Respiratory Symptoms,Pulmonary Function,and Reproductive History: Isparta Menopause and Health Study

Abstract

Objective:

We aimed to investigate the influence of reproductive factors on chronic respiratory symptoms and pulmonary function in a cross-sectional study of premenopausal and postmenopausal women 44–61 years of age.

Methods:

Self-reported data on respiratory symptoms were obtained from 1082 women. Usable forced expiratory volume in 1 second (FEV₁) (2.27 ± 0.44 L) and forced vital capacity (FVC) (2.75 ± 0.55 L) measurements were obtained from 1070 women. The influence of reproductive factors on FEV₁, FVC, and chronic respiratory symptoms was assessed by linear and logistic regression analyses.

Results:

Menopause is associated with lower FEV₁ (p < 0.0001, r = −0.181) and FVC (p < 0.0001, r = −0.198) only in univariate analysis. The association disappears when age is adjusted for. However, association of menopause with increased odds of chronic cough plus phlegm is significant in multivariate analysis (odds ratio [OR] 1.84, 95% confidence interval [CI] 1.16 - 2.92, p = 0.0088). Among reproductive factors, only age at first birth is independently associated with FEV₁ (p = 0.01, r = 0.008) and FVC (p = 0.004, r = 0.013). Results were similar when restricting analyses to those who never smoked and after exclusion of asthmatics.

Conclusions:

Postmenopausal status is independently associated with increased odds of chronic cough plus phlegm, whereas it does not appear to be independently associated with FEV₁ or FVC. Among the reproductive factors, age at first birth is the only independent predictor of FEV₁ and FVC. These findings suggest that reproductive factors may influence women's pulmonary health.

Introduction

There is compelling evidence that throughout a woman's reproductive life, her airways are subject to the influence of the cyclical variations in sex hormones occurring in relation to circadian rhythms, menstrual cycles, and use of oral contraceptives as well as pregnancy, menopause, and postmenopausal hormone replacement therapy (HRT).^1

–4 However, the literature provides very limited information on potential changes in respiratory symptoms and pulmonary function related to reproductive factors. Nilsson et al.⁵ reported that poor lung function was associated with early menopausal age. In a multicenter study from Europe, women not menstruating for the last 6 months were found to have significantly lower, forced expiratory volume in 1 second (FEV₁) and forced vital capacity (FVC) and higher frequency of respiratory symptoms.³ Krzyzanowski et al.⁶ reported that a greater number of children was a predictor for decline in pulmonary function. On the other hand, the results of the Respiratory Health in Northern Europe (RHINE) study showed no associations between asthma and menopause.⁷

Given our limited knowledge of the relation between reproductive factors and pulmonary function or symptoms, we sought to investigate the possible effect of a number of reproductive factors, including menopause, number of pregnancies and births, age at menarche, age at first and last birth, duration of breastfeeding, and duration of previous treatments with estrogenic or progestogenic medications on chronic respiratory symptoms and pulmonary function in a large sample of Turkish women from the Isparta Menopause and Health (IMH) Study.

Materials and Methods

Study design and population

The IMH Study is a cross-sectional study on determinants of age at natural menopause and other health measures in a convenience sample of 1106 women aged 44–61. The study was conducted in Isparta, Turkey, with a population of approximately 400,000.

Study participants were recruited by fliers. Eligibility criteria were as follows: born between 1946 and 1962 (aged 44–61 years during the survey); able to participate in all components of the study; absence of hysterectomy before menopause, oophorectomy, estrogen or progestogen use in the previous 3 months, hormone replacement therapy (HRT) use that may obscure the determination of age at natural menopause (>6 months in the 12 months after the final menstrual period), current pregnancy or lactation, amenorrhea due to anorexia, disease/surgery/radiotherapy involving the hypothalamopituitary axis, type 1 diabetes mellitus (DM), overt hypothyroidism or hyperthyroidism or history of thyroid surgery, adrenal gland dysfunction, systemic malignant disease, history of chemotherapy or radiotherapy, systemic autoimmune disease, disabling neurological conditions, and history of liver, heart, or renal failure. Subjects were also requested to report acute illnesses, including respiratory infections. Study procedures were postponed in the case of current infections.

After completing an interview and tests, participants were provided with test results and medical advice. The Institutional Review Board of Süleyman Demirel University School of Medicine approved the study, and a written informed consent was obtained from participants.

Study protocol

Each study participant underwent a 2-day evaluation. On the first day, interview data, including reproductive and lifestyle factors, were collected through a standardized questionnaire, and height and weight were measured. On the second day, a pulmonary function test was performed, and a pulmonary symptoms questionnaire was administered by a pulmonary disease specialist.

Outcome variables

Cough and phlegm production were evaluated using American Thoracic Society questionnaire items.⁸ “Chronic” was defined as occurring on most days for at least 3 months of the year, lasting for more than 2 consecutive years. Based on the questionnaire, participitants were categorized into chronic dry cough, chronic phlegm without cough, or chronic cough plus phlegm (chronic bronchitis) and a group with no cough, phlegm, or asthma. Chronic obstructive pulmonary disease (COPD) was defined as having an FEV₁/FVC of <70%, which corresponds to Global Initiative for Chronic Obstructive Lung Disease stage I or higher.⁹ Asthma was evaluated by the following question: Have you ever been diagnosed with asthma by a physician? The following question was used to inquire about wheezing or whistling: Have you had days with wheezing and/or whistling in your chest in the last 12 months?

FEV₁ and FVC were recorded using a standard spirometric (Schiller AG, Baar, Switzerland) method. Three acceptable and two reproducible maneuvers were required from up to five forced expirations, according to American Thoracic Society guidelines.¹⁰ The test was performed by specially trained nurses. Calibration of the spirometric instrument was performed at least once every testing day. FEV₁ and FVC are expressed in liters.

Predictor variables

Natural menopause is defined according to World Health Organization (WHO) criteria as 12 consecutive months of amenorrhea for which there is no other obvious pathological or physiological cause.¹¹ To investigate a possible influence of being in the earlier stages of menopausal transition on pulmonary outcomes, women with 3–11 months of amenorrhea were classified as experiencing menopausal transition. Other reproductive variables include age at menarche, age of menopause, number of births, number of pregnancies, age at first and last birth, duration of breastfeeding, and duration of previous treatments with estrogenic or progestogenic medications (including oral contraceptives and HRT).

