Hormonal factors and respiratory health in women – a review
Conflicts of interest
The authors have declared that they have no conflicts of interest.
Francisco Gómez Real, MD, Department of Gynecology and Obstetrics, Haukeland University Hospital, 5021
Tel: +47 55 974200
Fax: +47 55 974968
Sex hormones appear to play an important role in the lung health of women. This is, however, poorly understood and, in most aspects, poorly investigated; and the literature has been contradictory and confusing. This review presents recent research concerning the involvement of sex hormones in respiratory health of adult women, using the population surveys European Community Respiratory Health Survey and Respiratory Health in Northern Europe.
Respiratory health varied substantially according to hormonal and metabolic conditions. First, menopause was associated with lower lung function and more respiratory symptoms, especially among lean women. Second, hormonal replacement therapy (HRT) was associated with increased risk for asthma and wheeze; also, this association was particularly strong among lean women. Third, women with irregular menstruations in fertile age had more asthma, particularly allergic asthma, and reduced lung function, independently of body mass index (BMI) and physical activity. The findings were consistent across cultural and geographical borders. Our studies revealed that considering interplay between hormonal and metabolic factors is a clue to understand the effects of female sex hormones on the airways. A BMI of around 24–25 kg/m2 appeared to be optimal; women with this BMI had no increase in respiratory health problems when reaching menopause or using HRT, and women in fertile age with this BMI had optimal lung function independently of menstrual status.
In conclusion, female sex hormones appear to play a most important role for lung health in women. Further research on effects of sex hormones on the airways should take into account potential interplay with metabolic factors.
Please cite this paper as: Real FG, Svanes C, Macsali F and Omenaas ER. Hormonal factors and respiratory health in women – a review. The Clinical Respiratory Journal 2008; 2: 111–119.
Sex hormones appear to play an important role in the lung health of women (1). This is, however, poorly understood and, in most aspects, poorly investigated; and the literature is often contradictory and confusing. This is not surprising, given the complexity of this issue. Endogenous hormones vary according to factors directly related to reproduction such as menarche, menstrual cycle and its regularity, pregnancy, lactation, or proximity to the menopause. Exogenous hormones such as oral contraceptives, hormonal replacement therapy (HRT), or infertility therapy may play a role. In addition, sex hormone levels and effects may vary according to the underlying metabolic situation. Sex hormones and body fat mass are interrelated; the levels of estrogens are closely related to body fat mass (2) and the subject's metabolic status (3). This is plausible, since an optimal nutritional status is required for reproduction.
The menopausal transition illustrates the complexity of this issue, including a range of hormonal and metabolic changes as ovarian function declines and fertility disappears. Endogenous estrogen levels are first elevated and then decline (4). In addition, women in the perimenopausal transition become more insulin resistant (5), with increasing risk for cardiovascular diseases (6, 7). Moreover, with the appearance of climacteric symptoms, exogenous hormones (HRT) are widely recommended and used, as vaginal, oral (opposed or unopposed) or transdermal administration.
The literature on menopause and lung health is scarce. Based on reported cases, it has been postulated that asthma may start or worsen around the age of menopause (8). A large prospective and questionnaire-based study from the United States (The Nurses Health Study) found a significant lower risk of asthma in postmenopausal women not using HRT (9). Concerning HRT, clinical studies on lung function showed improvement or no change in lung function level in association with HRT (10–13); while in the large epidemiological Cardiovascular Health Study (14), significantly improved lung function was found among women using HRT. Studies on asthma have shown an association with HRT: the Nurses Health Study indicated that HRT was associated with increased risk for asthma (9), especially newly diagnosed asthma (15). Also, in the Copenhagen City Heart study (16), HRT was associated with more asthma. These studies were questionnaires-based and did not include objective data.
Although there is a growing awareness that sex hormones may play an important role in the development of airways pathology in women, the pathophysiological mechanisms remain unclear. A key to study the effects of sex hormones on the airways might be to take pre-existent metabolic status and changes of it into account. Such changes may be because of both endogenous and exogenous hormonal influences. The literature provides some intriguing pieces of information: asthmatic women may have abnormal sex-hormone levels (17, 18), and use of asthma medication is more common in women with infertility (19). Obesity is related to higher risk for asthma (20–23) and lower level of lung function (24). Subjects with low physical activity have shown increased bronchial responsiveness (25). Moreover, some evidence indicates that reduced lung function may be associated with increased insulin resistance (26–32). There could be a common denominator behind all these apparently unrelated conditions: insulin resistance. It is defined as impaired physiological response to insulin (33), and it is related to abnormal sex hormone levels, infertility, low physical activity and increased BMI (4).
