Proton pump inhibitor-related headaches: A nationwide population-based case-crossover study in Taiwan

Abstract

Background

Headaches resulting from proton pump inhibitor (PPI) use could cause discontinuation of PPI in as many as 40% of patients who experience such headaches. Previous studies focusing on acute headache risk from PPI use are rare and limited to clinical trials of a single PPI.

Objectives

To investigate the association between PPI use and headache with a nationwide population-based case-crossover study.

Methods

Records containing the first diagnosis of any headache, including migraine and tension-type headaches, were retrieved from Taiwan National Health Insurance Database (1998–2010). We compared the rates of PPI use for cases and controls during time windows of 7, 14, and 28 days. The adjusted self-matched odds ratios (ORs) and 95% confidence intervals (CIs) from a conditional logistic regression model were used to determine the association between PPI use and headache.

Results

Overall, 314,210 patients with an initial diagnosis of any headache during the study period were enrolled. The adjusted ORs for headache risk after PPI exposure were calculated for three time periods (within 7 days = 1.41, p = 0.002, 95% CI 1.14–1.74; within 14 days = 1.36, p < 0.001, 95% CI 1.16–1.59; within 28 days = 1.20, p = 0.002, 95% CI 1.07–1.35). Subgroup analyses showed female patients had an increased risk of headache. Among PPIs, lansoprazole and esomeprazole had the highest risks of headache incidence, which were similar to that of nitrates.

Conclusion

PPI usage is associated with an increased risk for acute headache. Female patients and use of lansoprazole or esomeprazole present the greatest risks of headache.

Keywords

Headache proton pump inhibitor incidence risk Taiwan

Introduction

Headache is a common adverse event associated with medications and may cause poor patient adherence to treatment regimens (1). Drug-induced headache complicates the diagnosis and treatment of patients who concurrently have other headache disorders.

Proton pump inhibitors (PPIs) are widely used to treat acid-related diseases, such as peptic ulcers and gastroesophageal reflux disease. Omeprazole was the first PPI and was introduced in 1989. There are currently five available PPIs (omeprazole, esomeprazole, lansoprazole, pantoprazole, and rabeprazole). Recently, some safety concerns about PPIs have emerged, such as osteoporosis, increased risk of certain infections, hypergastrinemia, hypomagnesemia, and reduced efficacy of clopidogrel (2 –5). However, short-term PPI treatment is still considered relatively safe and tolerable. In 2006, the total expenditure on brand-name PPIs was more than $10 billion in the USA (5).

Despite their widespread use, PPIs still have some adverse effects, such as nausea, diarrhea, and headache (6). A study showed as many as 40% of PPI-related headache sufferers discontinued use of PPIs (7). Headache incidence among PPIs users has been reported in randomized clinical trials (8 –12). In these trials, headaches were reported in 2.9–6.9% of omeprazole users, 3.8–8.8% of lansoprazole users, 1.3% of pantoprazole users, 2.4–6.0% of rabeprazole users, and 5.8–7.8% of esomeprazole users. In prescription-event monitoring (PEM) studies in England, headache was the fourth or fifth most frequently reported adverse event during PPI use. The incidence density of headaches per 1000 patient-months of exposure was 4.6 for lansoprazole, 3.7 for pantoprazole, 2.5 for omeprazole, and 2.1 for esomeprazole (13 –15).

Characteristics of PPI-related headache and predisposing factors for headache occurrence were reported in only one study (7), a nested case-control study that focused on lansoprazole users. The report showed that women, patients with previous use of analgesics, and patients reporting other adverse events were at risk of developing headaches. About two-thirds of PPI-related headaches were tension-type headaches, and approximately one-third were migraines. However, the response rate of this survey was low (44.6%).

In addition, little is known about the differences in headache risk among PPIs. Only one study indicated that lansoprazole and pantoprazole had greater headache risks than omeprazole (6). Therefore, we conducted a nationwide population-based case-crossover study to explore associations between the use of different PPIs and headaches.