Current smokers are those who smoked in the previous month, and ever smoking is defined as consuming ≥100 cigarettes in a lifetime.¹² Cumulative smoking was measured in pack-years. Alcohol consumption was defined as a woman being a user if she consumed any alcoholic drink at least once a year or as a nonuser otherwise. Cumulative alcohol consumption was defined as the product of years of consumption and weekly frequency of consumption.

Demographic and health-related variables include age, height (meters), body mass index (BMI) [weight (kg)/height (m²)], income, education, type 2 DM, hypertension, and family history of asthma and COPD.

Statistical methods

Participant characteristics for premenopausal and postmenopausal groups were compared by t test for continuous variables, with the assumption of nonconstant variance between groups, and chi-square test for categorical variables. Factors evaluated as continuous variables were age, height, BMI, pack-years of smoking, income, education, age at menarche, age of menopause, age at first and last birth, and duration of treatment with estrogenic or progestogenic medications. The dichotomous categorical variables included ever smoking, having smoked in the previous month, diagnosis of type 2 DM and hypertension, menopausal status, family history of asthma, and COPD. Both dichotomous (ever-never) and continuous variables were used for pregnancy, births, breastfeeding, and use of estrogenic or progestogenic medications. Age at first birth was also transformed into a categorical variable (1, age at first birth <20 years; 2, age at first birth between 20 and 22 years; 3, age at first birth >22 years) to estimate the FEV₁ and FVC variations in categories of age at first birth (assessed by a Kruskal-Wallis one-way ANOVA test).

Relationships between FEV₁/FVC and reproductive factors were evaluated in univariate (UVA) and multivariate (MVA) with linear regression analysis on 1070 women with technically adequate pulmonary function tests. A stepwise model selection algorithm is used in MVA. Variables that met the 0.15 p level were entered and remained in the model. Relations between respiratory symptoms, physician diagnosed asthma, spirometric COPD, and reproductive factors were evaluated in univariate and multivariate logistic regression analysis. A stepwise model selection algorithm was used in MVA. Variables that met the 0.05 level entered and remained in the model. Results are presented in terms of odds ratio (OR) and 95% confidence intervals (CI). For both linear and logistic regression, all predictor variables were entered simultaneously into the multivariate models.

Regression analysis investigating the effects of age at first and last birth was restricted to parous women (n = 1037). To minimize the possibility of residual confounding due to smoking and history of asthma, the analyses were also repeated after restricting the analyses to women who had never smoked (n = 829) and who do not have asthma (n = 999). Regression analysis investigating the effects of years since menopause and age at menopause was restricted to postmenopausal women (n = 495).

We stratified the study population by BMI categories according to WHO classification (1, 25; 2, 25–30; 3, 30–35; 4, > 35) to examine the possible interactions among reproductive factors, respiratory symptoms, and BMI.¹³ When performing the MVA in BMI categories, we forced the factors found to be significant for the whole population along with specific factors of interest, such as menopause and age at first birth or years since menopause, into the multivariate models.

Statistical analyses were conducted using SAS version 9.1 (SAS Institute, Cary, NC). A p value of <0.05 was considered significant.

Results

Of the original 1106 study participants, 1082 completed the pulmonary symptoms questionnaire, and 1070 provided usable spirometric measurements. Of 1082 women, 587 (54 %) were premenopausal, 495 (46 %) were postmenopausal, and 117 (10.8 %) were in the menopausal transition (3–11 months of amenorrhea). Median age at natural menopause was 52; mean age at natural menopause was 50.1 ± 0.2 years. Of the 1082 women who completed the pulmonary symptoms questionnaire, 79 had chronic dry cough, 49 had chronic phlegm without cough, 90 had chronic cough plus phlegm (chronic bronchitis), 258 had whistling, and 83 women had a history of physician-diagnosed asthma. Of 1070 women with usable spirometric measurements, 197 had COPD.

As shown in Table 1, postmenopausal women were older, shorter in height, had less years of education, reported a greater number of births and pregnancies and a longer duration of breastfeeding and estrogenic or progestogenic medication use, and had a more frequent history of asthma in their families compared with premenopausal women. Postmenopausal women also had lower levels of FEV₁/FVC and more frequent chronic cough plus phlegm.

Table 1.

Characteristics of Study Population (1082 Women Aged 44–61 Years) Overall and According to Menopausal Status

	Means ± SD or %
Variable	Whole study population (n = 1082)	Premenopausal women (n = 587)	Postmenopausal women (n = 495)
Age^*	51.12 ± 3.99	48.7 ± 2.6	54 ± 3.4
Education^*
0. Illiterate	3.79	2.04	5.86
1. Literate	4.53	2.56	6.87
2. Elementary school (5 years)	48.71	45.83	52.12
3. Secondary school (8–9 years)	5.91	6.47	5.25
4. High school (11–12 years)	15.99	21.47	9.49
5. Undergraduate (13–14 years)	12.20	11.93	12.53
6. Graduate/postgraduate (≥15 years)	8.87	9.71	7.88
Current monthly
income^*
1. < 420 USD	13.31	9.20	18.88
2. 420–840 USD	40.02	38.67	41.62
3. 840–1680 USD	30.87	33.56	27.68
4. ≥ 1680 USD	15.80	18.57	12.53
Current height (cm)^*	155.7 ± 5.50	156.7 ± 5.32	154.6 ± 5.4
Body mass index (kg/m²)	30.17 ± 4.66	29.9 ± 4.70	30.4 ± 4.59
Respiratory symptoms and disease
Chronic cough plus phlegm^**	8.32	6.30	10.71
Chronic phlegm without cough	4.53	5.11	3.84
Chronic dry cough	7.30	7.62	6.87
Wheeze or whistling	23.84	23.85	23.84
Self-reported physician-diagnosed asthma	7.67	9.03	6.06
Spirometric COPD	18.39	17.78	19.14
Pulmonary function
FEV₁(L) (n = 1070)^*	2.27 ± 0.44	2.84 ± 0.53	2.64 ± 0.55
FVC (L) (n = 1070)^*	2.75 ± 0.55	2.36 ± 0.42	2.17 ± 0.45
Reproductive factors
Age at menarche	13.53 ± 1.35	13.5 ± 21.37	13.4 ± 21.3
Age at first birth (among who ever gave birth, n = 1037)	21.51 ± 3.65	21.6 ± 23.53	21.3 ± 23.7
Age at last birth (among who ever gave birth, n = 1037)	28.2 ± 4.69	21.6 ± 3.53	21.3 ± 3.7
Ever pregnant	97.60	95.06	97.58
Number of pregnancies^*	3.98 ± 2.01	3.78 ± 1.87	4.2 ± 2.15
Ever gave birth	96.95	95.06	96.16
Number of births^*	2.55 ± 1.07	2.45 ± 0.95	2.68 ± 1.19
Ever breastfed	95.75	95.06	94.55
Total duration of breastfeeding (months)^*	30.20 ± 21.8	27.3 ± 19.4	33.5 ± 23.9
Ever use of estrogenic or progestogenic medications in past^*	62.29	58.60	66.67
Duration of previous treatments with estrogenic or progestogenic medications (months)^**	10.71 ± 25.79	8.98 ± 25.69	12.7 ± 25.7
Health-related factors
Diabetes mellitus^*	9.20	2.47	13.74
Hypertension^*	25.88	7.41	69.29
Family history of COPD	41.13	39.51	38.18
Family history of asthma^**	76.06	16.05	21.01
Ever use of alcohol (drink at least once a year)	2.68	2.73	2.63
Cumulative alcohol consumption	0.11 ± 1.41	0.058 ± 0.50	0.19 ± 2.01
Ever smoking	22.37	33.33	19.80
Current smoking	12.85	25.93	10.71
Cumulative smoking (pack-years)	1.90 ± 5.72	2.87 ± 5.64	1.93 ± 6.58

