Guideline Panel Members

Transcription

1 i Guideline Panel Members Saudi Expert Panel Dr. Adel Alhazzani Dr. Nasser Alotaibi Dr. Suleiman Mohammed Kojan Dr. Manal Abdulaziz Murad Dr. Mona Obaid McMaster University Working Group Dr. John J Riva, Dr. Ainsley E Moore, Dr. Jan L Brożek, and Dr. Holger J Schünemann, on behalf of the McMaster Guideline Working Group. Acknowledgements We acknowledge Dr. Ali M. Al Khathaami, Dr. Abdullah Al-Humaidan, Dr. Abdul-Hameed Ali Hassan, Dr. Fahmi M. Al-Senani for their contribution to this work. We gratefully acknowledge Dr Rajaa Alraddadi, from the Ministry of Health for peer reviewing this final report. Disclosure of potential conflict of interest: Dr. Suleiman Mohammed Kojan declares that he participated in a headache course sponsored by Janssen-Cilag and received honoraria for his contribution. Other co-authors declare that they have no conflict of interest related to this guideline. Funding: This clinical practice guideline was funded by the Ministry of Health, Saudi Arabia. Address for correspondence: Saudi Center for Evidence Based Health Care ebhc@moh.gov.sa Web:

2 ii Contents The Saudi Center for Evidence Based Health Care (EBHC)... iv Executive Summary... 1 Introduction... 1 Methodology... 1 How to use these guidelines... 2 Key Questions... 3 Recommendations... 4 Scope and purpose... 6 Introduction... 6 Methodology... 7 How to use these guidelines... 8 Key questions... 8 Recommendations... 9 Question 1: Should head MRI or CT (with or without contrast) rather than no imaging be done in patients with suspected migraine headache?... 9 Question 2: Should metoclopramide versus NSAIDs be used for acute migraine? Question 3: Should triptans versus metoclopramide be used for acute migraine? Question 4: Should triptans versus paracetamol be used for acute migraine? Question 5: Should triptans, in combination with NSAIDs, versus NSAIDs alone be used for acute migraine? Question 6: Should triptans, in combination with NSAIDs, versus triptans alone be used for acute migraine? Question 7: Should beta-blockers versus no beta-blockers be used for prevention of recurrence of migraine? Question 8: Should topiramate versus no antiepileptics be used for prevention of recurrence of migraine? Question 9: Should valproate versus no antiepileptics be used for prevention of recurrence of migraine? Question 10: Should topiramate versus valproate be used for prevention of recurrence of migraine? Question 11: Should antiepileptics other than topiramate or valproate versus no antiepileptics be used for prevention of recurrence of migraine? Question 12: Should topiramate versus beta-blockers be used for prevention of recurrence of migraine? Question 13: Should triptans versus no triptans be used for prevention of recurrence of menstrual-related migraine? Question 14: Should botulinum toxin A injections versus no injections be used for prevention of recurrence of episodic migraine? Question 15: Should botulinum toxin A injections versus no injections be used for prevention of recurrence of chronic migraine? Question 16: Should tricyclic antidepressants versus no antidepressants be used for prevention of recurrence of migraine?... 23

3 iii Question 17: Should SSRIs versus no antidepressants be used for prevention of recurrence of migraine? Question 18: Should education and self-management programs versus no such programs (usual care only) be used for prevention of recurrence of migraine? References Appendices Appendix 1: Evidence to Decision Frameworks : Should head MRI or CT (with or without contrast) rather than no imaging be done in patients with suspected migraine headache and no other indications? : Should metoclopramide versus NSAIDs be used for acute migraine? : Should triptans versus metoclopramide be used for acute migraine? : Should triptans versus paracetamol be used for acute migraine? : Should triptans, in combination with NSAIDs, versus NSAIDs alone be used for acute migraine? : Should triptans, in combination with NSAIDs, versus triptans alone be used for acute migraine? : Should beta-blockers versus no beta-blockers be used for prevention of recurrence of migraine? : Should topiramate versus no antiepileptics be used for prevention of recurrence of migraine? : Should valproate versus no antiepileptics be used for prevention of recurrence of migraine? : Should topiramate versus valproate be used for prevention of recurrence of migraine? : Should antiepileptics other than topiramate or valproate versus no antiepileptics be used for prevention of recurrence of migraine? : Should topiramate versus beta-blockers be used for prevention of recurrence of migraine? : Should triptans versus no triptans be used for prevention of recurrence of menstrualrelated migraine? : Should botulinum toxin A injections versus no injections be used for prevention of recurrence of episodic migraine? : Should botulinum toxin A injections versus no injections be used for prevention of recurrence of chronic migraine? : Should tricyclic antidepressants versus no antidepressants be used for prevention of recurrence of migraine? : Should SSRIs versus no antidepressants be used for prevention of recurrence of migraine? : Should education and self-management programs versus no such programs (usual care only) be used for prevention of recurrence of migraine? Appendix 2: Search Strategies and Results

