Menopause: The Journal of The North American Menopause Society Vol. 20, No. 10, pp. 000/000 DOI: 10.1097/gme.0b013e3182885e31 * 2013 by The North American Menopause Society Relationship between objectively recorded hot flashes and sleep disturbances among breast cancer patients: investigating hot flash characteristics other than frequency Marie-He ´le `ne Savard, PhD, 1,2,3 Jose ´e Savard, PhD, 1,2,3 Aude Caplette-Gingras, PhD, 1,2,3 Hans Ivers, PhD, 1,2,3 and Ce ´lyne Bastien, PhD 1,4 Abstract Objective: The aim of this study was to evaluate the relationship between various characteristics of objectively recorded hot flashes and sleep disturbances in breast cancer patients. Methods: Fifty-six women who had completed a similar treatment protocol for a first diagnosis of breast cancer within the previous 3 months wore ambulatory sternal skin conductance and polysomnography devices for a home- based nighttime recording of hot flashes and sleep. Results: Hot flash frequency was not associated with polysomnographic variables (r = j0.18 to 0.21) or beta-I and beta-II electroencephalographic activities (r = j0.01 and 0.03) but was significantly correlated with increased slow (r = 0.28) and delta (r = 0.32) electroencephalographic activities. A slower hot flash onset and a longer hot flash duration were associated with greater polysomnographic impairments (r = j0.50 to 0.48). Greater sleep distur- bances were found during hot flash onset or hot flash plateau as compared with the preYhot flash period (greater percentage of wake time, lower percentage of stage II sleep, and lower percentage of rapid eye movement sleep, all P values G 0.05). The probability that a stage change to a lighter sleep occurred was significantly greater during hot flash onset (11%) than during hot flash plateau (6%; P = 0.02). Conclusions: This study suggests that the speed and duration of hot flashes would contribute more importantly to sleep alterations than hot flash frequency. Sleep disturbances tend to occur simultaneously with hot flashes, suggesting that these two nocturnal symptoms are manifestations of a higher-order mechanism involving the central nervous system. Key Words: Breast cancer Y Hot flashes Y Quantitative electroencephalogram Y Sternal skin conductance Y Sleep disturbances Y Polysomnography. A s many as 65% of breast cancer survivors report having hot flashes. 1 This vasomotor symptom is triggered or exacerbated by the sudden decline in estrogen levels induced by chemotherapy and hormone ther- apy (eg, tamoxifen), as well as by the abrupt cessation of hormone therapy at cancer diagnosis for the purpose of low- ering the risk of cancer recurrence. 2 Recent work from our research group showed that during adjuvant treatments and at 3-month follow-up, breast cancer patients reported greater frequency and severity of hot flashes than 70% to 99% of individuals from a control group matched on age, level of education, and menopause status. 3 There is growing evidence of a relationship between the frequency and severity of self-reported hot flashes and subjec- tive sleep impairments among women with breast cancer. 4<10 Recently, a longitudinal study conducted by our research team showed that changes in self-reported vasomotor symptoms occurring between the end of initial adjuvant treatments and 3-month follow-up were significantly associated with parallel changes in insomnia complaints. 11 Together, these data sug- gest that nocturnal hot flashes may contribute significantly to explaining the higher prevalence rates of insomnia symptoms in breast cancer patients as compared with the general pop- ulation and patients with other cancer sites. 12<15 However, these findings are limited by the use of subjective measures (eg, questionnaires and diaries), which give very little in- formation about which specific hot flash characteristics and type of sleep impairments are interrelated. Moreover, a weak correlation between subjective and objective measures of noc- turnal hot flashes has typically been found. 16,17 For instance, Carpenter et al 18 showed that only 22% to 42% of objectively Received November 22, 2012; revised and accepted January 17, 2013. From the 1 School of Psychology, Universite ´ Laval, Que ´bec, Quebec, Canada; 2 Laval University Cancer Research Center, Quebec, Canada; 3 Centre de recherche du CHU de Que ´bec, Quebec, Canada; and 4 Centre de recherche Universite ´ Laval Robert-Giffard, Quebec, Canada. Funding/support: This study was supported by salary support awards from the Canadian Institutes of Health Research (M.H.S. and J.S.) and the Fonds pour la recherche en sante ´ du Que ´bec (M.H.S., J.S., and A.C.G.). This study was also supported in part by a grant from the Ca- nadian Breast Cancer Research Alliance (DEX 017529). Financial disclosure/conflicts of interest: None reported. Address correspondence to: Jose ´e Savard, PhD, Laval University Cancer Research Center, 11 Co ˆte du Palais, Que ´bec, Quebec, Canada G1R 2J6. E-mail: josee.savard@psy.ulaval.ca Menopause, Vol. 20, No. 10, 2013 1