Abstracts from the NIH Office of Research on Women's Health Eighth Annual Interdisciplinary Women's Health Research Symposium November 17,2011

1 Musculoskeletal Research Center, Department of Bioengineering, University of Pittsburgh 2 Magee-Womens Research Institute, Department of Obstetrics and Gynecology and Reproductive Sciences at Magee-Womens Hospital, University of Pittsburgh P-1: A New Method to Determine Vaginal Axis and Spatial Position Using a 3-D Anatomical Coordinate System

Background and Objective: The position and orientation of the vagina using sagittal images has been used previously to quantify the degree of pelvic organ prolapse. While useful, analyses based on 2-D sagittal images are subject to tilt errors due to patient misalignment. The objective is to establish a 3-D coordinate system of the female pelvis based on MRI and use it to describe the position and orientation of the centroidal axis of the vagina.

Methods: Nine subjects with symptoms of pelvic floor dysfunction were scanned in the supine position. Three-dimensional reconstructions and anatomical coordinate systems were established using custom software written in Mathematica^TM (Wolfram Research). Anatomical coordinate systems were established based on the location of the ischial spines and the length of the longitudinal axis of the pubic symphysis. Three-dimensional surface models of the vagina were created using custom segmentation software written in Mathematica^TM (Wolfram Research). The centroidial axis of the surface model was then utilized to calculate changes in vaginal angle as it crossed the plane of the levator hiatus.

Results: Vaginal angle became more acute (toward the hiatal plane) in all but two women with changes ranging from 32 degrees toward the plane to 5.4 degrees away. A less acute angle suggests compromised apical support.

Conclusions: By quantifying anatomy relative to each patient's skeleton, this approach accounts for patient misalignment and is not subject to tilt error. In the future, this approach will be utilized to quantify differences in vaginal position and orientation in patients with prolapse to nulliparous controls.

1 Division of Rehabilitation Sciences, The University of Texas Medical Branch at Galveston 2 Sealy Center on Aging, The University of Texas Medical Branch at Galveston 3 Preventive Medicine and Community Health, The University of Texas Medical Branch at Galveston P-2: Gender Differences in Body Mass Index and Physical Performance among Older Mexican Americans

Background and Objective: Obesity is a major public health problem associated with reduced physical function in older adults. The purpose of this study was to examine gender differences between body mass index (BMI) and decline of physical performance at 2-year follow-up among older Mexican Americans.

Methods: Two-year prospective study of Mexican-American men and women aged >75 residing in five southwestern states in the United States. Data were collected on socio-demographic variables, physical performance (tandem stand balance, timed chair stand, and timed 8-foot walk), medical conditions, and BMI. Multiple regression analysis was used to test the effect of BMI on physical performance at 2-year follow-up.

Results: Out of 668 subjects, 31.3% of women and 18.4% of men had a BMI ≥30 kg/m² at baseline. Mean score of physical performance at baseline was 5.1 (SD=3.2) in obese women and 7.1 (SD=2.7) in obese men. Multiple regression analysis showed that women with a BMI of 30<35 kg/m² showed a decline of 0.8 points (SE=0.37; p=0.04) and those with a BMI ≥35 kg/m² showed a decline of 0.9 points (SE=0.47; p=0.04), after controlling for medical conditions and physical performance at baseline. No significant association was found between BMI and physical performance in men.

Conclusions: The association between BMI and physical performance differed by sex in older Mexican Americans. Among women, high BMI (≥30 kg/m²) was associated with decline in physical performance after controlling for all covariates. Conversely, BMI does not seem to affect physical performance in men.

1 University of Maryland School of Medicine 2 National Institute on Aging P-3: Cohort differences in Body Composition in Women: A Comparison of the Baltimore Longitudinal Study of Aging and Old Order Amish

Background and Objective: It is unknown how age-related changes in body composition differ across birth cohorts in the context of rapidly rising obesity rates. The objective of this study is to compare cohort trajectories of body composition in adulthood (ages 20+) across birth cohorts and in two socioculturally distinct populations of women: the Old Order Amish (OOA, N=226) and non-Hispanic White women from the Baltimore Longitudinal Study of Aging (BLSA, N=413).

Methods: Mixed effects models predicted body mass index (BMI) and body composition (obtained from DXA scans conducted from 1991–2010) based on age, population (OOA vs. BLSA), birth cohort, and their interactions.

Results: More recent cohorts of women in the BLSA had higher BMI across the lifespan: Women in the most recent cohort (born after 1944) had an average BMI 4.5 kg/m² higher (p<0.001) than women in the earliest cohort (born before 1923). Percent body fat also differed, with the most recent cohort having an average percent body fat 5.2 percentage points higher than the earliest cohort (p<0.001). Cohort differences in body fat were not accounted for by increases in BMI. OOA women had higher BMI than BLSA women (p<0.05), but lower percent body fat (p<0.05) than BLSA women.

Conclusions: The association between BMI and body fat differs across birth cohorts and social environments. Although increases in body fat have exceeded those predicted by increases in BMI alone among women in Baltimore, analysis of Amish women suggests that high physical activity may result in lower percent body fat even at similar weights.

1 Betty Irene Moore School of Nursing, University of California, Davis 2 Department of Psychiatry, University of California, Davis P-4: Beyond the Dyad: Gender, Distress, and Dementia Caregiving Patterns in Mexican-Heritage Families

Background and Objective: Caregiver burden (i.e., stress and poor physical and mental health) is prevalent among chronic caregivers of individuals with dementia, if adequate caregiver support and resources are not available. Gender differences exist in caregiving. Women are more likely to be caregivers for persons with dementia and to incur greater caregiver burden than men. Compared to White Non-Hispanics, Latino caregivers (mostly women) report more depressive symptoms, smaller and less effective support networks, and more dissatisfaction with the support they receive. The objective of the study is to describe Mexican-heritage caregivers' perceptions of their social support network and explore gender differences in caregivers' perceptions of the nature of the support they receive.

Methods: Secondary qualitative analysis of in-depth interviews with 37 Mexican-heritage caregivers for persons with dementia from SALSA (Sacramento Area Latino Study on Aging, an NIH-funded longitudinal study).

Results: The total sample included 17 (46%) spouses (6 husbands) and 20 (54%) adult children (5 sons) as primary caregivers. All caregivers reported diverse distribution patterns of caregiving beyond the primary relationship (caregiver and informant). However, spousal caregivers (husbands and wives) reported having fewer kin involved in sharing the burden than non-spousal (adult children) caregivers. Further, among spousal caregivers, lower socioeconomic status (SES) and less acculturated wives expressed greater dissatisfaction, feelings of isolation and distress, discontent with available support, and reluctance to ask for assistance than their lower SES male spousal caregiver counterparts.

Conclusion: Social support networks differ for male and female caregivers of Mexican heritage, which should be considered in designing meaningful interventions.

1 Child Neurology, University of Colorado Denver School of Medicine 2 Hemophilia and Thrombosis Center, University of Colorado Denver School of Medicine 3 Center for Women's Health Research, University of Colorado Denver School of Medicine 4 Department of Pathology, University of Colorado Denver School of Medicine 5 Department of Obstetrics & Gynecology, University of Colorado Denver School of Medicine 6 Children's Hospital Colorado P-5: Inflammatory Patterns in Placentas from Mothers with Preterm Premature Rupture of Membranes

Background and Objective: Preterm premature rupture of membrane (PPROM) accounts for up to one-third of preterm births and is often associated with chorioamnionitis. Yet, studies of PPROM placental pathologies are limited and may give insight into neonatal consequences. Our objective is to determine patterns of inflammation in placentas of women with PPROM.

Methods: Placentas were collected from women enrolled in our prospective PPROM cohort study at the University of Colorado Hospital from July 1, 2010 to April 30, 2011. Placentas (n=19) underwent gross and microscopic examination for findings of fetal- and maternal-side inflammation.