4 iv The Saudi Center for Evidence Based Health Care (EBHC) The Saudi Centre for Evidence Based Health Care has managed and supported the coordination of the process of clinical practice guideline (CPG) development between the methodological team from McMaster University and the local clinical expert panel members in Saudi Arabia. The EBHC staff members recruited local clinical experts through contacting Saudi specialist societies and also independent experts interested in developing reliable and most up-to-date CPGs to harmonize the treatment and provide the highest quality of health care in the Kingdom of Saudi Arabia. These experts were health care professionals of multidisciplinary backgrounds. As much as possible, patient s representatives were also included in panels. In an effort to make national recommendations, the participating experts were professionals from the Ministry of Health (MoH), National Guard Hospitals, King Faisal Specialist Hospital and Research Centre (KFSHRC), University Hospitals, Security Forces Hospitals, Prince Sultan Military Medical City (PSMMC) and from some private hospitals. Based on a pre-selection of available evidence syntheses, the EBHC provided a list of potential topics to be addressed in CPGs after thorough consultations with the local stakeholders. These topics were further discussed with the McMaster team for important selection criteria and agreed on 12 topics for wave 2. The guideline panel meetings were held in Riyadh on 15 th -18 th March 2015 where about (96 local experts working in Saudi Arabia participated with the methodological support from 20 experts from McMaster University and its partners from the American University of Beirut, Lebanon, and the University of Freiburg, Germany, in providing high quality recommendations for common and important clinical conditions in the Kingdom. The Saudi Centre for EBHC supports the efforts for dissemination of the CPGs by publishing online the full reports of the CPGs, facilitates writing concise versions of the CPGs for publication in peer reviewed medical journals, sending hard copies to hospitals and health care centers. Finally, a mobile App has been introduced in KSA to facilitate the dissemination efforts of the completed practice guidelines. The staff members at the Saudi Centre for EBHC: Dr Zulfa Al Rayess, Consultant Family Medicine, Head of Saudi Center for EBHC Dr Yaser Adi, Scientific Advisor for the Saudi Centre for EBHC Miss Nourah Al Moufarreh, Project Manager, Saudi Center for EBHC

5 1 Executive Summary Introduction Migraine headache is very common in society with a prevalence and global burden that ranks third world-wide of all neurological disorders. 7 It also has been argued that current research underestimates the burden of disability associated with migraine. 8 In acute situations, most migrainers report trying more than three acute treatments with over-thecounter acute treatments more commonly reported than use of prescription treatments. 9 Those with more chronic migraine (15 or more migraines per month) tend to have even more debility, lost work time or decreased productivity. 10,11 This chronicity leads to higher headache-related disability, headache impact, typically worse socioeconomic status, worse health-related quality of life, and higher other medical and psychiatric conditions. 12,13 In addition, there is also increased healthcare resource utilization and higher direct and indirect costs. 14 In the context of Saudi Arabia, a 2010 systemic review found a high variability in prevalence across studies done in Arab countries and authors described a deficit in accurate epidemiological data in these countries. 15 This variability was described to pose a challenge in estimating the burden of headache, but also identifies a need and opportunity for research in this area. Prevalence of migraine headache in Saudi Arabia was originally reported from a survey of 22,630 individuals in 1997 at 5% (95% CI: ) with authors suggesting that the low figure relative to other populationbased surveys possibly being affected by a skew towards younger age of the sample, traditional lifestyles and cultural differences in the Kingdom. 16 More recently, a country wide cross-sectional survey with 2421 respondents found migraine headache to have 1-year prevalence of 32% with an odds ratio (OR) = 1.9 for female gender. Authors described poor sleep habits and hot weather as contributing factors. 17 Specific to the population of Saudi Arabia, situational contexts contribute to burden. For example, fasting for approximately one month during the month of Ramadan (i.e. first of Ramadan ) can contribute to headache exacerbation along with the effects of dehydration and caffeine withdrawal. 18 As well, up to 4.7% of traffic accidents were associated with migraine headaches in a survey of 1985 drivers in the United Arab Emirates. 19 While health utility data was not identified relative to Saudi Arabia, all levels of migraine pain severity in other countries have been found to have significantly reduced utility values. As example, in United Kingdom populations, severe migraine pain was considered a health state worse than death. 20 As well, the disutility in United States of America populations for mild migraine pain was estimated to be 0.14 (95% confidence interval (CI): ), with a disutility for moderate migraine pain of 0.19 (95% CI: ) and for severe migraine pain of 0.49 (95% CI: ). 21 Often the clinical approach to the diagnosis of migraine headache includes a decision around head imaging. Determining if the headache is a primary or secondary disorder is considered necessary as secondary headaches often result from underlying pathology such as thrombosis, tumour or abscess. Once a diagnosis of migraine headache is made, management strategies are introduced as either acute or prophylactic approaches based on episodic or chronic frequencies of migraine attacks. Given the importance of this topic, the Ministry of Health of the Kingdom of Saudi Arabia with the support of the McMaster University Working Group produced practice guidelines to assist health care providers in evidencebased decision-making on the diagnosis and management of migraine headache. Methodology This practice guideline is a part of the larger initiative of the Ministry of Health of the King-