Method

Data source

In 1995, Taiwan launched the National Health Insurance (NHI) program that covers approximately 99% of the Taiwanese population. In 1999, the Bureau of National Health Insurance made patient data available electronically through the National Health Insurance Research Database (NHIRD) project. Various extracted datasets are available to researchers, and hundreds of published studies have analyzed these data. The NHIRD contains comprehensive patient information, including demographic characteristic, diagnoses, medical expenditures, and prescription claims that include the date and duration of prescriptions, type of medications, dosages. Diseases are coded according to the International Classification of Diseases (ICD) -9-CM, 2001 edition. The diagnoses of several diseases, such as diabetes mellitus, acute coronary syndrome, and stroke, have been validated in addition to the accuracy of comorbidity by the Charlson score in the NHIRD (16 –20).

In this study, we used the Longitudinal Health Insurance Database (LHID) from 1995 to 2010 obtained from the NHIRD. The LHID contains 1,000,000 beneficiaries randomly sampled from the original NHIRD. The NHRI has reported that there were no statistically significant differences in age, sex, or healthcare costs between the sample group and all enrollees. The NHIRD consists of personally unidentifiable secondary data released for research purposes; therefore, this study did not require ethics approval.

Study subjects

In the current study, we defined patients with an incident headache event as those with an outpatient or emergency room visit for a primary diagnosis of headache that was noted under ICD-9-CM codes 346.x (migraine), 784.0 (headache), or 307.81 (tension headache) (21 –23). The index date for this dataset was defined as the date of the first visit when the patient was diagnosed with headache. Patients with an initial incident headache occurring between 1 January 1998 and 31 December 2010 were enrolled. Three groups of patients were excluded: those who had diagnoses of headache who were younger than 18 years of age; those who had diagnoses of acquired immune deficiency syndrome, human immunodeficiency virus, cancer, or cerebrovascular disease prior to the index date; and those who were hospitalized within 1 month before the diagnosis of headache because the dates of medication use during hospitalization could not be identified accurately.

Case-crossover design and exposure to proton pump inhibitors

The case-crossover design is a thoroughly validated research method to investigate the transient effects of acute events (24). The influence of unmeasured confounding factors, such as race, smoking status, and lifestyle, is reduced in the case-crossover design because each patient serves as his or her own control. The odds ratio (OR) for headache risk after exposure to a specific drug can be estimated by the ratio of patients who were exposed to that particular drug during the 7-day case period (1–7 days before the index date) to the patients who were exposed only during the 7-day control period (8–14 days before the index date). The ORs of other time windows, set at 14 days (1–14 days compared with 15–28 days before the index date for case and control periods, respectively) and 28 days (1–28 days and 29–56 days before the index date for case and control periods, respectively) were chosen for a sensitivity analysis and to determine the maximal risk window for headache after PPI exposure.

In this study, the exposure variable was PPI use. PPI exposure was defined as a prescription for PPIs for at least 1 day during the case or control time period. Concomitant medications were defined and accounted for in the same manner as PPIs. Two drugs were chosen as control drugs, nitrates as positive control and domperidone as negative control (1,25,26).

Data analyses

For case-crossover analyses, conditional logistic regression models were used to estimate the ORs and their 95% confidence intervals (95% CIs). Adjusted odds ratios (aORs) of headaches from PPI use were estimated after controlling for exposure to confounding medications, including angiotensin-converting-enzyme inhibitors, angiotensin II receptor blockers, beta-blockers, cilostazol, clopidogrel, dipyridamole, H2-receptor antagonists, sex hormones, nitrates, selective serotonin reuptake inhibitors, statins, ticlopidine, and warfarin.

Subgroup analyses were performed according to age and gender of the patients. Tests of interactions were performed for these subgroups by the likelihood ratio test. The p values of less than 0.05 were considered statistically significant. All analyses were performed using STATA statistical software (version 11.0; StataCorp., USA).

