I tend to go to bed freezing, especially so in the winter, so I pile our flannel sheet, blanket, and down comforter over me when I settle in to sleep. A few times each menstrual cycle, clustered together in the luteal phase between ovulation and menses, I wake up from sleep completely soaked in my own sweat – not a delightful sight or experience. Usually I get up, change pajamas, and try to find a dry spot on the bed to go back to sleep (I promise the sheets eventually get washed, but I’m not about to wake my husband – and sometimes daughter – to change the bed at 3am).
These night sweats started when I was still intensively breastfeeding my daughter and was marathon training, when she was under a year old. At first, I thought it was because we were co-sleeping and we slept next to each other. But I never experienced them next to my husband before that point, and he is a six foot four heat generating machine.
When the marathon was over and I returned to less strenuous activity, breastfeeding frequency was also starting to decline. I didn’t get any night sweats again for quite some time.
Then there was roller derby.
At first, roller derby was a pastime, a recreational activity where I got to learn something totally new and hang out with women I respected. But of course, being the competitive person I am, it became an obsession, and in addition to roller derby practices I was working out quite a lot on my own time. Over the last year I’ve made additional nutritional adjustments to further improve my performance, and I’ve increased the intensity of my off-skates workouts. I work out a minimum of five hours a week, but in the middle of the season it is usually a minimum of nine hours per week.
At about the same time as this increase in physical activity and nutritional improvement, and changes in my body composition, the night sweats came back. With a vengeance.
My advisor once joked that anthropology is a rather navel-gazing discipline, but reproductive ecology, our subfield, is gonad-gazing. There are many times that I have been driven to ask questions in my own research because of physiological phenomena occurring in me, my family, or my friends. My interest in the relationship between reproductive and immune function stemmed from my sister and I both being diagnosed as gluten intolerant almost six years ago. My interest in puberty and adolescence stemmed from me looking ahead to what my preschooler daughter will be dealing with in the next decade. I know many people who have had personal interests in their research, research that might not otherwise have happened if someone hadn’t said, I want to understand why this is happening to me or someone I love. It makes a real case for the importance of diversity among scientists.
Because, guess what? There is almost nothing in the literature on the prevalence of night sweats among healthy premenopausal women. And this sweaty lady wants to get to the bottom of it.
This ain’t my mama’s hot flash
The vast majority of the literature on night sweats is related to the general study of vasomotor symptoms associated with menopause – that is, hot flashes as well as night sweats. Vasomotor symptoms appear to have a pretty similar mechanism and it is a response to the body sensing it is overheated. Blood vessels near the skin’s surface dilate, and sweating can also ensue, to cool it back down again.
Anyone remember when The Cosby Show talked about menopause? That was my introduction to what a hot flash was.
(embed not working for you? video here)
These kinds of symptoms are said to increase in frequency in early menopause. Several studies have shown a negative association between hot flashes and estradiol concentrations, meaning that as estradiol goes down hot flashes go up (Deecher and Dorries 2007; Miller and Duckles 2008). Estradiol naturally declines as a woman gets older, until she reaches menopause when it remains quite low for the rest of her life, and comes from places other than the ovaries. Similar relationships have been found between progesterone and vasomotor symptoms among menopausal women (Hitchcock and Prior 2012; Spark and Willis 2012). Vasomotor symptoms are also associated with depression, panic attacks, and sleep disturbances (Mold et al. 2002). This is likely because vasomotor symptoms can signal some sort of dysregulation in the autonomic nervous system, which is the part of your nervous system that regulates visceral functions like heart rate and, you guessed it, sweating.
The relationship between vasomotor symptoms and exercise, however, is complicated. While I found a few studies that found improvement in menopausal women’s vasomotor symptoms with exercise (Elavsky and McAuley 2005), there were also studies that found no relationship between exercise and vasomotor symptoms (Sternfeld et al. 1999), and still others that found that exercise worsened symptoms (Aiello et al. 2004). The Sternfeld et al (1999) article even found that body mass index was negatively associated with vasomotor symptoms, so the heavier a woman was, the fewer hot flashes she had.
The effects of exercise operate at cross purposes when it comes to vasomotor symptoms. While exercise can improve circulation and be great for your general health, sustained exercise generally leads to at least slight reproductive suppression, and therefore a reduction in progesterone if not also estradiol. The primary way a postmenopausal woman gets endogenous estrogens is from her fat cells, and of course exercise can decrease fat mass. Finally, exercise raises core body temperature, so in some women this could actually trigger vasomotor symptoms.
So if circulatory health is protective, but so are progesterone and estradiol, exercise isn’t going to have the same relationship to vasomotor symptoms in every woman.
