Sage Journals: Discover world-class research

Abstract

Background

Chronic migraine has a well-documented association with increased insulin resistance and metabolic syndrome. The hypothalamus may play a role in the progression of insulin resistance in chronic migraine through the regulation of orexigenic peptides such as neuropeptide Y. Insulin resistance may lead to increased risk of future type 2 diabetes mellitus in patients with chronic migraine, which is more likely to occur if other pathogenetic defects of type 2 diabetes mellitus, such as impaired pancreatic β-cell functions and defects in intestinal glucagon-like peptide-1 secretion after meals. We studied the relationship of fasting neuropeptide Y with insulin resistance, β-cell function, and glucagon-like peptide-1 secretion in non-obese female chronic migraine patients. We also aimed to investigate glucose-stimulated insulin and glucagon-like peptide-1 secretions as early pathogenetic mechanisms responsible for the development of carbohydrate intolerance.

Methods

In this cross-sectional controlled study, 83 non-obese female migraine patients of reproductive age categorized as having episodic migraine or chronic migraine were included. The control group consisted of 36 healthy females. We studied glucose-stimulated insulin and glucagon-like peptide-1 secretion during a 75 g oral glucose tolerance test. We investigated the relationship of neuropeptide Y levels with insulin resistance and β-cell insulin secretion functions.

Results

Fasting glucose levels were significantly higher in migraine patients. Plasma glucose and insulin levels during the oral glucose tolerance test were otherwise similar in chronic migraine, episodic migraine and controls. Patients with chronic migraine were more insulin resistant than episodic migraine or controls (p = 0.048). Glucagon-like peptide-1 levels both at fasting and two hours after glucose intake were similar in chronic migraine, episodic migraine, and controls. Neuropeptide Y levels were higher in migraineurs. In chronic migraine, neuropeptide Y was positively correlated with fasting glucagon-like peptide-1 levels (r = 0.57, p = 0.04), but there was no correlation with insulin resistance (r = 0.49, p = 0.09) or β-cell function (r = 0.50, p = 0.07).

Discussion

Non-obese premenopausal female patients with chronic migraine have higher insulin resistance, but normal β-cell function is to compensate for the increased insulin demand during fasting and after glucose intake. Increased fasting neuropeptide Y levels in migraine may be a factor leading to increased insulin resistance by specific alterations in energy intake and activation of the sympathoadrenal system.

Keywords

Chronic migraine non-obese female glucagon-like peptide-1 insulin resistance neuropeptide Y β-cell function

Introduction

Migraine is a complex neurological disorder characterized by recurrent headache attacks (1,2). Based on the frequency of headache, migraine can be classified as episodic (EM) or chronic (CM) migraine (2). Premonitory symptoms of migraine attacks, including tiredness, mood changes, and increased appetite to certain foods, suggest a specific role for the hypothalamus in the pathophysiology of migraine (3). Recent imaging studies indicating increased blood flow in the hypothalamus during headache attack provide additional support for the prominent role of this brain region (4). The hypothalamus acts as one of the major modulators of food intake and energy expenditure by controlling a complex integrated system of orexigenic and anorexigenic pathways (5). One of the major hypothalamic orexigenic pathways is the neuropeptide Y (NPY)/agouti-related protein complex (5 –8). NPY has a stimulating role in food intake and drug-seeking behaviour as well as the pain pathway. Both insulin and NPY have pain modulating effects in animal models, and recent experimental data have demonstrated the antinociceptive actions of NPY on a migraine animal model (9,10). Experimental studies also suggest that the appetite stimulating effect of NPY may play a role in the development of human obesity and hypertension (11). Although results about the risk of metabolic complications in migraine are conflicting, patients with CM have a documented association with increased insulin resistance (IR) (12). In a study conducted in young, non-obese and non-diabetic migraine patients, IR was found to be significantly higher compared to non-migraineurs (13). According to a recent study, EM subjects were normal in terms of insulin sensitivity, but CM subjects had a significant association with increased IR (14,15).

The exact pathophysiological alterations that might be responsible for this complex association of migraine with IR are unknown. But IR is one of the major pathogenetic factors responsible for the development of type 2 diabetes mellitus (T2DM). Studies report conflicting results about the prevalence of T2DM in migraine (16 –18). The presence of IR associated with CM may lead to increased risk of T2DM in those patients, which is more likely if the other pathogenetic defects of T2DM are also present (19 –23). T2DM generally occurs due to a progressive pancreatic insulin secretion defect developed in the background of increased IR. An individual has already lost approximately 80% of their β-cell function by the time a diagnosis of T2DM is made (19,24). Therefore, CM patients with increased IR may have an increased future risk of developing T2DM if there is an associated impairment in β-cell insulin production (20 –24).

The other pathogenetic defect of T2DM is blunted intestinal glucagon-like peptide-1 (GLP-1) response to carbohydrate intake (25 –27). GLP-1 is the major incretin produced in intestinal L-cells after meal intake, and regulates post-meal blood glucose levels (26,27). Alterations in insulin and GLP-1 secretion as determinants for future diabetes risk have not been studied in CM and need to be documented to clarify the metabolic risk to those patients in relation to increased IR.

We hypothesized that alterations in NPY levels during migraine attacks and associated changes in food and energy intake may affect insulin sensitivity and glucose metabolism in the course of CM. We investigated the relationship of NPY with IR, β-cell function and GLP-1 secretion as determinants of T2DM in non-obese female CM patients and compared the results with those of EM subjects and healthy controls.