Significant difference between premenopausal and postmenopausal group.

p < 0.0001; **p < 0.05.

COPD, chronic obstructive pulmonary disease; FEV₁, Forced expiratory volume in 1 second; FVC, forced vital capacity.

Influence of menopause and other reproductive factors on FEV₁ and FVC

Menopause was associated with lower FEV₁ and FVC in UVA; however, it was not significant in MVA (Table 2). Models adjusting for a second variable in addition to menopausal status showed that menopausal status is no longer significant when age is in the model. When stratifying by BMI categories, the association of menopause with lower FEV₁ was significant only in women with a BMI of 25–30 (Table 3). Furthermore, years since menopause and age at menopause were not significant predictors of FEV₁ and FVC in UVA and MVA, although, when stratifying by BMI categories, there was an unexpected positive association between years since menopause and FVC in the lowest BMI category (<25 kg/m²) in MVA (p = 0.022, r = 0.04). In the group of premenopausal women, we also found no significant influence of being in the menopausal transition on FEV₁ or FVC (p = 0.25, r = 0.048; p = 0.43, r = 0.041; respectively).

Table 2.

Results of Univariate and Multivariate (Stepwise) Regression Analysis for Predictors of FEV₁ and FVC (n = 1070)

	Unadjusted coefficients				Adjusted coefficients ^a
	FEV₁		FVC		FEV₁		FVC
Variable	R ± SE	p value	R ± SE	p value	R ± SE	p value	R ± SE	p value
Age	−0.0291 ± 0.005	<0.0001	−0.0326 ± 0.004	<0.0001	−0.02 ± 0.003	<0.0001	−0.02 ± 0.004	<0.0001
Education	0.034 ± 0.008	<0.0001	0.035 ± 0.010	0.0002	–	>0.15	–	>0.15
Income	0.062 ± 0.014	<0.0001	0.069 ± 0.018	0.0009	–	>0.15	–	>0.15
Current height (cm)	0.0397 ± 0.002	<0.0001	0.0396 ± 0.002	<0.0001	0.02 ± 0.002	<0.0001	0.03 ± 0.003	<0.0001
Body mass index (kg/m²)	−0.0150 ± 0.002	<0.0001	−0.0225 ± 0.003	<0.0001	−0.007 ± 0.002	0.0037	−0.01 ± 0.003	<0.0001
Reproductive factors
Age at first birth^b (n = 1037).	0.013 ± 0.003	0.0003	0.016 ± 0.004	0.0005	0.008 ± 0.0005	0.0101	0.0134 ± 0.004	0.0041
Age at last birth^b (n = 1037).	0.0004 ± 0.003	0.888	0.0003 ± 0.003	0.916	–	>0.15	–	>0.15
Menopausal status	−0.181 ± 0.026	<0.0001	−0.198 ± 0.033	<0.0001	–	>0.15	–	>0.15
Age at menarche	0.008 ± 0.014	0.414	0.015 ± 0.012	0.227	–	>0.15	–	>0.15
Ever pregnant	0.041 ± 0.089	0.6394	0.049 ± 0.109	0.653	–	>0.15	–	>0.15
Number of pregnancies	−0.023 ± 0.006	0.0004	−0.025 ± 0.008	0.0022	–	>0.15	–	>0.15
Ever gave birth	0.052 ± 0.07	0.509	0.047 ± 0.097	0.626	–	>0.15	–	>0.15
Number of births	−0.056 ± 0.012	<0.0001	−0.062 ± 0.013	<0.0001	–	>0.15	–	>0.15
Ever breastfed	0.071 ± 0.067	0.643	0.053 ± 0.083	0.503	0.0009 ± 0.007	0.074	–	>0.15
Total duration of breastfeeding (months)	−0.002 ± 0.0006	<0.0001	−0.003 ± 0.0007	<0.0001	–	>0.15	–	>0.15
Ever use of estrogenic or progestogenic medications in past	−0.055 ± 0.02	0.0582	−0.044 ± 0.034	0.206	–	>0.15	–	>0.15
Duration of previous treatments with estrogenic or progestogenic medications	0.071 ± 0.067	0.289	−0.007 ± 0,0006	0.2232	–	>0.15	–	>0.15
Health–related factors
Ever smoking	−0.081 ± 0.032	0.012	−0.111 ± 0.04	0.005	–	>0.15	–	>0.15
Current smokers	0.044 ± 0.04	0.281	0.093 ± 0.05	0.062	−0.061 ± 0.037	0.10	–	>0.15
Cumulative smoking (pack–years)	0.001 ± 0.002	0.448	0.003 ± 0.002	0.212	–	>0.15	–	>0.15
Ever use of alcohol	0.097 ± 0.084	0.245	0.123 ± 0.102	0.123	–	>0.15	–	>0.15
Cumulative alcohol consumption	0.004 ± 0.008	0.649	0.007 ± 0.011	0.523	–	>0.15	–	>0.15
Family history of asthma	−0.088 ± 0.03	0.0058	−0.113 ± 0.039	0.003	−0.0594 ± 0.028	0.04	−0.103 ± 0.036	0.0049
Family history of COPD	−0.005 ± 0.02	0.845	0.028 ± 0.034	0.401	–	>0.15	−0.049 ± 0.031	0.1252
Diabetes mellitus	−0.121 ± 0.04	0.0084	−0.200 ± 0.052	0.0004	–	>0.15	–	>0.15
Hypertension	−0.073 ± 0.03	0.0197	−0.129 ± 0.031	0.0008	–	>0.15	–	>0.15

