SEARCH

SEARCH BY CITATION

Keywords:

  • migraine;
  • divalproate;
  • amitriptyline;
  • antiepileptic;
  • sodium valproate;
  • tricyclic antidepressant;
  • randomized controlled trial ;
  • prophylaxis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

Objective

This study compares efficacy and safety of divalproate extended release (DVA-ER) and amitriptyline (AMT) in migraine.

Materials and methods

Three hundred migraineurs having >4 attacks monthly were randomized into DVA-ER or AMT. The primary end points were >50% reduction in frequency, ≥1 grade improvement in the severity, and >50% improvement in a visual analogue scale (VAS). Secondary end points were functional disability, rescue medication, and adverse events.

Results

The median age was 32 years, and 241 were women. 150 patients each received DVA-ER and AMT. At 3 months, 74.7% in DVA-ER and 62% patients in AMT group improved in headache frequency (= 0.02) and at 6 months, 65.3% and 54%, respectively (= 0.90). At 3 months, the VAS score improved by >50% in 80.7% in DVA-ER and 64% in AMT (= 0.005). At 6 months, there was no significant difference between the two groups in VAS score (69.3% vs 56%; = 0.47) and other outcome parameters. The composite side effects were also not different between the two groups (68% vs 81%); however, hair fall, menstrual irregularity, polycystic ovary, and weight gain were commoner in DVA-ER group.

Conclusion

Divalproate extended release is more effective at 3 months than AMT; however, at 6 months, both are equally effective in migraine prophylaxis.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

Headache is a common health problem and affects more people than asthma and diabetes mellitus combined [1, 2]. Headache results in moderate to severe disability in more than 50% patients [3]. As per World Health Organization, migraine is one of the topmost disabling health disorders and leads to a lower quality of life compared with diabetes, heart disease, and depression [4-6]. Migraine is more than just an episodic disorder, which has led to emphasis on prophylactic therapy to prevent or reduce the frequency of acute attacks and improve the level of functioning by reducing disability. For migraine prophylaxis, a number of drugs of different classes are used and include antidepressants (amitriptyline, fluoxetine, and venlaflexine), antiepileptics (valproate, divalproate, and topiramate), beta-adrenergic blockers (propranolol, atenolol, and metoprolol), and calcium channel blocker (flunarizine).

Amitriptyline (AMT) was first used in the treatment of migraine in 1964, and in this study on 27 subjects, AMT resulted in significant improvement in 56% patients [7]. AMT was also found to be superior to placebo in reducing the headache frequency (55% vs 34%) [8]. In a randomized controlled trial (RCT), AMT resulted in significant improvement in headache frequency at 8 and 16 weeks compared with placebo in chronic daily headache [9]. AMT was found to be as effective as topiramate in a recent RCT [10]. Divalproate extended release (DVA-ER) has been recommended for migraine prophylaxis since 2005, and its efficacy has been reported in 45% to 75% of patients [11]. In a retrospective analysis of 642 patients on sodium valproate (SVA), there was 65% improvement, which was independent of age, gender, and duration of treatment [12]. The majority of studies on DVA/SVA are randomized placebo-controlled. There are only few studies comparing the efficacy of DVA/SVA with other migraine prophylactic drugs. In a study, SVA resulted improvement in migraine in 71% and flunarizine in 65% of patients [13]. Sodium valproate and topiramate were found to be equally effective in reducing the number of headache days [14]. Both DVA and amitriptyline have class 1 recommendation for migraine prophylaxis. There is, however, no study comparing the relative efficacy and tolerability of DVA and AMT. In this communication, we report the efficacy and tolerability of DVA-ER (extended release) vs AMT in migraine prophylaxis.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

Study design

This is a single-center prospective study with randomized controlled open-labelled design and evaluates the efficacy and safety of DVA-ER and AMT. The study was done in a tertiary care teaching hospital, and patients were enrolled from the neurology outpatient service. The protocol was designed and written by the first and third authors. The trial was not registered; however, it was approved by the Institute Ethics Committee. All the patients gave informed consent. The sample size was calculated keeping the type 1 error α = 0.05 and type II error β = 0.1 using Z-test of proportion. The critical response between the study drugs was considered to be significant as 20% considering the efficacy of DVA-ER as 70%. The sample size was calculated to be 134 on each arm with the power of test 90%. One hundred and fifty patients in each treatment arm were decided to be enrolled in this study.

Selection of the study patients and their evaluation

Inclusion criteria

Migraine patients between 15 and 60 years of age having more than four moderate to severe attacks were included in this study. The diagnosis of migraine was based on International Headache Society Criteria [15].

