Role of adipocytokines in the pathophysiology of migraine: A cross-sectional study

To the Editor:

We are writing with regard to the letter sent by Hyun Jin Min and Kyung Soo Kim (1) in connection with our recently published work in Cephalalgia, “Role of adipocytokines in the pathophysiology of migraine. A cross-sectional study” (2). First of all, we want to thank the authors for the attention and time devoted to the analysis of our work and all their valuable comments.

We acknowledge there were two minor transcription mistakes in our paper regarding mean age and prevalence of females in Table 1 and the acronym used for high-sensitivity C-reactive protein in Table 2.

In our study, there was no statistically significant difference in levels of leptin and adiponectin between patients suffering migraine with aura and migraine without aura. We did not expect to see a difference between these two groups, considering that we assume the same pathophysiology under both types of migraine and that this difference has not been previously evaluated or reported (3,4). Regarding the effect of BMI and hemodilution on adipocytokine levels, our results are already adjusted by BMI, age and sex by logistic regression analysis. Therefore, the possible effect of hemodilution on adypocitokine levels should be already controlled. Moreover, if hemodilution had an important effect on levels of leptin or adiponectin, levels of both molecules would be inversely correlated with BMI, and not positively correlated in the case of leptin and inversely correlated in the case of adiponectin. Finally, all the studies mentioned by the authors (5–7) refer to tumor markers and obesity; in our study we try to correlate adipocytokine levels with BMI, without comparing values in obese and non-obese subjects.

Last, levels of adiponectin in our sample are different in patients and controls, and also different when we analyze the subgroup of chronic migraineurs (CM) and the subgroup of episodic migraineurs (EM). Mean levels of adiponectin are lower in EM patients than in healthy controls, but this difference does not have statistical signification; therefore, we should assume these values as equal. Besides, values in EM in our sample are more scattered than in healthy controls if we look at standard deviation values. This variability suggests that there may be great clinical and physiopathological variations in the group of patients that we categorize as EM compared with the CM group and healthy controls. Previous studies also report very diverse levels of adipocytokine in migraineurs and controls: Bernecker et al. did not find statistically significant differences in ADP levels between migraineurs and controls(8,9); Peterlin et al. (10) analyzed results in EM and CM separately, but did not compare episodic migraineurs and controls; a later study by Duarte et al. (11) compared EM and CM with no significant differences, and more recently Dearborn et al. (9) also found higher levels in migraineurs, including CM and EM, but again did not compare levels in EM and controls. There is also a great variability in the levels of ADP reported in the mentioned studies, probably due to the different ADP multimers and its relation to inflammation, which make the study of this molecule and its role in migraine much more complex (12).