From stepwise linear regression model including all listed factors (age, height, BMI, education, income, smoking and alcohol, and health–related factors) as predictors. Variables that met the 0.15 p level were entered and remained in the model.

n = 1070 women had complete data on pulmonary function test.

COPD, chronic obstructive pulmonary disease; FEV₁, forced expiratory volume in 1 second; FVC, forced vital capacity; R ± SE, regression coefficient ± standard error.

Table 3.

Multivariate Associations of Menopause and Age at First Birth with FEV₁ and FVC Stratified by Body Mass Index Categories

	FEV₁ ^a				FVC ^b
	Menopause		Age at first birth		Menopause		Age at first birth
	R ± SE	P	R ± SE	P	R ± SE	P	R ± SE	P
BMI < 25 (n = 132)	0.06 ± 0.09	0.508	0.02 ± 0.009	0.031	0.13 ± 0.11	0.219	0.02 ± 0.01	0.013
BMI 25–30 (n = 409)	−0.11 ± 0.05	0.0407	0.006 ± 0.005	0.286	−0.11 ± 0.06	0.109	0.007 ± 0.006	0.304
BMI 30–35 (n = 361)	0.06 ± 0.05	0.276	0.01 ± 0.006	0.038	0.044 ± 0.066	0.504	0.017 ± 0.007	0.015
BMI > 35 (n = 135)	−0.18 ± 0.09	0.053	0.01 ± 0.01	0.212	−0.11 ± 0.12	0.376	−0.01 ± 0.01	0.164

Linear regression model for FEV₁ is adjusted for age, height, BMI, current smoking, family history of asthma, and total duration of breastfeeding.

Linear regression model for FVC is adjusted for age, height, BMI, current smoking, family history of asthma, and COPD.

BMI, body mass index; FEV₁, forced expiratory volume in 1 second; FVC, forced vital capacity; R ± SE, regression coefficient ± standard error.

Of the reproductive variables, only age at first birth showed a significant association with FEV₁ and FVC, which remained in the multivariate model (n = 1037) (Table 2). This result remained significant among nonsmokers (n = 829) and after excluding asthma (n = 987). When stratifying by BMI categories, the association of age at first birth with FEV₁ and FVC was significant in the first (<25 kg/m²) and third (30–35 kg/m²) BMI categories (Table 3). Using the categorical variable of age at first birth, mean FEV₁ and FVC values were significantly different between categories, with the earliest and the latest ages at first birth (p = 0. 0052 and p = 0. 0113 for FEV₁ and FVC, respectively) (Table 4).

Table 4.

Mean FEV₁ (L) and FVC (L) Values in 1037 Women Aged 44–61 Years Categorized According to Age at First Birth

Age at first birth (years)	n	Mean FEV₁ ^* liters (SD)	Mean FVC ^** liters (SD)
I (<20)	313	2.23 (0.44)	2.7 (0.54)
II (20–22)	382	2.26 (0.44)	2.72 (0.54)
III (>22)	342	2.34 (0.45)	2.82 (0.57)

p = 0.0052 (group I vs. group III); **p = 0.0113 (group I vs. group III).

FEV₁, forced expiratory volume in 1 second; FVC, forced vital capacity; SD, standard deviation.

The significant univariate associations of number of births, number of pregnancies, and duration of breastfeeding with FEV₁ and FVC were lost in the MVA (Table 2).

Influence of menopause and other reproductive factors on respiratory symptoms, physician-diagnosed asthma, and spirometric COPD

Postmenopausal women had increased odds of chronic cough plus phlegm in UVA (Table 5) and MVA (Table 6). Greater numbers of pregnancies were associated with increased odds of wheezing in UVA (Table 5) and MVA (Table 6), although this result did not remain significant among nonsmokers (n = 829) and after excluding asthma (n = 999). Total duration of breastfeeding was associated with increased odds of chronic cough plus phlegm and wheezing in UVA (Table 5) but not in MVA. Additionally, later age at first birth is associated with decreased chronic cough plus phlegm, wheezing, physician-diagnosed asthma, and spirometric COPD in UVA not in MVA (Table 5). When stratifying by BMI categories, we did not find any associations between reproductive factors and respiratory symptoms (analyses not shown).

Table 5.

Reproductive Factors as Predictors of Respiratory Symptoms, Asthma, and Spirometric COPD: Univariate Logistic Regression Analysis

