Plasma Estrogen Levels and Aneurysmal Subarachnoid Hemorrhage in Women.

The major limitation of this study is that we had a single measure of hormone levels with variable time to the development of aSAH. Participants are exposed to many years of hormonal exposure, and changes in reproductive lifestyles, such as smoking cessation and use of hormones during the study period, that we could not account for in a single hormone measure. We matched major determinants of hormone levels and resultantly, may not have detected subtle differences between case and control. Another major limitation of the study is that all patients are of Caucasian descent, which may limit the generalizability of the findings to other racial groups. Another potential bias is selecting cases and controls based on our matching criteria of age, smoking status, menopausal status and hormone therapy. While these factors were controlled in the study, we did not match other variables which may be significantly associated with aneurysmal risk, such as family history and genetic and connective tissue disorders. Furthermore, by matching these variables, we were unable to evaluate variations in estrogen levels that may be important in understanding their relationships with aneurysmal development. Furthermore, estrogen levels change over a woman’s menstrual cycle, and aSAH may occur many years after the initial estrogen levels were ascertained. We can only account for major differences in estrogen levels but not daily or monthly variability in hormonal changes. Despite the small size, to our knowledge, this is the largest prospective evaluation of estrogen levels and subsequent development of subarachnoid hemorrhages.

Requests to access the data set from qualified researchers trained in human subject confidentiality protocols may be sent to NHS (Nurses’ Health Study) at Brigham and Women’s Hospital. All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional Review Board of our hospital and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Participants gave informed consent to be included in the study. The Nurses’ Health Study (NHS) was established in 1976 when 121,700 female registered nurses aged 30–55 years were enrolled. The NHS II study was initiated in 1989 with 116,430 female nurses who were age 25–42. Participants complete follow up questionnaires every two years to update lifestyle factors, medical history and reproductive history. Blood samples were collected in 1989–1990 from 32,826 women in the NHS I study. For the NHSII study, blood samples were collected from 29,611 women from 1996–1999. Participants were sent a collection kit containing all supplies that were needed to have blood samples drawn. Samples were returned via overnight courier. Whole blood samples were centrifuged and plasma was aliquoted into labeled cryotubes. All samples were stored in liquid nitrogen freezers (temperature ≤ −130C) which are alarmed and monitored 24 hours a day. The blood samples were stored in the Channing/Harvard Cohorts Biorepository. Samples stored included plasma, white blood cells, and red blood cells. There is a standardized protocol where assays are processed the day after collection, and comparison of biomarker values in the samples processed immediately versus 24 or 48 hours after collection. Also, the protocol examined within-person stability over 1 to 3 years to confirm the analytes can be used to represent long-term status and are stable 5 . Medical records were requested from all participants who reported stroke and reviewed by physicians to classify the type of stroke including subarachnoid hemorrhage. To confirm aneurysmal origin of subarachnoid hemorrhage, documentation of aneurysm by computed tomography angiogram and magnetic resonance imaging was required. Subarachnoid hemorrhages without documented aneurysms were not included in this study. Women with confirmed aneurysmal subarachnoid hemorrhage from Nurses’ Health Study and Nurses’ Health Study II who had available blood samples were included in the current study. Eligible cases with aneurysmal subarachnoid hemorrhage were matched to controls by age (within ±12 months), menopausal status at blood collection, hormone therapy use (current, never/past), smoking status (current, past, or never), month of blood collection (± 3 months), fasting status (<8, ≥8 hours), and race (White, nonWhite). Age was matched within 12 months as age is closely associated with estrogen levels which may impact study’s outcomes. Additionally, the month of the blood collection was matched within three months to minimize potential variability in hormone levels due to temporal factors. Patients with confirmed cancer diagnosis and prior confirmed strokes were excluded. Plasma estrone (E1), estradiol (E2) were measured at the Mayo Clinic (Rochester, M) using turbulent flow liquid chromatography tandem mass spectrometry (LC-MS/MS) with intraassay coefficients of variation (CVs) of 8%. Sex hormone-binding globulin (SHBG) was measured by electrochemiluminescence immunoassay on an Elecsys 2010 immunoanalyzer (Roche Diagnostics, Indianapolis, IN), demonstrated a mean intraassay CV of 4%. Laboratory technicians were blinded to case-control status. Wilcoxon signed rank test was used to test the normality of the hormone levels. Sex hormones (E1, E2, and SHBG) were logarithmic transformed to normalize the data. Conditional logistic regression was used to determine the odds ratio (ORs) and 95% confidence intervals of aneurysmal subarachnoid hemorrhage for logarithmic transformed plasma estrogen levels and SHBG. The multivariable models were adjusted for additional risk factors, including hysterectomy, ovariectomy, body mass index (BMI; kg/m2), and history of hypertension (yes, no). P<0.05 was considered statistically significant. As a sensitivity analysis to test whether menopausal status has an impact on the association of subarachnoid hemorrhage and hormone levels, we performed independent analysis on pre-menopausal and post-menopausal women. An additional analysis was performed to evaluate the potential impact of smoking and hypertension on the association of hormone levels.