Long and/or irregular menstrual cycles (oligomenorrhea) are a common phenomenon among women in fertile age, with a prevalence of about 20% (34). Oligomenorrhea is usually a symptom of polycystic ovarian syndrome (PCOS) (35–37), accompanied by hormonal changes and subfertility. Today, we know that PCOS is a pro-inflammatory condition (38) and a manifestation of insulin resistance, that sometimes is treated with insulin-sensitising drugs (4, 39, 40). Furthermore, oligomenorrhea is associated with increased risk for diabetes mellitus (41), cardiovascular diseases (42, 43) and diabetes in pregnancy (44). No previous studies apart from those presented in this review have investigated a possible association between irregular menstruation and asthma or lung function.
The purpose of this review is thus to present recent research documenting the involvement of sex hormones in respiratory health of women in a wide age span. The menopausal transition, involving endogenous and exogenous sex hormones, will be discussed (45, 46), as well as menstrual irregularity in fertile age, a condition involving both metabolic disturbances and sex hormones (47, 48).
The analyses presented in this review are based on data from the European Community Respiratory Health Survey II (ECRHS II) and the Respiratory Health in Northern Europe (RHINE) studies:
The European Community Respiratory Health Survey I (ECRHS I) (49) is an international survey carried out in 1990–1994 including randomly selected individuals from general populations within administrative boundaries. In each of 36 centres in 16 countries, a random sample and a smaller symptomatic sample were invited for clinical investigation (ECRHS I stage II). The ECRHS II (50), which took place in 1998–2002, is the follow-up of this investigation. The examinations carried out at both surveys included an interviewer-led questionnaire, lung function measurements, and blood samples for measurement of serum IgE. There were 4259 women from 21 centres who answered a questionnaire concerning women-specific issues as part of ECRHS II. Serum hormones levels were measured in some of these centres.
The ECRHS is described in detail at http://www.ecrhs.org. Ethical approval was obtained for each study center from the appropriate institutional or regional ethics committee, and informed written consent was provided by each participant.
Among participants in the European Community Respiratory Health Survey II (2002), 4262 women from 21 centres responded to a questionnaire concerning women's health. Women aged 45–56 years not using exogenous sex hormones (n = 1274) were included in the study concerning lung function, respiratory symptoms and the menopausal transition (45). Lung function measurements (n = 1120) and serum markers of hormonal status (follicle-stimulating hormone, luteinising hormone, and estradiol; n = 710) were available. Logistic and linear regression analyses included adjustment for BMI, age, years of education, smoking status, centre and height. In the study about menstrual irregularity and asthma and lung function (48), women aged 28–44 years (n = 1631) participating in the ECRHS II were included. Women taking exogenous sex hormones, being pregnant or recently had given birth were excluded.
The Respiratory Health in Northern Europe (RHINE) (51) (http://www.rhine.nu) is a follow-up study of subjects from seven Northern European centres who participated in the ECRHS I stage 1 (http://www.ecrhs.org) (49). In stage I of the ECRHS I, men and women aged 20 to 44 years were randomly selected from population registers within specific boundaries of each participating centre. A postal questionnaire was sent to 3000–4300 subjects in each centre. The population included in the RHINE study was responders from Reykjavik in Iceland, Bergen in Norway, Umeå, Uppsala and Gothenburg in Sweden, Aarhus in Denmark and Tartu in Estonia (n = 21,802, response rate 83.7%). The eligible subjects were sent a postal questionnaire in 1999–2001. Subjects not responding to the first mailing were sent two reminders. In total, 16 191 subjects answered the questionnaire, including 8588 women (response rate 77%) born between 1945 and 1973. Local ethics committees approved the study.
In the study about hormone replacement therapy, body mass index and asthma in perimenopausal women (46), pregnant women, women using oral contraceptives, and women <46 years were excluded. This cross-sectional analyses included 2206 women aged 46–54 of which 884 were menopausal and 540 used HRT. Stratified analyses by BMI in tertiles were performed. In the study about association of asthma and hay fever with irregular menstruation (47), only non-pregnant and non-menopausal women not taking exogenous sex hormones were included in the analyses (n = 6137), with no age limitation.