	Chronic cough plus phlegm (n = 1082)		Chronic phlegm without cough (n = 1082)		Chronic dry cough (n = 1082)		Wheezing (n = 1082)		Physician-diagnosed asthma (n = 1082)		Spirometric COPD (n = 1070)
Reproductive factors	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value
Menopausal status	1.78 (1.15-2.76)	0.0097	0.74 (0.41-1.33)	0.31	0.88 (0.55-1.41)	0.61	0.999 (0.75-1.32)	0.99	0.65 (0.40-1.03)	0.06	1.09 (0.80-1.49)	0.56
Age at first birth (n = 1037)	0.919 (0.85-0.98)	0.0183	1.008 (0.93-1.08)	0.84	1.03 (0.97-1.09)	0.23	0.932 (0.89-0.97)	0.0013	0.93 (0.86-0.99)	0.04	0.94 (0.89-0.98)	0.014
Age at last birth (n = 1037)	0.98 (0.93-1.03)	0.45	0.97 (0.91-1.04)	0.47	1.002 (0.95-1.05)	0.94	0.987 (0.95-1.01)	0.387	0.97 (0.92-1.02)	0.24	0.97 0.94-1.012)	0.19
Age at menarche	1.09 (0.93-1.28)	0.25	1.05 (0.92-1.20)	0.41	0.90 (0.76-1.07)	0.25	1.024 (0.92-1.13)	0.64	1.03 (0.87-1.22)	0.68	1.05 (0.937-1.178)	0.39
Ever pregnant	0.68 (0.20-2.33)	0.54	0.8 (0.27-2.37)	0.69	> 999.9 (<0.001- >999)	0.97	1.044 (0.42-2.62)	0.92	0.62 (0.18-2.13)	0.45	0.49 (0.213-1.16)	0.10
Number of pregnancies	1.11 (1.007-1.22)	0.0356	1.04 (0.90-1.19)	0.56	1.01 (0.91-1.13)	0.74	1.126 (1.05-1.20)	0.0006	1.10 (0.99-1.22)	0.054	1.05 (0.951-1.13)	0.19
Ever gave birth	0.90 (0.27-3.02)	0.87	1.53 (0.20-11,4)	0.67	0.99 (0.26-1.03)	0.97	1.168 (0.50-2.72)	0.71	0.45 (0.16-1.19)	0.109	0.69 (0.309-1.56)	0.38
Number of births	1.07 (0.91-1,2)	0.10	1.07 (0.914-1,2)	0.38	1.08 (0.80-1.30)	0.59	1.181 (1.04-1.34)	0.0105	1.11 (0.90-1.35)	0.30	1.07 (0.936-1.24)	0.29
Ever breastfed	1.31 (0.39-4.32)	0.65	1.21 (0.47-3.13)	0.68	3.66 (0.49-26,9)	0.20	1.510 (0.69-3.27)	0.29	0.535 (0.22-1.30)	0.16	0.803 (0.39-1.64)	0.55
Total duration of breastfeeding	1.009 (1.001-1.01)	0.042	1.004 (0.99-1.017)	0.48	1.005 (0.99-1.01)	0.27	1.008 (1.001-1.01)	0.016	0.99 (0.98-1.007)	0.48	1.006 (1.0-1.01)	0.06
Ever use of estrogen or progestogen in past	1.167 (0.74-1.83)	0.505	1.146 (0.62-2.02)	0.65	0.988 (0.61-1.58)	0.95	1.254 (0.93-1.68)	0.130	1.357 (0.83-2.19)	0.21	1.179 (0.85-1.63)	0.31
Duration of treatments with estrogen and/or progestogen	0.99 (0.98-1.007)	0.71	0.99 (0.98-1.01)	0.66	0.99 (0.98-1.007)	0.58	1.001 (0.99-1.006)	0.79	0.99 (0.97-1.004)	0.17	1.00 (0.99-1.006)	0.88

CI, confidence interval; COPD, chronic obstructive pulmonary disease; OR, odds ratio.

Table 6.

Respiratory Symptoms, Physician-Diagnosed Asthma and Spirometric COPD in Relation to Reproductive, Anthropometric, and Health-Related Variables: Multivariate (Stepwise) Logistic Regression Analysis

	Odds ratio	95% confidence interval	p value
Chronic cough plus phlegm (n = 1082)
Menopause	1.848	1.167-2.927	0.0088
Education	0.767	0.648-0.908	0.0020
Family history of COPD	2.05	1.299-3.235	0.0020
Wheezing (n = 1082)
Number of pregnancies	1.039	1.012-1.072	0.0221^a
Physician-diagnosed asthma	3.157	1.965-5.073	<0.0001
Body mass index (kg/m²)	1.039	1.006-1.072	0.0197
Education	0.813	0.731-0.904	<0.0001
Current smoking	2.261	1.485-3.444	0.0001
Family history of COPD	1.646	1.228-2.205	0.0008
Chronic phlegm without cough (n = 1082)
Current smoking	2.599	1.342-5.034	0.0046
Chronic dry cough (n = 1082)
Current monthly income	0.641	0.430-0.955	0.0289
Spirometric COPD (n = 1070)
Income	0.652	0.545-0.781	<0.0001
Cumulative smoking (pack-years)	1.026	1.002-1.051	0.0356
Family history of COPD	1.541	1.125-2.111	0.0070
Physician-diagnosed asthma (n = 1082)
Education	0.802	0.681-0.943	0.0076
Family history of asthma	2.495	1.570-3.964	0.0001

From stepwise logistic regression models including reproductive factors (age at menarche, age of menopause, ever gave birth, number of births, ever pregnant, number of pregnancies, age at first and last birth, ever breastfed, duration of breastfeeding, ever use of estrogenic or progestogenic medications in past, duration of previous treatments with estrogenic or progestogenic medications including oral contraceptives and hormone replacement therapy), age, height, BMI, education, income, smoking and alcohol-related variables, diabetes mellitus, hypertension, family history of asthma, and COPD as predictor variables. Variables that met the 0.15 p value were entered and remained in the model.

This result did not remain significant after excluding smoking and asthma.

COPD, chronic obstructive pulmonary disease.

Influence of anthropometric and health-related factors on study outcomes

The influence of anthropometric and health-related factors on FEV₁ and FVC is shown in Table 2. The influence of anthropometric and health-related factors on respiratory symptoms, physician-diagnosed asthma, and spirometric COPD is shown in Table 6.

Discussion

Principal findings

To our knowledge, this is the first study to examine the impact of a variety of specific reproductive factors on pulmonary function and chronic respiratory symptoms in a large sample of women. The results of this study suggest that although many reproductive factors, such as menopause, number of pregnancies and births, breastfeeding, and age at first birth, seem to be related to FEV₁ and FVC, only younger age at first birth remains an independent predictor. Moreover, in our study, the relationship between lower pulmonary function and earlier age at first birth is independent of most factors known to influence pulmonary function, including smoking and asthma. Analyses stratified by BMI categories suggest that BMI is not a consistent effect modifier of the association between menopause and pulmonary function in this population, unlike findings for other populations.

Among the respiratory symptoms studied, menopause appears to be independently associated with increased odds of chronic cough plus phlegm. Earlier age at first birth appears to be associated with chronic cough plus phlegm, wheezing, physician-diagnosed asthma, and spirometric COPD in UVA but not in MVA. On the other hand, number of pregnancies was found to be an independent predictor of wheezing in MVA.