Methods

This study included the same subjects as in our previous study evaluating the impact of depression and ghrelin on body weight in migraineurs (28). Three hundred and seventy-four consecutive newly diagnosed migraine patients attending our institution for the first time were recruited according to the International Classification of Headache Disorders (ICHD)-II criteria (29). The exclusion criteria were pregnancy, diabetes mellitus (30), endocrinological disease, serious psychiatric disorders, chronic renal or hepatic disease, malignancy, acute cardiovascular events, neurological disease, intake of any migraine treatment drug, and use of any medication that might interfere with IR and glucose metabolism. Waist ratio (cm), body weight (kg) and height (cm) were measured, and body mass index (BMI) was calculated by dividing weight by height squared. Obese patients, that is, those with a BMI > 29.9 kg/m², were excluded. Out of 374 patients screened, 83 eligible female patients of reproductive age were included in the study to eliminate the effect of gender on carbohydrate metabolism. Patients underwent extensive face-to face-physical and neurological examinations by an expert endocrinologist and a neurologist specialized in headache disorders. Migraine patients were categorized as EM or CM (29). The control group consisted of 36 healthy female volunteers. All participants were Caucasian and were matched for smoking status and number of diabetic risk factors (30). The study was approved by the local Ethics Committee of our institution, and informed consent was obtained from the subjects.

The patients were studied during an attack-free period lasting at least two days. Blood samples were taken after a 10-hour overnight fasting period to determine fasting plasma glucose (FPG), insulin, NPY and GLP-1 levels. Participants underwent a 75 g oral glucose tolerance test (OGTT) (31). Blood samples were obtained at 60, 120, 180, 240 and 300 min after the oral glucose load. To study early insulin secretion, we used samples from the 60th and 120th minute to determine the glucose and insulin levels, and we measured GLP-1 levels in samples from the 120th minute. All blood samples were immediately centrifuged (4℃, 4000 rpm, 10 min), and the serum was stored at −70℃ until further analysis. Serum glucose levels were measured by enzymatic methods using a commercial kit (Architect system, Abbott, USA) and an automatic analyzer. The results are given in mg/dL. Serum insulin levels were obtained through a chemiluminescent microparticle immunoassay using commercial kits (Architect system, Abbott, USA). The sensitivity of the insulin assay was calculated to be 1.0 $μ U / ml$ , and the assay precision was ≤ 7% of the total coefficient of variation (CV). GLP-1 levels were determined during fasting and 120 min after the oral glucose load. Commercial enzyme immunoassay kits were used according to the manufacturer’s guidelines (Bachem, Peninsula Laboratories, LLC, USA). The intra- and inter-assay CVs were both 10%. The minimum detectable level for GLP-1 was 0.2 pg/mL. Serum NPY levels were determined by enzyme immunoassay according to the manufacturer’s guidelines (Phoenix Pharmaceuticals, Burlingame, CA, USA). Subjects were defined as diabetic and/or prediabetic according to the diagnostic criteria of the American Diabetes Association (30).

Insulin sensitivity

To estimate insulin sensitivity, we used the homeostasis model assessment for IR (HOMA-IR) index, calculated as [Fasting glucose (mmol/l) × fasting insulin (μU/mL)]/22.5 (32). We selected the 75th percentile value for HOMA-IR (2.37) as the cutoff point to define increased IR.

β-cell function

β-cell function was evaluated as insulin response to a glycemic stimulus as measured by the insulinogenic index, an OGTT-based measure that has been widely used (32,33). Plasma glucose increments and insulin responses to oral glucose intake were estimated by calculating the area under the curves (AUCs) for both glucose and insulin levels between 0 and 120 min of OGTT, obtained with the trapezoidal rule after removal of basal glucose and insulin concentrations. The insulinogenic index was calculated as AUC insulin_{0–120 min}/AUC glucose_{0–120 min}. The choice of 120 min is derived from the fact that insulin secretion occurs predominantly in the first two hours of the OGTT (33).

To estimate β-cell function, we also used the HOMA-β index, which is derived from a mathematical assessment of the balance between hepatic glucose output and insulin secretion as inferred from fasting levels of glucose and insulin. HOMA-β was calculated as 20 × fasting insulin (μIU/mL)/fasting glucose (mmol/ml) − 3.5 (34).

Statistical analysis

Statistical analyses were performed using SPSS software (version 20). All analyses were two-tailed with α = 0.05. The data are expressed as median ± standard deviation (minimum-maximum). Normality of the distribution of all variables was assessed by Kolmogorov-Smirnov test. ANOVA, Student’s t-test and the Pearson correlation coefficient were used for normally distributed variables. P values < 0.05 were considered statistically significant. Pearson’s chi-squared test was performed to compare observed data.

Results

Out of 374 migraineurs, 83 eligible female patients (mean age = 33.8 ± 8.6 years; EM = 59 patients; 71.9% and CM = 24 patients; 28.1%)) and 36 healthy female controls (mean age = 31.5 ± 8.0 years) were included. One CM patient with OGTT-diagnosed T2DM was excluded from the study, leaving 82 migraineurs (59 patients with EM and 23 with CM; 71.6 and 28.4%, respectively) in total. The baseline characteristics of migraine patients and control subjects are presented in Table 1. Migraineurs (n = 82) and healthy controls (n = 36) were similar in their mean age (33.8 ± 8.6 and 31.5 ± 8.0 years). Chronic migraineurs were slightly older than episodic patients (36.4 ± 8.8 and 32.8 ± 8.4 years), but the difference was not statistically significant. The mean BMI was 24.3 ± 3.3 kg/m² in migraineurs and 22.6 ± 3.1 kg/m² in controls. Although the subjects were all non-obese, BMI was significantly higher in migraine patients than controls (p = 0.015). The other parameters such as blood pressure, HDL cholesterol, triglycerides and waist circumference were statistically similar in migraineurs and controls (Table 1).

Table 1.

Baseline characteristics of migraine patients and control subjects.

	Groups
	Migraine	Controls	p-value	Episodic migraine	Chronic migraine	Controls	p-value
N Age	82 33.8 ± 8.6	36 31.5 ± 8.0	0.17	59 32.8 ± 8.4	23 36.4 ± 8.8	36 31.5 ± 8.0	0.1
WC¹	79.7 ± 10.6	77.8 ± 8.8	0.34	79.5 ± 11	80.3 ± 9.9	77.8 ± 8.8	0.61
BMI²	24.3 ± 3.3	22.6 ± 3.1	0.015	24.0 ± 3.1	24.3 ± 2.6	22.6 ± 3.1	0.051
SBP³	109.4 ± 12.7	110 ± 9.6	0.81	110 ± 16	110 ± 18	110 ± 9.6	0.91
DBP⁴	72.7 ± 9.7	71.7 ± 6.5	0.55	70 ± 11	70 ± 12	71.7 ± 6.5	0.78
TG⁵	95.7 ± 46.1	90.1 ± 40.7	0.57	92.5 ± 46.7	103.3 ± 44.6	90.1 ± 40.7	0.55
HDL-C⁶	52.9 ± 11.4	51.3 ± 12.7	0.54	52.5 ± 11.2	54.1 ± 12.1	51.3 ± 12.7	0.73
HbA1c⁷	4.9 ± 0.2	4.8 ± 0.2	0.81	5.0 ± 0.2	4.9 ± 0.2	4.8 ± 0.2	0.75

Waist circumference (cm); ²Body Mass Index (kg/m²); ³SBP: Systolic blood pressure (mmHg); ⁴DBP: Diastolic blood pressure (mmHg); ⁵TG: Triglycerides (mg/dl); ⁶HDL-C: High density lipoprotein cholesterol (mg/dl); ⁷HbA1c: HemoglobinA1c (%).

Glucose metabolism

Nine (10.9%) migraineurs (four with EM; 6.8%, five with CM; 21.7%), and three (8.3%) controls had impaired glucose tolerance (IGT; prediabetes). The difference between groups was found to be non-significant using the chi-squared test. Figure 1 represents plasma glucose and insulin responses to glucose load. Fasting glucose levels were significantly higher in migraine patients compared to controls (91.8 ± 8.2 and 83.4 ± 7.3 mg/dl; p < 0.001). When we performed ANOVA to compare the migraine groups separately, the difference was still significant than controls (p < 0.001). Plasma glucose and insulin levels obtained during OGTT were otherwise similar in CM, EM, and control subjects. The AUCs for glucose and insulin were also similar between patients with migraine and control subjects (Table 2).

Figure 1.

Plasma glucose and insulin concentrations after a 75 mg oral glucose load in migraine patients and controls.

Table 2.

Results in migraine subgroups and control subjects.

				Groups
	Migraine n = 82	Control n = 36	p-value	Episodic n = 59	Chronic n = 23	Control n = 36	p-value
Glucose₀¹	91.87 ± 8.21	83.44 ± 7.37	<0.001	91.34 ± 7.61	93.42 ± 9.64	83.44 ± 7.37	<0.001
Glucose₆₀²	128.10 ± 35.92	115.69 ± 36.76	0.089	126.9 ± 34.6	131.3 ± 39.8	115.69 ± 36.76	0.21
Glucose₁₂₀³	108.45 ± 27.22	103.94 ± 26.72	0.40	106.2 ± 23.1	114.2 ± 35.6	103.9 ± 26.7	0.35
Insulin₀⁴	9.56 ± 4.51	8.48 ± 3.20	0.19	9.5 ± 4.3	8.4 ± 5.1	8.48 ± 3.20	0.43
Insulin₆₀⁵	69.99 ± 44.72	57.21 ± 38.45	0.13	72.2 ± 46.35	64.2 ± 40.1	57.21 ± 38.45	0.25
Insulin₁₂₀⁶	45.44 ± 26.33	39.40 ± 22.70	0.28	46.1 ± 27.9	43.7 ± 22.3	39.40 ± 22.70	0.45
AUCglu⁷	7096 ± 1671	6589 ± 1735	0.14	6992 ± 1581	7364 ± 1933	6589 ± 1735	0.228
AUCins⁸	3827 ± 2096	3279 ± 2096		3820 ± 2061	3844 ± 2229	3279 ± 2096	0.41
HOMA-IR⁹	2.24 ± 1.23	1.73 ± 0.70	0.022	2.24 ± 1.23	2.26 ± 1.35	1.73 ± 0.70	0.07
HOMA-ßtblfn2¹⁰	33.82 ± 16.73	33.44 ± 14.29	0.90	33.9 ± 16.4	33.43 ± 17.89	33.44 ± 14.29	0.98
I Index¹¹	0.52 ± 0.22	0.49 ± 0.26	0.42	0.54 ± 0.238	0.51 ± 0.20	0.49 ± 0.26	0.58
GLP-1₀¹²	2.83 ± 0.41	2.71 ± 0.73	0.39	2.77 ± 0.39	3.01 ± 0.41	2.71 ± 0.73	0.21
GLP-1₁₂₀¹³	3.23 ± 0.54	3.05 ± 0.80	0.28	3.18 ± 0.56	3.38 ± 0.49	3.05 ± 0.80	0.29
ΔGLP-1¹⁴	0.40 ± 0.33	0.34 ± 0.57	0.62	0.41 ± 0.34	0.36 ± 0.29	0.34 ± 0.57	0.79
NPYtblfn2¹⁵	3.22 ± 0.52	2.18 ± 1.08	0.000	3.19 ± 0.54	3.31 ± 0.46	2.18 ± 1.08	0.000