Lung function, respiratory symptoms and the menopausal transition (45)
Women not menstruating for the last 6 months (n = 432, 34%) had significantly lower FEV (1) values [−120 mL (95% CI, −177 to −63)], lower forced vital capacity values [−115 mL (95% CI, −181 to −50)], and more respiratory symptoms [odds ratio (OR), 1.82 (95% CI, 1.27–2.61)] than those menstruating regularly. Results were similar when restricting analyses to those who never smoked. Associations were significantly stronger in women with BMIs of less than 23 kg/m(2) [respiratory symptoms: OR, 4.07 (95% CI, 1.88–8.80); FEV(1) adjusted difference: −166 (95% CI, −263 to −70)] than in women with BMIs of 23 to 28 kg/m(2) [respiratory symptoms: OR, 1.10 (95% CI, 0.61–1.97), P(interaction): .04; FEV(1) adjusted difference, −54 (95% CI, −151 to 43), P(interaction) = 0.06].
Hormone replacement therapy, body mass index and asthma in perimenopausal women: a cross sectional survey (46)
HRT was associated with increased risk for asthma [OR 1.57 (95% CI 1.07 to 2.30)], wheeze [OR 1.60 (1.22–2.10)] and hay fever [OR 1.48 (1.15–1.90)]. The associations with asthma and wheeze were significantly stronger among women with BMI in the lower tertile [asthma OR 2.41(1.21–4.77); wheeze OR 2.04 (1.23–3.36)] as compared with heavier women (asthma: pinteraction = 0.030; wheeze: pinteraction = 0.042). Increasing BMI was associated with more asthma [OR = 1.08 (1.05–1.12) per kg/m2]. This effect was only found in women not taking HRT [OR = 1.10 (1.05–1.14) per kg/m2]; no such association was detected in HRT users [OR = 1.00 (0.92–1.08) per kg/m2] (pinteraction = 0.046).
Association of asthma and hay fever with irregular menstruation (47)
Irregular menstruation was associated with asthma [OR = 1.54 (95%CI = 1.11–2.13)], asthma symptoms [OR = 1.47 (1.16–1.86)], hay fever [OR = 1.29 (1.05–1.57)], and asthma preceded by hay fever [OR = 1.95 (1.30–2.96)] among women aged 26–42 years. This was also observed in women not taking asthma medication, [asthma symptoms: OR = 1.44 (1.09–1.91); hay fever: OR = 1.27 (1.03–1.58); wheeze preceded by hay fever: OR = 1.76 (1.18–2.64)]. Irregular menstruation was associated with new-onset asthma in younger women [OR = 1.58 (1.03–2.42)], but not in women aged 42–54 years [OR = 0.62 (0.32–1.18)]. The results were consistent across centres.
Menstrual irregularity and asthma and lung function (48)
Long or irregular menstrual cycles were reported by 313 women (19%). Oligomenorrhea was significantly associated with more asthma symptoms [OR = 1.76 (95%CI = 1.29–2.40)], allergic asthma [OR = 2.46 (1.43–4.23)] and lower FVC [adjusted difference = 63 mL (95%CI = −124 to −1)]. When excluding women using asthma medication, very lean women, or women exercising daily, these associations remained significant. Effects of oligomenorrhea were additive to those of BMI on asthma and FVC. Asthma symptoms increased significantly with BMI. FVC and FEV1 increased with BMI until 25 kg/m2 and thereafter decreased with increasing BMI. Excluding women exercising daily, asthma symptoms increased significantly with decreasing physical activity [OR 1.09 (1.001–1.19)] per category of physical activity) independently of oligomenorrhea. Among women exercising daily, oligomenorrhea predicted very high risk for asthma symptoms [OR = 12.6 (3.7–43)].
Based on the studies presented in this paper, we conclude that respiratory health varies substantially according to common hormonal and metabolic conditions. First, menopause was associated with lower lung function and more respiratory symptoms, especially among lean women. Second, HRT was associated with increased risk for asthma and wheeze; also this association was particularly strong among lean women. Third, women with irregular menstruations in fertile age had more asthma, particularly allergic asthma, and reduced lung function, independently of BMI and physical activity. Further, they confirmed the association of obesity with asthma and lower lung function. All these findings were consistent across cultural and geographical borders, and the findings were consistent in never-smokers. Our analyses revealed that considering interplay between hormonal and metabolic factors is a clue to understand the effects of female sex hormones on the airways.