Exclusion criteria

The patients with history of drug allergy, severe hypertension, coronary artery disease, pregnancy, menstrual irregularity, liver or kidney dysfunction, polycystic ovary, systemic or psychiatric disease, malignancy, glaucoma, dysautonomia, and those unwilling were excluded.

Clinical evaluation

A medical history including duration of migraine, distribution, frequency and triggering factors, family history of headache, diabetes, hypertension, and other medical illness was enquired. The severity of headache was graded on a 0–3 scale (0 = none, 1 = mild, 2 = moderate, 3 = severe). The functional disability was graded on a 0–4 scale (0 = none, 1 = mild, 2 = moderate, 3 = severe impairment of daily activity, 4 = inability to perform daily activity requiring bed rest). The severity of associated symptoms such a nausea, vomiting, photophobia, and phonophobia was recorded on a 0–3 scale (0 = none, 1 = mild, 2 = moderate, 3 = severe). General medical and neurological examinations were performed in all. All the patients maintained a headache diary 1 month prior to the randomization and during the study period.

Investigations

Blood counts, hemoglobin, ESR, urinalysis, fasting and post-prandial blood sugar, serum creatinine, electrolytes, bilirubin, transaminases, and serum calcium were estimated. Electrocardiogram was performed in all, and abdominal ultrasonography was carried out in the females who had menstrual irregularities.

Randomization and treatment

The patients were randomized to amitriptyline or DVA-ER using the computer-generated random table numbers. The randomization, treatment, and evaluation of the patients were performed by different investigators. The identity of the drug was not concealed. Amitriptyline was prescribed in a dose of 12.5 mg per oral for 2 weeks, thereafter 25 mg daily which was increased to 50 mg if needed depending on the response or tolerability. Divalproate extended release was prescribed in a dose of 250 mg daily for 2 weeks followed by 500 mg daily and increased to 1000 mg depending on the response and tolerability. Ibuprofen 400 mg was used as rescue medication. The patients were advised to report if there was any adverse effect.

Follow-up

The patients were followed up at 3 and 6 months, and their headache frequency, severity, functional disability, number of rescue medication, and overall improvement on a 0–100 visual analogue scale (VAS) were recorded. Patients' body weight, liver function tests, and abdominal ultrasonography in the females who had menstrual irregularities were carried out.

Outcome

The primary outcome parameters were (i) more than 50% reduction in the headache frequency, (ii) more than 50% improvement in the overall headache on VAS score, and (iii) one or more grade reduction in the headache severity scale. The secondary outcome parameters included (i) one grade or more improvement in the functional disability, (ii) reduction in the number of rescue medication, and (iii) adverse events.

Statistical analysis

The demographic, clinical, and laboratory parameters in amitriptyline and DVA-ER groups were compared using various parametric and non-parametric tests. The severity and frequency of headache, number of rescue medication, and functional disability at baseline 3 months and 6 months following treatment within the group were compared using analysis of variance (ANOVA). The number of patients with the primary and secondary end points at 3 and 6 months was compared by chi-square test with Yates' correction. The intention to treat analysis was done for primary and secondary end points. The variables were considered significant if the two-tailed P-value was less than 0.05. The statistical analysis was performed using SPSS for Windows, version 12 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism for windows, Version 3 (GraphPad software Inc., Sandiego, CA, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

Recruitment of the patients

During the study period, 360 patients with migraine having more than four attacks per month were screened. Sixty patients were excluded because of associated tension-type headache in 14, deranged liver and renal function in 8, connective tissue disease in 6, pregnancy or lactation in 4, coronary artery disease in 3, enlarged prostate in 2, major depression in 2, acute close-angle glaucoma in 1, and lack of consent in 20 patients (Fig. 1). Our results are therefore based on 300 patients with migraine whose age ranged between 15 and 60 (median 32) years and 241 were women. The median duration of migraine was 5 years (6 months to 40 years). The family history of headache in the first-degree relatives was present in 86 patients. All the patients had moderate to severe headache. The median headache frequency was 6 per month prior to randomization. All the patients had migraine without aura except 19. The median severity of headache was 2.91 (range 2–3), and the functional disability was 3. Divalproate extended release and amitriptyline were prescribed to 150 patients each. The demographic and clinical features in both the groups were not significantly different (Table 1).

