Network Analysis Reveals That Headache-Related, Psychological and Psycho-Physical Outcomes Represent Different Aspects in Women with Migraine

Abstract

Evidence supports that migraine is a complex pain condition with different underlying mechanisms. We aimed to quantify potential associations between demographic, migraine-related, and psychophysical and psychophysical variables in women with migraine. Demographic (age, height, and weight), migraine-related (intensity, frequency, and duration), related-disability (Migraine Disability Assessment Scale, Headache Disability Inventory), psychological (Hospital Anxiety and Depression Scale), and psycho-physical (pressure pain thresholds -PPTs-) variables were collected from a sample of 74 women suffering from migraine. We calculated adjusted correlations between the variables by using a network analysis. Additionally, we also calculated centrality indices to identify the connectivity among the variables within the network and the relevance of each variable in the network. Multiple positive correlations (ρ) between PPTs were observed ranging from 0.1654 (C5-C6 and tibialis anterior) to 0.40 (hand and temporalis muscle). The strongest associations within the network were those between migraine attack frequency and diagnosis of chronic migraine (ρ = 0.634) and between the HDI-E and HDI-P (ρ = 0.545). The node with the highest strength and betweenness centrality was PPT at the second metacarpal, whereas the node with the highest harmonic centrality was PPT at the tibialis anterior muscle. This is the first study applying a network analysis to understand the underlying mechanisms in migraine. The identified network revealed that a model where each subgroup of migraine-related, psychological, and psycho-physical variables showed no interaction between each variable. Current findings could have clinical implications for developing multimodal treatments targeting the identified mechanisms.

Keywords: disability; migraine; network analysis; pressure pain.

Figure 1

Network analysis of the association between demographic, migraine-related, psychological, and psycho–physical/neuro-physiological measures. Edges represent connections between two nodes and adjusted for the remaining nodes. Direction of the partial correlations were expressed as the color green for positive associations. The color Grey represents the association for categorical variables which no sign is defined. The thickness and color saturation of an edge denotes its weight (the strength of the association between two nodes).

Figure 2

Centrality values (1: maximal importance; 0: no importance) of Strength and Betweenness of each node in the network.

Figure 3

Harmonic Centrality measure (1: maximal importance; 0: no importance) of each node in the network.

Figure 4

Average correlations between centrality indices of networks sampled with persons dropped and networks built on the entire input dataset, at all follow-up time points. Lines indicate the means and areas indicate the range from the 2.5th quantile to the 97.5th quantile.