Serotonin and Neuropeptides in Blood From Episodic and Chronic Migraine and Cluster Headache Patients in Case-Control and Case-Crossover Settings: A Systematic Review and Meta-Analysis

OBJECTIVE: The aim of this systematic review and meta-analysis (SR-MA) was to identify signaling molecule profiles and blood-derived biomarkers in migraine and cluster headache (CH) patients.

BACKGROUND: Currently no migraine and CH valid biomarkers are available. Blood tests based on biomarker profiles have been used to gather information about the nervous system. Such tests have not yet been established within the primary headache field.

METHODS: Case-control and case-crossover studies investigating whole blood, plasma, and serum were identified worldwide. The qualitative synthesis focused on 9 signaling molecules (serotonin [5-HT], calcitonin gene-related peptide [CGRP], endothelin-1 [ET-1], neurokinin A, neurokinin B, neuropeptide Y, pituitary adenylate cyclase-activating peptide 38 [PACAP-38], substance P (SP), and vasoactive intestinal peptide) and the quantitative synthesis on 5-HT and CGRP (≥5 comparisons available). The meta-analysis was conducted using standard and 3-level random effect models.

RESULTS: Fifty-four eligible studies were identified (87.0% migraine, 9.3% CH, 3.7% migraine, and CH), and 2768 headache patients and 1165 controls included. Comparable fluctuations of 5-HT, CGRP, ET-1, PACAP-38, and SP in blood were generally observed between migraine and CH. Significant findings were observed for some subgroups and strata, for example, higher interictal and ictal 5-HT venous blood levels (ratio of means = 1.32, 95% CI: 1.08; 1.61; ratio of means = 1.23, 95% CI: 1.01; 1.49) in episodic migraine with aura with a female-dominated case group, higher interictal CGRP blood levels in episodic migraine (ratio of means = 1.63, 95% CI: 1.18; 2.26), and chronic migraine (ratio of means = 1.89, 95% CI: 1.33; 2.68), and higher ictal CGRP blood levels (ratio of means = 1.35, 95% CI: 1.09; 1.68) in episodic migraine were observed. In most subgroups, the quantitative synthesis revealed a high degree of heterogeneity between studies in part explained by the blood sampling site, specimen source, blood specimen, and sex distribution. Other potential confounders were age, aura, study quality, menstrual cycle, and methodology (eg, storage temperature).

CONCLUSIONS: Potential migraine and CH signaling molecule profiles and biomarkers were revealed. Nevertheless, the high degree of heterogeneity between studies impedes identification of valid biomarkers but allowed us to assess the presence of confounders. Consideration of the potential confounders identified in this SR-MA might be of importance in the experimental planning of future studies. This consideration could be incorporated through establishment of specific guidelines.