Table 1. Baseline characteristics of migraine patients receiving amitriptyline and divalproate
ParametersDivalproate (= 150)Amitriptyline (= 150)P-value
Age (years)31.03 ± 10.532.80 ± 9.410.13
Education (years)10.22 ± 5.059.78 ± 5.110.45
Female123 (82%)118 (78.7%)0.56
Rural/urban65/8576/740.25
Family history of headache48 (32%)38 (25.3%)0.25
Duration of illness (years)7.23 ± 6.447.41 ± 6.540.81
Total number of triggers10.44 ± 2.8810.24 ± 2.970.55
Migraine with aura14 (9.3%)5 (3.3%)0.06
Frequency of migraine/month10.49 ± 9.0411.19 ± 9.650.51
Severity of headache2.95 ± 0.222.88 ± 0.380.06
Functional disability3.19 ± 0.623.05 ± 0.690.06
Rescue analgesics/month10.38 ± 6.9510.17 ± 7.950.81
image

Figure 1. Flowchart showing consort statement.

Download figure to PowerPoint

Outcome

Primary outcome

All the primary and secondary end points showed significant improvement at 3 and 6 months compared with baseline in the both DVA-ER and the AMT groups. The frequency of headache and number of rescue medications were not significantly changed at 6 months compared with 3 months in the DVA-ER group (Figs 2 and 3). On comparing the efficacy between the amitriptyline and the DVA-ER group at 3 months, there was significant improvement in the headache frequency in DVA-ER compared with AMT group (= 0.02). In the DVA-ER group, 112 (74.7%) patients had improved with respect to the headache frequency, whereas in the AMT group, 93 (62%) patients improved. At 6 months, however, there was no significant difference in the response between the DVA-ER and the AMT group (65.3% vs 54%, = 0.90). In the DVA-ER group, one patient at 3 months and nine patients at 6 months and in the AMT group two patients at 3 months and nine patients at 6 months were headache free. The severity of headache reduced in both the groups at 3 and 6 months compared with baseline, but it was not significantly different between DVA-ER and AMT group both at 3 months (= 0.09) as well as at 6 months (= 0.05). 138 (92%) patients in the DVA-ER group had improvement at 3 months and 102 (68.0%) at 6 months, whereas in the AMT group, these were 131 (87.3%) and 78 (52%), respectively. In the VAS score, 121 (80.7%) patients had more than 50% improvement at 3 months and 104 (69.3%) at 6 months in DVA-ER group. In the AMT group, these were 96 (64%) and 84 (56%), respectively. The details are presented in the Table 2.

Table 2. The primary and secondary outcome parameters at 3 and 6 month following divalproate and amitriptyline prophylaxis in patients with migraine using intention to treat analysis
ScaleDivalproate N (ITT%)Amitriptyline N (ITT%)P-value
  1. ITT, intention to treat analysis; VAS, visual analogue scale.

  2. [Baseline: Divalproate (= 150), Amitriptyline (= 150); 3 month: Divalproate (= 143), Amitriptyline (= 144); 6 month: Divalproate (= 107), Amitriptyline (= 89)].

Primary outcome
VAS (0–100)
0–3 months>50%121 (80.7)96 (64)

0.005

OR-2.74 (1.51–5)

<50%23 (19.3)50 (36)
0–6 months>50%104 (69.3)84 (56)

0.47

OR-2.06 (0.41–11.26)

<50%3 (30.7)5 (44)
Frequency of migraine/month
0–3 months>50% improvement112 (74.7)93 (62)

0.02

OR-1.9 (1.09–3.34)

<50% improvement31 (25.3)51 (38)
0–6 months>50% improvement98 (65.3)81 (54)

0.90

OR-1.08 (0.36–3.21)

<50% improvement9 (34.7)8 (46)
Severity of headache ≥1grade improvement
0–3 monthsImprovement138 (92)131 (87.3)

0.09

OR-2.74 (0.08–9.08)

No improvement5 (8)13 (12.7)
0 vs 6 monthsImprovement102 (68)78 (52)

0.05

OR-2.88 (0.88–9.96)

No improvement5 (32)11 (48)
Secondary outcome
Functional disability ≥1grade improvement
0–3 monthsImprovement141 (94)137 (91.3)

0.17

OR-3.6 (0.67–25.6)

No improvement2 (6)7 (8.7)
0–6 monthsImprovement106 (70.7)87 (58)

0.60

OR-2.44 (0.17–69.07)

No improvement1 (29.3)2 (42)
Average number of analgesic/monthBaseline10.38 ± 6.9510.17 ± 7.950.81
3 month2.92 ± 2.283.45 ± 2.440.06
6 month1.21 ± 0.941.43 ± 1.40.20
image

Figure 2. Error bar diagram shows significant improvement in frequency of migraine attack/month and its severity following divalproate (DVA) and amitriptyline (AMT) at 3 and 6 months compared with baseline. Patients on DVA had significant reduction in frequency of headache at 3 months compared with AMT.

