Pathophysiology and Treatment of Menopausal Hot Flashes

Robert R. Freedman, PhD


Semin Reprod Med. 2005;23(2):117-125. 

In This Article

Endocrinology of Hot Flashes

Given that hot flashes occur in most women with the estrogen withdrawal at natural and surgical menopause, there is little doubt that estrogens are involved in their initiation. This is supported by the fact that estrogen administration nearly eliminates hot flashes.[21] However, estrogen withdrawal alone does not explain the etiology of this symptom because there are no correlations between hot flash occurrence and plasma,[22] urinary,[23] or vaginal[23] levels of estrogens, nor are there differences in plasma levels between symptomatic and asymptomatic women.[19,24] Moreover, clonidine reduces hot flash frequency without changing circulating estrogen levels,[25] and prepubertal girls have low estrogen levels but no hot flashes. Thus, estrogen withdrawal is necessary but not sufficient to explain the occurrence of hot flashes.

The search for other factors began with the observation of a temporal correspondence between hot flashes and luteinizing hormone (LH) pulses.[26,27] However, subsequent research showed that women with hypothalamic amenorrhea had LH pulses but no hot flashes, and women with isolated gonadotropin deficiency had hot flashes but no LH pulses.[28] Furthermore, hot flashes occur in women with LH suppression from gonadotropin-releasing hormone analogs,[29,30] in women with pituitary insufficiency and hypoestrogenism,[31] and in hypophysectomized women, who have no LH pulses.[32] Thus, LH pulses are not the basis for hot flashes.

Next, an opioidergic system was postulated in the etiology of hot flashes. Lightman[33] found that naloxone infusion reduced hot flash and LH pulse frequencies in six symptomatic postmenopausal women. However, DeFazio et al[34] repeated these procedures and found no effects at all on objective or subjective hot flashes or on LH pulses. Tepper et al[35] found decreased plasma β-endorphin levels preceding hot flashes, whereas Genazzani et al[36] found the opposite effects. Thus, there is no consistent evidence of opiate involvement in menopausal hot flashes.

Norepinephrine (NE) plays a major role in thermoregulation, acting, in part, through α2-adrenergic receptors.[37] When injected into the preoptic hypothalamus, NE causes peripheral vasodilation and heat loss, followed by a decline in Tc. Furthermore, gonadal steroids modulate central NE activity.[38] Although plasma NE levels do not increase before or during hot flashes,[10,26] these do not reflect NE levels in brain.[39] We therefore measured plasma 3-methoxy-4-hydroxyphenylglycol (MHPG), the main NE metabolite, initially thought to derive primarily from the brain.[40] We obtained serial blood samples in 13 symptomatic and six asymptomatic postmenopausal women in thermoneutral and warm conditions. Baseline MHPG levels were significantly higher in the symptomatic women and increased significantly further during hot flashes. There were no hot flashes or significant MHPG changes in the asymptomatic women, and no differences in levels between the neutral and warm conditions.

Subsequent research showed that about half of plasma free MHPG is metabolized peripherally to vanillylmandelic acid (VMA), which competes with MHPG formation.[39,41] Thus, plasma MHPG levels could be affected by peripheral VMA. We therefore performed a second study in which we measured both compounds simultaneously before and after 29 hot flashes in 14 symptomatic women. We again showed that plasma MHPG increased significantly (3.7 ± 1.4 vs. 5.1 ± 2.3 μg/mL; p < 0.02) whereas VMA levels did not change (6.2 ± 1.8 vs. 6.1 ± 2.5 μg/mL). These data are consistent with the hypothesis that elevated brain NE is involved in the initiation of hot flashes.

This theory is supported by clinical studies showing that clonidine, an α2-adrenergic agonist that reduces brain NE, significantly reduces hot flash frequency.[42,43] In a controlled laboratory investigation, we then showed that yohimbine, an α2-adrenergic antagonist that elevates brain NE,[44] provoked hot flashes in symptomatic women, whereas clonidine ameliorated them.[45] These data support the theory that elevated brain NE acting at central α2-adrenoceptors is involved in the initiation of hot flashes. Postmortem studies have shown that most α2 receptors in the human brain are inhibitory, presynaptic receptors.[46] Blockade of these receptors with yohimbine would increase NE release, whereas clonidine would have the opposite effect.[47,48] Thus, the yohimbine provocation and clonidine inhibition of hot flashes may reflect a deficit in inhibitory α2-receptor function in symptomatic women. Given that estrogens modulate brain adrenergic receptors,[49,50] it is possible that the estrogen withdrawal at menopause is associated with this deficit.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: