PROSTAGLANDIN-E2 LEVELS OVER THE COURSE OF GLYCERYL TRINITRATE PROVOKED MIGRAINE ATTACKS 101 5 (Phenomenex, 50 × 2.1 mm, 1.7 µm). A gradient of 0.01% acetic acid in water (A) and 0.01% acetic acid in MeOH (B) was used to elute the components of interest from the column. The total flow rate was 400 mL/min. The column oven was set to 50°C. The mass spectrometer (MS) was equipped with an ESI source and operated in negative scheduled MRM mode.The needle voltage was set to -4,500 V, the drying temperature to 450°C, ion source gas 1/nebulizer gas (air) at 40 psi, ion source gas 2/drying gas (air) at 30 psi and the nebulizer gas (nitrogen) at 30 psi. For PGE2 the transition used was 351/271, for PGE2-d4 355/193. PGE2 was identified based on its tandem MS transition and relative retention time and, quantified using external calibration. Statistical analysis We aimed to investigate the role of PGE2 over the course of a provoked migraine attack, healthy controls were included to ensure that direct pharmacological effects of the provocation substance itself is not incorrectly labelled as a marker for provoked attacks. As we were primarily interested in the effect of different phases on PGE2 levels in blood, we distinguished three phases: interictal (outside a migraine-like attack), preictal (before a migraine-like headache of which the onset is ≤ 12 hours after GTN infusion), and ictal (migraine-like headache). To account for repeated measurements within each subject, we used a linear mixed model with a random effect per person and unstructured correlation, the same model was used previously.36 The outcome (dependent variable) was the measured PGE2 concentration. Predictors (independent variables) were age, diagnosis (migraine or control), time point (T0, T1, T2) and migraine phase (interictal, preictal, ictal). Controls were coded as “interictal” at all time points. Furthermore, we added the interaction between time point and diagnosis to account for subjects with migraine possibly reacting differently to GTN than controls, irrespective of migraine phase. Statistical analyses were performed using SPSS (version 25.0, IBM SPSS Statistics for Windows, IBM Corp, Armonk, NY). In this study, data was collected as part of an extensive larger study and, therefore, no a priori power calculations were performed for this sub-study. Results Clinical characteristics We initially included n = 37 participants with migraine and n = 25 healthy controls, of which five participants were excluded for further analyses. Two cases were removed as GTN infusion was not performed, both participants withdrew from participation after the baseline measurement. Two cases were excluded, because we were unable to classify the provoked headache attack (one not fully fulfilling a migraine-like headache nor classifying as a non-responder and the other developed a migraine-like attack, but already proceeded to a postdrome state during the study day). One healthy control was excluded due to a (first) provoked migraine-like headache. In total, data from n = 33 participants with migraine and n = 24 healthy controls were included in the
RkJQdWJsaXNoZXIy MTk4NDMw