Download figure to PowerPoint

image

Figure 3. Error bar diagram shows significant improvement in functional disability and reduction in number of analgesics use following divalproate (DVA) and amitriptyline (AMT) at 3 and 6 months compared with baseline. The improvement in different time points, however, was not significantly different between DVA and AMT groups.

Download figure to PowerPoint

The secondary end points were also improved in both the groups from the baseline, but the difference between the DVA-ER and AMT groups was not significant. The details of the response in the functional disability, number of rescue medications, and side effects are presented in Tables 2 and 3. On subanalysis intention to treat analysis (ITT) of the patients having migraine days ≤10 and >10 per months, DVA-ER resulted significant improvement in VAS score at 3 months (80% vs 62.7%, = 0.001) as well as at 6 months (69.3% vs 56%, = 0.02) compared with AMT. There was, however, no difference in the response in the other outcome parameters. The effect of AMT was not significantly different between those having <10 or more than >10 migraine days per month at 3 months (68.4% vs 63.2%, = 0.52) and 6 months (66.7% vs 52.9%, = 0.10) follow-up.

Table 3. Side effects of amitriptyline and divalproate at 3 and 6 months in migraine patients
Side effectsDivalproate N (%)Amitriptyline N (%)P-value
  1. PCOS, polycystic ovarian syndrome.

  2. [Baseline: Divalproate (= 150), Amitriptyline (= 150); 3 month: Divalproate (= 143), Amitriptyline (= 144); 6 month: Divalproate (= 107), Amitriptyline (= 89)].

Total side effects3 months68 (47.6)81 (56.3)0.16
6 months41 (38.3)30 (33.7)0.6
Drowsiness3 months7 (4.9)69 (47.3)0.001
6 months4 (3.7)9 (10.1)0.09
Hair fall3 months55 (38.5)2 (1.4)0.001
6 months10 (9.3)1 (1.1)0.01
Menstrual irregularity3 months6 (4.8)00.02
6 months9 (8.4)1 (1.1)0.001
Dry mouth3 months13 (9.1)78 (56.5)0.001
6 months4 (3.7)24 (27)0.02
G.I. symptoms3 months18 (12.6)12 (8.3)0.25
6 months16 (15)4 (4.5)0.001
Giddiness3 months3 (2.1)4 (3.6)1.0
6 months6 (5.6)3 (3.4)0.5
Vomiting3 months01 (0.7)1.0
6 months00NA
PCOS6 months6 (5.4)00.03
Weight gain3 months79 (61.7)71 (58.7)0.7
6 months66 (66.7)27 (37)0.001
Side effects

At 3 months, 68 patients in the DVA-ER and 81 in the AMT group had side effects (= 0.16). Majority of side effects were mild; DVA-ER was discontinued in six patients and AMT in four due to side effects. Drowsiness and dry mouth were more frequent in AMT, whereas hair fall, menstrual irregularity, gastro intestinal symptoms, weight gain, and polycystic ovarian syndrome (PCOS) were more common in DVA-ER group. Weight gain (66 vs 27, = 0.001) and PCOS (6 vs 0, = 0.03) were significant at 6 rather than 3 months. The side effects are summarized in the Table 3.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

In this study on migraine prophylaxis, both DVA-ER and AMT resulted in significant improvement in all the outcome parameters at 3 and 6 months compared with baseline. At 3 months, DVA-ER significantly reduced headache frequency (74.7% vs 62%) and improved in VAS score (80.7% vs 64%) compared with AMT. At 6 months, however, the difference in the response between the DVA-ER and the AMT group was not significant. In the available literature, there is no study comparing the efficacy of DVA-ER and AMT in migraine prophylaxis. The reported efficacy of DVA/SVA has ranged between 45% and 86.2% [11, 13-23]. Most of these studies have compared the efficacy of DVA or SVA with placebo. The head to head comparisons are available in small series in reference to topiramate [14, 23]. Most of these studies have compared the headache frequency as the primary end point except 2; one study compared duration, frequency, and severity of migraine [16], and the other study compared the severity, general pain VAS score, and frequency of headache [18]. In addition to headache frequency, we have also included severity, overall improvement in headache, functional disability, and number of rescue medications, which have lead to a more comprehensive evaluation.