Results

Study population

During the 12-year study period, 422,978 patients with incident headache were identified. We excluded patients who were younger than 18 years of age (67,626 patients), had antecedent acquired immune deficiency syndrome or human immunodeficiency virus (194 patients), cancer (14,515 patients), cerebrovascular disease (19,680 patients), or were admitted to a hospital less than 1 month before the diagnosis of headache (6753 patients). Consequently, we enrolled 314,210 patients with a first diagnosis of any headache in the study period.

The study group included 279,120 patients with headaches (ICD-9 784.0), 24,713 patients with migraines (ICD-9 346.x), and 17,130 patients with tension headaches (ICD-9 307.81). The mean age at diagnosis of any headache was 41.9 years (SD 15.9 years). The study population was composed of 41.2% males, and 55.6% of subjects had Charlson comorbid scores of zero. Chronic pulmonary disease (20.3%), hypertension (16.1%), and chronic liver disease (14.2%) were the most common comorbid diseases. The demographic data and comorbidities are summarized in Table 1.

Table 1.

Demographic data.

	Patients with an initial diagnosis of headache (N = 314,210)
Age (mean ± SD)	41.9 ± 15.8
Gender
Male	129,495 (41.2)
Charlson comorbid score
Score 0	174,631 (55.6)
Score 1	82,406 (26.2)
Score 2	36,300 (11.6)
Score 3	15,071 (4.8)
Score ≥4	12,555 (4.0)
Coronary artery disease	25,753 (8.2)
Myocardial infarction	1567 (0.5)
Heart failure	5266 (1.7)
Chronic pulmonary disease	64,313 (20.5)
Asthma	25,360 (8.1)
Peripheral vascular disease	2993 (1.0)
Diabetes mellitus	28,031 (8.9)
Diabetes mellitus with organ failure	4684 (1.5)
Hypertension	51,728 (16.5)
Dyslipidemia	35,473 (11.3)
Arrhythmia	32,124 (10.2)
Chronic renal disease	12,276 (3.9)
Chronic liver disease	44,750 (14.2)
Connective tissue disease	5426 (1.7)
Paraplegia or hemiplegia	614 (0.2)

Headache risks

PPI use was associated with an increased risk of headache after adjusting for confounding elements (Table 2). An increased risk of headache was identified for all three time windows in PPI users. The highest risk occurred during the 7-day window (aOR 1.41; 95% CI 1.14–1.74; p = 0.002) but risk was still present in the 14-day window (aOR 1.36; 95% CI 1.16–1.59; p < 0.001) and the 28-day window (aOR 1.20; 95% CI 1.07–1.35; p = 0.002). Regarding control drugs, nitrates increased the risk of headache (aOR 1.73; 95% CI 1.49–2.01; p < 0.001), but domperidone had no effect (aOR 1.05; 95% CI 0.98–1.12; p = 0.152) (Table S1, supplementary material). Of the five types of PPIs, esomeprazole (aOR 1.78; 95% CI 1.10–3.12; p = 0.046) and lansoprazole (aOR 1.73; 95% CI 1.21–2.47; p = 0.002) were associated with an increased risk of headache. The other three PPIs did not significantly increase the incidence of headache (Table 3). Regarding specific headache diagnoses, PPIs were not found to increase the risk of migraine or tension headaches significantly (Table 4).

Table 2.

Risk of headache associated with current use of proton pump inhibitor.

	Case	Control	Crude			Adjusted^a
	N = 314,210	N = 314,210	OR	95% CI	p	OR	95% CI	p
7 day window
Proton pump inhibitor	1037	975	1.43	1.15–1.77	0.001	1.41	1.14–1.74	0.002
14 day window
Proton pump inhibitor	1182	1073	1.40	1.20–1.64	<0.001	1.36	1.16–1.59	<0.001
28 day window
Proton pump inhibitor	1452	1332	1.24	1.10–1.39	<0.001	1.20	1.07–1.35	0.002

Adjusted for use of angiotensin-converting-enzyme inhibitors, angiotensin II receptor blockers, beta-blockers, cilostazol, clopidogrel, dipyridamole, H2-receptor antagonists, sex hormones, nitrates, selective serotonin reuptake inhibitors, statins, ticlopidine, and warfarin.