Results: Median gestational age at membrane rupture was 30 2/7 weeks (range 24 4/7–33 4/7). Average time to delivery after rupture was 8.3 days (range 1–36). There were 11 pathologically confirmed chorioamnionitis cases (58%). There was no difference in mean time to delivery after rupture between placentas with and without chorioamnionitis (7.5 vs. 9.3 days; p=0.68). Funisitis was also present (n=7; 37%), often with chorioamnionitis (n=6/11; 55%). There was no difference in mean time to delivery after rupture between cases with and without funisitis (8.9 vs. 7.9 days; p=0.82). Seven placentas had chorionic plate vasculitis (37%); all of which were associated with chorioamnionitis. Three placentas had chronic villitis (16%); one placenta had deciduitis reflecting sites of maternal inflammation.

Conclusions: In our PPROM cohort, more inflammation was seen on the fetal side of the placenta. Time to delivery after rupture did not predict pathological inflammation. Our findings supply groundwork to further delineate the relationship of PPROM, inflammation and newborn brain injury.

1 Department of Biochemistry & Molecular Biology, Virginia Commonwealth University School of Medicine 2 Department of Pathology, Virginia Commonwealth University School of Medicine P-6: The Transcriptional and Metabolic Response to Hypoxic Challenge

Background and Objective: The availability of oxygen is inherently tied to metabolism, where nutrients obtained through dietary input are balanced with O₂ levels to meet cell- and tissue-specific requirements of energy production and expenditure. In limiting oxygen conditions, the integrity of DNA, proteins, and lipids can be compromised—all are susceptible to attack by elevated levels of reactive oxygen species. The cellular adaptive response to hypoxia is multifaceted, and relies on shifts in transcription and metabolism. However, the processes underpinning conserved hypoxic-induced changes are not well understood. The objective of this study was to delineate the global transcriptional and metabolic changes that transpire in response to hypoxic insult and which are dependent upon the actions of HIF (hypoxia inducible factor) and ERR (estrogen-related receptor).

Methods: Drosophila display a robust hypoxic response that is mediated by conserved factors. We have used microarray analysis, GC/MS, and a variety of molecular genetic approaches (e.g., chromatin immunoprecipitation) to comprehensively define the hypoxic response in Drosophila larvae.

Results: Our approach has allowed us to determine, with high confidence and great resolution, those transcriptional and metabolic changes that occur in response to hypoxic challenge. Furthermore, we have defined the individual and collective roles of HIF and ERR in these transitions.

Conclusions: Our results indicate that HIF and ERR drive cellular behavior in hypoxia. Without either or both factors the adaptive hypoxic response cannot occur. Equally as important, our results suggest the existence of a novel, conserved hypoxic response pathway that is entirely HIF- and ERR-independent.

1 Section of Infectious Diseases, Boston University School of Medicine 2 Section of Endocrinology, Boston University School of Medicine P-7: Elevated Psychological Stress and Poor Physical Function among HIV+ women on HAART

Background and Objective: HIV infection is a risk factor that contributes to age-associated declines in physical function and psychological stress, despite effective HIV treatment. The objective of the study is to identify associations between physical function and psychological stress in women with HIV.

Methods: HIV-infected women aged 40 or older on stable HAART were enrolled. Psychological stress (Perceived Stress Scale [PSS]) and physical function testing (short physical performance battery [SPPB] and loaded stair climb) were assessed. Chi square, student T and correlation tests were performed (SAS version 9.1).

Results: Of 54 subjects enrolled to date, mean age is 50.5±6.9 years; 74% are Black, 50% are U.S. born, 31% are partnered, 40% are employed, and 52% graduated from high school. The median BMI is 27.4 (range 15.1–41.5); 28% have diabetes and 50% have hypertension. The median CD4 count is normal at 673 cells/mm³ (range 204–1320); while the median nadir CD4 count is low at 202 cells/mm³ (range 0–412); 94% have a HIV viral load ≤75 copies/mL. Seventy-six percent report higher levels of psychological stress (PSS score ≥14). While 33 of 41 (80%) had a high SPPB score ≥10, thirty-six completed the loaded stair climb demonstrating poor performance with mean power of 271.5±100.3 watts. Decreased power was associated with lower SPPB score and older age, but not higher psychological stress level.

Conclusions: HIV infection in these women was associated with a substantial burden of psychological stress despite viral suppression and CD4 restoration. Physical function as measured by power was poor, but not directly related to psychological stress level.

1 Division of Epidemiology, Vanderbilt University 2 Shanghai Institute of Preventive Medicine, Shanghai Cancer Institute P-9: Gene Expression Profile Selected Genetic Variants and Breast Cancer Survival

Background and Objective: Two commercially available assays measure gene expression in tumor tissues to predict breast cancer prognosis. Many genes and loci included in these assays represent novel biological mediators of cancer progression. The objective of the study was to evaluate associations between gene expression profile selected (GEPS) genetic variants and breast cancer survival.

Methods: Variants within and correlated to the expression of GEPS loci, were identified from genotype data for 2,918 Chinese women with breast cancer, followed for a mean of 5.7 years. Associations with breast cancer survival were evaluated using Cox proportional hazards regression.

Results: Genotype data was available for 2,308 variants within, and 433 variants correlated to the expression of, 80 GEPS loci. A total of 1,956 variants with minor allele frequencies above 5%, within 861 independent loci (r²<0.6), were evaluated. The observed number of nominally significant associations (p≤0.05) was greater than expected by chance for both genetic variants (256 vs. 98, p<0.001) and independent loci (163 vs. 43, p<0.001). Highly significant associations (p≤0.001) were seen for 10 variants in seven loci (within or correlated to the expression of AURKA, BCL6, BIRC5, C20orf108, COL4A2, DIAPH3, DHX58, EPR1, KAT2A, and TMEM235).

Conclusion: Associations for variants in genes known to influence cancer progression, as well as in genes and loci not previously studied in the context of breast cancer survival, were identified at a level greater than chance. These findings provide additional targets for investigations into the genetic mediators of breast cancer progression.

1 Department of Epidemiology and Public Health, University of Maryland School of Medicine 2 Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health P-12: Hot Flashes, Sleep Complaints, and Depressed Mood in Midlife Women

Background and Objective: One-fifth of midlife women report that sleepiness interferes with daily life and can influence depressed mood. Here we examined the relations among hot flashes (HF), sleep complaints, and depressed mood as part of a larger study on midlife women's genital health.

Method: Cross-sectional data were obtained from 179 women (ages 35–60, 79% white). HF experience was coded as ever/never and ever/never within the past year. Sleep complaints were assessed with the Pittsburgh Sleep Quality Index (PSQI); depressed mood was assessed with the Center for Epidemiological Studies-Depression scale (CES-D).

Results: The experience of HF was significantly related to the total PSQI score for both ever having HF (p<.001) and for HF within the past year (p=.001). Depressed mood was not significantly related to HF, but was correlated with PSQI total score (p<.001); 25% of the sample reported both elevated PSQI and CES-D scores (≥5 and ≥16, respectively). Ever experiencing HF was related to shorter sleep duration (p=.023), longer sleep latency (p=.041), more frequent daytime dysfunction (p=.008), and greater sleep disturbance (p=.015). Focusing on HF within the past year, sleep duration and daytime dysfunction were no longer significant (p-values >.05), whereas longer sleep latency and greater sleep disturbance remained. Among HF sufferers, those reporting <7 hours of sleep/night were more likely to report an elevated CES-D score, (p=.007).

Conclusion: Midlife women with HF reported numerous sleep complaints; these sleep complaints were not related to depressed mood. The combination of HF and insufficient sleep might trigger depressive symptoms.

