In our study we included twenty healthy subjects (9 females aged 37.0 ± 2.8 years, 11 males aged 41.8 ± 7.6 years) and 20 patients with migraine (15 females aged 41.9 ± 9.9 years, 5 males aged 38.2 ± 9.2 years). The number of the participants in study was driven from previous studies (4, 5, 6). We did not find significant differences in sex (p = 0.105) nor in age (p = 0.066) between the groups. The inclusion criteria for the group of healthy participants were negative family history for migraine and age 18 or more. The inclusion criteria for the migraine group were migraine with or without aura in accordance to the ICHD-3 criteria of the International Headache Society (9) and age 18 or more. The exclusion criteria for healthy group and migraine group were presence of cerebrovascular, renal, or liver diseases and uncontrolled endocrine diseases, uncontrolled hypertension, pregnancy, breast-feeding, abnormalities of somatic and neurological status and hemodynamic important atherosclerotic process of carotid or vertebral arteries.
The participants were free of tobacco, coffee, tea or any other food or beverages containing caffeine for at least 12 hours before the start of the measurements.
All participants were given written explanation regarding the experimental procedure. They all gave written consents to participate in the study. The study was approved by the National Medical Ethics Committee of the Republic of Slovenia.
Before the beginning of the experiment color-coded duplex sonography of the carotid and vertebral arteries was performed using standard procedure. The experiments took place at 9:00 am in a quiet and darkened room under constant conditions. During the experiment participants were resting in the supine position. The experiment lasted 40 min including a 10 min baseline period, a 20 min period during which an intravenous infusion of hαCGRP 1.5 mcg/min (Calbiochem, Merck4Biosciences, Darmstadt, Germany) was given and a 10 min period after the application of hαCGRP. The hαCGRP dose of 1.5mcg/min was chosen due to results of previous studies, which showed that it is safe and caused no profound hypotension.Transcranial Doppler (TCD) sonography with 2 MHz ultrasound probe was applied to measure mean flow velocity in PCA (vm PCA) through the right transtemporal acoustic window. The signal of the PCA was defined according to direction of the blood flow, typical depth of the signal and the response to closing eyes. A mechanical probe holder was used to ensure a constant probe position. During the entire experiment mean blood pressure (MAP) and heart rate (HR) were continuously measured using non-invasive plethysmography (Colin 7000, 12 Komaki-City, Japan). End –tidal carbon dioxide (Et-CO2) was measured in exhaled breath using a ventilation mask and an infrared capnometer (Capnograph, Model 9004, Smith Medical, USA).
TCD Multi-Dop X4 software (DWL, Sipplingen, Germany) was used to define and average integrals of 5 minutes values for vm PCA, MAP, HR and Et-CO2. TCD software enabled us to We calculated an average integral of 5 min values using the following equation for vm PCA: vmPCA ═ ∫ vdt/ (t0min – t5min). The same equation was used to calculate an average integral for the rest of variables (MAP, HR, Et-CO2).
For further statistical analysis we defined data of 5 min intervals as measuring points T0, T1, T2 and T3. Measurement point T0 represented an interval during baseline period before starting hαCGRP infusion (5–10 min of the experiment), T1 was 5 minutes interval in the first part of hαCGRP infusion (15–20 of the experiment), T2 represented 5 minutes interval in the last part of hαCGRP infusion (25–30 minute of the experiment) and T3 was the 5 minutes interval in the last part of the experiment after the end of hαCGRP infusion (35–40 min of the experiment).
In the next step we calculated the responses for vm PCA, Et-CO2, HR and MAP respectively as differences between measuring points. The response 1 represented the difference between measurement points T1 and T0, the response 2 represented the difference between measurement points T2 and T0 and the response 3 represented the difference between measurement points T3 and T0.
After that we performed statistical analysis of all three responses as one variable (vm PCAtot, Et-CO2tot, HRtot and MAPtot). Our assumption based on CGRP pharmacokinetic studies (10, 11) which confirmed that CGRP was active in time period of T1, T2 and T3 measurement points.
We determinate CGRP-IH by using ICHD-3 criteria (9). We didn’t calculated frequency of immediate CGRP-IH and delayed CGRP-IH separately but together as one variable.
For statistical analysis SPSS version 21 was used. Paired t-test and Student t-test was used to test the significance of differences between dependent and independent variables. Linear regression and logistic regressions were used to test the correlations between the variables. Normality of variability distribution was tested and all variables had value of the Shapiro-Wilk Test greater than 0.05. The results in the statistic tests were statistically significant if p < 0.005.