Stage 1. Prioritization of topics

The methodological group developed a first list of questions about the treatment of each of the four groups included in the definition of CDH. This list of questions was evaluated by a subgroup of the group of experts who rated the importance of including each question on a seven-point Likert scale (from nothing relevant to totally relevant).

The questions that obtained an interquartile range of 7 to 9 were considered for conducting systematic reviews ¹⁵.

After determining the questions related to each topic, the developer group created a search strategy based on the combination of the terms used for the denomination of the different types of chronic daily headache and the filter with the best balance of sensitivity and specificity for the identification of systematic reviews on the OVID platform. In the interventions that required searches for primary studies, no restrictions were used by type of study. All searches were carried out in the OVID version of the Medline, Embase and Cochrane databases and were limited to studies conducted in humans and published in English over the past 20 years (figure 1). Systematic reviews were evaluated by two independent evaluators who rated them using the Amstar methodology; those with a moderate and high evidence score were included in the evidence material considered for this project.

Stage 2. First round of consensus

The methodological group sent a series of preliminary recommendations and their respective tables of evidence individually to each expert via email. Each member was asked to rate their degree of agreement with each proposal, using a seven-point Likert scale. Additionally, each expert had the opportunity to openly write their comments on each preliminary recommendation. Proposals in which an interquartile range of 1 to 3 was obtained were considered as not approved by consensus and those that obtained an interquartile range of 7 to 9 were considered proposals approved by consensus. The proposals that obtained different interquartile ranges were taken to a second round of qualifications.

Stage 3. Second round of consensus

The proposals that obtained interquartile ranges between 4 and 7 in the first round of the consensus were sent back to each expert via email. Together with these proposals, each participant was sent a tabulation of the results of the qualifications in the first round and the comments made anonymously by the experts. Each of the new proposals from this group was rated individually by each participant. Proposals in which no consensus was obtained in the second round of ratings were taken to consensus through the nominal group method.

Stage 4. Nominal group

For the realization of the nominal group, the thematic leaders of each subgroup (one for each subtype of CDH) met to discuss each of the issues on which no consensus was reached. In this stage, arguments based on experience and theoretical sources were presented until reaching the consensus of the majority of the participants. At the end of the meeting, the group of experts evaluated the wording of each of the recommendations for inclusion in the final manuscript, including the general principles of the therapeutic approach of the CDH (table 2).

Table 2 General principles for the therapeutic approach of patients with CDH. Source: authors. 

Evidence analysis

Botulinum toxin A - onabotulinum toxin A (ONABOT A) was studied in two randomized controlled clinical trials, PREEMPT 1 and PREEMPT 2. In total, 1384 individuals between 18 and 65 years of age were studied, with a dose of 155-195 u, in 31-39 places of the scalp, compared with placebo and baseline frequency vs. week 24. In PREEMPT 1, the number of headache episodes was analyzed as primary outcome, and there was no statistically significant difference vs. placebo (5.2 vs. 5.3; p=0.344). However, secondary outcomes, headache days (p=0.006) and migraine days (p=0.002) showed a significant reduction compared to placebo ¹⁴. In PREEMPT 2 the number of days with headaches was chosen as primary outcome, and superiority of ONABOT A was evidenced vs. placebo (-9.0 vs. -6.7, respectively, p<0.001) ¹⁵. The secondary outcomes in PREEMPT 2; migraine days frequency, frequency of days of moderate to severe intensity, cumulative monthly hours of migraine, proportion of patients with severe HIT-6 and frequency of headache episodes showed favorable results that support ONABOT A compared to placebo; p<0.05 in all cases. Based on the data obtained from the pooled analysis of both studies ¹⁶ (onabotulinumtoxin A 47.1% vs. placebo 35.1%; p<0.001), an NNT of 8.3 was calculated to achieve 50% improvement. Both studies showed a favorable safety profile, with a low probability of withdrawal due to side effects, NNH: 38. These findings were confirmed in a pooled analysis that showed a decrease in the incidence of adverse effects when comparing cycle 1 with 5 ¹⁷. In the long-term follow-up, the COMPEL study ¹⁸ analyzed 716 patients aged between 18 and 76 years and showed a reduction of 9.2 and 10.7 days of headache at week 60 and 108, respectively, p<0.0001, from a record of 22 days/ month at baseline. These results match with the 56-week analysis based on the pivotal studies ¹⁴.

