Asthma and sex hormones
Giovanni Balzano, MD
Rehabilitation Institute of Telese Terme
Salvatore Maugeri Foundation
82037 Telese Terme
- NK cell
natural killer cell
- IL-1, -4, -5, -6
interleukin-1, -4, -5, -6
tumor growth factor-beta
peak expiratory flow rate
A chronic inflammatory state of the airways is considered the hallmark of asthma (1). However, the incidence, severity, and prognosis of asthma can be affected by a number of factors, including the patient's sex and age.
The sex of the fetus seems to influence the development of the respiratory apparatus. In fact, between weeks 28and 40 of gestation, the lungs are in a more advanced state in females than in males (2), a fact which could explain, at least in part, the higher incidence of respiratory distress in male infants (3). Clinical observations and epidemiologic studies agree that childhood asthma is more frequent in boys than in girls (4). Moreover, in some children, usually boys, asthmatic symptoms that started in infancy disappear around puberty (5), while many girls have the disease only during adolescence (6).
Age has been implicated in the severity of asthma. Asthma beginning during adulthood is generally more severe than childhood-onset asthma (7). Asthma that starts around the menopause or in old age is generally quite severe (5).
According to two recent epidemiologic investigations, the patient's sex and age affect the hospitalization rate for asthma (8, 9), which is a marker of severity. In the first study (a retrospective study of 33 269 patients), the admission rate of boys in the 0–10-year age group was almost twice that of girls. In the next decade, sex differences in the hospitalization rate were irrelevant, although there was a small, but significant prevalence of females. At all ages above 20 years, hospital admission rates were higher in women than men, with a female/male ratio ranging from 3:1 in subjects aged 20–50 years to 2.5:1 in those over 50 years of age (8). Similar data were reported by Elliasson (9). Both studies found a relationship between the age and sex of the patient and the severity of the asthma attack as indicated by duration of hospitalization. In fact, the duration of hospitalization increased with age in both sexes; moreover, hospital stays were longer in women over 30 years of age than in men of the same age (8, 9).
The influence of sex and age on asthma incidence and severity suggests that sex hormones could play a role in the pathogenesis of the condition. In particular, the change in hospitalization rates reported in the two epidemiologic studies described above (8, 9) could reflect the hormonal changes occurring in women around puberty and menopause. In the first decade of life, the hospitalization rate for asthma is higher in boys – an observation that can, at least in part, be attributed to differences in lifestyle between boys and girls. However, starting from the second decade of life, a progressive increase in the number of hospitalizations for asthma is seen in women. Serum levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol, progesterone, and dehydroepiandrosterone increase progressively in females from the age of 8–10 years, reaching adult values around the age of 16. Then, with the menopause, which generally occurs around the age of 50, a new hormonal pattern is established: high levels of FSH and LH, and low levels of estrogen and progesterone (10).
Serum levels of sex hormones have been directly correlated with the clinical and functional features of asthma. In a study of serum levels of estradiol, progesterone, and cortisol in asthmatic compared with healthy control women, the concentration of at least one hormone was outside the normal range in about 80% of the asthmatic women, indicating that asthma in women at reproductive ages is very frequently associated with alterations in the production and metabolism of steroid hormones (11). Girls with Turner's syndrome, which is characterized by low serum levels of estrogen, have increased airway responsiveness, which significantly improves after 6 months of estrogen treatment (12).
A body of compelling evidence supports the link between sex hormones and the pathogenesis of asthma. Danazol treatment in asthmatic women affected by endometriosis resulted in a significant improvement in the clinical and functional control of asthma (13). Moreover, in an in vitro animal study, estrogen treatment induced an increase in the number of β-adrenergic receptors in rabbit pulmonary tissues, and this effect was reversed by progesterone (14).
Sex hormones, the immune system, and inflammation
Airway inflammation in asthma is determined by a complex network of interactions between chemical mediators, cytokines, and cells, often initiated by exposure to an antigen (1). Similar immunologic mechanisms seem to operate even in those cases of so-called intrinsic asthma, in which sensitization to commonly inhaled allergens cannot be demonstrated (15).
There is compelling evidence of sex differences in some immunologic responses and that sex hormones affect several immunologic mechanisms. Taken together, these data suggest that sex hormones could play a role in the modulation of immunologic inflammation in asthma.
Compared to men, women seem to present more pronounced immune responses, as well as exaggerated reactions to autoantigens, and this could, at least in part, explain the greater predisposition of women to autoimmune diseases (16–18).
In animal models, sex steroid hormones have been shown to influence the immune system by various actions: by affecting the structure and function of the thymus; modulating the activity of B and T cells, natural killer (NK) cells, and phagocytic cells; interfering with cytokine production; and, finally, interacting with the effects of neuroendocrine hormones, such as growth hormone and prolactin (19).
