Abstract
Interictal serum C-reactive protein (CRP) was measured in 50 young adult patients with migraine and compared with 50 controls. The median CRP level was 1.42 mg/l in patients with migraine and 0.90 mg/l in controls (P = 0.03). This finding supports the role of inflammation in migraine, but needs confirmation in larger controlled studies. Prospective studies may establish whether measurements of CRP can identify patients with migraine at risk for cardiovascular events.
Introduction
Migraine is a risk factor for ischaemic stroke, in particular in young women suffering from migraine with aura (1). Migraine is also associated with silent brain infarcts and deep white matter lesions detected by magnetic resonance imaging (2, 3). Both types of brain lesion have been shown to increase stroke risk in the general population (4). The mechanisms underlying the association between migraine and ischaemic cerebrovascular disease are unknown.
Inflammation within certain brain tissues resulting from neuronal activation and the subsequent release of proinflammatory neuropeptides from perivascular nerve endings is supposed to occur during a migraine attack (5). Inflammatory processes within the vasculature are well recognized to play a central part in the pathogenesis of ischaemic stroke (6). Repeated episodes of perivascular inflammation during migraine attacks might therefore contribute to the increased stroke risk in migraine (7).
C-reactive protein (CRP) is a sensitive indicator of active systemic inflammation and an independent risk marker for cardiovascular morbidity, including ischaemic stroke (8). A small, uncontrolled retrospective review found abnormal CRP levels in migraineurs with complex clinical features referred to secondary or tertiary clinics (9). The aim of the present study was to assess structural and functional vascular properties in patients with migraine compared with individuals without a history of migraine. Here, we report the results of CRP measurements; other vascular measurements will be reported elsewhere.
Methods
Fifty patients suffering from migraine with aura (MA) or without aura (MoA), according to the diagnostic criteria of the International Headache Society, and 50 control subjects matched for gender, age, smoking status and current hormonal contraceptive use were recruited through advertisements from hospital and university staff, university students and the general population. Diagnosis was initially made by a general physician or neurologist and was confirmed using a validated questionnaire and structured interview. Major exclusion criteria were: age <18 and ≥35 years; body mass index <18 and ≥30 kg/m2; history of cardiovascular disease; hypertension; diabetes; hypercholesterolaemia; pregnancy or lactation and daily use of drugs (except hormonal contraceptives). The use of any acute non-specific or specific antimigraine drug was allowed. Patients with a history of migraine of <1 and >6 years, using prophylactic therapy or having ≥15 days of headache per month were excluded. Control subjects with more than one headache episode per month, a history of moderate or severe headaches or first-degree relatives with a history of migraine were excluded. Based on medical history, physical examination and routine laboratory tests, all subjects were in good health.
The study was approved by the Ethics Committee of the University Hospital. All subjects provided written, informed consent prior to participation.
Fasting blood samples were obtained from an antecubital vein during a headache-free period, i.e. >72 h headache free since last attack. If a migraine attack ensued within 24 h, sampling was repeated in another headache-free period. Serum was collected within 1 h and kept frozen at −70°C until analysis. Serum high-sensitivity CRP was measured by latex-enhanced immunoturbidimetry using commercial reagents with mouse monoclonal anti-CRP antibodies (Integra 800 analyser; Roche Diagnostics, Vilvoorde, Belgium).
Continuous variables were compared between groups by t-test for independent samples or Mann–Whitney U-test, as appropriate. Categorical variables were compared between groups by χ2 test. Smoking (never/ever) and current hormonal contraceptive use (yes/no) were modelled as dichotomous variables. CRP values were logarithmically transformed prior to statistical analysis. Multiple linear regression analyses were performed to adjust for known confounding factors (blood pressure, body mass index, smoking, high-density lipoprotein (HDL)-cholesterol, triglycerides and current hormonal contraceptive use).
Results
Clinical characteristics of the study participants are shown in Table 1. None of the study participants had clinical evidence of active infection. MA was diagnosed in 32 (64%) and MoA in 18 (36%) patients. As acute antimigraine therapy, analgesics were used by 33 patients (66%), triptans by nine (18%), ergotamine by one (2%) and no medication by seven (14%) patients.
Characteristics of study participants∗
Unless indicated otherwise, differences between groups were not statistically significant (P > 0.05).
Compared with controls: P = 0.046.
Serum CRP levels were increased in patients with migraine compared with controls and were highest in patients with MoA (Fig. 1; Table 2). After adjustment for potential confounding variables, the relationship between serum CRP and migraine remained significant (P = 0.045 and P = 0.003 for patients with migraine and patients with MoA vs. controls, respectively).