Clinical considerations

ONABOT A is effective in the treatment of chronic migraine with more evidence when comparing the reduction in headache days/month vs. migraine days/month. Studies show limitations that are represented in the risk of unmasking due to aesthetic effects in treated patients ¹⁹.

According to the recommendation of experts, therapy with ONABOT A should be initiated in patients who meet the ICHD 3 criteria for chronic migraine, after at least two medications at a maximum tolerated dose for a period of two months have failed. In those patients without response to the first cycle, the opportunity to perform cycle two and three can be considered, taking into account that the probability of obtaining 50% efficacy is 11.3% and 10.3% for each cycle, respectively ²⁰; individual time and clinical conditions should be part of the analysis factors when considering this choice. If a responding patient is considered, it is recommended to maintain stable doses and intervals of application every 12 weeks. After the first year it is possible to determine the following cycles at 16 and 20 weeks, respectively, with new adjustments according to the clinical response in each case. According to Grade recommendations, the evidence strongly favors the reduction in headache days and migraine days, with partial evidence in migraine episodes (table 3).

Table 3 Grade system, therapeutic options in chronic migraine. 

CCT: controlled clinical trial

1. Study sponsored by the pharmaceutical industry

2. Risk of inconsistency

3. Risk of inaccuracy

4. No clarity in the selective outcome report

Acta Neurol Colomb. 2020; 36(3): 150-167

5. Unclear generation of random sequences

6. Unclear allocation concealment

7. Unclear intervention masking

8. Significant losses to follow-up (> 20 %)

9. No intention-to-treat analysis was carried out

Source: authors.

Final Recommendation: ONABOT A is recommended for the treatment of patients with chronic migraine. Quality of evidence for headache days: high, recommendation: strong

2. Are anti CGRP drugs effective and safe to treat patients with chronic migraine?

Evidence analysis

Evidence analysis

Two clinical trials have proven the efficacy of erenumab in patients with chronic migraine. The first one is a phase 2 study with 667 patients and an average migraine day of 18.2, comparing placebo (n=286) vs. erenumab 70 mg (n=191), or erenumab 140 mg (n=190) with an age range of 18-65 years and a 12-week follow-up. Compared to placebo, erenumab 70 and 140 mg reduced migraine days/month by 2.5 days (-2.5, 95%, CI - 3.5 to -1.4) p<0.0001 in both cases. The probability of achieving improvement of 50% or greater was 40%, 41% and 23% in the doses of 70 mg, 140 mg, and placebo, respectively, p<0.001 in both cases. The calculated NNT was 5.8 and 5.5 for the doses of 140 and 70 mg, respectively, vs. placebo. Both doses also demonstrated a significant decrease in the number of days using analgesics. In the analysis of cumulative hours of headache per month there was only statistical significance for the dose of 140 mg, p=0.0296. The probability of withdrawal due to collateral effects was less than 1% ²¹.

The second study is part of a subgroup analysis in which the same doses are compared against placebo, using as a primary outcome the proportion of patients with a 50% reduction in the frequency of migraine days/month after 12 months of treatment in patients with failure to more than two medications, more than three, excluding failure to more than four therapeutic options. In the failure to two or more treatments group, the primary outcome was achieved in 34.7% and 40.8%, at doses 70 and 140 mg, respectively, vs. 17.3% for placebo. In the failure to three or more treatments group, this percentage was reached in 35.6% in doses of 70 mg and 41.3% in doses of 140 mg, vs. 14.2% for the placebo group (p=0.001 in all cases, in both doses of each group when compared against placebo). The comparison of both doses against placebo in naive patients showed no statistically significant differences. The report of side effects was low in both analyzes, with a probability of withdrawal of 1.6% or less in both groups for each dose ²².

In an observational study with 65 patients in real clinical practice, a decrease of 12.2 and 15 migraine days/month was documented at week 4 and 8 when compared with the baseline. The response >50% was reached in 68.2%, 87.5% for the same periods. The study reported no significant side effects ²³.

Final recommendation: erenumab is recommended for the treatment of patients with chronic migraine. Quality of evidence: moderate; recommendation: strong.