Glucose₀: Fasting plasma glucose (mg/dl); ²Glucose₆₀: First hour plasma glucose (mg/dl);³Glucose₁₂₀: Second hour plasma glucose (mg/dl); ⁴Insulin₀: Fasting plasma insulin (μIU/ml); ⁵Insulin₆₀: First hour plasma insulin (μIU/ml); ⁶Insulin₁₂₀: Second hour plasma insulin (μIU/ml); ⁷AUCglu: Area under the curve for glucose (mg/dl/min); ⁸AUCins: Area under the curve for insulin (μIU/ml/min); ⁹HOMA IR: Homeostasis model assessment of insulin resistance; ¹⁰HOMA-ß: Homeostasis model assessment of ß-cell function; ¹¹I index: Insulinogenic Index; ¹²GLP-1₀: Fasting glucagon-like peptide-1 (pg/mL); ¹³GLP-1₁₂₀: Second hour glucagon-like peptide-1 (pg/mL); ¹⁴ΔGLP-1: Increment in glucagon-like peptide-1 (pg/mL); ¹⁵NPY: Neuropeptide Y (ng/ml). Data were expressed as mean ± SD.

Insulin sensitivity and β-cell functions

Analyses related to IR and β-cell functions are summarized in Table 2. HOMA-IR levels were significantly higher in migraineurs than in controls (2.2 ± 1.2 and 1.7 ± 0.7, p = 0.022). There was a slight, but not significant, difference when we performed ANOVA to compare EM, CM, and control subjects (p = 0.07). Patients with CM had higher HOMA-IR in comparison to EM (2.2 ± 1.3 and 2.1 ± 1.0, respectively, p = 0.048). The HOMA-β index and the insulinogenic index (II) were comparable between EM, CM, and control subjects (Table 2).

GLP-1

GLP-1 levels both at fasting and two hours after glucose intake were similar between CM, EM, and control subjects. The increment in GLP-1 levels (ΔGLP-1) also did not show any significant difference between migraine subgroups.

NPY

Fasting NPY levels were significantly higher in migraineurs than in controls (3.2 ± 0.5 and 2.1 ± 1.0 ng/ml; p < 0.001) (Table 2). This difference remained significant between controls and each individual migraine group as well (EM: 3.1 ± 0.5 ng/ml, CM: 3.3 ± 0.4 ng/ml, and controls: 2.1 ± 1.0 ng/m, p < 0.001).

We compared the study parameters between migraine patients with and without increased IR using a HOMA-IR cut off level of 2.37 (75th percentile). The results are presented in Table 3. AUC values for glucose and insulin were significantly higher in insulin-resistant migraineurs. The HOMA-β (25.3 ± 7.9 vs. 53.2 ± 15.9; p < 0.001) and insulinogenic indexes (0.4 ± 0.1 vs. 0.6 ± 0.2 in controls, p = 0.001) were significantly higher in insulin-resistant migraine patients (HOMA-IR ≥ 75^th percentile). However, we did not observe any difference in GLP-1 or NPY levels between groups (Table 3). NPY was higher in insulin-resistant CM patients compared to those with normal insulin sensitivity, but the difference was not statistically significant (3.2 ± 0.4 vs. 3.5 ± 0.5, p = 0.21).

Table 3.

Results in migraine patients according to HOMA-IR levels.

	Groups
	Migraine with HOMA-IR ≥ 75^thpercentile; n = 59	Migraine with HOMA-IR < 75^th percentile; n = 23	p-value
AUCglu¹	6841 ± 1513	7600 ± 1889	0.09
AUCins²	3194 ± 1522	5242 ± 2559	0.001
HOMA-ß³	25.37 ± 7.91	53.25 ± 15.90	<0.001
I Index⁴	0.46 ± 0.18	0.67 ± 0.26	0.001
GLP-1₀⁵	2.79 ± 0.44	2.91 ± 0.31	0.25
GLP-1₁₂₀⁶	3.20 ± 0.63	3.31 ± 0.34	0.41
ΔGLP-1⁷	0.40 ± 0.36	0.39 ± 0.25	0.88
NPY⁸	3.22 ± 0.52	2.18 ± 1.08	0.6

AUC³glu: Area under the curve for glucose (mg/dl/min); ²AUCins: Area under the curve for insulin (μIU/ml/min); ³HOMA-ß: Homeostasis model assessment of ß-cell function; ⁴I Index: Insulinogenic Index; ⁵GLP-1₀: Fasting glucagon-like peptide-1 (pg/mL); ⁶GLP-1₁₂₀: Second hour glucagon-like peptide-1 (pg/mL); ⁷ΔGLP-1: Increment in glucagon-like peptide-1 (pg/mL); ⁸NPY: Neuropeptide Y (ng/ml). Data were expressed as mean ± SD.

