamate and substance P in the rat formalin pain model. Eur J Neuroscience 1994; 1: 1328–34. Hao J-X, Xu X-J. Treatment of chronic allodynia-like response in spinally injured rats: effects of systemically administered nitric oxide synthase inhibitors. Pain 1996; 6: 313–19. Mao J, Price DD, Zhu J, Lu J, Mayer DJ. The inhibition of nitric oxideactivated poly (ADP-ribose) synthase attenuates transsynaptic alteration of spinal cord dorsal horn neurons and neuropathic pain in the rat. Pain 1997; 72: 355–66. Rasmussen BK, Jensen R, Schroll M, Olesen J. Epidemiology of headache in a general population: a prevalence study. J Clin Epidemiol 1991; 44: 1147–57. Bendtsen L, Jensen R, Olesen J. Qualitatively altered nociception in chronic myofascial pain. Pain 1996; 65: 259–64. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988; 8 (suppl 7): 1–96. Russell MB, Rasmussen BK, Brennum J, et al. Presentation of a new instrument: the diagnostic headache diary. Cephalalgia 1992; 12: 369–74. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 56–62. Matthews JN, Altman DF, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990; 300: 230–35. Haynes WG, Noon JP, Walker BR, Webb DJ. Inhibition of nitric oxide synthesis increases blood pressure in healthy humans. J Hypertens 1993; 11: 1375–80. Olesen J. Analgesic headache. BMJ 1995; 310: 479–80. Woolf CJ, Doubell TP. The pathophysiology of chronic pain-increased sensitivity to low threshold A beta-fibre inputs. Curr Opin Neurobiol 1994; 4: 525–34. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993; 52: 259–85. Mizukawa K, Vincent SR, McGeer PL, McGeer EG. Distribution of reduced-nicotinamide-adenine-dinucleotide-phosphate diaphorasepositive cells and fibres in the cat central nervous system. J Comp Neurol 1989; 279: 281–311. Anderson CR. NADPH diaphorase-positive neurons in the rat spinal cord include a subpopulation of autonomic preganglionic neurons. Neurosci Lett 1992; 139: 280–84. Morris R, Southam E, Gittins SR, et al. The NO-cGMP pathway in neonatal rat dorsal horn. Eur J Neuroscience 1994; 1: 876-79. Roche AK, Cook M, Wilcox GL, Kajander KC. A nitric oxide synthesis inhibitor (L-NAME) reduces licking behaviour and Fos-labelling in the spinal cord of rats during formalin-induced inflammation. Pain 1996; 6: 331–41. Southan GJ, Szabo C. Selective pharmacological inhibition of distinct nitric oxide synthase isoforms. Biochem Pharmacol 1996; 4: 383–94. Kindgen-Milles D, Arndt JO. Nitric oxide as a chemical link in the generation of pain from veins in humans. Pain 1996; 64: 139–42. Holthusen H, Arndt JO. Nitric oxide evokes pain at nociceptors of the paravascular tissue and veins in humans. J Physiol 1995; 15: 253–58. Lassen LH, Ashina M, Christiansen I, Ulrich V, Olesen J. Nitric oxide synthase inhibition in migraine. Lancet 1997; 349: 401–02. Blitzer ML, Loh E, Roddy MA, et al. Endothelium-drived nitric oxide regulates systemic and pulmonary vascular resistance during acute hypoxia in humans. J Am Coll Cardiol 1996; 28: 591–96.
289