Evidence analysis

A phase III randomized clinical trial with participants between 18 and 65 years of age, with an average of 19,4 migraine days, treated as follows: 240 mg loading dose followed by 120 mg monthly dose (n:278); 240 mg monthly dose without loading dose (n:277); monthly placebo (n: 558). Both dosages of Galcanezumab resulted in a higher rate of success in reducing the migraine days per month, as follows: placebo -2,7; 120 mg Galcanezumab -4,8; 240 mg Galcanezumab -4,6 (p <0,001 for each dose relative to placebo). The probability of improving at a 50% rate was calculated to be 15,4% for placebo; 27,6% for 120 mg; and 27,5% for 240 mg ²⁴. The number needed to treat (NNT) calculated for 120 mg and 240 mg vs. placebo stood at 8,2 and 8,3 respectively. Furthermore, this trial evidenced statistically significant differences between the 120 mg and 240 mg doses in the reduction of days with headache, hours without pain and migraine, p under 0,005 in all cases. Quality of Life (QOL) scales and the likelihood of reduced use of analgesics were significant only at the 240 mg dose. There were 6 placebo-related withdrawals and 5 Galcanezumab-related withdrawals. There were no significant differences in laboratory tests when comparing Galcanezumab vs. placebo ²⁴.

One study compared Galcanezumab vs. placebo in patients with failure to > 2 and reported a reduction of 5.35, 2.77 and 1.01 migraine days per month for the 120mg, 240mg and placebo doses, respectively. A reduction of 5.35, 3.53 and 2.02 at doses of 120 mg, 240 mg and placebo, respectively, was observed in the group of failure to > 1 preventive, with p less than 0.05 in all cases. In the group without previous failure, the difference was statistically significant only at a dose of 240 mg ²⁵. There was a decrease in the number of migraine days per month at month three in patients with therapeutic failure to onabotulinum toxin as follows: -3.91, -5.27, -0.88 for Galcanezumab 120mg, Galcanezumab 240mg, and placebo, respectively ²⁶. Open-phase follow-up in 9 patients with an average of 18.38 ± 3.74 migraine days showed reductions of 5.38 ± 4.92 at weeks 1-4 (p = 0.001), 4.75 ± 4.15 at weeks 5-8 (p = 0.001), and 3.93 ± 5.45 at weeks 9-12 (p = 0.014). There was no statistically significant difference when comparing the migraine days per month 12 weeks after completing the open phase against the last 4 weeks of this phase ²⁷.

Final recommendation: Galcanezumab is recommended as a treatment for patients with chronic migraine. Quality of evidence: moderate; recommendation: strong.

Evidence analysis

A phase 3 clinical trial with 1130 participants suffering from chronic migraine having an average of 13.1 headache days per month was conducted to compare the following: 225 mg subcutaneous injection applied monthly and a loading dose of 675 mg (379); 225 mg subcutaneous injection applied monthly and a loading dose of 675 mg; and placebo (375).

The average reduction of headache days per month when comparing the baseline vs. week 12 was 4.6± 0.3 when administering the drug on a monthly basis, 4.6± 0.3 when administering the drug in a quarterly basis, and 2.5 ± 0.3 when administering the placebo (p<0.001 in both groups vs. placebo). The difference in days between the monthly dose and the placebo was 1.8, while the difference in days between the quarterly dose and placebo was 2.1. A similar result was found when comparing fremanezumab vs. placebo in patients with therapeutic failure to other preventative treatments including onabotulinum toxin ²⁸. Improvement at a rate of 50% was reported to be 38% for the quarterly injection, 41% for the monthly injection, and 18% for the placebo (p<0.001 in both groups vs. placebo) ²⁹. NNT: 5 for the quarterly injection, and NNT: 4.3 for the monthly injection. These differences have been documented from week 1 following the start of treatment with doses of 225 mg (p:0.048) maintaining efficacy in the second and third week up to month 3 (p:0.04, p:0.025 and p<0.001, respectively) ³⁰. This effect is maintained when fremanezumab is part of the adjuvant therapy ³¹. The analysis of secondary outcomes revealed significant differences in the number of migraine days per month, the number of days using acute treatments, and the disease disability score as measured by the Headache Impact Test (HIT) 6 (p<0.001 in all cases). Serious adverse effects were reported in 2% of patients who received placebo and 1% of those who received fremanezumab. Withdrawal for side effects due to treatment was reported in 3% of patients who received fremanezumab vs. 2% of patients who received a placebo ²⁹.