6 2 dom of Saudi Arabia (KSA) to establish a program of rigorous development of guidelines. The ultimate goals are to provide guidance for clinicians and other healthcare decision makers and reduce unnecessary variability in clinical practice across the Kingdom. The guideline panel selected the topic of this guideline and all healthcare questions addressed herein using a formal prioritization process. For all selected questions we updated existing systematic reviews on diagnosis and management of migraine headache that were previously published by the National Institute for Health and Care Excellence (NICE) to support the 2012 guideline, Headaches: Diagnosis and management of headaches in young people and adults. 6 We also conducted additional systematic searches for information that was required to develop full guidelines for the KSA, including searches for information about patients values and preferences, and costs and resource use specific to the Saudi context and for questions not specifically addressed in the previously published systematic review. Based on the systematic reviews we prepared summaries of available evidence supporting each recommendation following the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach. 1 We used this information to prepare the GRADE evidence-todecision frameworks that served the guideline panel to follow the structured consensus process and transparently document all decisions made during the meeting (see Appendix 1). The guideline panel met in Riyadh on March 15 & 16, 2015 and formulated all recommendations during this meeting. Potential conflicts of interests of all panel members were managed according to the World Health Organization (WHO) rules. 2 As a quality measure for any practice guideline prior to publication, the final report have been externally peer reviewed by a methodological expert who has not been involved in this guideline development. How to use these guidelines The guideline-working group developed and graded the recommendations and assessed the quality of the supporting evidence according to the GRADE approach. 3 Quality of evidence (confidence in the estimates of effects) is categorized as: high, moderate, low, or very low based on consideration of risk of bias, indirectness, inconsistency, imprecision and publication bias of the estimates as well as factors that lead to upgrading the quality of the evidence. High quality evidence indicates that we are very confident that the true effect lies close to that of the estimate of the effect. Moderate quality evidence indicates moderate confidence, and that the true effect is likely close to the estimate of the effect, but there is a possibility that it is substantially different. Low quality evidence indicates that our confidence in the effect estimate is limited, and that the true effect may be substantially different. Finally, very low quality evidence indicates that the estimate of effect of interventions is very uncertain, the true effect is likely to be substantially different from the effect estimate and further research is likely to have important potential for reducing the uncertainty. The strength of recommendations is expressed as either strong ( guideline panel recommends ) or conditional ( guideline panel suggests ) and has explicit implications (see Table 1). 4 Understanding the interpretation of these two grades is essential for sagacious clinical decision-making.

7 3 Table 1: Interpretation of strong and conditional (weak) recommendations Implications Strong recommendation Conditional (weak) recommendation For patients Most individuals in this situation would want the recommended course of action and only a small proportion would not. Formal decision aids are not likely to be needed to help individuals make decisions consistent with their values and preferences. The majority of individuals in this situation would want the suggested course of action, but many would not. For clinicians For policy makers Most individuals should receive the intervention. Adherence to this recommendation according to the guideline could be used as a quality criterion or performance indicator. The recommendation can be adapted as policy in most situations Recognize that different choices will be appropriate for individual patients and that you must help each patient arrive at a management decision consistent with his or her values and preferences. Decision aids may be useful helping individuals making decisions consistent with their values and preferences. Policy-making will require substantial debate and involvement of various stakeholders. Key Questions The following is a list of the priority clinical questions selected by the panel and addressed in this guideline and their respective recommendations. Diagnosis 1. Should head MRI or CT (with or without contrast) rather than no imaging be done in patients with suspected migraine headache and no other indications? Acute Pharmacological Management 2. Should metoclopramide versus NSAIDs be used for acute migraine? 3. Should triptans versus metoclopramide be used for acute migraine? 4. Should triptans versus paracetamol be used for acute migraine? 5. Should triptans, in combination with NSAIDs, versus NSAIDs alone be used for acute migraine? 6. Should triptans, in combination with NSAIDs, versus triptans alone be used for acute migraine? Prophylactic Pharmacological Management 7. Should beta-blockers versus no betablockers be used for prevention of recurrence of migraine? 8. Should topiramate versus no antiepileptics be used for prevention of recurrence of migraine? 9. Should valproate versus no antiepileptics be used for prevention of recurrence of migraine? 10. Should topiramate versus valproate be used for prevention of recurrence of migraine? 11. Should antiepileptics other than topiramate or valproate versus no antiepileptics be used for prevention of recurrence of migraine? 12. Should topiramate versus betablockers be used for prevention of recurrence of migraine? 13. Should triptans versus no triptans be used for prevention of recurrence of menstrual-related migraine?