Table 3.

Risk of headache associated with different proton pump inhibitor for 7-day window.

	Case	Control	Crude			Adjusted^a
	N = 314,210	N = 314,210	OR	95% CI	p	OR	95% CI	p
Esomeprazole	188	173	1.79	1.02–3.14	0.042	1.78	1.01–3.12	0.046
Lansoprazole	370	333	1.77	1.24–2.52	0.002	1.73	1.21–2.47	0.002
Omeprazole	273	270	1.05	0.74–1.50	0.785	1.04	0.73–1.49	0.824
Pantoprazole	156	146	1.43	0.84–2.44	0.184	1.46	0.86–2.49	0.162
Rabeprazole	72	71	1.10	0.47–2.59	0.827	1.06	0.45–2.51	0.897

Table 4.

Risk of different headache associated with current use of proton pump inhibitor for 7-day window.

		Using PPI		Crude			Adjusted^a
	No. of patients	Case	Control	OR	95% CI	P value	OR	95% CI	P value
Type of headache
Migraine (346.x)	24,447	109	110	1.00	0.48–2.10	1.000	1.01	0.48–2.13	0.973
Tension headache (307.81)	16,422	77	66	1.86	0.74–4.65	0.187	1.75	0.70–4.40	0.235
Headache (784.0)	273,341	1014	942	1.45	1.15–1.82	0.001	1.43	1.14–1.80	0.002

Subgroup analyses, stratifying by age and gender

In the subgroup analyses, the test of interaction was only significant between PPI use and gender (p = 0.028). Use of PPIs significantly increased the risk of headache in female patients (aOR 1.76; 95% CI 1.31–2.38; p < 0.001) (Table 5).

Table 5.

Subgroup analyses for risk of headache associated with current use of proton pump inhibitor for 7-day window.

		Using PPI		Crude			Adjusted^a
	No. of patients	Case	Control	OR	95% CI	p	OR	95% CI	p
Age^b
Age <45	191,631	422	381	1.68	1.22–2.32	0.001	1.68	1.22–2.31	0.002
Age ≥45	122,579	615	594	1.25	0.94–1.66	0.129	1.21	0.91–1.61	0.210
Gender^c
Male	126,490	532	523	1.12	0.82–1.52	0.481	1.08	0.80–1.48	0.595
Female	187,720	505	452	1.78	1.32–2.39	<0.001	1.76	1.31–2.38	<0.001

p interaction value = 0.137.

p interaction value = 0.028.

Discussion

The data in this study reveal several major findings. First, headache risk was temporarily increased after PPI use. The overall headache risk increased up to 1.41-fold within 7 days after PPI use. Second, the risk of headache differed between different PPIs, with esomeprazole and lansoprazole presenting the greatest risk that was similar to that of nitrates. Third, female patients were more susceptible to PPI-related headache. Fourth, the coding of PPI-induced headache was mainly nonspecific.

There are several strengths of this study. First, this is a population-based study that relied on a comprehensive medical utilization claims database that included all headache patients who visited physicians between 1998 and 2010 in Taiwan. The large sample size provided sufficient power for statistical analyses. Second, all five oral PPIs were included in the analysis and compared individually. We additionally controlled for the use of the most well-known headache-inducing drugs to prevent potential confounding effects. Third, the occurrence and reporting of headache are often influenced by other factors, such as personality traits, history of concurrent headache disorders, work stress, or values. These factors can be difficult to control for among groups of patients (27,28). The use of a case-crossover study design minimizes the confounding factors between subjects. Finally, we compared the risks of acute headache for PPIs to medications with known headache risks, such as nitrates (positive control) and domperidone (negative control). These controls demonstrate that the headache risk estimation for PPIs in our study was reliable.