Comparison with other studies

Only a limited number of studies have examined the relation between pulmonary function and reproductive factors: menopause (Real et al.)³, menopausal age (Nilsson et al.⁵), or parity (Krzyzanowski et al.⁶). Similarly, a limited number of studies examined the relation between respiratory symptoms and menopause. The study by Real et al.³ reported that menopause is associated not only with lower FEV₁ and FVC but also with more respiratory symptoms among lean women. On the other hand, the RHINE study showed no associations between asthma and menopause⁷ even though the authors later reported that menopause was associated with more respiratory symptoms among lean women based on results of unpublished analyses.

Because BMI was shown to be a significant factor that may alter the association between menopause and pulmonary functions or symptoms, we also performed an analysis stratified by BMI categories. According to this, although menopausal status does not appear to be independently associated with FEV₁ or FVC in the whole population, there is an association with lower FEV₁ in the BMI 25–30 category. On the other hand, there is an unexpected positive association between years since menopause and FVC in MVA in the leanest BMI category. These findings suggest that BMI is not a consistent effect modifier of the association between menopause and pulmonary function in this population, as unlike findings for other populations. Interestingly, the association between later age at first birth and better pulmonary function was particularly stronger in women in the leanest BMI category (<25 kg/m²) and the category with a BMI of 30–35 kg/m², which does not suggest a consistent effect modification by BMI either. When we examined the association between reproductive factors and pulmonary symptoms in specific BMI categories, we did not find any associations between reproductive factors and respiratory symptoms.

Although our findings on spirometric pulmonary function are not in line with the findings of Real et al.,³ both studies appear to suggest that menopause may be associated with an adverse respiratory symptom profile. It is difficult to explain these contradictory findings; several differences and similarities between our study and Real et al.'s study should be mentioned. First, the IMH Study investigates the influence of menopause on pulmonary function and symptoms, and the definition of menopause is based on WHO criteria¹¹ (12 consecutive months of amenorrhea). Real et al.³ examined the influence of 6 months of amenorrhea on FEV₁ and FVC and respiratory symptoms and, therefore, may have captured an influence attributable to being in the earlier stages of menopausal transition. To determine if specifically being in the earlier stages of menopausal transition may influence pulmonary function in the IMH Study, we defined another group of women with 3–11 months of amenorrhea as the menopausal transition group, but we found no significant differences with regard to FEV₁ and FVC between the menopausal transition and premenopausal groups. Even though Real et al.'s study population was somewhat larger than ours, both studies were conducted in populations with a sample size of >1000, with 46% postmenopausal in the IMH study and 34% with 6 months of amenorrhea in Real et al.'s study.

Second, our study population was a highly selected and relatively healthy group as a result of strict study eligibility criteria. Thus, many reasons for an early menopause and worse pulmonary function might have been eliminated, leaving only the pure effect of menopause. A selected study population was also a limitation of the IMH study, somewhat hindering the generalizability of current findings to the general population. Real et al.'s study population, however, was selected by random sampling.³ Third, the study by Real et al. excluded women using “current” exogenous hormones, with no particular specification for the time interval when women had been avoiding them, whereas the IMH Study excluded women with a history of exogenous estrogen or progestogen use in the previous 3 months and also aimed to investigate the cumulative effect of previous use of these hormones by considering the total duration of such hormone use in the past. Fourth, our population was more obese (mean BMI 30.17 vs. 25.1) than those in Real et al.'s study, in which the association of menopause with lower FEV₁ and FVC was especially pronounced among lean women (BMI < 23). This finding is different from our study, in which the association of menopause with lower FEV₁ was significant only in women with a BMI of 25–30. Our findings may be interpreted to suggest that BMI is not a consistent effect modifier of the association between pulmonary functions and menopause or the influence of BMI as an effect modifier may differ for different populations.

Finally, potentially different sociocultural and biological confounders between Western and Eastern populations may have a role in these differences. Our study population represents an Eastern population with a mix of rural and urban populations in Isparta. The majority of women in this sample had a low level of education (only 8.9% of women had a graduate or postgraduate education), low income (>50 % had a monthly income of <840 USD), never smoked (87.6%), and never used alcohol (97.3%). In contrast, Real et al.'s study population was selected from 16 countries and represents a heterogeneous Western population in which women had lower BMI (25.1 compared with 30.17 in the IMH sample), a high level of education (median age at completed education: 19), and smoked more (52%).³ What is more, Figure 1B from Real et al.'s study suggests that particular countries included in the study, such as the U.K., United States, Spain, and Iceland, may fail to show an association between amenorrhea (≥6 months) and FEV₁, as is the case in the IMH Study.

The mechanisms by which menopause contributes to chronic cough plus phlegm in women are unclear. Menopausal transition is marked by changes in hormonal balance, including a rapid decline of endogenous estradiol levels, leading to a period of relative androgen excess.¹⁴ This shift in hormonal balance contributes to an increase in visceral adiposity that is associated with insulin resistance in postmenopausal women.¹⁵ A proinflammatory condition related to insulin resistance may cause pulmonary inflammation and, hence, airway damage, and this could explain the increase in respiratory symptoms in the postmenopause.^16,17 On the other hand, menopause has been reported to be associated with increased oxidative stress due to loss of estrogens.^18
–20 Given the important role of oxidative stress in the pathogenesis of COPD, the increased prevalence of chronic cough plus phlegm in our postmenopausal women may be explained by diminished antioxidant defense.

A most interesting result from this study is that among all reproductive factors, age at first birth is the most significant predictor for FEV₁ and FVC. Women whose age at first birth was <20 years had significantly lower FEV₁ and FVC than those whose age at first birth was >22years. When stratifying by BMI categories, the association of age at first birth with FEV₁ and FVC was significant in the leanest category (BMI < 25). The associations of age at first birth with FEV₁ and FVC remained significant when excluding the subgroup of women reporting asthma and smoking. Additionally, later age at first birth is associated with decreased respiratory symptoms in UVA but not in MVA.