A total of 38 cases with aneurysmal subarachnoid hemorrhage, and 38 case-matched controls were included in the study, including 54 individuals from NHSI and 22 from NHSII. At baseline, cases and controls were similar for matched factors including age, smoking status, menopausal status, and hormone therapy ( Table 1 ). The mean age for cases and cohorts at the time of blood draw was 54.6 +/− 7.6 years. Cardiovascular risk factors, including BMI, hypertension, hypercholesterolemia, and diabetes were similar. As shown in Table 2 , mean estrone and estradiol levels were similar for cases and controls, 162±155pg/mL and 44.7±67pg/mL, vs 158.3±247pg/mL and 46.7±60 pg/mL, respectively. The mean SHBG level for cases was 92.0±60 nmol/L and controls was 105±61nmol/L. To determine whether hormone levels were associated with risk factors associated with cerebral aneurysms (hypertension and smoking), subsequent regression analyses were performed. Current or former smoking history was associated with lower E1 levels (β = −0.35±0.12, p=0.0042), and lower E2 levels (β = −0.63±0.16, p=0.0002), but was not associated with SHBG level (p=0.20) ( Figure 1 ) ( Supplementary Table 1 ). There was no association of hypertension and with E1 (p=0.97), E2 (p=0.38), and SHBG (p=0.80) levels ( Supplementary Table 2 ). Conditional logistic regression was performed with logarithmically transformed hormone biomarkers. There was no significant association for E1, E2 or SHBG with aneurysmal subarachnoid hemorrhage ( Table 3 ). Results were similar after multivariable adjustment for demographic factors, stroke risk factors, and reproductive history. When results were stratified by pre-menopausal and post-menopausal status, E1, E2 and SHBG remained not associated with aneurysmal subarachnoid hemorrhage ( Supplementary Table 3 ).

In conclusion, we found no association between estrone, estradiol or SHBG levels and risk of aSAH. We found that smoking is associated with lower estrone and estradiol levels, which may contribute to increased risk of aneurysmal rupture. Understanding the hormonal influences on cerebral aneurysms may have significant clinical implications, especially in tailoring prevention and treatment strategies for at-risk populations, and in counseling women on smoking cessation.

Due to its vasodilatory effects in the vasculature, 6 , 7 higher plasma estrogens have been proposed to be associated with a decrease in the development of cerebral aneurysms due to estrogen’s vasodilatory and anti-inflammatory effects on arteries. In in vitro and rodent studies, estrogens have been found to promote vascular wall vasodilation through estrogen receptor-mediated stimulation of nitric oxide production, and the re-introduction of estrogens in ovariectomized animals have found a reduced rate of aneurysmal subarachnoid hemorrhage 3 , 6 – 8 . However, clinical studies on the effects of reproductive factors and hormone therapy in women and its associations with aSAH have been mixed. While some studies have found a protective effect with post-menopausal hormone therapy use in women, others have found an increased rate of aSAH 9 – 11 . In our recent work, we found shorter reproductive lifespan and early menopausal age were associated with a higher risk of cerebral aneurysmal rupture in this population 12 . We hypothesized that this association is due to lifetime estrogen exposure’s effect on aneurysmal development. The goal of the current study was to determine, in a prospective approach, if baseline levels of estradiol and estrone are associated with the subsequent development of aSAH. In a prospective, nested case-control study from the Nurses’ Health Study, we did not find a significant association of plasma estrogens with the risk of aneurysmal subarachnoid hemorrhage in women. The absence of significant association may be explained by the true lack of association between baseline estrogen levels and the development of aSAH, or due to inability to detect a difference due to small sample size and temporal variation in hormonal levels. The mean estrone and estradiol levels in our cases were 162 pg/mL and 44.7 pg/mL. Due to significant differences in pre- and post-menopausal levels of estrogens, we performed a subsequent analysis on menopausal level and did not find a difference. Cigarette smoking 13 – 15 and hypertension 6 , 16 , 17 are two established major risk factors associated with the development and rupture of cerebral aneurysms. In this study, we did not find an association with hypertension and estrogen levels. Interestingly, we found former and current smokers had lower levels of estrone and estradiol, but not SHBG levels. This is consistent with existing literature that cigarette smoking significantly reduces endogenous levels of estrogens in both women and men 18 , 19 . Women who smoke have a lower risk of uterine fibroids, and endometriosis, but have higher rates of some osteoporotic fracture, which may be partially explained by their lower estrogenic state 19 , 20 . Smoking also changes the vasculature architecture, weakening arterial walls 21 . In this study, because smoking is a risk factor for aSAH and has a known effect on estrogen levels, we matched on smoking status so we could not examine differences between case and controls. Perhaps cigarette smoking increases aneurysmal rupture rate partially through its suppression of estrogens, resulting in remodeling of arterial wall and reduction in vessel vasodilation. Further studies to understand this relationship between estrogens and cigarette smoking, and their effects on cerebral aneurysms are needed.

Post-menopausal women have a higher incidence of cerebral aneurysm rupture, which can result in devastating clinical outcomes 1 . In vitro and in vivo studies have proposed that estrogens play a role in the development of cerebral aneurysms through their vasodilatory effects on the vasculature 2 – 4 . While rodent studies have found a protective association of estrogens with cerebral aneurysms, it remains controversial in humans whether such association between estrogens and aneurysmal subarachnoid hemorrhage (aSAH) exists. The rarity of aSAH and retrospective biases have made it difficult to assess the association of hormones with the disease. Estradiol is the primary form of estrogen during the pre-menopausal period while estrone is the primary form of estrogen after menopause. In this study, we investigated the association of baseline estradiol and estriol levels with incidence of aSAH among women in the Nurses’ Health Study (NHS) to determine whether levels of estrogen are predictive of subsequent cerebral aneurysm rupture. The NHS is the longest-running cohort study in understanding epidemiologic studies in history, focusing on the health and risk factors in women. The study has maintained a high follow-up with >90% of participants responding to each follow up cycle since 1988 and is a unique prospective cohort for understanding factors affecting aneurysm rupture.