1 General Internal Medicine and Public Health, Vanderbilt University Medical Center 2 Department of Obstetrics and Gynecology, Vanderbilt University Medical Center 3 Department of Biostatistics, Vanderbilt University 4 Department of Hematology/Oncology, Vanderbilt-Ingram Cancer Center 5 Department of Rheumatology and Immunology, Vanderbilt University Medical Center 6 Institute of Medicine and Public Health, Vanderbilt University Medical Center 7 Medical Social Sciences, Northwestern University Feinberg School of Medicine P-14: Arthralgia and Patient-Reported Outcomes in Postmenopausal Women with Early Breast Cancer Taking Aromatase Inhibitors

Background and Objective: To prevent cancer recurrence, each year over 100,000 U.S. women begin a 5-year course of aromatase inhibitors (AIs) for early-stage breast cancer. AI-related joint pain (arthralgia) may interfere with patients' well-being, leading to nonadherence. More information is needed about arthralgia and patient-reported outcomes (PROs) in these women. We sought to assess relationships of arthralgia with other PROs over the first 12 weeks of AI therapy in a longitudinal cohort study.

Methods: Postmenopausal female oncology outpatients with early-stage breast cancer (n=52, pilot cohort) completed paper surveys prior to AI initiation, and every 2 weeks thereafter for 12 weeks. Pain was measured in 16 joint locations using a 0–10 numeric rating scale. Baseline covariates included age, comorbidities, major depressive disorder, social support, performance status, and menopausal symptoms. Time-varying PROs examined included physical function, sleep disturbance, pain interference, and emotional distress (depression). We used mixed models to analyze arthralgia and PROs.

Results: Mean age was 62±10 years. The majority had active performance status (n=46) and no major depression (n=47) at baseline. Median worst pain in any joint prior to AI initiation was 1/10 (interquartile range=0–5). Adjusting for baseline covariates, greater arthralgia at 12 weeks was associated with worse physical function (β=−0.09[-0.13,-0.04]), greater pain interference (β=0.15[0.09,0.20]), and greater emotional distress (β=0.06[0.01,0.11]) but not with sleep disturbance (β=−0.01[-0.03,0.004]).

Conclusions: Our findings contribute to understanding how arthralgia relates to PROs over the first 12 weeks of AI therapy. We will build targeted AI adherence interventions from more comprehensive longitudinal PRO information.

1 Department of Obstetrics and Gynecology, University of Pittsburgh 2 Department of Epidemiology, University of Pittsburgh 3 Department of Epidemiology, University of Texas Health Science Center at Houston 4 School of Public Health, University of Texas Health Science Center at Houston P-15: Race Disparities in the Association between Preterm Birth and Maternal Cardiovascular Disease

Background and Objective: Women with preterm births (PTB) have excess cardiovascular disease (CVD) risk perhaps related to metabolic, vascular, and lipid aberrations. How risk may differ according to race is unknown. We considered that both White and Black women with PTB would have elevated CVD risk, but risk factor profiles may vary.

Methods: Eight years after pregnancy (mean age 38) we measured total, LDL and HDL cholesterol, triglycerides, apolioprotein-B, and glucose in women with prior PTB (<37 weeks, White n=135; Black n=46) or term births (≥37 weeks, White n=203; Black n=103). Women also underwent a standardized evaluation of carotid intima media thickness (IMT), brachial flow mediated dilation (FMD), and pulse wave velocity (PWV). Women with preeclampsia or small for gestational age births were excluded.

Results: White women with PTB versus term births had higher total and LDL cholesterol, Apo-B, and glucose concentrations, adjusted for age, body size and smoking (p for all <0.01). Black women with PTB versus term births had higher adjusted systolic blood pressure (beta 8.11 mmHg, p=0.04). No lipid or metabolic markers were different in Black women according to PTB history. White women with PTB<34 weeks versus term births had higher adjusted IMT (beta 0.03 mm, p=0.03).

Conclusions: White women with a history of PTB had a proatherogenic profile in the decade after pregnancy compared to those with term births, and they had evidence of early atherosclerosis. Black women with PTB had higher blood pressure, but no other vascular or metabolic aberrations were detected.

1 Department of Biomedical Engineering (Pelvic Floor Research Group), University of Michigan 2 Department of Obstetrics and Gynecology (Pelvic Floor Research Group), University of Michigan P-16: Relationship between Exposed Vaginal Length, Genital Hiatus Size and Levator Ani Muscle Status

Background and Objective: Birth-induced muscle injury is highly associated with pelvic organ prolapse but the fundamental mechanism linking them remains unknown. Theoretical work suggests that muscle damage could lead to an enlarged hiatus exposing the vagina to pressure differentials favoring cystocele development. This study tests the hypothesis that exposed vaginal length, hiatus size and muscle injury are associated.

Methods: Thirty women were selected to represent a full spectrum of AVW support including women with normal support and AVW prolapse. Each underwent supine, static and dynamic MR imaging. On midsagittal slice at maximum Valsalva, the curved length of the exposed AVW from the distal urethra to the point of perineal body contact and hiatus size from the pubic bone to the perineal body were measured. Levator ani muscle injury was scored into “No Defect,” “Minor Defects,” and “Major Defects” based on the axial MR scans. Polynomial regression analyses and ANOVA were used for statistical analysis.

Results: There is a strong quadratic relationship between the exposed vaginal length and hiatus size (R²=0.87, p<0.001). Women with major defects have significantly larger hiatus (major: 4.7±1.5 cm vs. no defect: 2.6±1.2 cm, p=0.007; major vs. minor: 3.0±1.4 cm, p=0.021) and longer exposed vaginal length (major: 4.2±2.6 cm vs. no defect: 1.2±1.7 cm, p=0.019; major vs. minor: 1.4±2.3 cm, p=0.025) than women with minor and no defects.

Conclusion: Exposed AVW length increases with increasing hiatus size. Major defect results in larger hiatus size and longer exposed vaginal length at the maximum Valsalva.

1 Boston University 2 University of New England 3 Washington University in St. Louis 4 University of Alabama in Birmingham 5 University of Utah P-17: Association between Low Activity Levels and Falling in Women and Men with Parkinson's Disease

Background and Objectives: Falling contributes to greater levels of disability. Low activity levels may contribute to falling. Women with Parkinson's are less active than men, potentially contributing to falling. The objective is to examine (1) whether lower activity was associated with falls among adults with Parkinson's and (2) whether the association between activity and falls differed among men and women.

Methods: Eighty-eight women and 121 men with Parkinson's participated in a multi-center longitudinal study. Fall history and activity level were evaluated at baseline and 6 months later. Mann-Whitney tests (α=0.05) were used to compare activity levels between fallers and non-fallers. Logistic regression analysis was used to assess associations between activity level and gender with falls.

Results: Forty-one women (49%) and 53 men (44%) fell over a 6-month period. Fall rates did not differ by gender (p=.58). Baseline median activity levels were significantly lower (p<.05) among fallers than non-fallers among women (101.2 vs. 147.0) and men (109.1 vs. 166.9). Men had higher levels of activity than women for both fallers (p=.08) and non-fallers (p=.10). Women with low baseline activity levels were three times as likely to fall compared to women with high activity levels (OR=2.91, 95% CI=1.2–7.1). This relationship was not significantly different (p=.69) in men (OR=2.31, 95% CI=1.1–4.9).

Conclusion: Approximately half the sample with Parkinson's fell over a 6-month period. Both women and men fallers were less active than non-fallers. These results suggest that low activity levels contribute to falling.

Acknowledgments: Funding for this project was provided by NIH K12 Building Interdisciplinary Research Careers in Women's Health (HD43444), the Davis Phinney Foundation, and the Parkinson's Disease Foundation.