8 4 14. Should botulinum toxin A injections versus no injections be used for prevention of recurrence of episodic migraine? 15. Should botulinum toxin A injections versus no injections be used for prevention of recurrence of chronic migraine? 16. Should tricyclic antidepressants versus no antidepressants be used for prevention of recurrence of migraine? 17. Should SSRIs versus no antidepressants be used for prevention of recurrence of migraine? Prophylactic Non-Pharmacological Management 18. Should education and selfmanagement programs versus no such programs (usual care only) be used for prevention of recurrence of migraine? Recommendations Diagnosis Recommendation 1: The panel recommends that clinicians do not use head MRI or CT imaging in patients with migraine or suspected of migraine that do not have other indications for imaging. (strong recommendation, very low quality evidence) Acute Pharmacological Management Recommendation 2: The panel suggests either metoclopramide or a NSAID in patients with acute migraine. (conditional recommendation, very low quality evidence) Recommendation 3: The panel suggests metoclopramide rather than a triptan in patients with acute migraine. (conditional recommendation, low quality evidence) Recommendation 4: The panel suggests a triptan rather than parcetamol in patients with acute migraine. (conditional recommendation, very low quality evidence) Recommendation 5: The panel suggests a combination of a triptan with a NSAID rather than a NSAID alone in patients with acute migraine. (conditional recommendation, low quality evidence) Recommendation 6: The panel suggests a combination of a triptan with a NSAID rather than a triptan alone in patients with acute migraine. (conditional recommendation, very low quality evidence) Prophylactic Pharmacological Management Recommendation 7: The panel suggests using beta-blockers for the prevention of migraine attacks. (conditional recommendation, low quality evidence) Recommendation 8: The panel suggests topiramate 50 to 100 mg daily for the prevention of migraine attacks. (conditional recommendation, moderate quality evidence) Recommendation 9: The panel suggests valproate 500 to 1000 mg daily for the prevention of migraine attacks. (conditional recommendation, low quality evidence) Remark: The panel determined that there is not enough evidence to favor one over the other.

9 5 Recommendation 10: The panel suggests that clinicians use either topiramate or valproate for the prevention of migraine attacks. (conditional recommendation, very low quality evidence) Remark: The panel found not enough evidence to favor one over the other. Recommendation 11: The panel suggests that clinicians do not use antiepileptics other than topiramate or valproate for the prevention of migraine attacks until more research about their efficacy and safety is available. (conditional recommendation, low quality evidence) Recommendation 17: The panel suggests that clinicians do not use SSRIs for the prevention of migraine attacks until more evidence is available. (conditional recommendation, low quality evidence) Prophylactic Non-Pharmacological Management Recommendation 18: The panel suggests that more research is done on effectiveness and cost-effectiveness of education and self-management programs. (conditional recommendation, very low quality evidence) Recommendation 12: The panel suggests that clinicians use either topiramate or beta-blockers for the prevention of migraine attacks. (conditional recommendation, low quality evidence) Recommendation 13: The panel suggests triptans for the prevention of menstrual-related migraine attacks. (conditional recommendation, low quality evidence) Recommendation 14: The panel recommends that clinicians do not use botulinum toxin A for the prevention of migraine attacks in patients with episodic migraine. (strong recommendation, moderate quality evidence) Recommendation 15: The panel suggests botulinum toxin A injections for the prevention of chronic migraine in patients who have not responded to other prophylactic treatments. (conditional recommendation, low quality evidence) Recommendation 16: The panel suggests tricyclic antidepressants for the prevention of migraine attacks. (conditional recommendation, low quality evidence)

10 6 Scope and purpose The purpose of this document is to provide guidance about the diagnosis and management of migraine headache. The target audience of these guidelines includes primary and specialist care clinicians in the Kingdom of Saudi Arabia (KSA). Other health care professionals and policy makers may also benefit from this guideline. Given the importance of this topic, the Ministry of Health (MoH) of Saudi Arabia with the support of the McMaster University Working Group produced practice guidelines to assist health care providers in evidence-based decision-making. This practice guideline is a part of the larger initiative of the Ministry of Health of Saudi Arabia to establish a program of rigorous adaptation and de novo development of guidelines in the Kingdom; the ultimate goal being to provide guidance for clinicians and other healthcare decision makers and reduce unnecessary variability in clinical practice across the Kingdom. This guideline does not encompass all aspects of diagnosis and management of migraine headache. Due to feasibility, priority questions were identified and addressed in this project. Introduction Migraine headache is very common in society with a prevalence and global burden that ranks third world-wide of all neurological disorders. 7 It also has been argued that current research underestimates the burden of disability associated with migraine. 8 In acute situations, most migrainers report trying more than three acute treatments with over-thecounter acute treatments more commonly reported than use of prescription treatments. 9 Those with more chronic migraine (15 or more migraines per month) tend to have even more debility, lost work time or decreased productivity. 10,11 This chronicity leads to higher headache-related disability, headache impact, typically worse socioeconomic status, worse health-related quality of life, and higher other medical and psychiatric conditions. 12,13 In addition, there is also increased healthcare resource utilization and higher direct and indirect costs. 14 In the context of Saudi Arabia, a 2010 systemic review found a high variability in prevalence across studies done in Arab countries and authors described a deficit in accurate epidemiological data in these countries. 15 This variability was described to pose a challenge in estimating the burden of headache, but also identifies a need and opportunity for research in this area. Prevalence of migraine headache in Saudi Arabia was originally reported from a survey of 22,630 individuals in 1997 at 5% (95% CI: ) with authors suggesting that the low figure relative to other populationbased surveys possibly being affected by a skew towards younger age of the sample, traditional lifestyles and cultural differences in the Kingdom. 16 More recently, a country wide cross-sectional survey with 2421 respondents found migraine headache to have 1-year prevalence of 32% with an odds ratio (OR) = 1.9 for female gender. Authors described poor sleep habits and hot weather as contributing factors. 17 Specific to the population of Saudi Arabia, situational contexts contribute to burden. For example, fasting for approximately one month during the month of Ramadan (ie. first of Ramadan ) can contribute to headache exacerbation along with the effects of dehydration and caffeine withdrawal. 18 As well, up to 4.7% of traffic accidents were associated with migraine headaches in a survey of 1985 drivers in the United Arab Emirates. 19 While health utility data was not identified relative to Saudi Arabia, all levels of migraine pain severity in other countries have been found to have significantly reduced utility values. As example, in United Kingdom populations, severe migraine pain was considered a health state worse than death. 20 As well, the disutility in United States of America populations for mild