Final recommendation: fremanezumab is recommended as a treatment for patients with chronic migraine. Quality of evidence: moderate; recommendation: strong

Clinical considerations

The consensus analysis recommends prescribing Erenumab, Galcanezumab or Fremanezumab to patients with chronic migraine following the failure of two first-line drugs at maximum tolerated doses for a minimum period of two months ³². The efficacy analyses in real life mirror those reported in pivotal trials (table 4). In these observations, super-responder rates (70-100%) have been reported in 14 to 18% of cases ³³. In case of previous treatment with ONABOT A, it is recommended to apply at least two cycles before considering these therapeutic alternatives.

Table 4 Comparative efficacy preventive treatments in chronic migraine. Source: authors. 

*monthly

**quarterly

Although the available results for Eptinezumab revealed statistically significant differences compared to the placebo, the source trials were designed for dose exploration, thus Eptinezumab has not been included as part of the recommendations derived from this consensus ³⁴.

In terms of safety, no significant cardiovascular effects have been reported in phase 2 and 3 clinical trials. These findings are consistent with safety data of Erenumab in individuals with stable angina ³⁵ and high blood pressure in combination with Sumatriptan ³⁶. However, the length of the observation periods in these trials is not enough to extrapolate conclusions to clinical practice, which is why this type of medication is not recommended for patients with uncontrolled cardiovascular risk factors. Additionally, these drugs are not to be prescribed for pregnant or lactating women or individuals with severe mental disorders ³². Caution should be exercised in patients with rigid rheumatic diseases and in cases of recent bone injury ³⁷. There is still no information on the effect of anti CGRP on the HPA (Hypothalamic pituitary axis) axis, and therefore these drugs are not recommended for individuals who have not reached hormonal maturity ³⁷.

Regarding tolerability, real-life patient follow-up has revealed an incidence of constipation, fatigue, and depression at 6-20%, 4-6% and 3-6%, respectively ³³. Erenumab, Galcanezumab and Fremanezumab may be prescribed along with preventive oral medications, although treatment with ONABOT A has been recommended before starting an anti CGRP ³². The complementary mechanism of action ³⁸ along with the increase in efficacy reported without impact on safety could allow this combination in selected cases ³³. Based on expert recommendations, this consensus suggests maintaining therapy for 6-12 months. Further recommendations will be considered as clinical developments and new evidence come to light ³².

3. Is topiramate effective and safe for the treatment of patients with chronic migraine?

Evidence analysis

Two multicenter, randomized, double-blind placebo-controlled trials have shown topiramate's efficacy in the treatment of patients with chronic migraine. The first of these, which took into account 306 individuals (topiramate, n=153; placebo, n=153), showed a reduction in migraine headache days (topiramate -6.4 vs. placebo -4.7, p=0.010) (39). The second study included 59 patients and reported a decrease of 3.5 migraine days/month from a baseline of 15.5 days/month, compared with an increase of 0.2 days/ month from 16.4 in the baseline (p=0.02), the percentage of patients with 50% improvement was 22% for patients with topiramate and 0% for patients with placebo (40). In both studies, paresthesias were reported as the most frequent side effects.

Clinical considerations

The recommended dose, according to clinical trials, should be 100 mg. This dose is based on the results obtained in the clinical trials in episodic migraine in which a balance of greater efficacy and safety was demonstrated, however, according to the clinical response it is possible to reach 200 mg ⁴¹. The results obtained in both clinical trials included patients that presented excessive use of analgesics, a variable in which no significant efficacy was demonstrated when compared with placebo, which may be due to the calculation of power in the sample studied. This factor should be considered relevant in clinical practice in those patients in whom it is not possible to significantly reduce the frequent consumption of analgesics. The efficacy of topiramate has been reported equivalent to that obtained with ONABOT A in the global determination of improvement, headache-free days and MIDAS scores at week 12 of treatment ⁴², Table 4. Other factors to take into account regarding safety are memory impairment, which occurs more frequently with doses greater than 100 mg, urolithiasis, glaucoma and weight loss ⁴¹.

Final recommendation: topiramate is recommended for the treatment of patients with chronic migraine. Quality of the evidence: low; recommendation: weak.

4. What other pharmacological options can be considered as part of preventive treatment in patients with chronic migraine?

Evidence analysis

Amitriptyline: in a dose of 25-50 mg, amitriptyline in an open study showed therapeutic equivalence to ONABOT A 250 UI, in the probability of improvement 50% at frequency, 67.8% vs. 72% (p=0.78; RR=0.94; CI=0.11-8). Nor were there significant differences in pain intensity and use of analgesics. Weight gain was described in 11.8% of patients treated with ONABOT A vs. 58.3% of those treated with amitriptyline (p=0.0001). Drowsiness was reported in 4% of patients with ONABOT A vs. 52.7% of those who received placebo (p=0.0001). Constipation occurred in 0% of the ONABOT A group vs. 38.8% of the amitriptyline group ⁴³.