In human leukocyte studies (20, 21) and in animal studies (22), estrogens have been shown to possess some anti-inflammatory properties. In healthy, postmenopausal women, the administration of estradiol reduces both the skin delayed immune responses and mixed lymphocyte reactions, although it has no effect on lymphocyte subpopulations (23). In a murine in vitro model, estrogen treatment enhanced production of specific autoantibodies, although the number of B cells was not affected (24). In other studies, estrogens have been shown to reduce leukocyte production in bone marrow (25), and to inhibit chemotactic activity in circulating polymorphonuclear leukocytes and monocytes; in contrast, progesterone enhances chemotactic activity (26). Physiologic doses of estradiol significantly reduce the inflammatory responses in castrated mice, although the percentage of polymorphonuclear cells in peripheral blood does not change (22).
In vitro, estradiol and progesterone reduce the leukocyte oxidative capacity induced by a phagocytic stimulus (20), and inhibit neutrophil degranulation (27). Both testosterone and progesterone, but not estradiol, inhibit the mitogen-induced lymphocyte proliferation in vitro by enhancing the activity of T-suppressor cells (28). High-affinity receptors for androgens have been detected in human thymocytes during maturation, but not in peripheral blood T cells (29). Data from in vivo and in vitro studies indicate that these receptors could play a role in thymus involution by exerting antiproliferative effects on thymocytes (30, 31). Estrogen-specific receptors have been found in T cells that have a suppressor/cytotoxic phenotype (16, 32). Activation of these receptors is thought to be responsible for the inhibition of suppressor/cytotoxic T cells and for the subsequent increased antibody production by human spleen cells induced by estradiol (16). Estrogen and progesterone receptors are overexpressed in the human allergic mucosa, but not in the normal one, where they appear mainly on activated eosinophils (33).
In vitro, sex hormones modulate the synthesis, release, and action of several cytokines involved in immune responses, such as IL-1, IL-4, IL-5, IL-6, interferon-gamma, and TGF-β (18). Interestingly enough, progesterone favors the switch from a Th0 to a Th2 cytokine profile (34).
Finally, the lipopolysaccharide-induced prostaglandin E2 production by human peripheral monocytes is reduced by testosterone and enhanced by progesterone, whereas, in the same model, estradiol has an inhibitory effect at a low concentration (0.4 ng/ml), and a stimulatory effect at a higher concentration (20 ng/ml) (35).
In summary, although the above reported data do not allow a straightforward interpretation in terms of positive or negative effects by different sex hormones on allergic and asthmatic symptoms (these effects also depend on the experimental model and dose of sex hormones), there is no doubt that the allergic inflammation of asthma can be significantly modulated by sex hormones.
Asthma and the menstrual cycle
Reproductive-age women experience cyclic variations in the serum concentration of sex hormones (36). During the 4 days after menstruation, FSH, LH, 17-β-estradiol, and progesterone serum levels are low. During the follicular phase of the menstrual cycle (days 12–16), progesterone serum levels remain low, while levels of FSH, LH, and 17-β-estradiol reach a peak. Finally, during the luteal phase (days 24–28 of the cycle), FSH and LH levels are low, whereas progesterone and 17-β-estradiol serum levels are moderately high (36). Hormonal fluctuations during the menstrual cycle influence cutaneous responses to histamine, morphine, and allergen, as demonstrated by a significant increase of wheal and flare reactions at days 12–16 of the cycle, which corresponds to the peak estrogen serum level (37).
Clinical observations indicate that the hormonal fluctuations described above might be responsible for the cyclic changes in asthmatic symptoms reported by patients (38–41). In fact, about one-third of asthmatic women report an increase in asthma symptoms during the premenstrual period (40, 41), although these symptoms were associated with a reduction in peak expiratory flow rate (PEFR) recordings in one study (40), but not in another (41). The lack of a constant association between increased asthma symptoms and impaired airway function during the premenstrual period suggests that the patient-reported occurrence of symptoms may be related to an actual increase in airway obstruction, or, alternatively, to an enhanced perception of symptoms due to an altered psychological state shortly before menstruation. However, although a positive correlation has been found between symptoms ofpremenstrual asthma and symptoms of premenstrual tension in one study (42), no correlation was found during the premenstrual period between symptoms of asthma and symptoms of tension, use of aspirin, use of oral contraceptives, or duration of menstrual cycle, in another study (41).
Investigations of the origin of premenstrual asthmatic symptoms are also hampered by variations in the phase of the menstrual cycle, during which asthmatic symptoms appear to worsen. In fact, asthmatic symptoms are reported to worsen only during the premenstrual period in 10% of patients, and only during the menstrual period in 16% of patients, whereas as many as 74% of patients report impaired daily activities in both periods (42).