11 7 migraine pain was estimated to be 0.14 (95% confidence interval (CI): ), with a disutility for moderate migraine pain of 0.19 (95% CI: ) and for severe migraine pain of 0.49 (95% CI: ). 21 Often the clinical approach to the diagnosis of migraine headache includes a decision around head imaging. Determining if the headache is a primary or secondary disorder is considered necessary as secondary headaches often result from underlying pathology such as thrombosis, tumour or abscess. Once a diagnosis of migraine headache is made, management strategies are introduced as either acute or prophylactic approaches based on episodic or chronic frequencies of migraine attacks. Methodology To facilitate the interpretation of these guidelines; we briefly describe the methodology we used to develop and grade recommendations and quality of the supporting evidence. The guideline panel selected the topic of this guideline and all health care questions addressed herein using this formal prioritization process. Based on the questions addressed in the 2012 NICE systematic review, a pool of 52 potential research questions was created specific to migraine headache diagnosis and management. Through an online surveybased rating and ranking exercise from July- August 2014, panel members selected their top priority questions from the question pool and also identified new questions to inform a priority list consisting of 15 research questions. In November 2014, panel members through an online survey-based exercise also ranked importance of outcomes as they related to each of the selected priority research questions. As well, panel members provided estimates on KSA specific costs related to specific interventions in their clinical environment. For the selected questions we updated existing systematic reviews on diagnosis and management of migraine headache that were previously published by NICE to support the 2012 guideline, Headaches: Diagnosis and management of headaches in young people and adults. 6. For each question, the McMaster Guideline Working Group updated the search strategy to identify new studies and/or new systematic reviews. When relevant, the metaanalyses were updated. We also conducted systematic searches for information that was required to develop full guidelines for the KSA, including searches for information about patients values and preferences, and costs and resource use specific to the Saudi context (see Appendix 2). Next, we developed for each question an evidence profile and an evidence-to-decision (EtD) table following the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach and shared them with the panel members (see Appendix 1). 1,5 The guideline panel was invited to provide additional information, particularly when published evidence was lacking. The final step consisted of an in-person meeting of the guideline panel in Riyadh on March 15 & 16, 2015 to formulate the final recommendations. During the in-person meeting 2 questions were collapsed into one more general question and 3 questions were divided into more specific questions, based on clinical relevance, for a total of 18 questions. We used the GRADE evidence-to-decision frameworks to follow a structured consensus process and transparently document all decisions made during the meeting. Potential conflicts of interests of all panel members were managed according to the World Health Organization (WHO) rules. 2 Grading of the quality of evidence The GRADE working group defines the quality of evidence as the extent of our confidence that the estimate of an effect is adequate to support a particular decision or recommendation. 3 We assessed the quality of evidence using the GRADE approach.

12 8 Quality of evidence is classified as high, moderate, low, or very low based on decisions about methodological characteristics of the available evidence for a specific health care problem. The definition of each category is as follows: High: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. Grading of the strength of recommendations The GRADE working group defines the strength of recommendation as the extent to which we can be confident that desirable effects of an intervention outweigh undesirable effects. According to the GRADE approach, the strength of a recommendation is either strong or conditional (also known as or called weak) and has explicit implications. 4 Understanding the interpretation of these two grades either strong or conditional of the strength of recommendations is essential for sagacious clinical decision-making. (see Table 1) As a quality measure for any practice guideline prior to publication, the final report have been externally peer reviewed by a methodological expert who has not been involved in this guideline development. How to use these guidelines The Ministry of Health of Saudi Arabia and McMaster University Practice Guidelines provide clinicians and their patients with a basis for rational decisions about the diagnosis and management of migraine headache. Clinicians, patients, third-party payers, institutional review committees, other stakeholders, or the courts should never view these recommendations as dictates. As described in other guidelines following the GRADE approach, no guideline or recommendation can take into account all of the often-compelling unique features of individual clinical circumstances. Therefore, no one charged with evaluating clinicians actions should attempt to apply the recommendations from these guidelines by rote or in a blanket fashion. Statements about the underlying values and preferences, resources, feasibility, equity, acceptability as well as other qualifying remarks accompanying each recommendation are its integral parts and serve to facilitate an accurate interpretation. They should never be omitted when quoting or translating recommendations from these guidelines if they influence the strength or direction of the recommendation. Key questions The following is a list of the priority clinical questions selected by the panel and addressed in this guideline. Diagnosis 1. Should head MRI or CT (with or without contrast) rather than no imaging be done in patients with suspected migraine headache and no other indications? Acute Pharmacological Management

