You are on page 1of 9

European Journal of Neurology 2013, 20: 655–662


Overuse of paracetamol caffeine aspirin powders affects cerebral
glucose metabolism in chronic migraine patients
W. Dia,†, X. Shib,†, Y. Zhuc, Y. Taoa, W. Qia, N. Luoa, Z. Xiaoa, C. Yib, J. Miaoa, A. Zhanga,
X. Zhangb and Y. Fanga

Department of Neurology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou; bDepartment of Nuclear Medicine, the First
Affiliated Hospital, Sun Yat-sen University, Guangzhou; and cZhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China


analgesic, chronic
migraine, glucose
metabolism, insula,
medication overuse,
Received 27 May 2012
Accepted 18 September 2012

Background and purpose: Overuse of analgesic plays a prominent role in migraine
chronification. Paracetamol caffeine aspirin (PCA) powders are commonly used in
Chinese migraineurs. This study investigated the effects of the specific combination
analgesic on cerebral glucose metabolism in chronic migraine (CM).
Methods: 18F-FDG-PET was used to measure regional metabolism in all subjects.
Brain metabolisms of CM patients with analgesic overuse (AO-CM; n = 10), no
analgesic overuse (NAO-CM; n = 10), and no regimen (NR-CM; n = 10) and 17
age- and gender-matched normal controls (NC) were compared using statistical
parametric mapping. Then, all patients underwent brain MRI analysis within 7 days
after PET scans, as well as MMSE and MoCA scale for cognitive function tests.
Results: Glucose metabolic changes in CM patients taking different dosage of analgesic during headache-free periods and clear distinctions in several brain regions
were observed. Patients with AO-CM exhibited significant metabolic reductions in
thalamus, as well as increased metabolism in middle temporal gyrus and insula relative to NR-CM and NAO-CM. However, in these regions, no difference was
observed in AO-CM except for increased metabolism in the right insula relative to
NC group.
Conclusions: Overusing PCA powders affects regional brain glucose metabolism in
CM. Increased metabolism in the right insula may be associated with recurrently
overusing of PCA powders.

Chronic migraine (CM) is a common and disabling
complication of migraine. Its global prevalence is
approximately 0.9–5.1% [1]. CM has a severely negative effect on patients’ quality of life and a higher economic burden on society than other forms of migraine
[2]. Treatment for CM is generally unsatisfied and
largely not evidence based [3]. Up to 14% of patients
with episodic migraine in tertiary clinic populations
are at risk of developing CM over 1 year [4]. The
probable risk factors for CM were subdivided into
Correspondence: Y. Fang, Department of Neurology, the First
Affiliated Hospital, Sun Yat-sen University, No. 58 Zhongshan
Road 2, Guangzhou 510080, China (tel.: +86 20 87755766;
fax: +86 20 87335935; e-mail: and
X. Zhang, Department of Nuclear Medicine, the First Affiliated
Hospital, Sun Yat-sen University, No. 58 Zhongshan Road 2,
Guangzhou 510080, China (tel.: +86 20 87755766; fax: +86 20
87338319; e-mail:

These authors contributed equally to this work.

© 2012 The Author(s)
European Journal of Neurology © 2012 EFNS

non-modifiable (such as female, age, low education
level, low socioeconomic status, and head injury) and
modifiable categories (such as stressful event, sleep
disturbance, obesity, depression, increased caffeine
consumption, baseline headache frequency, and analgesic overuse) [5–7]. Amongst these factors, analgesic
overuse plays a prominent role [8,9]. A recent research
reported that the medication classes were classified
based on the correlation between analgesic and
migraine progression (opioids and barbiturates > triptans >NSAIDs) [10].
Analgesic is commonly used to treat migraine in
China [11]; especially, paracetamol caffeine aspirin
(PCA) powders, which are compound preparation of
three ingredients (aspirin 230 mg, paracetamol
126 mg, and caffeine 30 mg per packet). Non-prescription drugs and low price of $ 0.016 per packet contribute to easily accessible to patients. The fact that quite
a few migraineurs frequently take PCA powders by
themselves when headaches attack was observed in the
Headache Center of our hospital. Unfortunately,