Our study showed that PPIs were associated with an increase in headaches (ICD 784.0). Although not significant, the risk of tension headaches was also increased. In contrast, the risk of migraines was not increased. Our results somewhat agree with a previous study that reported that PPI-induced headaches were mainly tension-type headaches (75%) or unclassified headaches (7). However, the diagnoses of headaches in our study were based on ICD coding, and the proportion of non-specific headaches was high. These non-specific headache patients might fulfill the diagnosis of migraine or tension-type headache (29). The lack of statistical significance in tension headache or migraine might be a result of small sample size. Our data also showed that female patients are more susceptible to PPI-induced headache. This finding agrees with previous clinical trials and post-marketing surveys (7). The reasons for gender discrepancies remain unknown, but women are more susceptible to certain headache disorders, such as migraines, tension-type headaches, and nonspecific headaches (30,31). Moreover, the nocebo effect is more common in female patients and should be considered in medication-related side effects (32,33).

Compared with the other PPIs, lansoprazole and esomeprazole were more likely to induce headache. This result is similar to one cohort study that included only omeprazole, lansoprazole, and pantoprazole (6). The reason for the discrepancy in headache risk among different PPIs remains speculative, but certain biological and physiological characteristics are different between PPIs, such as bioavailability, metabolism, and drug interactions (34,35). For example, lansoprazole and esomeprazole were metabolized mainly by liver enzyme cytochrome P450(CYP) 3A4, whereas omeprazole and pantoprazole PPIs were metabolized mainly by CYP 2C19, and rabeprazole by both renal and liver (CYP 2C19 and 3A4) pathways (8,34 –36). Lansoprazole and esomeprazole also have the highest bioavailability. Another study showed lansoprazole had the highest incidence of adverse effects stemming from drug interactions among the five PPIs (37). Therefore, further studies are needed to determine the mechanisms of PPI-induced headaches.

Our study has clinical implications. First, as PPIs increased the risk of acute headache, especially in female patients, physicians should be cautious when prescribing these medications to susceptible patients. Second, headache patients are more likely to be comorbid with peptic ulcers or gastroesophageal reflux disease and need PPI treatment (38,39). PPI-induced headache should be considered when headaches worsen after treatment. Third, certain PPIs are more likely to induce headache than the others. Physicians may select different PPIs to prevent or improve headaches in susceptible patients.

Some methodological issues should also be addressed. First, the occurrence of headache in this study was based on physician visits in the administrative database. The headache incidence could be underestimated because patients with headaches might not visit physicians. Nevertheless, this bias might not be clinically significant because patients who did not seek medical treatment for headaches might have milder or tolerable headaches because the accessibility to the healthcare system is high in Taiwan. Second, the NHIRD dataset is an administrative database that lacks thorough clinical data, such as headache characteristics, severity, neuroimaging, or other laboratory results. Therefore, the precise diagnosis of headache disorders was not available. Besides, there was no validation study of these diagnostic codes for headache disorders in the NHIRD in Taiwan. However, the estimation of overall headache should be reliable because the diagnosis of headache was based on patient complaint rather than a diagnostic algorithm. In this study, most headache diagnoses were unclassified. Third, there may be deviations in the claims data, as coding by physicians may purposefully correspond to their prescriptions or other factors. However, the data remain representative. The Bureau of the NHI routinely and randomly samples a fixed percentage of claims from every contracted medical institution, and the accuracy of coding is audited by an independent group of physicians. Fourth, the case-crossover design does not account for within-person confounding such as the acute indication of PPI use or fluctuations in disease severity. As mentioned above, migraine and GERD are comorbid disorders to each other. Fifth, our study could not delineate whether the clinic visits for headache were because of headache exacerbation or new onset headache based on the claim data. Finally, although we controlled for the most well-known and frequently used medications that are known to induce headaches, there are still unidentified headache-inducing drugs that would not have been included in this study.