13 9 2. Should metoclopramide versus NSAIDs be used for acute migraine? 3. Should triptans versus metoclopramide be used for acute migraine? 4. Should triptans versus paracetamol be used for acute migraine? 5. Should triptans, in combination with NSAIDs, versus NSAIDs alone be used for acute migraine? 6. Should triptans, in combination with NSAIDs, versus triptans alone be used for acute migraine? Prophylactic Pharmacological Management 7. Should beta-blockers versus no betablockers be used for prevention of recurrence of migraine? 8. Should topiramate versus no antiepileptics be used for prevention of recurrence of migraine? 9. Should valproate versus no antiepileptics be used for prevention of recurrence of migraine? 10. Should topiramate versus valproate be used for prevention of recurrence of migraine? 11. Should antiepileptics other than topiramate or valproate versus no antiepileptics be used for prevention of recurrence of migraine? 12. Should topiramate versus betablockers be used for prevention of recurrence of migraine? 13. Should triptans versus no triptans be used for prevention of recurrence of menstrual-related migraine? 14. Should botulinum toxin A injections versus no injections be used for prevention of recurrence of episodic migraine? 15. Should botulinum toxin A injections versus no injections be used for prevention of recurrence of chronic migraine? 16. Should tricyclic antidepressants versus no antidepressants be used for prevention of recurrence of migraine? 17. Should SSRIs versus no antidepressants be used for prevention of recurrence of migraine? Prophylactic Non-Pharmacological Management 18. Should education and selfmanagement programs versus no such programs (usual care only) be used for the prevention of recurrence of migraine? Recommendations Patient Values and Preferences: Across the 18 recommendations, all panel members agreed the problem of migraine was a priority. Health utility data was not identified specific to Saudi Arabia, yet all levels of migraine pain severity in other countries have been found to have significantly reduced utility values. As example, in United Kingdom populations, severe migraine pain was considered a health state worse than death. 20 As well, the disutility in United States of America populations for mild migraine pain was estimated to be 0.14 (95% confidence interval (CI): ), with a disutility for moderate migraine pain of 0.19 (95% CI: ) and for severe migraine pain of 0.49 (95% CI: ). 21 Cost Effectiveness: Across the 18 recommendations, no cost effectiveness studies were identified specific to Saudi Arabia. Studies from various other countries and studies presented in the 2012 NICE systematic review were identified, 6,57-60,,137,138 but the panel felt these were not relevant to the local context. For each recommendation, panel members provided their estimate of average unit costs for the specific intervention - in Saudi Arabian Riyal (SAR). I. DIAGNOSIS Question 1: Should head MRI or CT (with or without contrast) rather than no imaging be done in patients with suspected migraine headache? Summary of Findings: The MRI imaging section of the question was based on the 2012

14 10 NICE systematic review including 4 retrospective cohort studies The updated literature search identified no new relevant studies. The CT imaging section of the question was primarily based on the 2012 NICE systematic review including 3 retrospective cohort studies The updated literature search identified one new retrospective cohort study. 39 Benefits and harms of the Option: For MRI, the 4 retrospective cohort studies (total of 1139 participants) were very low quality evidence that could not estimate the absolute effect of head MRI with or without contrast on: identifying a tumour/neoplasm compared to no imaging (5 events per 1000); identifying an abscess compared to no imaging (0 events per 1000); identifying a subdural haematoma compared to no imaging (2 events per 1000); identifying hydrocephalus compared to no imaging (3 events per 1000); identifying an arteriovenous malformation compared to no imaging (2 events per 1000); identifying a stroke (i.e. venous thrombosis) compared to no imaging (0 events per 1000). For CT, the 4 retrospective cohort studies (total of 2287 participants) were very low quality evidence that could not estimate the absolute effect of head CT with or without contrast on: identifying a tumour/neoplasm compared to no imaging (7 events per 1000); identifying an abscess compared to no imaging (0 events per 1000); identifying a subdural haematoma compared to no imaging (0 events per 1000); identifying hydrocephalus compared to no imaging (1 event per 1000); identifying an arteriovenous malformation compared to no imaging (0 events per 1000); identifying a stroke (i.e. venous thrombosis) compared to no imaging (0 events per 1000); identifying venous thrombosis in pregnant women compared to no imaging (63 events per 1000). Resource Use: Panel members estimated a MRI unit cost range of SAR. Using MRI alone carried an imaging resource cost between 181,800 SAR and 409,050 SAR per serious abnormality detected. Panel members estimated CT unit cost range of SAR. Using CT or MRI combinations carried an imaging resource cost between 133,312 SAR and 499,950 SAR per serious abnormality detected. Other factors such as outpatient and hospital provider visits, x-rays and labs would also need to be considered. Other considerations: The panel judged MRI or CT imaging to likely be acceptable to patients but uncertain whether also acceptable for policymakers and payers. The panel judged MRI or CT imaging to probably not be a feasibility intervention to implement; and the impact on health inequities to probably increase. Balance between desirable and undesirable consequences: Prevalence of potentially treatable serious head lesions is likely very low. The proportion of false negative and false positive results is not known, but likely to result in substantial anxiety and unnecessary cost of further testing. The resources required would be large. Recommendation 1: The panel recommends that clinicians do not use head MRI or CT imaging in patients with migraine or suspected of migraine that do not have other indications for imaging. (strong recommendation, very low quality evidence) II. ACUTE PHARMACOLOGICAL MANAGMENT Question 2: Should metoclopramide versus NSAIDs be used for acute migraine? Summary of Findings: The question was based on the 2012 NICE systematic review including one randomized controlled trial in only pediatric patients. 45 The updated literature search identified one new trial in adults. 46 Benefits and harms of the Option: Findings from one trial (total of 208 participants) found very low quality evidence that did not rule out a reduction or an increase in headache relief