1 Department of Psychiatry and Psychology, Mayo Clinic 2 Department of Neurology, Emory University School of Medicine P-18: Effects of Behavioral Intervention on Caregivers of Individuals with Mild Cognitive Impairment

Background and Objectives: Caregivers of individuals with mild cognitive impairment (MCI) experience significant distress and decreased quality of life (QOL) similar to that experienced by caregivers of dementia patients, but few studies have examined the effects of intervention programs on caregiver burden in MCI. Women are disproportionately affected, i.e., 65% of persons with Alzheimer's disease (AD) and 60% of caregivers for persons with AD are women. The objectives were to determine if a memory compensation program, Healthy Action to Benefit Independence and Thinking (HABIT), could increase self-efficacy in individuals with amnestic MCI that could lead to improvement in functional status and less dependence on caregivers, and to see if men and women caregivers are differentially affected.

Methods: HABIT, a multicomponent, 2-week, behavioral intervention program, was delivered to individuals with MCI and their caregivers. Self- and informant-based mood and perceived burden survey data were collected at baseline, at intervention end, and 8 weeks post-intervention in 38 caregivers (18 women, 20 men; mean age 69) and MCI cases (26 women, 12 men; mean age 75).

Results: MCI cases reported significant improvement in QOL (p=.001) through 8-week follow-up even though an initial improvement in functional status was not sustained. Trends toward improvement in perceived burden (p=.056) and QOL (p=.081) were noted by caregivers. Further, while men reported no change from baseline to 8 weeks post-intervention (t(19)=1.56, p=.14), women reported decreased burden (t(17)=2.89, p=.01).

Conclusions: Programs such as HABIT can help MCI cases and their caregivers, particularly women, adjust to the disease and its consequences on well-being.

1 Department of Public Health Sciences, University of California Davis 2 Department of Epidemiology, University of Pittsburgh 3 Rush University Medical Center, University of Minnesota P-19: The Relationship of Violence to Weight Change, Pre-diabetes and Diabetes: Longitudinal Analyses from the Study of Women's Health Across the Nation (SWAN)

Background and Objective: Several studies indicate that violence against women is associated with poorer health, as well as death. However, few studies have examined the longitudinal effects of violence during midlife on the risk of obesity and diabetes. The objective is to investigate the associations of violence with changes in weight, waist circumference, incident pre-diabetes and diabetes.

Methods: Ten years of data from the longitudinal cohort of women enrolled in the Study of Women's Health across the Nation (SWAN) were collected. An ethnically/racially and socially diverse group of 2,208 women between the ages of 42 and 52 years at baseline who did not have diabetes or pre-diabetes at baseline comprised the study sample. Data were collected annually for 10 years on violence, health outcomes and confounding variables, and annually collected fasting blood samples analyzed for blood glucose levels to detect the incidence of pre-diabetes and diabetes.

Results: At baseline more African-American women reported violence than Caucasian, Chinese and Japanese women. In bivariate analyses, being separated or divorced, smoking, reporting depressive symptoms and having no health insurance were positively associated with violence. Baseline violence was positively associated longitudinally with incident pre-diabetes (p=.03) and nonsignificantly with incident diabetes (p=.06); changes in weight and waist circumference were not associated with violence.

Conclusion: Violence in middle-aged women was associated with incident pre-diabetes; and while non-significant, results indicated that incident diabetes may be related to violence. Our findings suggest that violence in midlife may have an adverse effect on women's health, specifically related to pre-diabetes and diabetes.

1 Obstetrics, Gynecology, and Reproductive Sciences, Division of Urogynecology, University of Pittsburgh 2 Center for Research on Healthcare, Department of General Internal Medicine, University of Pittsburgh P-20: The Impact of Pelvic Organ Prolapse on Emotional Well-Being

Background and Objective: Women's emotive responses to having pelvic organ prolapse are poorly studied. Better understanding may lead to improved patient-provider communication and outcomes. The objective of the study was to describe emotions experienced by women with prolapse before and after treatment.

Methods: Forty-two women with symptomatic prolapse participated in semi-structured focus groups or one-on-one phone interviews. The facilitator stimulated discussion about prolapse and emotional well-being by asking open-ended scripted questions. All groups and interviews were audiotaped and transcribed. Two members of the study team utilized a grounded theory approach to code the transcripts. Major themes were identified throughout this process. The codebook was amended until thematic saturation was achieved.

Results: Subject mean age was 60.7±10.5 years. Twenty-five participated in focus groups and 17 in phone interviews. Women expressed annoyance, frustration, anger, fear, uneasiness, sadness, depression, anxiety, and feelings of getting older or “falling apart” as emotions associated with prolapse. Women expressed hope at the thought of surgery, and improvement in their emotional well-being after surgery. Other women struggled with “negative” emotions associated with recovering from surgery, fear of recurrence, or with persistent symptoms.

Conclusions: Women experience significant “negative” emotions associated with prolapse that appear to impact emotional well-being. Planned surgery can provide a sense of hope and improvement for some women. Postoperative followup will help us determine if surgical correction has a sustained improvement on emotional well-being related to prolapse. A better understanding of the range of women's emotions surrounding prolapse may help providers better meet patients' needs and improve patient-provider communication.

1 Division of Maternal-Fetal Medicine, Duke University School of Medicine 2 Division of Cardiology, Duke University School of Medicine 3 Department of Cell Biology, Duke University School of Medicine 4 Department of Molecular Genetics and Microbiology, Duke University School of Medicine P-23: MSH-releasing Inhibitor Factor (pro-leu-gly-NH2) Acts as an Allosteric Activator for Oxytocin-induced MAPK Signaling but Does Not Affect Uterine Contractility

Background and Objective: MSH-releasing inhibitor factor (pro-leu-gly-NH2) is an oxytocin fragment, containing the last three amino acids of the hormone oxytocin. The oxytocin receptor (OXTR) is a G-protein coupled receptor that is capable of both G-protein and β-arrestin-mediated signaling. Oxytocin-induced β-arrestin signaling occurs through the MAPK pathway. We determined whether the peptide, pro-leu-gly-NH2 affects oxytocin-mediated G-protein and β-arrestin-mediated signaling.

Methods: HEK-293 cells expressing the OXTR were stimulated with oxytocin at concentrations of 1nM to 1μM with or without pretreatment with pro-leu-gly-NH2. MAPK activation was measured with Western blot for phospho-ERK following oxytocin stimulation for five minutes at each concentration. To determine if pro-leu-gly-NH2 increased oxytocin-induced uterine contractions, uterine muscle strips from wild-type mice were suspended in a tissue organ bath and stimulated with oxytocin with or without pretreatment with pro-leu-gly-NH2 and the contraction responses measured.

Results: Pretreatment of HEK-293 cells expressing the OXTR with pro-leu-gly-NH2 increased oxytocin-induced phospho-ERK signaling compared to cells not receiving pro-leu-gly-NH2. Pretreatment of uterine muscle strips from wild-type mice did not affect oxytocin-induced uterine contractions. Stimulation of cells or uterine muscle strips alone with pro-leu-gly-NH2 did not activate MAPK signaling or induce uterine contractions.

Conclusions: The oxytocin fragment pro-leu-gly-NH2 acts as an allosteric activator for oxytocin-induced MAPK signaling, but does not increase G-protein-mediated oxytocin-induced contractions. Further work is needed to understand the molecular conformations of the OXTR that produce these findings.

1 Division of Surgical Oncology, Virginia Commonwealth University School of Medicine 2 Office of Academic Affairs and Department of Community Health, Boonshoft School of Medicine, Wright State University 3 Careers in Medicine, Association of American Medical Colleges P-24: Women Physicians: Choosing a Career in Academic Medicine

Background and Objective: The development of opportunities and support of women researchers is important in women's health. Academic women physicians are an important component of women researchers. The pathway to a career in academic medicine (AM) is poorly understood despite recent attempts to understand the complex process of physician career development. This study attempted to better define how, when and why women choose a career in AM.