Our findings in relation to international literature
Lower lung function among menopausal women in the ECRHS II is a novel finding. Similarly, potential interactions between BMI and menopausal status have not been investigated earlier. Increased respiratory symptoms in association with menopause as observed in the ECRHS study was not supported by previous literature (45); the RHINE study (46) showed a weak association between asthma and menopause while the Nurses Health Study (9) showed lower asthma risk in menopausal women. However, these studies did not stratify menopausal women by BMI. Repeated analysis of RHINE revealed that menopause was associated with more respiratory symptoms among lean women, in accordance with the present analysis (unpublished data). The estimate for the association of asthma with menopause as reported by Troisi et al. in Nurses Health Study (RR = 0.66) was similar to the corresponding estimate in the presented analysis for women with BMI 23–28 kg/m2 (OR = 0.66)(9). Discordance between the studies may be related to different selection criteria when excluding women using HRT, different BMI distributions and different age spans.
The presented higher prevalence of asthma among HRT users in RHINE was in agreement with findings from two large cohort studies, the Copenhagen City Heart Study (16) and the Nurses Health Study (9, 15). The interaction between BMI and HRT in effects on asthma was supported by Nurses Health Study showing a RR for HRT on asthma of 3.09 among lean women as compared to 1.58 among heavier women (15). The association of HRT with hay fever has not been reported previously. Unpublished data from ECRHS showed that postmenopausal women using HRT had significantly higher FEV1 and indicated higher FVC when compared to postmenopausal women not using HRT (52), in accordance to findings from Carlson et al.(14) in the Cardiovascular Health Study.
Results concerning irregular menstruations from RHINE (47) were in accordance with previous studies reporting that asthmatic women more often have abnormal levels of sex steroid hormones (17, 18), and that women taking asthma medication before age 21 years more often are referred for infertility (19). The study from ECRHS concerning irregular menstruation (48) confirmed the association with asthma symptoms and allergy. This study was the first to show an association of irregular menstruation with reduced lung function. The observed associations of lung function with BMI, with an optimum at about 25 kg/m2 and lower lung function among the very lean and the overweight was to our knowledge not been described previously. The association of asthma with lower physical activity agrees with a previous analysis of exercise and bronchial responsiveness in the ECRHS (25).
The analyses were based on women's reporting of their hormonal status. The literature shows that women report reproductive data with high reliability (53, 54). Attempts to validate the information provided by the women in our studies also suggest that the women give accurate information in spite of their coming from different cultural and lingual settings. Measured levels of FSH, LH and estradiol in serum in a subgroup of women in the ECRHS II corresponded very well with the women's reporting of their menopausal status (45). The answers to a question on irregular menstruation in the postal questionnaire of RHINE corresponded very well with the answers to a corresponding question provided by the same women two years later in a face-to face interview in the ECRHS (48). In conclusion, our analyses give further support to previous results suggesting a high reliability in women's reporting of reproductive factors.
The ECRHS and the RHINE both have extensive information on potential confounding factors, thus residual confounding by unknown factors is likely to be limited. Most important in the above analyses is proper adjustments, and in some cases stratification, for smoking, as smoking influences effects of estrogens. The main findings were confirmed in analyses of never-smokers.
Definition of the reference group is most important in analyses of reproductive factors in women. Considering HRT, inconsistencies in the literature may partly be related to differences in reference groups: as the effects of HRT on the airways appear to differ with regard to phase in the menopausal transition (unpublished analyses) (52), and menopause appears to have independent effects, the definition of the reference group is thus critical.
The effects of sex hormones are complex. Estradiol has effects on inflammation (55–58), immunity (59, 60) and metabolism (5, 61–64). These effects are not clear-cut and sometimes not even unidirectional (5, 65). The effects may depend on dose and therapeutic window (4, 5, 64), timing in relation to menopause, and route of administration (66). Information about direct effects of estrogens on the airways is scarce (57). However, estrogens may have effects on the airways through the mechanisms mentioned above. There is even less knowledge about potential effects of other sex hormones on the airways. Sex hormones are closely interrelated with body fat mass and this is well documented for estrogens (2, 67, 68). The metabolic status contributes to determine levels of sex hormones (4, 69, 70). As known, insulin resistance implies a general inflammatory condition and pathophysiological mechanisms involved have been well described with regard to cardiovascular disease, breast cancer and polycystic ovarian syndrome. It seems plausible that the airways as well could be affected by such general inflammation, and there are several pieces of indirect evidence indicating that this might be the case (20–30). A role of insulin resistance in respiratory health might be further supported by the literature showing that balance of oxidants and antioxidants in the diet is of importance for obstructive lung disease (71, 72), as oxidative stress is an important aspect of insulin resistance (73–75).