Correlation analyses

To study possible correlations, we performed linear regression analyses. In migraineurs, there was significant positive correlation between BMI and insulin, glucose, and HOMA-IR, HOMA-β and insulinogenic indexes.

After controlling for BMI, we documented a negative correlation of AUC glucose with GLP-1₁₂₀ (r = −0.27, p = 0.05) but there was no correlation with ΔGLP-1 (r = −0.25, p = 0.07). HOMA-β had a positive correlation with HOMA-IR (r = 0.84, p < 0.001) and ΔGLP-1 (r = −0.32, p = 0.03). NPY levels did not show any significant correlation with parameters related to IR or β-cell function. When we performed linear regression analyses specifically in CM patients, there was a slight positive correlation between NPY and ΔGLP-1 levels (r = 0.57, p = 0.04), but the correlation between NPY and HOMA-IR (r = 0.49, p = 0.09) and HOMA-β (r = 0.50, p = 0.07) were not significant.

Discussion

In this cross-sectional, controlled study, we investigated characteristics of glucose-stimulated insulin and GLP-1 secretion to clarify future diabetes risk in non-obese female migraine patients. We also evaluated the effects of orexigenic NPY on ß-cell functions and insulin sensitivity in study subjects. We did not observe any significant difference in dynamic insulin or GLP-1 secretion between the groups. Chronic migraineurs, however, were more insulin resistant compared to patients with EM or control subjects. NPY levels were significantly higher in migraineurs compared to controls, and we observed a slight significant correlation between NPY and ΔGLP-1 levels in CM. Data about the prevalence of T2DM in migraine are controversial. A large cross-sectional population-based study documented a moderate increase of diabetes frequency in subjects with migraine compared with headache-free controls (12.6% vs. 9.4%; OR = 1.4, 95% CI: 1.2–1.4) (16). The prevalence of diabetes did not differ between episodic and chronic migraine (16,17). Another analysis using data from the Women’s Health Study could not evaluate the association between migraine and T2DM in middle-aged and older female subjects (18).

In our study, subjects were all non-obese and comparable in terms of diabetes risk factors consistent with the general population. Indeed, the prevalence of prediabetes was 10.9%, which is similar to the prevalence in the general population of Turkey (35). Fasting plasma glucose (FPG) was significantly higher compared to healthy controls, but HbA1c and all other glucose levels measured in the OGTT were similar in the two groups. Although HbA1c is a reliable indicator of long term glycemic control in diabetics, it has low sensitivity in detecting variations in FPG, and small changes in FPG levels observed in our non-diabetic subjects may not be expressed by HbA1c. This observation may explain the similarity of HbA1c between the groups despite the difference in FPG (36).

There are a number of studies evaluating IR in migraine, and the results are different between subjects with EM and CM. Guldiken et al. found a significant association between IR and migraine in non-obese patients using the HOMA index (37). Favaa et al. reported that women with CM were three times more likely than those with EM to have IR (15). Our patients with CM also had higher HOMA-IR in comparison to EM (2.2 ± 1.3 and 2.1 ± 1.0, respectively, p = 0.048). In subjects with increased IR, β-cell function is the major determinant of progression from normal glucose tolerance to T2DM (38). In patients with diabetes, insulin secretion is insufficient to compensate for increased IR (19), and an individual has already lost approximately 80% of their β-cell function by the time a diagnosis of T2DM is made (39).

For the estimation of β-cell function, we used the HOMA-β and insulinogenic indexes. The HOMA-β index has been proposed to be a good measure of β-cell function, and several prospective studies have shown either HOMA-IR or HOMA-β may predict future risk of T2DM in diverse populations (40). Another parameter was the insulinogenic index, which is an OGTT-based measure examined by Stadler et al. for evaluation of β-cell secretory activity. Stadler calculated the insulinogenic index (II) as the ratio between the AUCs for C-peptide and glucose levels (33). C-peptide and insulin are released into circulation from β-cells in equimolar amounts, and use of plasma C-peptide as an index of β-cell function depends on the assumption that the mean clearance rate of C-peptide is constant. However, the plasma half-life of the molecule is long and does not reflect short-term changes in plasma insulin levels. Therefore, we used OGTT-derived insulin levels instead of C-peptide to calculate the II (41).

There is no clear threshold for the HOMA-β or II in the definition of impaired insulin secretion. We observed significantly higher HOMA-β and insulinogenic indexes in migraine patients with IR. After controlling for BMI, we documented a significant correlation of HOMA-β with HOMA-IR (r = 0.84, p < 0.001). This led us to speculate that although our patients with migraine have higher IR compared to controls, their β-cell function was normal in its ability to compensate for increased insulin demand in fasting as well as after glucose intake, as assessed by the HOMA-β and insulinogenic indexes (13,42).