In conclusion, PPI use was associated with an increased risk of acute headache. The headache risks differed between populations based on gender and specific PPI. Clinicians should consider the association of PPIs and headache when prescribing PPIs.

Clinical implications

PPIs increased the risk of acute headache, especially in female patients.

Lansoprazole and esomeprazole had the highest risks of headache, similar to that of nitrates.

PPI-induced headache should be considered in patients with other headache and acid-related disorder concurrently.

Footnotes

Funding

This study was supported in part by grants from the National Science Council of Taiwan (100-2314-B-010-019-MY2, 100-2314-B-010-018-MY3, 102-2321-B-010 -030 -), Taipei-Veterans General Hospital (VGHUST102-G7-6-1, V102C-118, V102E9-001), NSC support for Center for Dynamical Biomarkers and Translational Medicine, National Central University, Taiwan (NSC 102-2911-I-008-001-), Brain Research Center, National Yang-Ming University and a grant from Ministry of Education, Aim for the Top University Plan. No additional external funding was received for this study.

Conflict of interest

None declared.

References

Young

. Drug-induced headache. Neurol Clin 2004; 22: 173–184.

Hess

Hoenderop

Bindels

. Systematic review: Hypomagnesaemia induced by proton pump inhibition. Aiment Pharmacol Ther 2012; 36: 405–413.

Chen

Yuan

Leontiadis

. Recent safety concerns with proton pump inhibitors. J Clin Gastroenterol 2012; 46: 93–114.

McCarthy

. Adverse effects of proton pump inhibitor drugs: Clues and conclusions. Curr Opin Gastroenterol 2010; 26: 624–631.

Moayyedi

. Economic reflections on proton pump inhibitor therapy for non-erosive reflux disease. Digestion 2008; 78(Suppl 1): 61–69.

Martin

Dunn

Freemantle

. The rates of common adverse events reported during treatment with proton pump inhibitors used in general practice in England: Cohort studies. Br J Clin Pharmacol 2000; 50: 366–372.

Claessens

Heerdink

Van Eijk

. Determinants of headache in lansoprazole users in The Netherlands: Results from a nested case-control study. Drug Saf 2002; 25: 287–295.

Langtry

Wilde

. Lansoprazole. An update of its pharmacological properties and clinical efficacy in the management of acid-related disorders. Drugs 1997; 54: 473–505.

Wilde

McTavish

. Omeprazole. An update of its pharmacology and therapeutic use in acid-related disorders. Drugs 1994; 48: 91–132.

10.

Fitton

Wiseman

. Pantoprazole. A review of its pharmacological properties and therapeutic use in acid-related disorders. Drugs 1996; 51: 460–482.

11.

Langtry

Markham

. Rabeprazole: A review of its use in acid-related gastrointestinal disorders. Drugs 1999; 58: 725–742.

12.

Scott

Dunn

Mallarkey

. Esomeprazole: A review of its use in the management of acid-related disorders in the US. Drugs 2002; 62: 1091–1118.

13.

Freemantle

Pearce

Wilton

. The incidence of the most commonly reported events with 40 newly marketed drugs − A study by prescription-event monitoring. Pharmacoepidemiol Drug Saf 1997; 6(Suppl. 1): 1–52.

14.

Wilton

Key

Shakir

. The pharmacovigilance of pantoprazole: The results of postmarketing surveillance on 11 541 patients in England. Drug Saf 2003; 26: 121–132.

15.

Davies

Wilton

Shakir

. Safety profile of esomeprazole: Results of a prescription-event monitoring study of 11 595 patients in England. Drug Saf 2008; 31: 313–323.

16.

Lin

Lai

Syu

. Accuracy of diabetes diagnosis in health insurance claims data in Taiwan. J Formos Med Assoc 2005; 104: 157–163.