5 ± 5. although orbitofrontal cortex is involved in MOH evolving from episodic migraine [13]. other chronic systemic diseases. written informed consents were obtained. A total of 56 patients were recruited from the Headache Clinic. cognitive impairment. which commonly chose massage. patients) without migraine or family migraine. thus.45 vs.0-Tesla scanner. Netherlands) in threedimensional acquisition mode during headache-free © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology . scraping. and Montreal Cognitive Assessment (MoCA) Beijing Version (www. (episodic) tension type were used to detect the possible changes in brain structure and cognitive function. Inclusion criteria for the patients were as follows: complied with diagnostic criteria for CM or CM with MOH. known morphological brain abnormality on MRI or CT.656 W. Thus. Headache intensity was assessed using a visual analog scale (VAS) from 0 (no pain) to 10 (most headache imaginable). and persistent orbitofrontal cortex hypometabolism were observed after the withdrawal of analgesics in patients with MOH [13]. 10 were CM without overusing PCA powders (21 packets mean monthly). long-term headache. and thus help to understand migraine pathophysiology [14–16]. Neuroimaging facilitates researchers to explore brain areas or neural networks involved in the generation and mediation of migraine. Seventeen age. The reversible hypometabolism in insula. topiramate. and such metabolic abnormalities might be associated with medication dosage. substance abuse. other 10 cases with CM had no regimen including analgesic. and All CM patients were evaluated by two experienced neurologists of Headache Clinic. the Chinese version of Mini-Mental State Examination (MMSE). or sleeping to cope with headache. and detailed clinical data were collected. taking PCA powders or no regime. three were taking traditional Chinese medicine. P = 0. All had no identifiable abnormality on brain MRI. diabetes mellitus. three were taking paracetamol. The VAS scores refer to the mean intensity of the most painful part of their typical untreated or unsuccessfully treated attacks experienced in recent 3 months. T2WI. two were taking ergotamine–caffeine. thalamus. and substance abuse were recruited after signing written informed consents. 10 patients were CM with overusing PCA powders (100 packets mean monthly). in which one was overusing triptan. It is unclear whether the chronification was associated with brain structural and biochemical alterations. we performed 18F-fluorodeoxyglucose-PET (18F-FDG-PET) scans in CM patients [analgesic overuse (AO). abnormal scores (>50) of the Chinese version of Self-Rating Depression Scale (SDS) and Self-Rating Anxiety Scale (SAS). Amongst 56 patients. smearing ‘Qing Liang You’.5. neurocranial MRI. We hypothesized that interictal metabolic abnormalities in brain regions involved both in pain processing and in drug dependence might be found in CM patients overusing PCA powders. positron emission tomography (PET) studies have demonstrated that CM is associated with cerebral abnormal metabolism [17. and had transformed to CM for 3 months or more. overusing triptans or other drugs. 23 patients complied with the diagnosis criteria of CM with MOH. and other neurological or psychiatric disorders.18].and gender-matched normal volunteers (6 males and 11 females. and were devoid of taking drugs or foods affecting cerebral metabolism for 72 h before PET scanning. no analgesic overuse (NAO). had a history of migraine without aura (MOA) for more than 10 years. notably. taking prophylactic medications including valproate. CM with MOH was diagnosed according to the criteria proposed by Silberstein and Lipton [20]. the remaining 10 patients who were taking PCA powders took part in the study. Germany) within 7 days after PET scan. Second Edition (ICHDII) criteria [19]. aged 40–60 years. many individuals have transformed to CM or medication overuse headache (MOH) for many years and inclined to continue overusing analgesic despite knowing its side effects such as gastrointestinal tract hemorrhage [12]. All patients in headache-free periods received brain MRI (3. chronic pain disorder. 18F-FDG-PET/CT was performed using a Gemini GXL 16 scanner (Philips. The diagnosis of CM was based on the revised International Classification of Headache Disorders. Materials and methods Methods Participants The study received the approval from the Ethics Committee of Sun Yat-sen University. mocatest. Exclusion criteria for the patients were as follows: had a history of hypertension. Erlangen. According to the selection criteria. mean age 48. Siemens. and propranolol. and no regimen (NR)] during headache-free periods and in 17 normal controls (NC). the First Affiliated Hospital of Sun Yat-sen University between May 2011 and May 2012. T1WI. anterior cingulates. four were excluded (>60 or <40). Di et al. cupping. and FLAIR sequences were analyzed. Whether these metabolic alterations are associated with analgesic dosage remains unknown. stroke. Besides. heart disease.