15 11 at 2 hours (change in severe or moderate headache to mild or none) for metoclopramide use compared to NSAID use in patients with acute migraine (risk ratio (RR) 1.17; 95% CI 0.93 to 1.47; absolute effect: 91 more events per 1000). Findings from one trial (total of 220 participants) found very low quality evidence that did not rule out a reduction or an increase in headache freedom at 2 hours for metoclopramide use compared to NSAID use in patients with acute migraine (RR 1.15; 95% CI 0.74 to 1.79; absolute effect: 37 more events per 1000). Findings from one trial (total of 218 participants) found very low quality evidence that did not rule out a reduction or an increase in Sustained freedom from headache at 24 hours for metoclopramide use compared to NSAID use in patients with acute migraine (RR 0.71; 95% CI 0.35 to 1.4; absolute effect: 45 fewer events per 1000). Findings from one trial (total of 205 participants) found very low quality evidence that did not rule out a reduction or an increase in Sustained relief from headache at 24 hours (change from severe to moderate to mild or none) for metoclopramide use compared to NSAID use in patients with acute migraine (RR 1.27; 95% CI 0.89 to 1.82; absolute effect: 88 more events per 1000). Findings from one trial (total of 219 participants) found low quality evidence that did not rule out a reduction or an increase in adverse events (dizzy, too drowsy to function, dyspepsia, heartburn, bloating) for metoclopramide use compared to NSAID use in patients with acute migraine (RR 0.71; 95% CI 0.31 to 1.63; absolute effect: 36 fewer events per 1000). Resource Use: Panel members estimated an average monthly cost of SAR for NSAIDs and metoclopramide. Other factors such as outpatient and hospital provider visits, hospital stays, x-rays and labs would also need to be considered. Other considerations: The panel judged the use of metoclopramide in patients with acute migraines to be both acceptable and feasible. The panel judged there to be no impact on health inequity. Balance between desirable and undesirable consequences: No evidence of difference in benefits, adverse effects and cost has been identified. Recommendation 2: The panel suggests either metoclopramide or a NSAID in patients with acute migraine. (conditional recommendation, very low quality evidence) Remark: The panel determined that there is not enough evidence to favor one over the other. Research Priorities: The panel suggested that it might be beneficial to conduct more headto-head RCTs that investigate commonly used NSAIDs and other medications used for acute migraine and measure patient-important outcomes. Question 3: Should triptans versus metoclopramide be used for acute migraine? Summary of Findings: The question was based on the 2012 NICE systematic review including one randomized controlled trial. 55 The updated literature search identified no new studies. Benefits and harms of the Option: Findings from one trial (total of 77 participants) found very low quality evidence that: did not rule out a reduction or an increase in freedom from pain at 2 hours (change in severe or moderate headache to mild or none) for triptan use compared to metoclopramide use in patients with acute migraine (RR 0.59; 95% CI 0.35 to 0.97; absolute effect: 246 fewer events per 1000); did not rule out a reduction or an increase in sustained freedom from pain at 24 hours for triptan use compared to metoclopramide use in patients with acute migraine (RR 0.68; 95% CI to 1.3; absolute effect: 128 fewer events per 1000); did not rule out a reduction or an increase in adverse events (weakness) for triptan use compared to metoclopramide use in patients with acute migraine (OR 2.25; 95% CI 0.67 to 7.48; abso-