17.

Chan

. Histamine-2-receptor antagonists are an alternative to proton pump inhibitor in patients receiving clopidogrel. Gastroenterology 2010; 139: 1165–1171.

18.

Cheng CL, Kao YH, Yang CY, et al. Validation of claims record on acute myocardial infarction and ischemic stroke in the National Health Insurance Research Database in Taiwan. In: ISPE 3nd Asian Meeting, Seoul, Korea, 3–5 November 2008. Oral presentation. Bethesda, USA: ISPE.

19.

Cheng

Kao

Lin

. Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf 2011; 20: 236–242.

20.

Chang L. A study of validation on comorbidity derived from claims data, Master Thesis, National Yang-Ming University, Taiwan, 2004.

21.

Zipp

Weil

Einhäupl

. No increase in demyelinating diseases after hepatitis B vaccination. Nat Med 1999; 5: 964–965.

22.

Friedman

Feldon

Holloway

. Utilization, diagnosis, treatment and cost of migraine treatment in the emergency department. Headache 2009; 49: 1163–1173.

23.

Bailey

Wan

Mabry

. Does health information exchange reduce unnecessary neuroimaging and improve quality of headache care in the emergency department? J Gen Intern Med 2013; 28: 176–183.

24.

Maclure

. The case-crossover design: A method for studying transient effects on the risk of acute events. Am J Epidemiol 1991; 133: 144–153.

25.

Iversen

Olesen

. Headache induced by a nitric oxide donor (nitroglycerin) responds to sumatriptan. A human model for development of migraine drugs. Cephalalgia 1996; 16: 412–418.

26.

Akerman

Goadsby

. Dopamine and migraine: Biology and clinical implications. Cephalalgia 2007; 27: 1308–1314.

27.

Sjösten

Nabi

Westerlund

. Influence of retirement and work stress on headache prevalence: A longitudinal modelling study from the GAZEL Cohort Study. Cephalalgia 2011; 31: 696–705.

28.

Passchier

Schouten

Van der Donk

. Headache. The association of frequent headaches with personality and life events. Headache 1991; 31: 116–121.

29.

Blumenthal

Weisz

Kelly

. Treatment of primary headache in the emergency department. Headache 2003; 43: 1026–1031.

30.

Jensen

Stovner

. Epidemiology and comorbidity of headache. Lancet Neurol 2008; 7: 354–361.

31.

Rasmussen

. Epidemiology of headache. Cephalalgia 2001; 21: 774–777.

32.

Klosterhalfen

Kellermann

Braun

. Gender and the nocebo response following conditioning and expectancy. J Psychosom Res 2009; 66: 323–328.

33.

Papadopoulos

Mitsikostas

. A meta-analytic approach to estimating nocebo effects in neuropathic pain trials. J Neurol 2012; 259: 436–447.

34.

Baldwin

Keam

. Rabeprazole: A review of its use in the management of gastric acid-related diseases in adults. Drugs 2009; 10: 1373–1401.

35.

Shi

Klotz

. Proton pump inhibitors: An update of their clinical use and pharmacokinetics. Eur J Clin Pharmacol 2008; 64: 935–951.

36.

Hagymási

Müllner

Herszényi

. Update on the pharmacogenomics of proton pump inhibitors. Pharmacogenomics 2011; 12: 873–888.

37.

Salgueiro

Rubio

Hidalgo

. Safety profile of proton pump inhibitors according to the spontaneous reports of suspected adverse reactions. Int J Clin Pharmacol Ther 2006; 44: 548–556.

38.

Chen

Tang

. Comorbidity profiles of chronic migraine sufferers in a national database in Taiwan. J Headache Pain 2012; 13: 311–319.

39.

Katić

Golden

Cady

. GERD prevalence in migraine patients and the implication for acute migraine treatment. J Headache Pain 2009; 10: 35–43.