net/xjview8/). previously monthly headache hours. Given brain regions possibly related to overusing PCA powders.020 0. Statistical analysis The spatially transformed and smoothed PET images were used to perform the voxel-based statistical analysis based on the general linear model. and trilinear interpolation.592 0.7 ± ± ± ± ± ± ± ± ± ± 6. 657 alivelearn.0 (SPSS Inc. the small volume-corrected P-value (P < 0. three groups of CM patients were compared with each other by the full factorial analysis of variance model (ANOVA) and post hoc multiple comparisons.6 9. Besides. weight.9 54.9 56.1 3/7 47. 10 no analgesic overuse – NAO-CM. stopping drugs for 24 h and fasting for over 8 h were necessary. There were no statistical differences in gender.7 4.535 0.3 5.7 © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology ± ± ± ± ± ± ± ± ± ± 4.1 40.0 42. Data were analyzed using statistical parametric mapping 5 (SPM5).6 1. We found a significant hypometabolism in bilateral thalamus and hypermetabolism in bilateral middle temporal gyrus and right insula in AO-CM group relative to NAO-CM and NR-CM.736 0. and 10 no regimen – NR-CM) are summarized in Table 1.7 2.2 22. Chicago. and the anatomical location of each cluster was reported on the MNI coordinate.412 0.2 19. Noteworthy.765 0.916 0.Overuse of PCA powders affects brain metabolism periods in subjects.6 42.4 39. Second. All images were spatially normalized onto the stereotaxic Montreal Neurological Institute (MNI) template using a 12-parameter affine transformation. USA).8 6.702 0.6 0.1 56.9 6.6 30.6 139.3 29.8 4.1 0. migraine history. and between each CM group and NC group. Other variables were analyzed by ANOVA and post hoc multiple comparisons and indicated as mean ± standard deviation.0 3. PET imaging data Amongst the three groups of patients (AO-CM.8 20.673 0.5 127.568 0.8 34.001 and voxel value > 50.8 20.8 . Specific imaging protocol for brain was selected with field of view 180 mm..9 1.5 9.310 0.7 4.8 29. IL.7 7. The normalized scans were then smoothed with an 8 mm full width at half-maximum isotropic Gaussian kernel to compensate for interindividual anatomical variability and increase the signal-to-noise ratio. they rested in a quiet.000 0.0 4.6 5.573 0.2 Fisher = 0. NAO-CM.401 0.374 0.9 20.203 0.3 27.3 142. and VAS score amongst the three groups. and NR-CM).5 0.4 1. et al. All the individual global counts were normalized by proportional scaling to the mean voxel value of 50.6 28.358 0. Clinical data were analyzed using SPSS 17. Significance was set at P < 0.1 42. significant differences in cerebral glucose metabolism are presented in Table 2. age.271 1.05) of each cluster was compared (8-mm-diameter sphere centered on peak voxel) amongst three groups of CM.817 0.3 28.409 1. Results Clinical data The clinical characteristics of the 30 CM patients (10 with analgesic overuse – AO-CM. Based on these Table 1 Comparison of clinical characteristics amongst three groups of CM patients Items AO-CM (n = 10) NAO-CM (n = 10) NR-CM (n = 10) F-value (unless specified) P-value Gender (male/female) Age (years) Weight (kg) Migraine history (years) Headache h/month (MOA) Headache days/month (CM) Visual analog scale score SAS (normal value < 50) SDS (normal value < 50) MMSE (normal value > 24) MoCA (normal value > 26) 4/6 49.9 4/6 46. non-linear transformation.0 29.6 6. no abnormality was observed in brain MRI examination in all patients.4 18. dim room with eyes closed.1 0.14 mCi/ kg) was injected intravenously in the morning. MMSE and MoCA scores were normal and showed no significant differences amongst three groups. Fisher’s exact test was used to compare gender variable.406 0..2 6. http://www. Statistical results were presented by xjView (Xu Cui. a 10-min acquisition started and yielded 90 axial slices with a defined voxel size of 2 mm3.01 (false discovery rate corrected) with minimal cluster of 50.6 9.6 41. The PET images were reconstructed by the line-of-response row-action maximum likelihood algorithm with lowdose CT images for attenuation correction. First. Before PET scanning. 45 min later.0 5. Once 18F-FDG (0. monthly headache days in CM situation.3 1.668 ± ± ± ± ± ± ± ± ± ± 7.9 0.3 41. comparisons between each group of patients and NC were made using two-sample t-test at a threshold of P < 0. and the transaxial spatial resolution was 2 mm full width at halfmaximum at center of the field of view.0 6.