To our knowledge, the impact of age at first birth on long-term pulmonary health has not been recognized to date. A significant increase in risk of coronary heart disease has been reported for women whose age at first birth is <20 years.²¹ Mirowsky²² mentioned age 22 as a critical age and suggested that the correlation of motherhood with health problems and mortality switches from detrimental to beneficial when age at first birth is >22. Much of the risk to the mother and fetus of early first birth may result from incomplete development of the reproductive system. However, some of it may reflect poor preparation for motherhood. Early parenthood may reflect a disordered or disadvantaged transition from adolescence into adulthood and may itself disrupt that transition, with lifelong consequences that influence pulmonary health. Perhaps the pregnancies and births at younger ages initiate pulmonary pathologies that grow over time. It is unclear, however, how early age at first birth might set women on a course toward worse pulmonary function in later life.

To obtain a hormonally well-defined study population, women with a history of estrogen or progestogen use in the previous 3 months were excluded from our study population. However, we had data to investigate the cumulative effect of previous use of these medications. We did not find any association between duration of previous estrogen or progestogen use and pulmonary function or respiratory symptoms. Some reports have suggested that postmenopausal women taking HRT are more likely to have asthma and asthma-like symptoms.^23
–25 Others have found that postmenopausal women who use HRT have higher levels of FEV₁ and less obstruction and also improvement in asthma or lung function related to HRT intake.^26,27 The influence of current or previous use of these hormonal treatments on pulmonary function and symptoms may not be equivalent at all. Our findings suggest that previous cumulative use of these hormones may not alter pulmonary functions and symptoms in later life.

In the current study, higher parity was found to be associated with worse pulmonary function in UVA but not in MVA. In a follow-up of the Cracow study, a faster decline of FEV₁ was observed in women with higher parity, although this was suggested to be due to worse socioeconomic conditions.⁶ On the other hand, Hank-Khan et al.²⁸ showed that parity was associated with higher FEV₁ in younger women (aged 18–50 years) but not in older women (aged 51–92). Discordance between the studies might be related to different population characteristics, such as different age span and underlying disease. Also, the results of our study suggest that greater number of pregnancies is associated with increased respiratory symptoms, such as chronic cough plus phlegm and wheezing in UVA. Number of pregnancies remained an independent predictor of wheezing in MVA. To our knowledge, a similar finding had not been reported before. The existence of such an association should be confirmed, and whether hormonal or other physiological changes during pregnancy may cause the observed increase in respiratory symptoms in later life should be investigated in future studies.

We did not find any studies in the literature investigating the relationship between ever (or duration of ) breastfeeding and pulmonary function or respiratory symptoms in later life. In our study, a highly significant inverse association between duration of breastfeeding and pulmonary function (FEV₁ and FVC) was observed in UVA. Although, ever breastfeeding remained in the multivariate model as a predictor of FEV₁, it did not reach statistical significance. We also observed that increased duration of breastfeeding and greater number of pregnancies were associated with increased respiratory symptoms, such as chronic cough plus phlegm and wheezing, in UVA but not MVA. Parity, gravidity, and duration of breastfeeding are highly related reproductive factors and deserve to be studied simultaneously as confounders for each other in future studies of pulmonary function and symptoms.

Strengths and limitations

A major limitation of this study is its cross-sectional design. Errors due to recall and misclassification are a major concern in cross-sectional studies. However, we particularly selected an age group so that the study population would be close enough to the end of their reproductive life but not at a far advanced age in order to accurately remember past reproductive factors. Another disadvantage of the cross-sectional design is that unmeasured confounders may influence the differences between premenopausal and postmenopausal groups. Following a group of women in a longitudinal design may overcome this limitation by measuring intraindividual changes during menopause.

Several factors may limit the generalizability of findings from this study. The self-selected nature of the participants (i.e., by fliers) and the strict study eligibility criteria make this a healthy select population. As mentioned, this may be advantageous in terms of avoiding the influence of several morbidities on pulmonary function and symptoms. However, findings may differ in a randomly selected sample of women. In addition, this population represents a mix of rural and urban populations in Southwestern Turkey and may have unique characteristics of its own, such as a very low rate of alcohol consumption and a greater number of pregnancies, compared with most Western populations. Reproductive predictors of pulmonary function might, therefore, be different in other populations. Any inference from the study should be considered in comparative terms within the considered sample.

A possible limitation of the present study is that the pulmonary symptoms questionnaire is a translation of the American Thoracic Society questionnaire to the Turkish language and was not specially designed for this population. However, the questions were quite simple and easy to understand.

There are several reasons to support the reliability of these data. First, the interview data were collected through standardized instruments; all the women were interviewed face to face by two trained nurses who were supervised by the principal investigator of the study, and the pulmonary symptoms questionnaire was administered by pulmonary disease specialists. Second, age, height, and BMI were strong predictors of pulmonary function, and smoking, low SES, family history of asthma, and COPD were also strong predictors of chronic respiratory symptoms, as shown in previous studies.^29

–33

Conclusions

This cross-sectional study of healthy select women shows that reproductive factors may influence women's pulmonary health. Our data suggest that among the reproductive factors, menopause is independently associated with increased chronic respiratory symptoms (specifically chronic cough with phlegm), even though we could not demonstrate an association between menopausal status and FEV₁ or FVC. The observed association of increased respiratory symptoms with menopause may be explained by increased oxidative stress or insulin resistance triggered by menopause which results in a proinflammatory condition, and, hence, airway damage. In addition, early age at first birth is associated with lower FEV₁ and FVC in this population, which may result from difficult pregnancies and births at early ages, initiating pathologies that grow over time. Further investigation is needed to confirm these findings in other groups of women.

Footnotes

Acknowledgments

This work was partially funded by the Technical and Scientific Research Council of Turkey (TUBITAK) research grant 106S079 (SBAG-3357) and Süleyman Demirel University research grant (21-DES-06) to Z.D.A.

Disclosure Statement

The authors have no conflicts of interest to report.

References

Kauffmann

, Becklake

. Maladies obstructives pulmonaires: Un paradigme de la complexity des différences de santé entre hommes et femmes. Saurel-Cubizolles

, Blundell

. La santé des femmes. Paris: Flammarion, Medicine et Sciences, 1996; 209–233.

Becklake

, Kauffmann

. Gender differences in airway behavior over the human life span. Thorax, 1999; 54:1119–1138.

Real

, Svanes

, Omenaas

et al. Lung function, respiratory symptoms, and the menopausal transition. J Allergy Clin Immunol, 2008; 121:72–80.