Methods: Fifty-three women physicians in AM were interviewed regarding how, when, and why they chose this career path. Transcripts were thematically analyzed by project investigators individually and then collectively. Emerging themes in the data were objectively identified.

Results: Academic medicine: Why? Five themes—Aspects of Academic Health Center Environment (14 subthemes); Fit (5 subthemes); People (7 subthemes); Exposure (4 subthemes) and Clinical Medicine. How? Five themes—Change in Specialty; Parental Influence; Dissatisfaction with Former Career; Decision-Making Styles (6 subthemes); Emotionality (2 subthemes). When? Four themes—Practicing Physician; Fellowship; Resident; Medical Student.

Conclusion: Choosing a career in AM is greatly influenced by the environment in which one trains and by people—be they faculty, mentors, role models, or family. An interest in teaching is a primary theme associated with women choosing a career in academic medicine. Many women physicians entering AM chose this career path in fellowship or after. Many became more aware of AM as a career path during residency and their decision was solidified during fellowship. For many women, choosing AM was not necessarily an active, planful decision; rather it was serendipitous or circumstantial.

1 Center for Women's Infectious Diseases, Washington University in St. Louis 2 Infectious Diseases Division, University of Miami P-25: An Integrated Metabolomic Profiling Approach to E.coli Urinary Tract Infections

Background and Objectives: Urinary tract infections (UTIs) caused by uropathogenic E. coli (UPEC) disproportionately affect women. Metabolomic profiling of patients offers the prospect of differentiating disease states and better understanding pathophysiology based upon metabolic pathway activities. To date, few studies have been performed to characterize infectious disease-associated pathway changes. The objective of this study was to identify metabolic pathway changes in a population with recurrent urinary tract infection.

Methods: The optimized liquid chromatography-mass spectrometry method involves minimal sample preparation and uses a Fused-Core^® phenyl-hexyl column interfaced with a hybrid triple quadrupole/linear ion trap instrument operating in negative ion ESI mode. PCA analysis of the approximately 2,300 features revealed this method to be reproducible and capable of distinguishing patients by diagnostic group. Multiple supervised chemometric analyses revealed substantial metabolomic shifts between control and urinary tract infection patients.

Results: A consensus scoring strategy, while conservative, identified nine metabolites that distinguish E. Coli UTI from control patient populations. These metabolites are biologically plausible and suggest multiple infection-associated biochemical processes.

Conclusions: This metabolomic approach suggests a robust strategy for distinguishing metabolomic shifts among patients with urological infectious diseases. Pathways identified by this method suggest new diagnostic, therapeutic and prophylactic approaches to UTI in women.

1 Division of Cardiovascular Medicine, Oregon Health & Science University 2 Portland VA Medical Center P-28: Sexual Dimorphism of Cardiomyocyte Size in Adult but Not Fetal Sheep

Background and Objective: The capacity for cardiomyocyte enlargement (necessary for adaptive cardiac growth) influences progression of cardiac disease, a process that differs between men and women. The objective is to determine how cardiomyocytes differ between the sexes in adulthood, and if those differences are present near birth.

Methods: Adult (male n=4; female n=4) and near-term fetal (138±1 of 145 days of gestation; male n=8; female n=9) sheep hearts were enzymatically dissociated. Cellular dimensions were measured from fixed isolated cardiomyocytes. In cells from fetal hearts, proliferative index (measured by Ki-67 immunoreactivity) and maturational index (measured by binucleation) were quantified.

Results: Adult male sheep were significantly larger than adult females (105±2 vs. 69±8 kg) and had larger hearts (444±50 vs. 301±40 g). In adult sheep, males had significantly longer (171±9 vs. 139±7 μm) and wider (45±3 vs. 38±4 μm) left ventricular cardiomyocytes, and longer (176±15 vs. 142±20 μm) right ventricular cardiomyocytes. In contrast, neither body weight, heart weight, nor cardiomyocyte dimensions were found to be different between male and female fetuses. Additionally, the proliferative index and cardiomyocyte maturation were similar between male and female fetuses.

Conclusions: Differences in cardiomyocyte dimensions in adult sheep reflect sexual divergence in body (and consequently heart) weights. In contrast, body weight, heart weight, and cardiomyocyte parameters are similar between males and females in late gestation. These results suggest that cardiomyocyte growth in the fetus does not presage adult sexual dimorphisms. Sex differences in cardiomyocyte morphology that are important for adaptive growth in the adult may arise during postnatal maturation.

1 Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, Beijing, China 2 Tulane University School of Medicine, Beijing, China 3 Cardiovascular Institute and Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, and Chinese National Center for Cardiovascular Disease Control and Research, Beijing, China 4 Department of Epidemiology, University of Texas School of Public Health 5 Division of Biostatistics, Washington University School of Medicine, St. Louis 6 Academy of Medical Sciences, Shandong, China 7 Shandong Center for Diseases Control and Prevention, Shandong, China 8 Department of Medicine, Xi'an Jiaotong University, Shanxi, China 9 Office of the President, Loyola University Health System and Medical Center, Maywood, IL P-29: The Estrogen Receptor 1 (ESR1) Gene and Blood Pressure Salt Sensitivity

Background and Objective: Previous reports have documented increased blood pressure (BP) salt sensitivity in women compared to men. The objective of the study is to examine the association between 799 single nucleotide polymorphisms (SNPs) in 44 genes involved in sex hormone biosynthesis, bioavailability and metabolism for their association with BP responses to sodium intervention in men and women.

Methods: A 7-day low-sodium (51.3 mmol sodium/day) followed by a 7-day high-sodium feeding study (307.8 mmol sodium/day) was conducted among 1,906 Han, Chinese participants. Nine BP measurements were obtained at baseline and the end of each intervention period using a random-zero sphygmomanometer.

Results: Among men, absolute BP responses to sodium interventions decreased with the number of minor alleles of ESR1 markers rs9340844, rs9397453 and rs9383951. For example, men with genotypes C/C, C/T, and T/T of rs9397453 had respective mean DBP responses (95% CI) of: −2.67 (−3.13, −2.22), −1.23 (−1.98, −0.48), and 0.08 (−2.31, 2.47) mmHg to low-sodium intervention (p=0.0001; adjusted-p=0.04); and 1.46 (1.03, 1.89), 0.19 (−0.54, 0.91), and −1.10 (-2.82, 0.61) mmHg to high-sodium intervention (p=0.0002; adjusted-p=0.04). In addition, mean SBP responses (95% CI) were: −5.70 (−6.19, −5.20), −4.34 (−5.37, −3.31), and −2.65 (−5.15, −0.16) mmHg, respectively, for low-sodium intervention (p=0.002; adjusted-p=0.17); and 4.56 (4.12, 4.99), 3.47 (2.63, 4.30), and 1.97 (−0.49, 4.43) mmHg, respectively, for high-sodium intervention (p=0.003; adjusted-p=0.40). Similar, non-significant trends for these SNPs were observed among women.

Conclusions: We identified strong, consistent relationships between genetic variants in the ESR1 gene and salt sensitivity in men. These findings support a role for sex-hormones in the etiology of this complex trait.

1 Breast Cancer Survivorship Center, University of Kansas Cancer Center 2 Physical Therapy and Rehabilitation Sciences, University of Kansas Medical Center 3 Mid-America Cardiology, University of Kansas Hospital, University of Kansas Medical Center P-30: Cardiopulmonary Testing and Correlates of Breast and Cardiac Biomarkers in Breast Cancer Survivors

Background and Objective: Cardiac events are the leading cause of death in long-term breast cancer (BrCa) survivors next to primary malignancy recurrence. BrCa treatment may increase the risk of cardiac dysfunction including other factors such as weight gain and genetics. Low cardiorespiratory (CR) fitness level (measured as VO_{2 peak}) is a measure of cardiac dysfunction and contributes to cardiovascular, BrCa, and all-cause mortality. Strategies preventing cardiovascular disease (CVD) in this high-risk population are necessary. CR testing may be a useful biomarker to identify asymptomatic cardiac dysfunction. The objective of the study is to explore biomarkers of cardiovascular and BrCa risk in cancer survivors.