Night sweats are about as common in reproductive as perimenopausal women
Here’s the next interesting thing. Hot flashes occur in a slightly higher frequency among menopausal women (32 versus 19%, Mold et al. 2002), which explains why research has tended to focus on them as a menopausal phenomenon. But night sweats occur at a more similar frequency in menopausal and premenopausal women (29 versus 22%, Mold et al. 2002). Check out this figure I made from Table 1 of Mold et al (2002):
The frequencies were statistically significantly different here (p < 0.001). But the practical difference is quite small – I mean, night sweats are occurring in almost a quarter of premenopausal women in this sample! To me, this suggests that the idea that night sweats are part of a suite of vasomotor symptoms that appear almost exclusively at menopause is very likely wrong. The question is whether the things that drive all vasomotor symptoms, particularly hot flashes, in menopausal women are also the things that drive night sweats in premenopausal women.
Premenopausal women aren’t all the same
The evidence I mentioned above regarding exercise and vasomotor symptoms certainly corresponds to my experience of night sweats resuming when I changed my diet and increased the duration and intensity of my workouts. Are the ways in which variation in estradiol and progesterone influence menopausal vasomotor symptoms similar in premenopausal women?
To figure this out, I turned to a family planning method. The Fertility Awareness Method uses a combination of basal body temperature and cervical mucus consistency to determine ovulation. It also tends to do a good job delineating the follicular (menses to ovulation) and luteal (ovulation to menses) phases, because many (but not all) ovulatory cycles have a biphasic body temperature pattern, with temperature higher in the luteal phase.
While it was difficult to find evidence linking hormone concentrations and body temperature, one study did seem to suggest a dose-response relationship, with higher temperature positively correlated with progesterone (Biller et al. 1999). There is also some indication that this change is driven by luteinizing hormone at ovulation, which is certainly correlated with estradiol and progesterone.
In addition to the fact that body temperature is higher in the luteal phase for many women, skin conductance may be higher at this time. Skin conductance measures electrical conductance, which gives one an idea of the moisture of the skin, and thus the activity of the sweat glands (Little and Zahn 1974). Both melatonin and temperature circadian rhythms are a bit different in the luteal phase: the nighttime increase in melatonin, and concurrent temperature increase, are delayed by 90 minutes in women in the luteal phase (Cagnacci et al. 1996). These changes in the timing of melatonin and body temperature could influence the sleep cycle and nighttime vasomotor symptoms.
Women in the luteal phase may also sweat more during exercise (Garcia et al. 2006). In this study, participants were allowed to have water during exercise, which may be why they sweat more, where in other studies where water is restricted core temperature increased in luteal phase women instead. Study authors did not find increased sweating in the luteal phase among women who likely did not ovulate, based on low serum progesterone concentrations.
It makes sense, then, that in some premenopausal women the increase in basal body temperature in the luteal phase could lead to night sweats, particularly if there are other factors that increase their chances. One I found almost by accident is that antihistamine use can increase the incidence of night sweats – I take loratadine almost every day to quell my allergy-induced asthma.
When are we going to pay more attention to reproductive state?
My field works pretty hard to measure global variation in reproductive function, but one understudied group is lactating women. There are some great papers about variation in resumption of ovarian activity with weaning and supplemental feeding – but what about continued ovarian or endometrial activity during lactation? Since breastfeeding is usually part of our exclusion criteria, we know little of the effect of lactation on reproductive function except in the most basic ways, and even less on vasomotor symptoms. I did find a single letter to the editor in American Family Physicians from a physician saying that he found many of his breastfeeding patients report night sweats (Taylor 2003).
Two factors drive the resumption of reproductive function in lactating women: prolactin levels and energy balance(Valeggia and Ellison 2001). Each breastfeeding bout brings on a spike in prolactin, which suppresses ovarian function. Closely spaced breastfeeding bouts can then impact ovarian function. However, the factor that seems to multiply the prolactin effect is whether the lactating woman is also in any kind of energetic deficit – either because she is exercising or doing physical work, or not getting enough food to replace her work and the 400-600 calories she burns a day making milk. Prolactin, then, was where I wanted to look next.
Prolactin concentrations vary through the ovulatory menstrual cycle, where it decreases from the late follicular into the midluteal phase (Bckstrm et al. 2008), meaning prolactin is lower in the luteal phase. High prolactin concentrations, or hyperprolactinemia, are a signal of infertility. Hyperprolactinemia is most often caused by a small mass on the pituitary gland, and there are therapeutic interventions that resolve it easily. But it is possible to have higher prolactin concentrations within the range of normal just by performing a lot of physical activity (Rojas Vega et al. 2012). Similar to a breastfeeding bout, prolactin increases with physical activity, stays high for a short period during recovery, and then declines again.