001 0. Significant hypometabolism in bilateral thalamus and hypermetabolism in bilateral middle temporal gyrus and right insula in AO-CM group compared with NAO-CM and NR-CM were observed.005 0.003 <0.002 0.06 3.003 0. Relative to NR-CM. 1.001 <0.62 3. with 8-mm-diameter sphere centered on peak voxels from SPM analyses in NR-CM.008 0.001 0.06 0.003 Left rectus (BA47) Right middle frontal gyrus (BA9) Left post-central gyrus (BA5) Right superior frontal gyrus (BA6) Right rectus (BA11) Left middle temporal gyrus (BA21) Right superior temporal gyrus (BA42) Right cuneus (BA19) 13 33 26 25 8 60 60 8 27 23 42 19 22 20 28 93 16 37 64 62 28 0 16 24 2.001 <0.001 <0.05).001 <0.001 <0.002 0.001 0.88 3.001 0.001 <0.001 0.19 3.76 3.19 3 <0. © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology .001 0. AO-CM.001 0.001 <0.006 <0.24 3.001 0.001 0. Thalamic metabolism was increased above the normal in NAO-CM or NR-CM (P < 0. patients with AO-CM exhibited significant hypometabolism in thalamus (P < 0. although the latter group did not differ significantly from NR-CM.002 0. but there was no difference between NAO-CM and NR-CM in middle temporal gyrus.001 <0.5 3. metabolism in AO-CM was higher than in NAO-CM (P < 0.001 0.13 2.05). and right insula.002 The coordinates and statistics for decreased metabolism (AO-CM < NR-CM. and metabolic change between each CM patient group and NC group is shown in Fig.002 <0.57 3.002 0. NAO-CM. AO-CM < NAO-CM) and increased metabolism (AO-CM > NR-CM.05).56 3.51 3.001 0.54 4.17 3.001 0.001 0.001 0. NAO-CM > NR-CM. Patients with AO-CM exhibited significant hypermetabolism relative to NR-CM in middle temporal gyrus (P < 0.001 0.004 0.001 <0.001 <0.corr Left cerebellum posterior lobe Right thalamus Left thalamus Right caudate nucleus Right medial superior frontal gyrus (BA10) Left supramarginal gyrus (BA40) Left pre-central gyrus (BA6) Right middle temporal gyrus (BA21) Left middle temporal gyrus (BA21) Left superior temporal gyrus (BA22) Right insula (BA13) 30 3 4 12 4 64 33 62 55 64 50 56 14 24 14 67 53 21 21 24 32 23 20 8 6 6 0 23 69 8 4 10 21 3.01) group.04 3.45 3.002 <0. metabolism of the right insula was increased in AO-CM compared with NC group (P < 0.001 <0. Figure 3 shows the metabolic differences in regions.003 <0.001 <0.01 and minimal cluster of 50.001 0.05).001 <0.001 <0.59 0. However.004 <0.001 <0.39 3.04 3.001 0.001 <0. metabolism in AO-CM group was lower than in NAO-CM (P < 0.001 0. All three groups did not differ significantly from the NC group in middle temporal gyrus.87 3. Table 2 Comparison of brain glucose metabolism amongst three groups of CM patients MNI coordinates Group comparison AO-CM vs.63 2.66 3.001 0.001 0.56 3. In this region. but not in AO-CM group.001 0. 2. Z score represents the peak voxel.53 3. such as thalamus.77 4. NAO-CM > NAO-CM) regions were shown with significance level P < 0. NR-CM NAO-CM< NR-CM NAO-CM> NR-CM AO-CM vs.001 <0. NAO-CM < NR-CM.84 3. results.002 <0.05).2 2. PFDR-corr = P-value false discovery rate corrected. the visual analysis of 18F-FDG-PET/CT amongst the three groups is shown in Fig.75 3. middle temporal gyrus.001 0.001 <0.001 Right cerebellum Left cerebellum Right thalamus Left thalamus Right cuneus (BA18) Left rectus (BA25) Right superior temporal gyrus (BA21) Left middle temporal gyrus (BA21) Right middle temporal gyrus (BA22) Right insula (BA13) Right superior frontal gyrus (BA6) Right post-central gyrus (BA5) Left angular gyrus (BA39) 48 48 5 4 4 4 58 55 59 42 22 19 45 70 66 24 24 10 26 20 24 44 28 23 30 68 32 34 2 6 12 16 6 4 6 18 60 80 33 3.001 <0. NR-CM AO-CM < NR-CM AO-CM> NR-CM NAO-CM vs.001 0.05) and insula (BA13) (P < 0. and NC group.87 3. NAO-CM AO-CM< NAO-CM AO-CM> NAO-CM Brain regions x y z Z score Uncorrected P PFDR.001 0. MNI = Montreal National Institute template.001 0.005 0. In these regions.16 3.658 W.001 <0. Di et al.