Methods: The study included 30 postmenopausal female BrCa survivors, 6 months to 10 years post treatment with cardio-toxic chemotherapy, who had +/− left chest wall radiation along with ≥2 CVD risk factors, completed maximal and submaximal CR testing (treadmill), along with an echocardiogram (ECHO) or Multigated Acquisition Scan (MUGA), serum biomarkers, and quality of life measures.

Results: BrCa survivors were 50.5±5.6 year old and approximately 4.5 years from initial treatment. Maximal and submaximal CR tests were VO_{2 peak}=25.4 +/− 5.3 ml/kg/min and VO_{2 peak}=20.5 +/− 4.3 ml/kg/min, respectively. Eighty-seven percent of participants were in a poor CR fitness category. ECHO/MUGA were normal (LVEF=60.5 +/− 5.0). Poor CR fitness was associated with higher weight, BMI, and percent body fat, use of anti-hypertensive and statin medications, and exposure to and duration of an aromatase inhibitor.

Conclusion: CR fitness may serve as a reliable, modifiable biomarker of CVD and BrCa risk and be used as a target in future cancer survivorship trials.

1 Beth Israel Deaconess Medical Center, Harvard Medical School 2 Massachusetts General Hospital, Harvard Medical School 3 Howard Hughes Medical Institute 4 Gulbenkian Program for Advanced Medical Education, Portugal P-31: Natural Killer Cells and Preeclampsia

Background and Objectives: Preeclampsia is a leading cause of maternal and perinatal morbidity and mortality worldwide. Recent evidence suggests that decidual Natural Killer cells (dNKs) at the maternal-fetal interface may play an important role in the regulation of maternal spiral artery remodeling. This process is compromised in preeclampsia and believed to be important for its pathogenesis. In contrast to peripheral blood NK cells (pNKs), which are cytotoxic and non-angiogenic, first trimester human dNKs have been shown to secrete proangiogenic molecules and to have reduced cytotoxicity. The objective is to convert human pNKs into dNK-like cells using specific factors.

Methods: pNKs were cultured under hypoxia in the presence TGF-β1 and tested for expression of dNK cell surface markers, for secretion of proangiogenic molecules by ELISA, for angiogenic potential in tube formation assays and for cytotoxic activity on K562 target cells.

Results: pNKs cultured under hypoxia secreted VEGF, a potent proangiogenic molecule, and induced tube formation by Human Umbilical Vein Endothelial Cells. Under hypoxia in the presence of TGF-β1 they also expressed the dNK cell marker CD9 and showed reduced cytotoxicity. An animal model is being established to study in vivo the spiral artery remodeling capacity of these in vitro generated dNK-like cells.

Conclusions: A combination of hypoxia and TGF-β1 converts pNKs into proangiogenic non-cytotoxic cells with similarities to dNKs. These observations may have applications in the development of novel therapeutic approaches for preeclampsia and related disorders.

1 Oregon Health & Science University School of Nursing 2 Oregon Health & Science University School of Medicine P-34: Gender Differences in Advanced Heart Failure Symptom Profiles

Background and Objective: Advanced heart failure (HF) is characterized by symptoms that limit daily life despite optimal therapy. The objective of the study is to identify gender differences in symptom profiles in adults with advanced HF.

Methods: We analyzed data on 39 women and 69 men in an ongoing study of advanced HF symptoms. Physical symptoms were measured with the HF Somatic Perception Scale, daytime sleepiness was measured with the Epworth Sleepiness Scale, depression was measured with the Patient Health Questionnaire, and hostility and anxiety were measured with the Brief Symptom Inventory. Latent class mixture modeling was used to identify distinct profiles among these five symptoms scores in separate models stratified by gender.

Results: Two symptom profiles were identified for women. Women with a high symptom burden profile (36%) reported worse daytime sleepiness, more depressive symptoms, and greater anxiety and hostility compared with women with the low symptom burden profile (64%) (all p <.01); there was no difference in physical symptoms between women with high vs. low symptom burden. Three symptom profiles were identified for men (21.7% had high, 17.4% had medium, and 60.9% of men had low symptom burden). There was a graded increase in severity of physical symptoms, daytime sleepiness, depressive symptoms, anxiety, and hostility across the three symptom profiles for men (all p<.01).

Conclusions: Physical symptoms, often viewed as hallmarks of HF, are not helpful in identifying women with greater symptom burden in HF. Psychological distress and sleep disturbances are indices of greater distress in women with advanced HF and therefore warrant assessment and treatment.

1 Department of Obstetrics and Gynecology, Boston University 2 Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical School 3 Department of Epidemiology, Harvard School of Public Health 4 Channing Laboratory, Brigham and Women's Hospital, Harvard Medical School 5 Department of Medicine, Brigham and Women's Hospital, Harvard Medical School 6 Department of Statistics, North Carolina State University 7 Department of Biostatistics, Harvard School of Public Health 8 Department of Environmental Health Sciences, University of Michigan School of Public Health 9 Department of Environmental Health, Harvard School of Public Health 10 Department of Environmental and Occupational Health, School of Public Health and Health Services, The George Washington University 11 Vincent Memorial Obstetrics and Gynecology Service, Andrology Laboratory and, In Vitro Fertilization Unit, Massachusetts General Hospital, Harvard Medical School P-36: Association of Hexachlorobenzene (HCB), Dichlorodiphenyltrichloroethane (DDT), and Dichlorodiphenyldichloroethane (DDE) with In Vitro Fertilization (IVF) Outcomes.

Background and Objective: Hexachlorobenzene (HCB), dichlorodiphenyltrichloroethane (DDT), and dichlorodiphenyldichloroethane (DDE), are persistent chlorinated pesticides with endocrine activity that may adversely affect the early stages of human reproduction. The objective is to determine the association of serum levels of HCB, DDT, and DDE with implantation failure, chemical pregnancy, and spontaneous abortion (SAB) in women undergoing IVF from 1994 to 2003.

Methods: Levels of HCB and congeners of DDT and DDE were measured in serum collected during the follicular phase. Multivariable-adjusted statistical models accommodating multiple outcomes and multiple cycles per woman were used to estimate the relation between serum pesticide levels and IVF outcomes.

Results: A total of 765 women with a mean age (SD) of 35.4 years (4.2) at enrollment contributed 827 IVF cycles. All samples had detectable levels of HCB, DDT, and DDE, with median levels of 0.87 ng/g serum for HCB, 1.22 ng/g serum for total DDT, and 1.09 ng/g serum for p,p′-DDE. Compared to the lowest quartile of HCB, the lipid- and multivariable-adjusted odds for failed implantation was significantly elevated for those with higher HCB quartiles [Adjusted Odds Ratio (aOR) for Q2: 1.71, 95% confidence interval (CI): 1.03, 2.82, aOR for Q3: 2.30, 95% CI: 1.39, 3.81; aOR for Q4: 2.32, 95% CI: 1.39, 3.90) and showed a significantly increasing trend (p=0.001). No statistically significant associations were observed between DDT/DDE and IVF outcomes or between HCB and chemical pregnancy or SAB.

Conclusions: Serum HCB concentrations were significantly associated with failed implantation among women undergoing IVF.

1 Department of Medicine, Penn State College of Medicine 2 Department of Public Health Sciences, Penn State College of Medicine P-39: “I Just Keep My Antennae Out”—How Primary Care Physicians Respond to Intimate Partner Violence in Rural Communities

Background and Objectives: In rural communities, limited health services, poverty and social isolation compromise healthcare access, producing unique challenges for healthcare providers caring for women exposed to intimate partner violence (IPV). Using qualitative methods, we assessed the opinions of primary care physicians caring for rural women with regard to IPV identification, the scope and severity of IPV as a health problem, physician's response to IPV, and barriers to optimized IPV care.