My weekday roller derby practices are late at night, one not ending until 10:30pm. Further, normal body temperature gets set a little higher after exercise (Haight and Keatinge 1973). The post-exercise drop in prolactin that is occurring while I’m asleep, along with my elevated exercise and luteal basal body temperature while I’m huddled under all my blankets, could trigger night sweats.
Changes in body temperature and prolactin from late night exercise, combined with higher basal body temperature in the luteal phase and antihistamine administration, could all contribute to the hot mess that has led to my night sweats. What I enjoyed about this foray into blatant navel/gonad/sweat gland-gazing was that it got me thinking about the ways in which the menstrual cycle affects the autonomic nervous system, and that I ended up learning a lot more about prolactin than I’d known before. I may need to measure prolactin and basal body temperature in future research projects, now that I know the kinds of factors that affect their variation, and their downstream effects.
Knowing the factors that may be contributing to my night sweats doesn’t necessarily mean I can prevent them, since I won’t be quitting roller derby night practices or allergy meds any time soon. But I can play around with ways to lower my body temperature on nights when I know it’s most likely – which may mean giving up my beloved down comforter.
Aiello EJ, Yasui Y, Tworoger SS, Ulrich CM, Irwin ML, Bowen D, Schwartz RS, Kumai C, Potter JD, and McTiernan A. 2004. Effect of a yearlong, moderate-intensity exercise intervention on the occurrence and severity of menopause symptoms in postmenopausal women. Menopause 11(4):382-388.
Bckstrm C, McNeilly A, Leask R, and Baird D. 2008. Pulsatile secretion of LH, FSH, prolactin, oestradiol and progesterone during the human menstrual cycle. Clinical Endocrinology 17(1):29-42.
Biller B, Luciano A, Crosignani P, Molitch M, Olive D, Rebar R, Sanfilippo J, Webster J, and Zacur H. 1999. Guidelines for the diagnosis and treatment of hyperprolactinemia. The Journal of reproductive medicine 44(12 Suppl):1075.
Cagnacci A, Soldani R, Laughlin GA, and Yen S. 1996. Modification of circadian body temperature rhythm during the luteal menstrual phase: role of melatonin. Journal of Applied Physiology 80(1):25-29.
Deecher D, and Dorries K. 2007. Understanding the pathophysiology of vasomotor symptoms (hot flushes and night sweats) that occur in perimenopause, menopause, and postmenopause life stages. Archives of Women's Mental Health 10(6):247-257.
Elavsky S, and McAuley E. 2005. Physical activity, symptoms, esteem, and life satisfaction during menopause. Maturitas 52(3–4):374-385.
Garcia A, Lacerda M, Fonseca I, Reis F, Rodrigues L, and Silami-Garcia E. 2006. Luteal phase of the menstrual cycle increases sweating rate during exercise. Brazilian Journal of Medical and Biological Research 39:1255-1261.
Haight JSJ, and Keatinge WR. 1973. Elevation in set point for body temperature regulation after prolonged exercise. The Journal of Physiology 229(1):77-85.
Hitchcock CL, and Prior JC. 2012. Oral micronized progesterone for vasomotor symptoms—a placebo-controlled randomized trial in healthy postmenopausal women. Menopause 19(8):886-893.
Little BC, and Zahn TP. 1974. Changes in mood and autonomic functioning during the menstrual cycle. Psychophysiology 11(5):579-590.
Miller VM, and Duckles SP. 2008. Vascular actions of estrogens: functional implications. Pharmacological reviews 60(2):210-241.
Mold JW, Mathew MK, Belgore S, and Dehaven M. 2002. Prevalence of night sweats in primary care patients. J Fam Pract 51:452-456.
Rojas Vega S, Hollmann W, and Strder HK. 2012. Influences of Exercise and Training on the Circulating Concentration of Prolactin in Humans. Journal of Neuroendocrinology 24(3):395-402.
Spark MJ, and Willis J. 2012. Systematic review of progesterone use by midlife and menopausal women. Maturitas 72(3):192-202.
Sternfeld B, Quesenberry Jr CP, and Husson G. 1999. Habitual physical activity and menopausal symptoms: a case-control study. Journal of women's health 8(1):115-123.
Taylor RD. 2003. Common causes of night sweats in various populations. Am Fam Physician 68(7):1264.
Valeggia CR, and Ellison PT. 2001. Lactation, energetics, and postpartum fecundity. In: Ellison PT, editor. Reproductive ecology and human evolution. New York: Aldine de Gruyter. p 85-106.