Functional imaging studies have revealed the activation of the right insula during drug urges including cigarettes. (a) Showed significant hypometabolism in bilateral thalamus. Besides regionally decreased metabolism. Few studies were available for CM [13. However.01). and metabolism in NAO-CM or NR-CM group was increased above the normal. as they were controlled across the study populations. on cerebral glucose metabolism in 30 CM patients using voxel-based 18F-FDG-PET. thalamic metabolism in NR-CM or NAO-CM group was higher than in NC group. although AO-CM in our study differs from MOH. However. unlike the changes in the thalamus. NAOCM. but not different from NC group. we can conclude that there is a strong relationship between analgesic and thalamus. during exacerbation periods of CM (mean VAS 9. The thal- Figure 2 Comparison of brain metabolism between each CM patient group and normal controls (NC) group using voxelbased SPM analysis (P < 0. gender.Overuse of PCA powders affects brain metabolism 659 (a) (b) (c) Figure 1 Comparison of brain metabolism amongst three groups of CM patients using voxel-based SPM analysis (P < 0. PCA powders. and NR-CM) during headache-free periods and found that glucose metabolic changes in several brain regions in AO-CM were different from those in NAOCM and NR-CM group. respectively. suggesting that insular hypermetabolism is specific to AO-CM. possibly could not recover by proper dosage of PCA powders. We observed glucose metabolic changes in three groups of patients (AO-CM. and only AO-CM group exhibited abnormally increased metabolism. This finding is consistent with our result to some extent. Thus. One possible reason was that patients in the current study. colored bar represents T-values. The first row showed no differences in thalamus and middle temporal gyrus between (AO-CM. and PCA powders may play analgesic effect through reducing thalamic metabolism.22]. NAO-CM.0) and hypermetabolism in thalamus. These findings could not be attributed to the differences in pain levels or other factors such as age. These results indicated that AO could reduce glucose metabolism in the thalamus of CM. and monthly headache days. Positron emission tomography imaging revealed significant hypometabolism in the thalamus of patients with AO-CM relative to NAO-CM and NR-CM. blue-green represents hypometabolism. amus is a major center for processing and integration of nociceptive inputs.23].001). (b) and (c) showed significant hypermetabolism in bilateral middle temporal gyrus and right insula. and cocaine [24–26]. we also found hypermetabolism in the right insula in AO-CM compared with NAO-CM and NR-CM group. duration of migraine history. For example. alcohol. second row showed significant hypermetabolism in the right insula (AO-CM) relative to NC group. Red-yellow represents hypermetabolism. which underlies abnormal decision making that leads © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology . no difference was observed except for increased metabolism in the right insula between AO-CM and NC group. NAO-CM or NR-CM group revealed no difference from NC group. Noteworthy. pronounced hypometabolism in thalamus in MOH was confirmed [13]. Discussion This study investigated the influence of the specific combination analgesic. There were evidences from functional imaging of activation in the thalamus in spontaneous migraine [21. The insula has increasingly become the focus of attention for its role in body representation and subjective emotional experience. NR-CM) and NC group.