Moreover, the severity and response to treatment of premenstrual asthmatic exacerbations probably vary. A small group of patients with serious premenstrual asthma exacerbations did not respond to high doses of systemic steroids (43). These patients generally had severe falls in PEFR measurements which seemed to be reversed only by treatment with high doses of progesterone (43). Moreover, two cases of fatal asthma have been reported in two sisters aged, respectively, 13 and 14 years. In these patients, the fatal asthma attack occurred 1 day before the expected beginning of menstruation (44).
Patients presenting premenstrual asthma exacerbations are frequently affected by alterations of the cyclic changes in progesterone serum levels (11, 36). The mechanism by which progesterone serum levels would interfere with asthma remains to be clarified. In contrast to the Th2 effect of the hormone (34), progesterone has been shown in different experimental models to exert a number of effects that could benefit asthma patients. In vitro, progesterone has been demonstrated to exert some immunosuppressive effects on maternal lymphocyte activation during pregnancy (45). In an in vitro animal model, sex steroid hormones, including progesterone, potentiated the isoprenaline-mediated relaxation of bronchial smooth muscle (46). Finally, in man, in vivo, progesterone is capable of reducing smooth-muscle contractility, as well as increasing the central respiratory drive in alveolar hypoventilation associated with obesity (47).
Several studies have investigated the variations in bronchial responsiveness during the menstrual cycle and its relationship to clinical asthma (48–50). The results generally argue that the hormonal changes occurring during the menstrual cycle are not associated with cyclic variations in bronchial responsiveness (48–50).
In summary, objective changes in lung-function measurements in relation to the menstrual cycle are rare in most asthmatic women. However, patients affected by premenstrual asthma, in comparison to asthmatic women without premenstrual exacerbation, generally have more severe symptoms and more frequent hospitalizations (42). This in itself represents a marker of disease severity; therefore, further studies are urgently needed to clarify the mechanisms responsible for asthma occurring in relation to the menstrual cycle.
Asthma and pregnancy
The reported incidence of asthma in pregnancy ranges from 0.4% to 4% (51, 52). Several metabolic and immunologic modifications accompanying pregnancy can remarkably influence the asthmatic state (5). During pregnancy both progesterone and cortisol serum levels increase (5). In particular, in pregnant women, serum concentration of total cortisol is threefold, that of globulin-associated cortisol is twofold, and that of free cortisol, which represents the biologically active form of cortisol, is about twofold greater than in women not pregnant (53). However, during pregnancy, other steroid hormones increase in the serum, including aldosterone, desoxycorticosterone, and, as mentioned above, progesterone (54–57), and these hormones can exert an antagonistic effect on cortisol receptors (58). Although the increases in progesterone and cortisol serum levels should be considered in combination with the other numerous hormonal modifications accompanying pregnancy, both progesterone and cortisol, in vitro, can exert some beneficial effect on asthma (46, 54). In addition, mild immunodepression has been shown to occur in pregnancy, particularly in terms of reduced cell-mediated immunity (5), and this could favorably influence the asthmatic state. On the other hand, successful pregnancy is considered to be a typical Th2 condition, while abortion is associated with Th1 predominance (55).
The clinical course of asthma during pregnancy is variable, also depending on mechanical changes in lung function apart from endocrine effects (52, 53). The few retrospective studies available indicate that asthma worsens in about one-third, and improves in about one-quarter of pregnant women (59, 60). Although the mechanisms responsible for these variations are not clear, it appears that mild asthma is likely to improve, whereas more severe forms of the disease frequently worsen (52, 59, 60). Moreover, asthma generally improves during the last 4 weeks of gestation, and asthma attacks are very infrequent during labor and delivery (52, 61, 62). This may, at least in part, be due to increases in prostaglandin E and cortisol serum levels that occur in the final phases of pregnancy (62, 64).
In patients presenting an improvement in asthma during pregnancy, a parallel reduction in bronchial hyperresponsiveness has been documented in two studies (65, 66). Both studies show that the more severe the initial hyperresponsiveness, the more pronounced its reduction. However, no correlation was found between reduction in airway hyperresponsiveness and either progesterone or estradiol serum levels (65, 66).
Changes in the asthmatic state during pregnancy generally disappear within 3 months after delivery (61). Repeated pregnancies usually exert the same effect on the asthma course; thus, these effects could be attributed to patient-related factors that remain constant during successive pregnancies (61). However, in as many as 40% of women, the asthma course is different in successive pregnancies, suggesting that other pregnancy-specific factors may play a role in the interaction between asthma and pregnancy (61).