Methods: Based on thematic saturation, 19 internist, family practice, and OB-GYN physicians were interviewed. Using grounded theory, interview transcripts were analyzed for major themes.

Results: Most physicians did not practice routine inquiry for IPV due to competing time demands, lack of provider training, low confidence in referral services, and concern that inquiry would harm the patient-doctor relationship. They identified IPV largely among patients with symptoms of mood, anxiety or somatic disorders. Response to IPV included reassurance, validation, danger assessment, safety planning, referral to services and follow-up planning. Physicians perceived that barriers preventing rural women from seeking care for IPV included traditional gender roles, lower education, economic dependence on an abusive partner, low self-esteem, and patient reluctance to discuss IPV. They also perceived inadequate mental health and other referral services for IPV. To overcome barriers, physicians created a “safe sanctuary” to discuss IPV and suggested improved public health education and referral services.

Conclusions: Interventions to improve IPV-related care in rural communities should address physician and patient comfort discussing IPV, through both provider training and community education programs.

1 Department of Epidemiology and Community Health, Virginia Commonwealth University School of Medicine 2 Department of Social and Behavioral Health, Virginia Commonwealth University School of Medicine P-40: Are Depression and Frailty Overlapping Syndromes in Later Life? A Latent Variable Analysis

Background and Objective: Depression and frailty are common geriatric conditions. However, it is unclear to what extent these represent independent rather than overlapping syndromes. The objective of the study was to examine the degree to which depression and frailty indicate overlapping constructs.

Methods: Data come from 2004/5 wave of the Baltimore Epidemiologic Catchment Area Study and analysis is limited to participants aged 40 and older with complete data on frailty and depression indicators (N=683). Depression was measured using the Diagnostic Interview Schedule and frailty was indexed by modified Fried criteria. Exploratory and confirmatory latent class analyses (LCA) were used to assess the degree to which depression and frailty syndromes co-occur within individuals. A latent Kappa coefficient was calculated from joint class distributions to measure chance-corrected latent agreement between the two constructs.

Results: LCA analyses supported a three-class solution for depression, with 3.9% categorized as severe depression, 19.6% as mild/moderate depression, and 76.5% as not depressed. A two-class solution was the best fit for frailty, with 8.8% classified as frail. Modeling the indicators of depression and frailty as a single latent construct provided comparable fit to treating them as separate but correlated latent constructs. There was a high degree of correspondence between the latent depression and frailty constructs (Kappa=0.66).

Conclusions: Results suggest that depression and frailty are interrelated concepts, and that the operational criteria of these constructs identify highly overlapping subpopulations. This construct overlap has implications for understanding factors that contribute to the etiology and prognosis of depression and frailty in later life.

1 Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina 2 Department of Anesthesia, Stanford University P-41: Effects of Psychosocial Stress on Brain Activity in Cocaine-Dependent Men and Women

Background and Objective: Sex/gender differences in substance-dependent individuals are well documented. However, little is known about the neural correlates that underlie stress-induced relapse. Brain activity was compared between cocaine-dependent men and women during the Montreal Imaging Stress Task (MIST). The purpose of the study was to compare brain activity between cocaine-dependent men and women during a psychosocial stressor.

Methods: Subjects (n=10 men; n=10 women) were placed in a 3T scanner and exposed to the MIST. A block design of three conditions (rest/control/stress) was utilized. Results are based on data from 16 subjects. Images were thresholded using clusters determined by Z>2.3 and a corrected cluster significance threshold of p=0.05. Mean group images were thresholded using clusters determined by a corrected threshold of p=0.05. Subjective ratings of stress and craving were obtained at the end of each run.

Results: The MIST produced a significant increase in stress in cocaine-dependent men. However, the magnitude of subjective stress following the second run of the MIST was greater in women than men. There was no effect of the MIST on craving. The math task elicited activity in both men and women in the prefrontal, anterior cingulate and parietal cortices. The stress condition increased activity in the precuneus, anterior and posterior cingulate cortices of women, with no change observed in men.

Conclusion: The MIST produces sex/gender-specific changes in measurable brain activity related to stress, suggesting that it may be an ideal paradigm for exploring the neural correlates of stress in substance-dependent individuals.

1 Public Health & Preventive Medicine 2 Molecular Biology & Immunology, Oregon Health & Science University P-42: CMV Seropositivity as a Risk Factor for Frailty and Osteoporosis among Older Adults in the U.S. NHANES III Study

Background and Objective: Frailty and osteoporosis are significant predictors of morbidity and mortality in the elderly. Associations with inflammatory markers implicate chronic inflammation as a risk factor for both outcomes, and cytomegalovirus (CMV) may be a source of this inflammation. The objective was to evaluate CMV seropositivity as a risk factor for frailty and bone mineral density (BMD) and to determine whether associations differ by sex.

Methods: We included 5,161 participants≥60 years old from the U.S. National Health and Nutrition Examination Survey III who had CMV IgG serology and data on the outcome measures of interest. Frailty categories were based on the number of the following criteria: BMI ≤18.5 kg/m², slow walking, and self-reported weakness, exhaustion, or low physical activity. Hip BMD was measured using dual-energy X-ray absorptiometry. For frailty, prevalence ratios (PR) and 95% confidence intervals (CI) were calculated. For BMD, linear regression was used. Many potential confounders were considered.

Results: CMV seroprevalence was 91%, increased with age, and was higher in women (93%) than in men (88%). CMV seropositivity was associated with frailty in men (age-adjusted PR for frail vs. robust: 2.3 [CI: 1.1–4.9]) but not women (PR: 1.6 [CI: 0.8–3.3]). Mean BMD was 4% lower for CMV seropositive men (p=0.01) and women (p=0.02) than for seronegative participants and was consistent after multiple adjustment.

Conclusions: CMV seropositivity was highly prevalent but not independently associated with frailty in this large survey. However, mean BMD among CMV seropositive participants was lower in both sexes. CMV may play a role in osteoporosis.

1 Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh School of Medicine 2 Magee-Womens Research Institute 3 Department of Epidemiology, University of Pittsburgh 4 Epidemiology Data Center, University of Pittsburgh 5 Department of Medicine and Center for Research on Health Care, University of Pittsburgh 6 Department of Psychiatry, University of Pittsburgh 7 Department of Epidemiology, University of Pittsburgh 8 Department of Psychology, University of Pittsburgh 9 Department of Women's Studies, University of Pittsburgh 10 Womens' Behavioral HealthCARE, Western Psychiatric Institute and Clinic, University of Pittsburgh 11 Department of Obstetrics and Gynecology, Albert Einstein College of Medicine P-43: Symptoms of Depressed Mood, Disturbed Sleep and Sexual Problems in Midlife Women: Cross-sectional Data from the Study of Women's Health across the Nation (SWAN)

Background and Objective: Depression is associated with sleep disturbance and sexual problems in midlife women. These symptoms are commonly reported, but little has been reported on their covariance in individual women. The objective was to evaluate the interrelatedness of depressed mood, disturbed sleep, and sexual problems in midlife women and to characterize these women's demographic, psychosocial, and clinical characteristics.

Methods: SWAN is a multi-site, multi-ethnic observational cohort study of the menopausal transition across the United States. Depression was assessed using the Center for Epidemiological Studies Depression Scale (CES-D). Sleep disturbance was defined as nighttime or early awakenings, or difficulty falling asleep in the past 2 weeks. Sexual problems were defined as self-reported problems in any one of five domains: desire, arousal, satisfaction, orgasm and vaginal dryness. Logistic regression models were used to estimate the association of the demographic, psychosocial and clinical characteristics with the symptom complex.