Real

, Svanes

, Omenaas

et al. Menstrual irregularity and asthma and lung function. Allergy Clin Immunol, 2007; 120:557–564.

Nilsson

, Möller

, Köster

, Hollnagel

. Social and biological predictors of early menopause: A model for premature aging. J Intern Med, 1997; 242:299–305.

Krzyzanowski

, Jedrychowski

, Wysocki

. Factors associated with the change in ventilatory function and development of chronic obstructive pulmonary disease in a 13-year follow-up of the Cracow study. Am Rev Respir Dis, 1986; 134:1011–1119.

Gómez Real

, Svanes

, Björnsson

et al. Hormone replacement therapy, body mass index and asthma in perimenopausal women: A cross-sectional survey. Thorax, 2006; 61:34–40.

Ferris

. Epidemiology Standardization Project (American Thoracic Society) Am Rev Respir Dis, 1978; 118:1–120.

Pauwels

, Buist

, Calverley

, Jenkins

, Hurd

. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Pulmonary Disease (GOLD) Workshop summary. Am J Respir Crit Care Med, 2001; 163:1256–1276.

10.

American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med, 1995; 152:1107–1136.

11.

World Health Organization Scientific Group. Research on the menopause in the 1990s. WHO Technical Services Department series No. 866. Geneva, Switzerland: WHO, 1996.

12.

Centers for Disease Control Prevention (CDC). Cigarette smoking among adults—United States, 2006. MMWR, 2007; 56:1157–1161.

13.

WHO. Physical status: The use and interpretation of anthropometry. Report of a WHO Expert Committee. WHO Technical Report Series 854. Geneva: World Health Organization, 1995.

14.

Liu

, Ding

, Bush

et al. Relative androgen excess and increased cardiovascular risk after menopause: A hypothesized relation. Am J Epidemiol, 2001; 154:489–494.

15.

Khan

, Prigeaon

, Schwartz

et al. Obesity, body fat distribution, insulin sensitivity and islet β-cell function as explanations for metabolic diversity. J Nutr, 2001; 131:354S–360S.

16.

Ghanim

, Aljada

, Hofmeyer

, Syed

, Mohanthy

, Dandona

. Circulating mononuclear cells in the obese are in a proinflammatory state. Circulation, 2004; 110:1564–1571.

17.

Arnalich

, Hernanz

, Lopez-Maderuel

et al. Enhanced acute-phase response and oxidative stress in older adults with type II diabetes. Horm Metab Res, 2000; 32:407–412.

18.

Arteaga

, Villaseca

, Rojas

, Arteaga

, Bianchi

. Comparison of the antioxidant effect of estriol and estradiol on low density lipoproteins in post-menopausal women. Rev Med Chil, 1998; 126:481–487.

19.

Tamir

, Izrael

, Vaya

. The effect of oxidative stress on ERalpha and ERbeta expression. J Steroid Biochem Mol Biol., 2002; 81:327–332.

20.

Shrivastava

, Singh

, Sapre

. Status of antioxidant enzymes and trace metals in postmenopausal women. J Obstet Gynecol, 2005; 55:64–66.

21.

Beard

, Fuster

, Annegers

. Reproductive history in women with coronary heart disease: A case-control study. Am J Epidemiol, 1984; 120:108–114.

22.

Mirowsky

. Age at first birth, health and mortality. J Health Soc Behav, 2005; 46:32–50.

23.

Troisi

, Speizer

, Willett

et al. Menopause, postmenopausal estrogen preparations, and the risk of adult onset asthma. A prospective cohort study. Am J Respir Crit Care Med, 1995; 152:1183–1188.

24.

Barr

, Wentowski

, Grodstein

et al. Prospective study of postmenopausal hormone use and newly diagnosed asthma and chronic obstructive pulmonary disease. Arch Intern Med, 2004; 164:379–386.

25.

Lange

, Parner

, Prescott

et al. Exogenous female sex steroid hormones and risk of asthma and asthma like symptoms: A cross-sectional study of the general population. Thorax, 2001; 56:613–616.

26.

Carlson

, Cushman

, Enright

et al. Hormone replacement therapy is associated with higher FEV₁ in elderly women. Am J Respir Crit Care Med, 2001; 163:423–428.

27.

Kos-Kudla

, Ostrowska

, Marek

et al. Effects of hormone replacement therapy on endocrine and spirometric parameters in asthmatic postmenopausal women. Gynecol Endocrinol, 200; 15:304–311.

28.

Harık-Khan

, Wise

, Lou

, Morrell

, Brant

, Fozard

. The effect of gestational parity on FEV₁ in a group of healthy volunteer women. Respir Med, 1999; 93:382–388.

29.

Quanjer

. Standardized lung function testing of the European Community for Coal and Steel. Bull Eur Physiopathol Respir, 1983; 19:7–10.

30.

Falaschetti

, Laiho

, Primateste

, Purdon

. Prediction equations for normal and low function from the health survey to England. Eur Respir J, 2004; 23:456–463.

31.

Chen

, Horne

, Dosman

. Body weight and weight gain related to pulmonary function decline in adults: A six-year follow-up study. Thorax, 1993; 48:375–380.

32.

Islam

, Schottenfeld

. Declining FEV₁ and chronic productive cough in cigarette smokers: A 25-year prospective study of lung cancer incidence in Tecumseh, Michigan. Cancer Epidemiol Biomarkers Prev, 1994; 4:289–298.

33.

Mannino

, Buist

. Global burden of COPD: Risk factors, prevalence, and future trends. Lancet, 2007; 370:765–773.

Respiratory Symptoms,Pulmonary Function,and Reproductive History: Isparta Menopause and Health Study

Abstract

Objective:

Methods:

Results:

Conclusions:

Introduction

Materials and Methods

Study design and population

Study protocol

Outcome variables

Predictor variables

Statistical methods

Results

Influence of menopause and other reproductive factors on FEV1 and FVC

Influence of menopause and other reproductive factors on respiratory symptoms, physician-diagnosed asthma, and spirometric COPD

Influence of anthropometric and health-related factors on study outcomes

Discussion

Principal findings

Comparison with other studies

Strengths and limitations

Conclusions

Footnotes

Acknowledgments

Disclosure Statement

References

Influence of menopause and other reproductive factors on FEV₁ and FVC