Impaired GLP-1 response to meal intake is another pathogenetic defect of T2DM (25). A limited number of studies have evaluated fasting GLP-1 levels in migraine, but no data exist concerning the interaction of postprandial GLP-1 secretion, IR, and β-cell functions. Bernecker et al. (43) investigated fasting leptin, adiponectin, resistin, GLP-1 and GLP-2 levels in non-obese female migraineurs and controls. Their patients were more insulin resistant and had higher GLP-2 levels compared to controls, but fasting GLP-1 levels did not show any significant differences. The study included women between 18–64 years of age with a HOMA index of 1.8 ± 1.6. To eliminate the effect of sex hormones on diabetes risk, we only included premenopausal patients, and we studied not only fasting but also glucose-stimulated GLP-1 levels to evaluate the role of postprandial GLP-1 secretion. Our patients were more insulin resistant than the study subjects of Bernecker et al. (HOMA-IR 2.2 ± 1.2), but GLP-1 levels in fasting and after glucose intake were similar in EM, CM, and controls in relation to IR. Plasma glucose is a potent stimulator of GLP-1 secretion (25).

We observed a positive correlation between NPY and ΔGLP-1 levels (r = 0.57, p = 0.04), but we observed no significant correlation between NPY and HOMA-IR (r = 0.49, p = 0.09) and HOMA-β (r = 0.50, p = 0.07)in CM. The significantly higher FPG in migraineurs led to a slight but not significant increase of ΔGLP-1 (Table 2), which could be a compensatory mechanism to keep postprandial glucose levels in the normal range. The positive correlation between NPY and ΔGLP-1 levels may also reflect the metabolic adaptive behavior to increased NPY secretion and/or IR.

NPY is a potent orexigenic peptide present in hypothalamic nuclei, basal ganglia, the limbic system, and the peripheral nervous system (10), and has a stimulating role in food intake. Both insulin and NPY have pain modulating effects and antinociceptive actions in animal models (9,10). According to experimental data, the appetite stimulating effect of NPY may play a role in the development of human obesity and hypertension (11). In our study, NPY levels were significantly higher in both EM and CM subjects than controls, and, in CM subjects only, we observed a slight but not significant positive correlation between HOMA-IR (r = 0.49, p = 0.09) and NPY. IR has a documented association with increased NPY levels (44,45). Migraine-associated activation of the sympathoadrenal system may also cause NPY elevation (10). NPY is widely distributed throughout the sympathetic nerve endings, where it is co-stored and co-secreted with noradrenaline (46). Goadsby et al. highlighted the role of NPY- and noradrenaline-containing sympathetic nerve fibers in cerebral circulation during migraine headache (47). In another study, Gallai and Sarchielli reported that plasma NPY levels tended to increase during attacks in migraineurs with aura, and this was regarded as an expression of sympathetic activation (48). In animal models, it has been shown that chronically-increased NPY secretion induces hyperphagia and leads to increased weight gain (6,10,49).

The limitation of our study was the small sample size. Future studies should include a bigger sample size and evaluate NPY-associated alterations in appetite and caloric intake in our patients. To clarify the possible relationship of increased NPY with attack frequency in our patients, we must further study NPY secretion during migraine attacks.

The complex orexigenic network is also modulated by peripheral signals such as insulin, and increased fasting insulin levels in CM may play a role in the alterations of food intake by acting as an orexigenic hormone on hypothalamic ion channels (15,37,42,50). However, as a limitation of study, we did not evaluate alterations of food intake in relation to frequency of migraine attacks.

Insulin is also an important regulator of brain glucose metabolism, and hypoglycemia following prolonged fasting may trigger migraine attacks in CM. Dalkara and Kilic wrote a review on brain glycogen metabolism during migraine (51). Plasma glucose is stored as glycogen in astrocytes and is rapidly metabolized for uptake of glutamate and potassium during intense synaptic activity in headache attacks. The authors suggested that during long lasting fasting, extended periods of low blood glucose and prolonged sympathetic activity may reduce available glycogen-derived glucose in perisynaptic astrocytes, and this may trigger aura and/or headache (51). We hypothesize that peripheral IR in fasting and increased NPY levels in CM may be protective mechanisms to maintain plasma glucose levels and brain glycogen levels during migraine attacks.

In conclusion, the major metabolic alteration in our non-obese, non-diabetic female migraine patients was increased IR. We did not observe any pathology in β-cell functions or GLP-1 secretion, which are major determinants of future diabetes risk. In light of the association between NPY and IR, increased NPY levels suggest a role for the hypothalamus in the impairment of insulin sensitivity in migraine. Another explanation may be the antinociceptive actions of NPY and increased demand in CM with frequent headache attacks. Migraine-associated activation of the sympathoadrenal system may also cause NPY elevation.

However, we cannot provide enough evidence to establish that NPY is a cause of IR rather than a consequence of it, and further studies are needed to clarify this association (52,53). IR is linked to a number of diseases including diabetes, hypertension, obesity, dyslipidemia, coronary artery disease and stroke, all of which are risk factors for migraine (54), and it is clinically relevant to clarify the pathophysiology of IR to aid in the search for preventive strategies. This study was mainly focused on the metabolic aspects of migraine, but we were not able to provide data on NPY- or GLP-1-related appetite alterations, food preferences or total daily caloric intake. We performed a large number of statistical analyses on a rather small group of participants without correcting for multiple comparisons. Therefore, these preliminary findings would need to be confirmed in a further study with a larger number of patients with CM. This future study should include nutrition data to provide evidence for the role of hypothalamus-associated metabolic alterations. The challenge at present is to understand the complexity of these pathways and try to find ways of modulating them in order to find therapeutic targets.

