Polycystic Ovary Syndrome and the Risk of Asthma in Reproductive-Age Women: Insights from two real-world large cohorts

Asthma is more prevalent and severe in women of reproductive age, often worsening with hormonal fluctuations during the menstrual cycle. This indicates that ovarian sex hormones contribute to asthma pathophysiology.^1,2 However, there is limited understanding of how these hormonal imbalances contribute to asthma risk. Polycystic ovary syndrome (PCOS) is a common disease affecting 4 to 20% of women of reproductive age.³ PCOS has several common risk factors associated with asthma prevalence and severity, including sex hormone disturbances, insulin resistance, anovulatory infertility, central obesity, and chronic inflammation.⁴ This shared clinical presentation has led to the proposition of a distinct entity termed “asthma-PCOS overlap syndrome”.⁵ Nonetheless, the exact incidence of asthma among patients with PCOS remains controversial, and previous data have been limited to small cohorts and selection bias. Here, we determine the risk of asthma in women with PCOS in two real-world population-based datasets: IBM-Explorys^® and TriNetX.

The IBM-Explorys^® (IBM, Cleveland, OH) is a large clinical registry that encompasses de-identified in and out-patient data extracted from electronic health records (EHR) across 39 healthcare organizations in the United States.⁶ We identified women of child-bearing age from 20 to 50 years between 1999–2016, with or without a diagnosis of PCOS (SNOMED: 237055002). To mitigate confounding factors such as hormonal changes during puberty and perimenopausal transition, we excluded individuals younger than 20 years and older than 50 years of age. Asthma (SNOMED: 195967001) risk in individuals with PCOS was adjusted for age, ethnicity, smoking, and body mass index (BMI) using a logistic regression model. Moreover, we conducted a sensitivity analysis by excluding current or former smokers and patients with the concurrent diagnosis of chronic obstructive pulmonary disease (COPD) (SNOMED: 13645005) to minimize bias of smoke-related airway disease.

To replicate our results, we analyzed live data from TriNetX, the largest global federated EHR network, encompassing more than 120 healthcare organizations across 29 countries.⁷ We selected women between the ages of 20 and 50 years included in TriNetX as of March 21, 2024, with or without a diagnosis of PCOS (IC-10-CM: E28.2). When assessing the risk for asthma (ICD10-CM: J45.xx), we used a 1:1 propensity score matching method without replacement to adjust for age, race, and obesity.

Out of 5,486,620 women in Explorys^®, 126,000 (2.3%) had PCOS. The prevalence of asthma was higher in women with PCOS than controls (21.6% vs. 13.7%, p<0.001). (Table 1) This higher prevalence was also seen after subgroup stratification by age and BMI (see Figure 1). PCOS was associated with an increased risk for asthma (adjusted odds ratio (aOR)= 1.59 [1.57; 1.61], p<0.001). Among women with PCOS, asthma was associated with White race (aOR= 1.08 [1.04; 1.12], p<0.001), BMI 25–30 compared to BMI <25 (aOR= 1.14 [1.09; 1.19], p<0.001), BMI >30 compared to BMI <25 (aOR= 1.37 [1.31; 1.42], p<0.001), and smoking (aOR= 2.07 [2.01; 2.14], p<0.001). The risk of asthma is even higher once patients with COPD or smoking history were excluded (aOR= 1.82 [1.79; 1.85]).

Table 1:

Clinical Characteristics of women with asthma stratified by their history of polycystic ovarian syndrome in Explorys^® and TriNetX.

Datasets	Explorys®		TriNetX
Category	PCOS	No PCOS	PCOS	No PCOS
n with asthma (%)	27,190(21.6)	734,070 (13.7)	66,523 (17.1)	1,404,677 (6.4)
Age in years (±SD)	-	-	35 (7.7)	3.5 (8.8)
Age category (%)
20–29	8,960 (33.0)	256,620 (35.0)	-	-
30–39	11,960 (44.0)	243,150 (33.1)	-	-
40–49	6,270 (23.1)	234,300 (32.0)	-	-
Race (%)
White	21,520 (79.2)	523,000 (71.2)	43,097 (64.8)	795,570 (56.6)
Others	5,670 (20.8)	211,070 (28.8)	23,426 (35.2)	609,107 (43.4)
Weight Category (%)¶
Normal	3,820 (14.1)	218,780 (29.8)	-	-
Overweight	6,420 (23.6)	225,690 (30.8)	-	-
Obese	16,950 (62.3)	289,600 (39.4)	17,554 (26.0)	44,414 (8.0)
Smoker (%)	23,760 (18.9)	906,680 (16.9)	-	-
Outcomes: Risk of Asthma
Unadjusted OR [95% CI]	1.73 [1.71; 1.76]		1.75 [1.72; 1.78]
Adjusted OR [95% CI]	1.59 [1.57; 1.61]		1.73 [1.70; 1.77]

Data are presented as n (%) for categorical variables and mean (standard deviation) for continuous variables.