Fhinarizine: an open study that compared the efficacy measured in days/month of headache in subjects with and without excessive use of analgesics, showed no differences between patients receiving 10 mg of flunarizine vs. TPM 50 mg during an observation period of eight weeks (-4.9 ± 3.8 vs. -2.3 ± 3.5; difference -2.6, 95%; CI -4.5 to -0.6; p=0.012), respectively. The safety and tolerability profile did not show statistically significant differences between the two groups (44). An open study in 40 patients also showed no statistically significant differences at the fourth month of observation but showed a reduction in days of severe pain by 75% for TPM vs. 70% for flunarizine (p=0.6236), in doses of TPM 100 mg vs. 5 mg ⁴⁵.

Divalproex sodium: the comparison of ONABOT A 100 Ul vs. divalproex sodium 500 mg/day in a double dummy design with 59 patients with evaluations at months 1, 3, 6, and 9. In the analysis of the chronic migraine subgroup, no statistically significant intergroup differences were observed in the reduction of the number of migraine days, 50% efficacy rate and disability indices ⁴⁶.

Naratriptan: in an observational study in patients with refractory chronic migraine with a dose of 2.5 mg twice a day, a reduction in headache frequency was reported after two-month treatment (15.3 days vs. 24.1 days on baseline, p<0.001), six months (9.1 days vs. 24.1, p<0.001), and after one year (7.3 days vs. 24.1, p<0.001) ⁴⁷.

In a similar study, statistically significant reduction to one month (20.4, p<0.001), two months (18.9, p<0.001), and three months (19.0, p<0.001), with better HIT 6 score in every aspect (p<0.05 in all cases) was demonstrated 27.1 days/month after baseline by ITT analysis ⁴⁸.

Clinical considerations

The data obtained for amitriptyline, flunarizine, and naratriptan are based on open studies, a consideration that limits the probability of generating evidence-based recommendations; however, obtaining therapeutic alternatives prior to the indication of high-cost medications establishes an option of interest in daily clinical practice. Despite the methodological limitations of the observational studies, amitriptyline, flunarizine, and divalproex sodium- valproic acid have similar degrees of efficacy to botulinum toxin A and topiramate; however, the presence of side effects that must be adjusted to the profile of each patient should be considered in order to maintain therapeutic adherence. The indication of naratriptan should only be considered in highly refractory patients. Blood pressure monitoring and informed consent are recommended in all cases and should not be indicated in patients with uncontrolled cardiovascular risk factors.

Final recommendation: amitriptyline, divalproex sodium, and flunarizine are recommended as part of the first-line therapeutic options in chronic migraine or as part of polytherapy. Naratriptan on a daily basis may be a treatment option in highly refractory patients. Consensus.

5. Is prednisolone effective in transitional therapy in chronic migraine and excessive use of analgesics?

Evidence analysis

Prednisolone. The comparison of prednisolone 100 mg vs. placebo in 20 patients showed significant reduction in headache hours of moderate or severe intensity in the first 72 hours after the withdrawal of analgesics (18.1 vs. 36.7 h, p=0.031, and 27.22 vs. 42.67 h, p= 0.05) ⁴⁹. A second randomized, double-blind study carried out in 100 patients used a 60 mg dose of prednisolone, with progressive reduction to day 5, compared with placebo, and calculated the average days of headache, including frequency and intensity. The primary outcomes for each group did not show statistically significant differences (1.48 [CI 1.281.68] vs. 1.61 [CI 1.41-1.82]) ⁵⁰. In a similar analysis, the comparison of 100 mg of prednisolone for five days showed statistically significant differences in the rate of analgesic consumption compared to placebo. This analysis found no differences in the reduction of hours with moderate and high intensity headache attacks ⁵¹. Comparison between prednisolone 75 mg with celecoxib 400 mg found reduction in pain intensity favoring celecoxib (p<0.001). This same study did not show differences in the reduction in headache frequency and consumption of analgesics (p=0.115, p=0.175, respectively) ⁵² (table 3).