Clinical implications

Non-obese pre-menopausal female patients have higher insulin resistance, but their β-cell function and glucose stimulated GLP-1 secretions were normal and there is no increased risk for future type 2 diabetes.

The association between NPY and insulin resistance in CM suggest a role for the hypothalamus in the impairment of insulin sensitivity in migraine. Increased secretion of NPY by the hypothalamus may lead to alterations in food intake and impairment in insulin sensitivity in migraineurs.

Peripheral insulin resistance in fasting and increased NPY levels in CM may be protective mechanisms maintaining plasma glucose levels and brain glycogen levels during migraine attacks.

IR is linked to a number of diseases including diabetes, hypertension, obesity, dyslipidemia, coronary artery disease and stroke, all of which are risk factors for migraine, and it is necessary to clarify the pathophysiology of IR and the role of NPY to aid in the search for preventive strategies.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Charles

. The evolution of a migraine attack – a review of recent evidence. Headache 2013; 53: 413–419.

Headache Classification Committee of the International Headache Society (IHS). The international classification of headache disorders, 3rd edn (beta version). Cephalalgia 2013; 33: 629–808.

Denuelle

Fabre

Payoux

et al.

Hypothalamic activation in spontaneous migraine attacks. Headache 2007; 47: 1418–1426.

Sprenger

Maniyar

Monteith

et al.

Midbrain activation in the premonitory phase of migraine: A H2O-positron emission tomography study. Headache 2012; 52: 863–864.

Simpson

Martin

Bloom

. Hypothalamic regulation of appetite. Expert Rev Endocrinol Metab 2008; 3: 577–592.

Kalra

Dube

Sahu

et al.

Neuropeptide Y secretion increases in the paraventricular nucleus in association with increased appetite for food. Proc Natl Acad Sci 1991; 88: 10931–10935.

Stanley

Kyrkouli

Lampert

et al.

Neuropeptide Y chronically injected into the hypothalamus: A powerful neurochemical inducer of hyperphagia and obesity. Peptides 1986; 7: 1189–1192.

Harris

Wimmer

Aston-Jones

. A role for lateral hypothalamic orexin neurons in reward seeking. Nature 2005; 437: 556–559.

Rajendran

Thirugnanasambandam

Parvathavarthini

et al.

Modulation by insulin rather than blood glucose of the pain threshold in acute physiological and flavone induced antinociception in mice. Indian J Exp Biol 2001; 39: 1009–1016.

10.

Oliveira

Akerman

Tavares

et al.

Neuropeptide Y inhibits the trigeminovascular pathway through NPY Y1 receptor: Implications for migraine. Pain 2016; 157: 1666–1673.

11.

Baltatzi

Hatzitolios

Tziomalos

et al.

Neuropeptide Y and alpha-melanocyte-stimulating hormone: Interaction in obesity and possible role in the development of hypertension. Int J Clin Pract 2008; 62: 1432–1440.

12.

Casucci

Villani

Cologno

et al.

Migraine and metabolism. Neurol Sci 2012; 33: S81–S85.

13.

Rainero

Limone

Ferrero

et al.

Insulin sensitivity is impaired in patients with migraine. Cephalalgia 2005; 25: 593–597.

14.

Sachdev

Marmura

. Metabolic syndrome and migraine. Front Neurol 2012; 3: 1–8.

15.

Favaa

Pirritanob

Consolic

et al.

Chronic migraine in women is associated with insulin resistance: A cross-sectional study. Eur J Neurol 2014; 21: 267–272.

16.

Bigal

Kurth

Santanello

et al.

Migraine and cardiovascular disease: A population-based study. Neurology 2010; 74: 628–635.

17.

Buse

Manack

Serrano

et al.

Sociodemographic and comorbidity profiles of chronic migraine and episodic migraine sufferers. J Neurol Neurosurg Psychiatry 2010; 81: 428–832.

18.

Burch

Rist

Winter

et al.

Migraine and risk of incident diabetes in women: Prospective study. Cephalalgia 2012; 32: 991–997.

19.

DeFronzo

. The triumvirate: Beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 1988; 37: 667–687.

20.

Etminan

Takkouche

Isorna

et al.

Risk of ischaemic stroke in people with migraine: Systematic review and meta-analysis of observational studies. BMJ 2005; 330: 63–66.

21.

Scher

Terwindt

Picavet

et al.

Cardiovascular risk factors and migraine: The GEM population-based study. Neurology 2005; 64: 614–620.

22.

Merikangas

Fenton

Cheng

et al.

Association between migraine and stroke in a large-scale epidemiological study of the United States. Arch Neurol 1997; 54: 362–368.

23.

Hamed

Ezz Eldin

et al.

Vascular risk factors, endothelial function, and carotid thickness in patients with migraine: Relationship to atherosclerosis. J Stroke Cereb Dis 2010; 19: 92–103.

24.

DeFronzo

Eldor

Abdul-Ghani

. Pathophysiologic approach to therapy in patients with newly diagnosed type 2 diabetes. Diabetes Care 2013; 36: 127–138.

25.

Nauck

Vardarli

Deacon

et al.

Secretion of glucagon-like peptide-1 (GLP-1) in type 2 diabetes: What is up, what is down?

Diabetologia 2011; 54: 10–18.

26.

Burns

Bacha

Jung Lee

et al.

Declining β-cell function relative to insulin sensitivity with escalating OGTT 2-h glucose concentrations in the nondiabetic through the diabetic range in overweight youth. Diabetes Care 2011; 34: 2033–2040.