16 12 lute effect: 118 more events per 1000); did not rule out a reduction or an increase in adverse events (feeling of heaviness / dizziness) for triptan use compared to metoclopramide use in patients with acute migraine (OR 6.09; 95% CI 0.68 to 54.85; absolute effect: 110 more events per 1000); did not rule out a reduction or an increase in adverse events (stiffness or abnormal movements) for triptan use compared to metoclopramide use in patients with acute migraine (OR 2.88; 95% CI 0.69 to 12.09; absolute effect: 114 more events per 1000). Resource Use: Panel members estimated an average monthly cost of SAR for metoclopramide and SAR for triptans. Other factors such as outpatient and hospital provider visits, hospital stays, x-rays and labs would also need to be considered. Other considerations: The panel judged the use of triptans to be both acceptable and feasible. They judged the impact on inequity to probably increase. Balance between desirable and undesirable consequences: The panel determined that undesirable effects were likely small and favored the control intervention. The quality of available evidence supporting this recommendation is low leaving much uncertainty about the balance of desirable and undesirable consequences. However, use of metoclopramide is less costly. In settings where resources are available, an alternative option (i.e. using triptans) may be equally reasonable. Recommendation 3: The panel suggests metoclopramide rather than a triptan in patients with acute migraine. (conditional recommendation, low quality evidence) Subgroup Considerations: Available evidence is for intravenous administration in emergency departments. It is less clear whether it is applicable to less severe cases and oral administration of these medications. Note: Evidence is available for sumatriptan only. Research Priorities: The panel thought that more research on comparative safety and effectiveness for all routes of administration and types of triptans would be desirable. Question 4: Should triptans versus paracetamol be used for acute migraine? Summary of Findings: The question was based on the 2012 NICE systematic review including one randomized controlled trial. 64 The updated literature search identified no new studies. Benefits and harms of the Option: Findings from one trial (total of 86 participants) found very low quality evidence that: did not rule out a reduction or an increase in headache response at up to 2 hours for triptan use compared to paracetamol use in patients with acute migraine (RR 1.1; 95% CI 0.85 to 1.42; absolute effect: 70 more events per 1000); did not rule out a reduction or an increase in freedom from pain at up to 2 hours for triptan use compared to paracetamol use in patients with acute migraine (RR 1.54; 95% CI 0.82 to 2.9; absolute effect: 138 more events per 1000); did not rule out a reduction or an increase in sustained headache response at up to 24 hours for triptan use compared to paracetamol use in patients with acute migraine (RR 1.19; 95% CI 0.82 to 2.9; absolute effect: 138 more events per 1000); and did not rule out a reduction or an increase in sustained freedom from pain at up to 24 hours for triptan use compared to paracetamol use in patients with acute migraine (RR 1.43; 95% CI 0.6 to 3.4; absolute effect: 70 more events per 1000). Resource Use: The panel estimated that triptans would carry an average monthly cost estimated at SAR and paracetamol would carry an average monthly cost estimated at SAR. Both outpatient and hospital

17 13 provider visits would also need to be considered. Other considerations: Based on their unsystematic clinical observations, panel members considered paracetamol as not very effective in acute migraine. However, panel members acknowledged that most of them have experience from tertiary care centers with patients who likely already tried and failed other treatments (including paracetamol). The panel also felt that those with access to specialty clinics are more likely to receive a triptan, while in family care settings it is less likely. The panel judged the use of triptans to be both acceptable and feasible. They judged the impact on inequity to probably increase, as again those with access to specialty clinics are more likely to receive a triptan while in family care setting it is less likely. Balance between desirable and undesirable consequences: Available evidence, though very uncertain, and unsystematic observations in panel members practice suggest larger benefit from triptans compared to paracetamol. One panel member felt that the balance between desirable and undesirable consequences is uncertain because available evidence is of too low quality to be able to make the judgment. Recommendation 4: The panel suggests a triptan rather than parcetamol in patients with acute migraine. (conditional recommendation, very low quality evidence) Subgroup Considerations: Patients with hemiplegic migraine should not receive triptans. Note: Evidence is available only for risatriptan. Question 5: Should triptans, in combination with NSAIDs, versus NSAIDs alone be used for acute migraine? Summary of Findings: The question was based on the 2012 NICE systematic review including 3 randomized controlled trials The updated literature search identified no new studies. Benefits and Harms of the Option: The metaanalysis of 3 trials (total of 1944 participants) found low quality evidence that showed: an increase in freedom from pain at 2 hours for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 2.03; 95% CI 1.71 to 2.4; absolute effect: 165 more events per 1000); an increase in headache relief at up to 2 hours for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 1.41; 95% CI 1.3 to 1.54; absolute effect: 180 more events per 1000); an increase in sustained freedom from pain at 24 hours for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 2.25; 95% CI 1.82 to 2.78; absolute effect: 134 more events per 1000); an increase in sustained headache relief at 24 hours for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 1.64; 95% CI 1.45 to 1.85; absolute effect: 179 more events per 1000); a reduction in rescue medication use for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 0.61; 95% CI 0.54 to 0.7; absolute effect: 164 fewer events per 1000); and an increase in adverse events for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 1.77; 95% CI 1.47 to 2.13; absolute effect: 112 more events per 1000). The meta-analysis of 2 trials (total of 1352 participants) found low quality evidence that showed an increase in relief of functional disability at 2 hours for the combination of triptan and NSAID use compared to NSAID use alone in patients with acute migraine (RR 1.43; 95% CI 1.35 to 1.97; absolute effect: 84 more events per 1000). Resource Use: Panel members estimated an average monthly cost of SAR for NSAIDs and SAR for triptans. Other factors such as outpatient and hospital pro-