During the menopausal transition, women undergo important hormonal and metabolic changes. Endogenous estrogens are first elevated and then decline (4). In addition, women become more insulin resistant (6, 76). It is well documented from breast cancer research that lack of estrogens may contribute to the development of the metabolic syndrome (3, 63). A role of insulin resistance may explain the association of lower lung function and increased respiratory symptoms with oligomenorrhea as shown in analyses of RHINE and ECRHS II (47, 48). Other studies have observed a fall in lung function in association with insulin resistance (26–32). Thus, lower lung function in menopausal women could be explained by increased insulin resistance in menopause. Further, as insulin resistance is a pro-inflammatory condition (77–79), this may help explain the increase in respiratory symptoms associated with menopause. The effect modification of BMI on effects of menopause may be explained by a role of the individual's level of available estrogen and metabolic status (80, 81), both of them partly determined by the adipose tissue (81) and reflected by BMI (2). Estrogens are closely related to BMI, which is the strongest marker of estrogen levels in post-menopausal women (67, 68). Insulin resistance is intimately involved in the regulation of local estrogen production (70, 76).
There are several factors to consider when analysing HRT: First, in the perimenopausal transition, menstrual status changes from regular to irregular menstruations and then amenorrhea. This is accompanied by changes in estradiol: there is a rise in circulating levels at the beginning of the perimenopausal transition because of increasing FSH, and then a fall caused by ovarian failure; there is as well a shift in estrogen production from the ovaries to peripheral tissues. Second, BMI: it is related to estrogen production especially in the postmenopause, it is also related to insulin resistance (77, 82) and to asthma and lower lung function. Third, HRT: is HRT given before the menopause, when endogenous estrogens are still present or after menopause when endogenous production is minimal and HRT acts as a precursor for peripheral estrogen production? Last, but not the least, smoking status: smoking has an anti-estrogen effect; it is related to early menopause and to deteriorating lung health. We suggest that the estrogens in HRT may have direct inflammatory effects (55), and indirect anti-inflammatory effects mediated by improved insulin resistance (63), depending on the individual's menopausal or metabolic status. As insulin resistance is prevalent in healthy postmenopausal women, a moderate dose of estrogen appears to increase insulin sensitivity but higher doses may attenuate this benefit and progestins may cause a decrease in insulin sensitivity (5).
Our study reveals that the association of HRT with asthma appears to be modified by BMI (46). This is biologically plausible, considering the close interplay between sex hormones, fatty tissue and metabolic status. A similar interaction is described for breast cancer, where the increase in relative risk among HRT users was greatest in lean women (83, 84). The association of asthma with BMI is well documented although not fully understood (21, 51, 85). We speculate that the association of asthma with BMI is because of the inflammatory effects of insulin resistance (26, 77, 86). There is no direct evidence linking asthma with insulin resistance, although there are several studies showing associations of lung function with insulin resistance (26, 28, 30). We hypothesise that estrogens and BMI interact on the airways in part through common pathways.
Oligomenorrhea, BMI and physical activity were in these studies all independently related to asthma. It seems plausible that insulin resistance may be a common denominator for these findings: lower lung function, particularly FVC, appears to be associated with insulin resistance and diabetes (26–32), in agreement with the lower FVC observed in women with oligomenorrhea. The association of asthma and lung function with BMI may be explained by insulin resistance and inflammation related to obesity (4, 77, 87). The observed association of physical activity with lower prevalence of asthma symptoms may be explained by increased physical activity leading to improved insulin sensitivity, and thus to less inflammation.
Sex hormones appear to play a most important role for lung health in women. The effects of sex hormones on the airways cannot be considered without also taking into consideration the interplay with metabolic factors (88). Clinicians should be aware that respiratory health may vary with reproductive factors in women, and that this may differ according to BMI, with risks being higher for lean and obese women. Our studies identified the following three groups of women at increased asthma risk: (i) lean women entering the menopause; (ii) lean women using HRT; and (iii) women in fertile age with irregular menstruation. Menopause and irregular menstruation were also associated with lower lung function. Our studies further identified an optimal BMI of around 24–25 kg/m2; women with this BMI had no increase in respiratory health problems when reaching menopause or using HRT, and women in fertile age with this BMI had optimal lung function independently of menstrual status. These findings are worth to investigate further, as menopause affects most women, HRT being a widely used remedy, irregular menstruation affecting 20% of the female population in reproductive age, and BMI increasing in the general population.