Predictors of hot flushes in postmenopausal women who receive raloxifene therapy

Abstract

In a previous report, we described the results of a randomized, controlled trial that evaluated the potential of raloxifene to induce or exacerbate hot flushes. Here, we provide additional analyses that were undertaken to identify potential predictors of hot flushes and to assess the clinical usefulness of various therapeutic strategies for the reduction of hot flushes in postmenopausal women who receive raloxifene therapy. In this randomized, double-blind, placebo-controlled study, 487 unselected postmenopausal women were assigned randomly to receive treatment for 8 months with raloxifene, which was administered either at a dose of 60 mg/d every other day for 2 months followed by 60 mg/d (slow-dose escalation) or 60 mg/d throughout (raloxifene), or placebo. Data on the number, duration, intensity, and severity of hot flushes and awakenings because of night sweats were collected. Logistic regression models were used to examine the predictive value of various demographic and menopausal factors on the development or worsening of hot flushes. At baseline, 40.4% of all randomly assigned patients had hot flushes. The mean number of hot flushes (3-5 per week) was low. Fewer years postmenopause, surgical menopause, and previous estrogen or estrogen/progestin therapy were significant predictors of hot flushes at baseline but were not predictive of incident hot flushes during treatment with raloxifene. Of the women who received raloxifene therapy who had pre-existing hot flushes at baseline, 36% women had none at the end point. Early postmenopause and surgical menopause were significant predictors of a biologically relevant increase in hot flushes (>/=14 flushes/week). Early postmenopause, previous estrogen/progestin therapy, high body mass index, and greater duration of hot flushes at baseline were significant predictors of the need for symptomatic treatment. After 2 months of treatment, women in early postmenopause had significantly more hot flushes with raloxifene therapy than with slow-dose escalation ( P = .042), whereas there was no significant difference between raloxifene therapy and slow-dose escalation among women in later postmenopause. In the 50 patients who requested symptomatic treatment during the study, phytohormones or veralipride did not reduce the number of hot flushes markedly. A shorter time since menopause and surgical menopause are important predictors of hot flushes both before and during treatment with raloxifene. Previous estrogen/progestin therapy also increases the risk of hot flushes at baseline. For women in early postmenopause, slow-dose escalation of raloxifene therapy may be a suitable therapeutic strategy for the reduction of the risk of hot flushes.