Amitriptyline

In a placebo-controlled clinical trial, slow-release amitriptyline 75 mg (25 mg the first week, 50 mg the second week, and 75 mg starting from the third week) significantly reduced the average daily duration of headache in patients with chronic tension-type headache between weeks 1 and 6; the effect of amitriptyline began to be significant at week 3 of treatment ⁵⁵. In a controlled clinical trial with 60-90 mg amitriptylinoxide, 50-75 mg amitriptyline and placebo, no significant differences were found in the headache duration x frequency index or in the 50% reduction in headache intensity during 12 weeks of treatment ⁵⁶. Amitriptyline was studied with citalopram in a 32-week placebo-controlled crossover clinical trial for chronic tension-type headache prophylaxis without depression. This medication reduced the area under the headache curve by 30%, compared to placebo, with a significant reduction in the frequency and duration of headache ⁵⁷. In another controlled clinical trial of patients with chronic tension-type headache, 203 patients were assigned to amitriptyline 100 mg or nortriptyline 75 mg, placebo; stress management therapy and placebo; or stress management therapy and antidepressant. The three treatment groups improved, compared to placebo, but the improvement was faster with antidepressant medication than with stress management therapy with a response one month after treatment ⁵⁸. In a meta-analysis with 387 patients, the amitriptyline group had 6.2 headache days less compared to placebo at week 4, the result was maintained at weeks 8, 12 and 24. Additionally, amitriptyline reduces the amount of analgesics and the headache index, in addition to improving the quality of life, with an evidence quality evaluated as high ⁵⁹. The most frequent side effects of amitriptyline are dry mouth and drowsiness.

Venlafaxine

The efficacy and safety of venlafaxine was assessed in a controlled clinical trial in the treatment of chronic tension-type headache in patients without anxiety or depression ⁶⁰. Thirty-four patients were treated daily with venlafaxine XR 150 mg, while 26 patients were provided with placebo for 12 weeks. Venlafaxine was taken once a day after breakfast; the first week they took 75 mg and then the dose was raised to 150 mg. The venlafaxine group showed a significant reduction in the number of headache days (from 14.9 to 11.7 days) compared to the placebo group, which did not decrease headache days (from 13.3 to 14.2). The number of responders (a 50% or more reduction in the headache days) was significantly higher in the venlafaxine group (44%) than in the placebo group (15%); NNT was 3.48. In the venlafaxine group, six patients discontinued treatment due to adverse effects (vomiting, epigastralgia, nausea, loss of libido/anorgasmia) and none of the patients were reported to withdraw in the placebo group. The NNH for a side effect was 5.58.

Imipramine

In a controlled clinical trial, the efficacy of imipramine 25 mg was evaluated every 12 hours in patients with chronic tension-type headache, compared to transcutaneous electrical nerve stimulation. Imipramine significantly reduces headache intensity, with a score of 6.71 to 2.49 on the visual analog scale, with a significantly better reduction compared to transcutaneous electrical nerve stimulation ⁶¹.

The efficacy of mirtazapine in chronic tension-type headache was evaluated in a controlled crossover clinical trial ⁶². Mirtazapine 15 to 30 mg or placebo was administered, two to three hours before bedtime for eight weeks separated by a two-week wash period. The area under the headache curve (duration x intensity) was 34% lower during treatment with mirtazapine than with placebo in 22 patients who participated for the whole study. During the last four weeks of treatment with mirtazapine, 45% of patients improved the area under the curve by at least 30% compared to placebo. The number of headache days in the four weeks decreased from 28 during the placebo period to 25.5 during the period with mirtazapine. Two patients withdrew during active treatment due to the effects.

Indomethacin

Indomethacin has proven to be an effective medication in the acute and preventive treatment of hemicrania continua, even since the initial reports ^63,64. This efficacy has been demonstrated with doses between 25-150 mg/day with total symptom control after 72 hours of treatment initiation ^65,66.

Table 5 Pharmacological options in the hospital treatment of patients with refractory chronic migraine.  

* The recommended dose is lower than that used in clinical trials, in order to reduce the likelihood of side effects.

Source: authors.

In a cohort of 36 patients, 33 with criteria for HC who received placebo and indomethacin at different times achieved total pain control in 89% of the cases when they were exposed to the medication (average dose 176 mg/ day, with a minimum of 25 and maximum of 500 mg/ day). None of the patients in the placebo group achieved improvement ⁶⁷. Another series of 36 patients exposed to oral indomethacin test with a maximum dose of 250 mg/day reported pain control in all patients; however, poor tolerance to high doses of the molecule was described ⁶⁸. In a retrospective case series, three out of ten patients diagnosed with hemicrania continua reported efficacy in doses of 50-300 mg with limited gastrointestinal tolerability in all cases ⁶⁹.

In a series of 26 patients with an average dose of 75 mg, 16 with a diagnosis of hemicrania continua, 3.8 years were followed up, and improvement was reported within the first three days of treatment. During follow-up, 42% of patients maintained clinical control, which allowed the dose to be reduced up to 60% ⁶⁶.

Melatonin

In 2006, the case of a 42-year-old patient was reported, with HC and indomethacin intolerance (but with complete response to it), who achieved total pain control with the use of 7 mg of melatonin during the 5-month-follow-up time ⁷⁰. Subsequently, three women with indomethacin intolerance used melatonin in ascending dose every fifth day, starting with 3 mg to achieve total pain control (maximum allowed 24 mg/day), two of them at 9 mg and the third at 15 mg reported total pain control; in cases of pain attack, control was reported with 6 mg rescue dose ⁷¹.

In 2013, the case of a 60-year-old patient with a response to indomethacin was reported, but with limited tolerability, in which the administration of melatonin 9 mg/day produced complete pain control ⁷². The largest case series presented the retrospective analysis of 11 patients (nine women, two men). Of these, six showed no response to the use of melatonin in doses of 9-27 mg/ day, two achieved total pain control at low doses of 3-6 mg/day, while the remaining three obtained only partial control, which allowed the decrease of the daily dose of this medicine ⁷³.

Topiramate

In 2006, two cases were reported with hemicrania continua and a positive response to indomethacin, but with adhe- rence limitations due to non-tolerability. Both patients, who received topiramate between 100 and 200 mg/day, being intolerant as well though, achieved pain control similar to that obtained with indomethacin ⁷⁴. Afterwards, based on the previous reference, two similar cases were reported with indomethacin intolerance due to gastrointestinal effects. After the start of topiramate at 150-200 mg/day, sustained pain control was achieved after six and eight months of treatment suspension ⁷⁵. Recently, pain control was reported in two patients, combining indomethacin 75 mg/day and topiramate 50 to 75 mg/day, given the non-tolerance at higher doses in one or both molecules, which shows synergy between them and total pain control ⁷⁶.

Celecoxib

In a series of 14 patients, nine with rofecoxib and five with celecoxib, the latter group, in weekly ascending dose up to a maximum of 800 mg/day, produced total or partial response in 80% of cases ⁷⁷. In a report of four patients with hemicrania continua and indomethacin intolerance, total pain control was documented in doses ranging from 200-400 mg/day and with persistence of the clinical effect to the continuous use of the molecule in a range of 6 to 18 months later ⁷⁸.

Gabapentin

Gabapentin, in doses of 1800 and 2700 mg per day has reported efficacy in series of patients with new daily persistent headache ^79,80; the main adverse effects include sedation, ataxia and fatigue.

Doxycycline

In an open study that included four patients with new daily persistent headache and elevated tumor necrosis factor in the open CSF, doxycycline was used at a dose of 100 mg every 12 hours for three months; at two months of treatment, partial or total improvement was reported; the main adverse effects were nausea, vomiting and epigastralgia ⁸⁰.

Clinical considerations

It is recommended in patients without response to outpatient treatment despite optimal doses of acute and preventive medications, presence of comorbidities that limit adherence to therapeutic indications and excessive use of opioids. This procedure is justified by the use of parenteral medications and comprehensive assessment including psychiatry, neuropsychology, nutrition and other specialties considered according to the basic profile of each patient ¹¹. The recommended options in hospital treatment vary according to the comorbidities of each patient and the options available in each institution (table 4).

10. What procedures are indicated in the treatment of patients with chronic daily headache?

Clinical considerations

Nerve blocks seek to control the phenomenon of peri-cranial hypersensitivity common to several subtypes of headaches, including those of the primary type, and the entities that make up the CDH. Although most of the reports are described with bupivacaine, it is possible to use lidocaine in equivalent doses to 1-2 ml per point in the occipital region and 0.1-0.3 ml in the temporal region and facial points. Its use is safe in pregnant women and a short-term clinical effect is expected, which can be useful in the transition of the effect of the chosen preventive medications and in the detoxification due to excessive use of analgesics.