The advantage of DVA-ER over AMT at 3 months was not sustained at 6 months. The efficacy with respect to 50% reduction in headache frequency in AMT and DVA-ER was similar at 6 months. This may be due to a quicker response of DVA-ER and slower and sustained response of AMT. Most of the earlier studies have evaluated the response at 4–12 weeks [8, 24]. Intravenous sodium valproate has also been used in a dose of 900 mg to 1200 mg as an abortive therapy in 36 patients with severe migraine. It resulted in pain reduction or relief in 75% patients at 1 h without any adverse event [25]. In a recent study, the efficacy and tolerability of topiramate and AMT were compared in 331 patients with migraine. The mean monthly number of migraine attacks was not significantly different between the two groups. The secondary outcome measures such as severity of migraine-associated symptoms, mean monthly migraine days, and total headache days were also not different between the topiramate and the AMT group. Patients on topiramate, however, had improvement in functional disability and reduction in body weight compared with those on AMT [10]. In this study, amitriptyline was titrated up to 100 mg, which is double the dose in comparison with our study. A large trial of AMT was performed in 1976–1979 [8]. The report was delayed due to loss of patent protection in 1980, and reanalysis of these data was published in 2011 [9]. There is no large study on smaller dose of AMT. Even with smaller dose of AMT in our study, 81 patients developed side effects necessitating withdrawal of drug in four patients. The improvement in headache parameters in our study, however, was comparable with other studies using higher dose of amitriptyline. The efficacy of smaller dose of amitriptyline in our study may be due to smaller body configuration of Indians, different metabolic rate, or genetic basis.

In a study, AMT has been reported to reduce the migraine frequency by 42% at 8 weeks, 47% at 12 weeks, 48% at 16 weeks, and 51% at 20 weeks, suggesting progressive improvement in the response. The severity and the duration of migraine also had a sustained effect, and the maximum response was reported at 3 months [9]. In our study, however, the improvement in the frequency and severity of headache, VAS score, and functional disability was significant even at 6 months compared with 3 months, confirming ongoing improvement.

Tricyclic antidepressants (TCAs) upregulate the GABA-B receptor, downregulate histamine receptor, and reduce the neuronal sensitivity to substance-P. Tricyclic antidepressants also interact with endogenous adenosine systems in the central nervous system by inhibiting neurogenic uptake of adenosine and augmenting the action of adenosine. The enhanced availability of adenosine and adenosine receptors contributes to antinociception [26]. The mechanism of action of SVA in migraine prophylaxis includes facilitation of GABA-ergic neurotransmission, decrease in the activity of serotonergic cells, and reduction in neurogenic inflammation. Thereby, SVA/DVA acts both at the stimulus and at the site of pain resulting in reduction of neurogenic inflammation [27]. The cumulative side effects in the patients receiving DVA-ER and those receiving AMT, although, were not different, but anticholinergic side effects and sedation were common in the AMT group, and hair fall, menstrual irregularity, and PCOS were common in the DVA-ER group. In a study on 194 patients with migraine who were on AMT; the adverse events were dry mouth in 68, constipation in 23, urinary retention in 6, dizziness in 21, and somnolence in 53 patients which were significantly more compared with placebo. At least one side effect was noted in 111 patients in the AMT and in 53 patients in the placebo group [9]. In a long-term study of DVA in chronic daily headache, adverse events occurred in 35% patients. Weight gain was noted in 7.1% patients. None had severe adverse events necessitating drug withdrawal [12]. In another study, no significant difference in side effects was noted between the DVA and the placebo group [17]. The common side effects leading to drug withdrawal were anorexia, nausea, and disturbance in attention [22]. Higher frequency of side effects in our study especially in females may be due to greater susceptibility and vigor of monitoring, as in the abovementioned study, and pelvic ultrasonography was not discussed.

We have not used a placebo arm in our study. Placebo effect, however, is reported in variable proportion of migraine patients, both in abortive and in prophylactic treatment. In a meta-analysis of 22 placebo-controlled trials, the percentage of placebo responders (>50% improvement) was 23.5% ± 8.0% [28]. The response has been reported to be more pronounced in invasive placebo treatment [29]. The analysis of pooled data of 2 randomized placebo-controlled trial on onabotulinum toxin A revealed placebo effect in 35.1% at 24 weeks [30]. In view of available drug options, it may not be ethical to follow placebo orthodoxy. Our aim was to compare the efficacy and tolerability of these two class I drugs recommended for migraine prophylaxis. The dose of the AMT was comparatively lower in our study than in the reported literature; although, even with this dose, the efficacy of AMT was comparable to DVA at 6 months.

It can be concluded from this class I study that DVA-ER is more effective at 3 months, but AMT is equally effective at 6 months and seems to be safer and more desirable in females.

Author's Contribution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References

Jayantee Kalita and Usha K Misra performed Planning, interpretation, and writing of manuscript. Sanjeev K Bhoi carried out data collection and analysis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. Source of funding
  10. Author's Contribution
  11. Ethics approval
  12. References