Is there a hypothalamic–pituitary–adrenal axis dysfunction in patients with familial Mediterranean fever?

Abstract

Sirs, I read with great interest the article by Tansu et al. entitled “Adrenal axis functions in patients with familial Mediterranean fever” [1]. They reported that the response of cortisol to adrenocorticotropic hormone (ACTH) stimulation in familial Mediterranean fever (FMF) patients is significantly higher in the attack period when compared to that in the remission period. However, they reported that they could find no difference in terms of cortisol response in comparison with normal individuals. Based on these results, they assumed that hypothalamic–pituitary–adrenal (HPA) axis is regulated normally in FMF patients. As is well known, FMF attacks are characterized by increased cytokines stimulation, such as interleukin (IL)-1, IL-6, and tumor necrosis factor alpha (TNF-α), along with brisk acute phase response [2]. FMF attacks are known to be painful and stressful. A stressful condition alone may stimulate cytokines [3]. Physiological stimuli and stressors can normally stimulate the HPA axis to produce its main glucocorticoid products. The proinflammatory cytokines, e.g., IL-1, IL-6, and tumor necrosis factor-alpha, are soluble products of the activated immune system, which can also stimulate the HPA axis [3]. Considering the effect of both cytokine stimulation during the attack period and exogenous administration of ACTH, significantly high cortisol response would theoretically be expected in FMF patients with attack. However, Tansu et al. could find no difference between patients stricken with FMF attacks and normal individuals in terms of cortisol response. This result may show the presence of defective HPA axis in FMF patients, which is in disagreement with the comments of the study by Tansu et al. I believe these authors can refer to some rheumatoid arthritis (RA) studies in terms of HPA axis that can help them better comment on their own study results. RA patients are characterized by normal or subnormal cortisol response to ACTH and corticotropinreleasing hormone (CRH) in spite of increased IL-6 levels and TNF-α [4, 5]. Based on these results, the existence of a defective HPA axis in RA patients was considered [4, 5]. We have shown early blunted cortisol response to insulin-induced hypoglycaemia (IIH) in FMF patients before [6]. In our study, the patients' cortisol levels to IIH were not different from those of normal control individuals in attackfree periods, which is in agreement with the results of Tansu et al. However, unlike theirs, our interval-specific delta analysis revealed a lower cortisol response to IIH in FMF patients, which was significant for the 0to 30-min interval of hypoglycaemia stimulus [6]. This may at least suggest that cortisol response during attack-free periods in FMF is blunted. Although there are not enough studies in the literature into the behavior of HPA axis in FMF patients for us to obtain solid evidence as yet, the results of these few studies could lead us to think of the presence of a defective HPA axis in FMF patients. It is a fact that further studies are needed to clarify this curious point.