31]. and drug dosage may also contribute to the difference in regionally cerebral metabolism. to drug abuse despite the awareness of negative consequence [27]. we speculated that the right hypermetabolic insula in AO-CM patients was associated with dependence on the analgesic compound. no significant metabolic change in brainstem was observed in our study. NAO-CM. ns = not significant. at the threshold of P < 0. *Represents the significance of each patient group compared with NC group.05. Besides its role in drug dependence. and caffeine. Besides. metabolism in each patient group was not statistically different from NC group. suggesting a weak effect of overusing PCA powders on metabolism in this region. Interestingly. which is not consistent with a previous finding of persistent hypometabolism after withdrawal of analgesics [13]. Similar to the report in MOH during interval headaches [13]. the insula was recognized as being involved in central pain processing and frequently activated during pain attack [21. Although brain dysfunction might vary in severity between analgesic overuse and drug addiction. Figure 3 Means. Also. An imaging study demonstrated that hyperperfusion in the right middle temporal gyrus was strongly predictive of poor response to gabapentin [29].660 W.28]. Di et al. drug metabolism. no abnormality in MMSE and MoCA scores was found in all CM patients. suggesting no influence of PCA powders on metabolism in brainstem. we speculated that hypermetabolism in middle temporal gyrus might be functional manifestation of subjective evaluation of unreasonable therapy. Medication overuse is recognized as violating medication principle. we did not detect dysmetabolism in orbitofrontal cortex. Other factors including race. but a combination analgesic. intractable AO-CM is possibly associated with DMN function. which is composed of aspirin. These findings confirmed that the insula could play a pivotal role both in drug dependence and in pain processing. Hypermetabolism in the bilateral middle temporal gyrus was observed in AO-CM relative to NAO-CM and NR-CM group. the middle temporal gyrus is involved in the visual subjective evaluation of action rationality and showed increased activity when the expected pattern of rational goal attainment is violated [32]. providing the evidence that CM or PCA powders perhaps would not cause cognitive dysfunction. indicating that hypermetabolism in this region was possibly associated with analgesic overuse. There is no evidence currently that cognitive function is affected by © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology . this region is involved in the default mode network (DMN). Noteworthy. thus predisposing these patients to recurrently overuse PCA powders. and significance level of metabolism with 8 mm diameter sphere centered on peak voxel in NR-CM.*P < 0. It is possible that these patients with AO-CM in our study did not take ergotamine–caffeine or other preventive drugs. standard errors.05. Thus. AO-CM and normal controls (NC) group. Moreover. Given the present findings. which appears to provide a balance of opposing forces and maintain the brain resting state [30. which was realized by AO-CM patients. paracetamol.