Box and whisker plots of serum C-reactive protein levels in patients with migraine (n = 50) and controls (n = 50). MA, Patients with migraine with aura (n = 32); MoA, patients with migraine without aura (n = 18). The line within the box marks the median and the boundaries of the box closest and farthest to zero indicate the 25th and 75th percentile, respectively. Whiskers below and above the box indicate the 10th and 90th percentiles, respectively. ∗P = 0.03; †P = 0.0002.
Serum levels of C-reactive protein
MA, Patients with migraine with aura; MoA, patients with migraine without aura.
Compared with controls: P = 0.03.
Compared with controls: P = 0.0002.
Discussion
In this cross-sectional study, interictal serum CRP levels were increased in young adult patients with migraine compared with matched control subjects without a personal or family history of migraine. These findings support the putative role of inflammation in migraine.
In a retrospective study in patients with migraine who presented to a specialist headache or general neurology clinic with complex clinical features, an elevated serum CRP level, defined as >3.0 mg/l, was present in a substantial number (26 out of 60, 43%) of patients (9). The main limitations of this previous study were the absence of a control group, the highly selected study population and possible confounding by coexisting disease and classical cardiovascular risk factors or other known confounders that were not considered, such as current use of statins or hormonal contraceptives. In contrast, the present study included a control group of individuals without a personal or family history of migraine. In addition, study participants were recruited from the general population and free of coexisting disease, did not take concomitant medication, had a low cardiovascular risk profile and were matched for known confounders, including blood pressure, body mass index, smoking, HDL-cholesterol, triglycerides and current use of hormonal contraceptives.
Serum CRP is an established inflammatory risk marker for ischaemic stroke (8) and coronary heart disease (10). Recently, large prospective studies have convincingly shown a small increased risk for ischaemic stroke (11–13) and myocardial infarction (13) in patients with MA. In addition, increased serum CRP levels are associated with the presence and progression of white matter lesions (14). Interestingly, a population-based study has suggested that, among women, migraine increases the risk of white matter lesions (2). The findings from the present study therefore suggest a possible mechanism underlying the relation between migraine and ischaemic cardiovascular disease. Although prospective studies suggest a concentration-dependent relationship between serum CRP and the risk of ischaemic stroke, it is still unclear which CRP values identify different grades of ischaemic stroke risk (8). In contrast, relative risk categories for coronary risk have now been established based on aggregation of population studies (10). When applying these cut-offs to our findings, 48% of the control subjects vs. 66% of migraine patients had an intermediate to high coronary risk. Future prospective studies may establish whether measurements of serum CRP can identify patients with migraine at risk for ischaemic cardiovascular disease.
In the present study, CRP levels tended to be higher in patients with MoA compared with patients with MA. This concurs with the study by Welch et al., who observed that elevated CRP levels were more frequent in patients with MoA (9). If confirmed by larger studies, this finding suggests a real difference in pathophysiology between MoA and MA. Interestingly, in prospective studies, an increased cardiovascular risk was observed exclusively in patients with MA (11–13).
It is important to note several limitations of the present study. First, alcohol consumption and physical activity, which are associated with decreased serum CRP levels, were not recorded and might have introduced a bias. Indeed, a population-based study has suggested that migraine patients are less likely to drink alcohol and participate in sport (15). Second, no repeated measurements were made to account for within-individual variability of serum CRP. Third, because of the cross-sectional design of the study, the temporal relationship between the increase in serum CRP and the onset of migraine is unknown. Fourth, the large proportion of migraine patients with aura is not consistent with the known predominance of migraine patients without aura and suggests a sampling bias. The exclusion of patients with a history of migraine >6 years could explain why more patients with aura were recruited, as this diagnosis is probably made earlier due to the easily recognizable aura symptoms. Thus, we urge caution in extrapolating these findings to the general population of migraine patients. Finally, the number of patients is relatively small, increasing the risk of a false-positive finding, especially when performing analyses according to migraine subtypes.
Competing interests
None declared.
Acknowledgements
This work was supported by a grant from the Research Foundation–Flanders (F.W.O.–Vlaanderen). F.H.V. was a Junior Research Fellow of the F.W.O.–Vlaanderen at the time of the study. We are indebted to Dr Michel Langlois (Department of Clinical Chemistry, Ghent University Hospital, Ghent, Belgium) for expertly performing the serum CRP determinations.