27.

Tfayli

Lee

Arslanian

. Declining b-cell function relative to insulin sensitivity with increasing fasting glucose levels in the nondiabetic range in children. Diabetes Care 2010; 33: 2024–2030.

28.

Turan

Siva

Uluduz

et al.

The impact of depression and ghrelin on body weight in migraineurs. J Headache Pain 2014; 15: 15–23.

29.

Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. Cephalalgia 2004; 24: 9–160.

30.

American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2009; 32: 62–67.

31.

American Diabetes Association. Clinical Practice Recommendations. Diagnosis and classification of diabetes mellitus. Diabetes Care 2010; 33: 62–69.

32.

Matthews

Hosker

Rudenski

et al.

Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419.

33.

Stadler

Pacini

Petrie

et al.

B cell (dys)function in non-diabetic offspring of diabetic patients. Diabetologia 2009; 52: 2435–2444.

34.

Wallace

Levy

Matthews

. Use and abuse of HOMA modeling. Diabetes Care 2004; 27: 1487–1495.

35.

Satman

Omer

Tutuncu

et al.

Twelve year trend in the prevalence and risk factors of diabetes and prediabetes in Turkish adults. Eur J Epidemiol 2013; 28: 169–180.

36.

Bonora

Tuomilehto

. The pros and cons of diagnosing diabetes with A1c. Diabetes Care 2011; 34: S5184–S5209.

37.

Guldiken

Demir

et al.

Insulin resistance and high sensitivity C-reactive protein in migraine. Can J Neurol Sci 2008; 35: 448–451.

38.

Mitrakou

Kelly

Mokan

et al.

Role of reduced suppression of glucose production and diminished early insulin release in impaired glucose tolerance. N Engl J Med 1992; 326: 22–29.

39.

DeFronzo

Eldor

Abdul-Ghani

. Pathophysiologic approach to therapy in patients with newly diagnosed type 2 diabetes. Diabetes Care 2013; 36: 127–138.

40.

Bonora

Kiechl

Willeit

et al.

Population-based incidence rates and risk factors for type 2 diabetes in white individuals: The Bruneck study. Diabetes 2004; 53: 1782–1789.

41.

Shapiro

Tillil

Rubenstein

et al.

Peripheral insulin parallels changes in insulin secretion more closely than C-peptide after bolus intravenous glucose administration. J Clin Endocrinol Metab 1988; 67: 1094–1099.

42.

Cavestro

Rosatello

Micca

et al.

Insulin metabolism is altered in migraineurs: A new pathogenic mechanism for migraine?

Headache 2007; 47: 1436–1442.

43.

Bernecker

Pailer

Kieslinger

et al.

GLP-2 and leptin are associated with hyperinsulinemia in non-obese female migraineurs. Cephalalgia 2010; 30: 1366–1374.

44.

Baranowska

Wolinska-Witort

Wasilewska-Dziubinska

et al.

The role of neuropeptides in the disturbed control of appetite and hormone secretion in eating disorders. Neuro Endocrinol Lett 2003; 24: 431–434.

45.

Jaakkola

Kakko

Seppala

et al.

The Leu7Pro polymorphism of the signal peptide of neuropeptide Y (NPY) gene is associated with increased levels of inflammatory markers preceding vascular complications in patients with type 2 diabetes. Microvasc Res 2010; 80: 433–439.

46.

Morris

Russell

Kapoor

et al.

Increases in plasma neuropeptide Y concentrations during sympathetic activation in man. J Auton Nerv Sys 1986; 17: 143–149.

47.

Goadsby

Edvinsson

. Neuropeptides in migraine and cluster headache. Cephalalgia 1994; 14: 320–327.

48.

Gallai

Sarchielli

Trequattrini

et al.

Neuropeptide Y in juvenile migraine and tension type headache. Headache 1994; 34: 35–40.

49.

Baltatzi

Hatzitolios

Tziomalos

et al.

Neuropeptide Y and alpha-melanocyte-stimulating hormone: Interaction in obesity and possible role in the development of hypertension. Int J Clin Pract 2008; 62: 1432–1440.

50.

Porte

Baskin

Schwartz

. Insulin signaling in the central nervous system. Diabetes 2005; 54: 1264–1276.

51.

Dalkara

Kılıç

. How does fasting trigger migraine? A hypothesis. Curr Pain Headache Rep 2013; 17: 368–374.

52.

Overem

van Vliet

Lammers

et al.

The hypothalamus in episodic brain disorders. Lancet Neurol 2002; 1: 437–444.

53.

Peres

Sanchez del Rio

Seabra

et al.

Hypothalamic involvement in chronic migraine. J Neurol Neurosurg Psychiatry 2001; 71: 747–751.

54.

Sacco

Pistoia

Degan

et al.

Conventional vascular risk factors: Their role in the association between migraine and cardiovascular diseases. Cephalalgia 2015; 35: 146–164.

Determinants of glucose metabolism and the role of NPY in the progression of insulin resistance in chronic migraine

Abstract

Background

Methods

Results

Discussion

Keywords

Introduction

Methods

Insulin sensitivity

β-cell function

Statistical analysis

Results

Glucose metabolism

Insulin sensitivity and β-cell functions

GLP-1

NPY

Correlation analyses

Discussion

Clinical implications

Footnotes

Declaration of conflicting interests

Funding

References