and a combination of flunarizine and topiramate in migraine prophylaxis. 27. Headache 2006. 661 9. 3. 20. Cephalalgia 2007. Matharu MS. et al. Sanchez DRM. Magis D. or caffeine. et al. 8. Several limitations of the study should be considered. Lipton RB. Brain 2005. Neurology 2008. Manack A. Burstein R. Cephalalgia 2006. Kim JH. Moulton E. Brainstem activation specific to migraine headache. Differential brain activity in alcoholics and social drinkers to alcohol cues: relationship to craving. Bigal ME. 77: 236–242. Functional neuroimaging of headaches. Headache 2008. 31: 837–850. animal experiment will be needed to clarify the pathogenesis of analgesic dependences. Goadsby PJ. 318: 602–606. et al. Borsook D. 22. First. Evaluation and proposal for optimization of neurophysiological tests in migraine: part 2–neuroimaging and the nitroglycerin test. All normal brain MRIs seem to provide some information that there are no abnormal brain structure in many patients with AO-CM. Luo N. 158: 86–95. Midgette LA. 128: 932–939. Mudge M. 15. Headache 2008. Khodavirdi A. focusing on chronic migraine with or without medication overuse: a systematic review. et al. Serrano D. Headache 2007. Quality of life impairment. Duru G. Zhang A. Am J Psychiatry 2001. 46(Suppl 1): S32–S38. Myrick H. et al. tDCSinduced analgesia and electrical fields in pain-related neural networks in chronic migraine. Bigal ME. Koh SB. 13: 277–283. Pain Med 2012. 23. Acknowledgements This study was supported by a grant from the National Natural Science Fund of China (grant number 81171101). Wang Z. one-year clinical trial comparing the efficacy of topiramate. 27: 1339–1359. 11. Incidence and predictors for chronicity of headache in patients with episodic migraine. Zaghi S. Excessive acute migraine medication use and migraine progression. Tepper SJ. 48: 1157–1168. 19. comparison in cerebral metabolism or structure between PCA powders and other analgesics or migraine medications should be explored in future clinical trials. Barrodale PM. Suh SI. Lancet 2001. Lanteri-Minet M. 26: 742–746. Fulbright RK. Anton RF. 27: 14035–14040. Curr Opin Neurol 2000. Decision-making dysfunctions in psychiatry –altered homeostatic processing?. 357: 1016–1017. Bahra A. flunarizine. 13. 29: 393–402. Dasilva AF. A randomized. Faith M. Mendonca ME. 47: 996–1007. Science 2007. Frackowiak RS. even though several researches have been performed [33–35]. Paulus MP. Tipton RL. Lancet Neurol 2004. 129: 543–550. J Neurosci 2007. 13: 80– 86. 22: 269–276. 26. Kim S. A PET study exploring the laterality of brainstem activation in migraine using glyceryl trinitrate. et al. 24. 21. 78: 940–941. Bigal ME. What predicts the change from episodic to chronic migraine? Curr Opin Neurol 2009. et al. Bigal ME. metabolic changes could be due to the overuse of any of the aspirin. Oh K. Scher A. Functional imaging of the trigeminal system: applications to migraine pathophysiology. Goadsby PJ. Katsarava Z. Wexler BE. Dean B. Lee L. Combination therapy for chronic migraine: bad news but not the last word. Afridi SK. Tepper DE. Laureys S. 2. Interictal metabolic changes in episodic migraine: a voxelbased FDG-PET study. Olesen J. Lipton RB. Breaking the cycle of medication overuse headache. Buse D. et al. Buchel C. 30: 53–61. et al.Overuse of PCA powders affects brain metabolism migraine and migraine-related drugs [33]. 3: 645–651. 48: 16–25. Di Clemente L. Neuropsychopharmacology 2004. Park KW. Schneeweiss S. Alvarez LJ. 10. Natoli JL. 52: 1283–1295. Global prevalence of chronic migraine: a systematic review. New appendix criteria open for a broader concept of chronic migraine. 62: 788–790. Lipton RB. Diener HC. Bendtsen L. Third. et al. © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology . Cephalalgia 2011. Gottschalk CH. paracetamol. 7. Aurora SK. Fumal A. 16. Orbitofrontal cortex involvement in chronic analgesic-overuse headache evolving from episodic migraine. Cleve Clin J Med 2010. 14. Lipton RB. 4. Acute migraine medications and evolution from episodic to chronic migraine: a longitudinal population-based study. a large population-based study is very necessary. Chronic daily headache. 25. Duan Xu at University of California San Francisco for his assistance in writing this manuscript. 13: 301–307. Functional magnetic resonance imaging of cocaine craving. 5. 71: 1821–1828. Matharu MS. Lipton RB. this is a small case–control study investigating the effects of PCA powders on cerebral glucose metabolism. Second. et al. Headache 2012. Becerra L. Neurology 2004. Cephalalgia 2010. and NR-CM. Patterson F. References 1. Risk factors for headache chronification. Li X. 30: 599–609. Is medication-overuse headache a distinct biological entity? Nat Clin Pract Neurol 2006. Cottrell S. Brain 2006. Scher AI. disability and economic burden associated with chronic daily headache. Cephalalgia 2010. Neurology 2012. 18. Overuse of acute migraine medications and migraine chronification. Kurth T. Stewart WF. Bousser MG. Finally. Curr Pain Headache Rep 2009. NAO-CM. Brainstem dysfunction in chronic migraine as evidenced by neurophysiological and positron emission tomography studies. 6. Disclosure of conflict of interest The authors declare no financial or other conflict of interests. Neural substrates of abstinence-induced cigarette cravings in chronic smokers. Goadsby PJ. Di W. Pascual J. Lipton RB. Of course. Silberstein SD. 17. Appreciation is expressed to Prof. 2: 401. 12.