^¶Weight category is stratified into: Normal weight for a body mass index (BMI) 18.5 to 24.9 kg/m²; Overweight for BMI 25 to 29.9 kg/m², and Obese for BMI above 30 kg/m²

Figure 1:

Bar plot of the prevalence of asthma comparing women from Explorys^® with and without PCOS in different age and BMI subgroups. Across all age groups and BMI categories, women with PCOS have a higher prevalence of asthma compared to women without PCOS irrespective of age and weight categories.

In TriNetx, 389,024 (1.7%) women were diagnosed with PCOS. Asthma was diagnosed in 66,523 (17.1%) of women with PCOS and in 1,404,677 (6.4%) of individuals without PCOS. In this cohort, the risk of asthma was also higher in women with PCOS compared to women without PCOS (aOR=1.73 [1.70; 1.77]). (Table 1)

The main finding of this analysis of two large population-based cohorts confirmed previous findings from smaller studies. A previous meta-analysis established that PCOS is an independent risk factor for asthma and is associated with lower lung function.⁴ Among asthmatics, PCOS is associated with increased asthma severity and worse asthma-related outcomes and healthcare utilization.^4,8 Furthermore, metabolic syndrome is well documented in both asthma and PCOS individuals. As previously reported, women with PCOS and asthma were more likely to be overweight or obese compared to asthmatics without PCOS.^4,5,8 Our findings also suggest a possible synergistic effect between PCOS and BMI on the development of asthma, which warrants further research.

In PCOS, androgen levels are converted to excessive estrogen by adipose tissue aromatase. This high level of estrogen causes suppressive feedback of gonadotropin secretion and failure of ovarian follicle development. Consequently, anovulation occurs and results in persistently high estrogen and insufficient progesterone production.³ Previous studies have shown that individuals with irregular menstruation and early menarche were more prone to develop asthma, have reduced lung function, and have more frequent asthma attacks. In addition, asthma is associated with infertility and longer time to pregnancy.^4,5,8 Premenstrual asthma (PMA), characterized by high estradiol levels and lower androgen levels, is seen in 11 to 45% of women with asthma. PMA is also associated with 20 to 40% decrease in peak expiratory flow and reduced vital capacity.^4,5,8 All this suggests a potential link between asthma and ovulatory dysfunction. Although epidemiologic data support the role of ovarian sex hormones with asthma, the role of hormone-based contraception in asthma has not been well studied in randomized-controlled trials, especially among women with PCOS. Cumulative evidence supports the importance of studying the role of such intervention in preventing and improving asthma outcomes among this high-risk population of women with hormone imbalance.

This study has several limitations. One major limitation is that patients might receive diagnoses, treatments, or observations from a hospital outside Explorys^® and TriNetX network, resulting in incomplete patient histories within the system. In addition, selection bias could emerge due to the inability to verify the criteria used for diagnosing asthma or PCOS. Despite these limitations, we found a large effect size in the association of PCOS and asthma, supporting the validity of our findings. To the best of our knowledge, this is the largest investigation of the prevalence of asthma among patients with PCOS using real-world data. We were able to externally validate and replicate our results in the national Exlporys^® data by analyzing a national and international TriNetX cohort. Of note, such differences between the geographic scopes of these cohorts explain the variability in asthma and PCOS prevalences.

In conclusion, individuals with PCOS are at increased risk for asthma. Future studies are needed to determine the underlying mechanisms that contribute to asthma risk in patients with PCOS, and whether targeted interventions or hormonal therapy improve asthma symptoms in PCOS patients.

Clinical Implication:

Clinicians should consider screening women of childbearing age with Polycystic Ovary Syndrome (PCOS) for asthma symptoms to avoid delays in diagnosis and management. In addition, weight management and obesity prevention in PCOS patients should be prioritized to reduce the risk of asthma. Future studies should assess the role of hormonal supplementation/therapy in this patient population to improve asthma severity and outcomes.