Asthma and the menopause
Clinical observations indicate that the menopause is generally associated with exacerbation of pre-existing asthma. In addition, the menopause can also coincide with the clinical beginning of asthma – a finding indirectly supported by epidemiologic studies, which have recorded a peak in the frequency of asthma beginning in women around the age of 50, the mean age of the onset of the menopause (4). When asthma begins at the menopause, it is frequently characterized by such features as absence of a family history of asthma, absence of atopy, association with recurrent sinusitis and/or urticaria/angioedema, high severity, and need of systemic steroids for control of symptoms (5, 39).
In a study of women with asthma and/or urticaria that appeared around the menopause, low serum levels of FSH and LH, and particularly high serum levels of 17-β-estradiol were observed (67). This hormonal pattern is not typical of the menopause, and resembles that of women at a fertile age. Similar results have been obtained in another study of the hormonal profile in women with nonallergic asthma developed after the menopause as compared to age-matched women with pre-existing allergic asthma (68).
The mechanism by which this hormonal pattern can facilitate asthma may involve a high serum level of 17-β-estradiol. In fact, this sex steroid can enhance both the formation of prostaglandin F2-α, through the activation of α-9-ketoglutarase, and arachidonic acid metabolism (69). Thus, the well-known bronchoconstrictive and proinflammatory actions of prostaglandin F2-α and arachidonic acid metabolites can explain the asthmagenic potential of persistently high serum levels of 17-β-estradiol during the menopause. In addition, estrogens are likely to increase gap junction formation in bronchial smooth muscle, even though this activity has experimentally been reported only in myometrium (70). These hypotheses have indirectly been confirmed by the study of Della Torre et al. that demonstrated a clinical and functional improvement of both asthma and urticaria after normalization of the hormonal profile (67). Thus, in summary, an altered hormonal pattern, particularly higher than expected serum levels of 17-β-estradiol, can play a role in the genesis and/or maintenance of asthma and/or urticaria during the menopause in predisposed individuals.
A recent study found postmenopausal women to present a significantly lower risk of developing asthma than premenopausal women of similar age (71). In the same study, replacement therapy with estrogens increased, in a dose-dependent manner, the risk of asthma (71). In another study, replacement estrogen treatment caused bronchospasm in some individuals (72). Taken together, these observations suggest that during the menopause, physiologically low levels of estrogens may have protective effects against asthma, whereas abnormally high levels of estrogens – either naturally or iatrogenically occurring – can increase the risk of asthma.
Significant advances have been made in recent years in our understanding of the pathogenesis of asthma. The inflammatory nature of the disease is now established, and the diagnosis and treatment are well defined (1). However, the clinical presentation and evolution of asthma vary widely in terms of severity, and the determinants of severity in individual cases remain largely unknown.
Clinical observations and epidemiologic studies indicate that the sex and age of the patient may exert an important influence on the incidence, time of appearance, severity, and progression of asthma (Fig. 1). This, in turn, suggests the possibility of manipulating the pathogenetic mechanisms modulated by sex hormones. In fact, a number of in vitro and in vivo studies have shown that sex hormones can influence the allergic inflammation of asthma (14, 16–35) (Fig. 1). However, the most telling evidence that sex hormones are involved in asthma emerges from natural models, i.e., menstrual cycle, pregnancy, and menopause (Fig. 1). In these conditions, clinical and functional variations of asthma can be measured in parallel with well-known fluctuations of the hormonal profile.
It is not yet clear whether the asthma exacerbation occurring in many women during the premenstrual period are due to objectively measurable intensification of the disease or to increased perception of symptoms caused by the particular psychological state occurring shortly before menstruation.
In general, pregnancy is characterized by amelioration of both clinical and functional indices of asthma, including a reduction in airway hyperresponsiveness. In particular, asthma attacks in the last period of gestation and during labor are quite rare. The increased production of protective hormones, particularly progesterone and cortisol, may exert a protective effect, although the reasons that many asthmatic patients, particularly those with the most severe forms of the disease, may present exacerbation of asthma with pregnancy remain unclear.
The menopause can coincide with the appearance of asthma. On the other hand, during the menopause pre-existing asthma often becomes more severe. Most asthmatic menopausal patients have an abnormal hormonal profile, particularly an unexpectedly high serum level of estradiol. This, in conjunction with the lower risk of asthma generally presented by postmenopausal women than premenopausal women of similar age, suggests that serum levels of estradiol in excess of those expected for the age can increase the risk of asthma, or can increase the severity of pre-existing asthma. This hypothesis is supported by the increased risk of asthma associated with use of estrogen-replacement therapy.
In conclusion, sex hormones should be considered among the possible determinants of asthma severity (Fig. 1). Further studies on this subject should increase our knowledge of the pathogenetic mechanisms involved in the interaction between sex hormones and asthma, and improve our ability to manage the disease.
We thank Jean Gilder for revising the text.