Jolles J. Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. 32. Cognitive and psychiatric effects of topiramate monotherapy in migraine treatment: an open study. Reiss AL. Marciani MG. J Neurol Sci 2009. Greicius MD. 13: 255–262. 50: 176–184. van Boxtel MP. 33. 1097–1099. 30. Brain perfusion in fibromyalgia patients and its differences between responders and poor responders to gabapentin. 29. 100: 253–258. Raichle ME. 34. 283: 134–136. Headache 2010. Ferraro S. Pain processing in medication overuse headache: a functional magnetic resonance imaging (fMRI) study. 15: 190–195.662 W. Hatta K. © 2012 The Author(s) European Journal of Neurology © 2012 EFNS European Journal of Neurology . 21: 318–329. Pain Med 2011. Arthritis Res Ther 2010. Di et al. Proc Natl Acad Sci USA 2003. Romigi A. Gergely G. Cervellino A. Clavagnier S. 12: R64. Usui C. Jastorff J. Neuroimage 2007. Neural mechanisms of understanding rational actions: middle temporal gyrus activation by contextual violation. et al. Doi N. 35. Krasnow B. Baars MA. Grazzi L. 28. Orban GA. Brain lesions and cerebral functional impairment in migraine patients. et al. Migraine does not affect cognitive decline: results from the Maastricht aging study. Menon V. Snyder AZ. Cereb Cortex 2011. 31. Mandelli ML. Eur J Neurol 2008. et al. A default mode of brain function: a brief history of an evolving idea. Paemeleire K. 37: 1083–1090.

However. or email articles for individual use. download.Copyright of European Journal of Neurology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. . users may print.