Results: Five percent (N=90 of total N=1,716) experienced all three symptoms. In multivariable models, these women were more likely to have lower incomes, less education, be surgically postmenopausal (OR 3.37 [95% CI: 1.56, 7.26]) or late peri-menopausal (OR 1.99 [95% CI: 1.06, 3.75]), have lower self-reported health status, more stressful life events, and lower social support. No effect was noted for race/ethnicity or hormone therapy.

Conclusions: In this cross-sectional analysis of the SWAN cohort, 5% of women were affected by the complex of symptoms of depressed mood, disturbed sleep and sexual problems. Late-perimenopause, surgical menopause, poor social support, and stressful life events were associated with having the symptom complex.

1 Division of Pediatric Rheumatology, Medical University of South Carolina 2 Division of Pediatric Genetics, Medical University of South Carolina 3 Division of Rheumatology and Immunology, Medical University of South Carolina 4 Division of Rheumatology, Medical University of South Carolina 5 Division of Pediatric Rheumatology and Immunology, Medical University of South Carolina P-46: Association of NMDA Receptor Antibodies and Neurocognitive Dysfunction in Pediatric Lupus Patients

Background and Objective: Systemic Lupus Erythematosus (SLE) is an episodic, multisystem autoimmune disease that is recognized worldwide. The incidence of SLE is dramatically higher in women than in men. Approximately 1/5 of all systemic lupus starts in childhood and central nervous system (CNS) dysfunction is more common in childhood-onset SLE (cSLE). Despite these findings, the diagnosis of CNS disease in SLE remains difficult. The objective is to assess the association between elevated anti-NMDA-NR2 subunit receptor antibodies and neurocognitive dysfunction in pediatric lupus patients.

Methods: Patients diagnosed with SLE prior to age 18 were recruited. Each patient underwent formal neurocognitive testing, assessing a range of cognitive domains. The patients also underwent NMDA receptor-NR-2 subunit antibody testing.

Results: Thirteen lupus patients, ages 10–19, completed neurocognitive testing and NMDA receptor antibody testing (two males, eight African Americans, three Caucasians, two others). SLEDAI scores ranged from 0–19. Means for intellectual functioning were low (FSIQ=87.8) compared to the general population (50th percentile FSIQ=100). All other cognitive score means were in the average range, except working memory and math fluency (both low average). There were no significant correlations between neurocognitive functioning and NMDA receptor antibodies, although verbal fluency approached significance (p=.097).

Conclusions: A larger sample size with control group is necessary to make more definitive conclusions about the association of NMDA receptor antibodies and neurocognitive dysfunction in pediatric lupus patients. This study is ongoing and will include: 30 patients with cSLE, 30 patients with JIA, and 30 children born to mothers with lupus (first degree relatives).

Supported by NIH grant: 5K12JD055885-04

1 Department of Obstetrics & Gynecology, University of Texas Medical Branch 2 Department of Biochemistry and Molecular Biology and the Sealy Center for Structural Biology and Molecular Biophysics, University of Texas Medical Branch 3 Center for Biomedical Engineering, University of Texas Medical Branch 4 Department of Pharmacology & Toxicology, University of Texas Medical Branch P-47: Transport of Dexamethasone-loaded Nanoparticles across an In Vitro Model of Placental Trophoblast Cells

Background and Objectives: The administration of dexamethasone during pregnancy when treating fetal congenital adrenal hyperplasia can result in severe side effects in the mother. Nanoparticles for targeted fetal therapy may improve in utero treatment of congenital adrenal hyperplasia and reduce maternal dexamethasone exposure. These studies were carried out to compare the influence of particle size and nanoparticle surface modification on the transport of dexamethasone-loaded nanoparticles across an in vitro model of human placental trophoblast cells (BeWo b30 cells).

Methods: Nanoparticle characterization included x-ray diffraction, transmission electron microscopy, particle size, zeta potential, encapsulation efficiency, and drug release determinations. Transport studies were performed across BeWo cell monolayers cultured on Transwell^® inserts, with drug concentrations measured by HPLC. Coumarin-6 was incorporated as a fluorescent marker in nanoparticle tracking experiments.

Results: The transport of dexamethasone across BeWo cells was greater in nanoparticle formulations as compared to dexamethasone dissolved alone in the transport medium (apparent permeability 4.4×10⁻⁶ cm/s). Dexamethasone transport was dependent on nanoparticle size, as the percent of drug added to the maternal compartment reaching the fetal compartment after 2 hours was 53.8±4.8% when dexamethasone was formulated in smaller particles (146±2 nm), compared to 13.3±3.4% with larger particles (227±5 nm). Experiments with fluorescent particles confirmed the size dependence of nanoparticle transport. Increasing nanoparticle surface PEGylation (to 10% density) resulted in reduced transport rates.

Conclusions: These results highlight the feasibility of nanoparticle-based targeted fetal dexamethasone therapy as a strategy to reduce maternal side effects.

1 Department of Medicine, Section of Gastroenterology and Hepatology, University of Wisconsin School of Medicine and Public Health 2 Department of Curriculum & Instruction, University of Wisconsin-Madison 3 Division of Information Technology, University of Wisconsin-Madison 4 Division of Gastroenterology, Oregon Health & Sciences University P-48: Scenario-Based Electronic Learning to Teach Women's Gastrointestinal Health

Background and Objective: Scenario-based electronic learning (SBeL) immerses learners in a contextualized situation. SBeL can complement learning through direct patient care, especially in topics in which appropriate patient exposure is limited and training is deficient. The objective is to assess user satisfaction, perceived value, and knowledge gained in women's gastrointestinal (GI) health using SBeL.

Methods: An interactive case illustrating a woman's GI health topic was developed using the Case Scenario Builder authoring platform. Second-year medical students were recruited to play the case. Pre-play surveys assessing prior experience with SBeL, its perceived value, and knowledge of the subject of the case were administered. Post-play, satisfaction and usability of the tool was assessed. Perceived value of SBeL and subject knowledge were reassessed. Responses were recorded using Likert Scales. Descriptive statistics were calculated.

Results: Pre- and post-surveys were completed by 38 and 31 subjects, respectively. Prior experience with SBeL was low (mean score 1.49±0.6, 1=no prior use and 6=daily use). Subjects were initially neutral about its value in learning and/or becoming proficient in the following: medical facts (3.8±0.9), patient-provider interactions (3.1±1.1), trainee-supervisor interactions (3.3±1.0), 1=not valuable and 5=very valuable. Post-play surveys found subjects to be very satisfied with the tool. They also reported being more knowledgeable and prepared to handle a similar case. Ninety-four percent reported being likely to complete additional cases.

Conclusions: SBeL can be used to teach women's GI health. Although prior experience with SBeL among students is low, once exposed, they perceive it to be valuable and are interested in using it more.

1 Department of Medical Microbiology & Immunology, University of California, Davis 2 Department of Obstetrics and Gynecology, University of California, Davis 3 Department of Internal Medicine, University of California, Davis 4 Kaiser Permanente Medical Group, Sacramento P-49: Role of Gender in HIV-1 Associated Decline in Gut Mucosal Integrity

Background and Objective: Human Immunodeficiency Virus (HIV) infection is one of the leading causes of death in women in Africa. HIV infection results in loss of CD4+ T-cells in Gut-Associated Lymphoid Tissue (GALT). HIV disease progression correlates with immune activation which is higher in women compared to men. Antiretroviral therapy (HAART) effectively suppresses viral replication and stimulates immune reconstitution. HIV+ women on HAART have higher levels of T-cell activation than men on HAART. We hypothesize that compromised mucosal integrity allows for increased microbial translocation in women which stimulates systemic immune activation, which has implications for increased morbidity and disease progression in women. The objective is to compare the effects of gender on mucosal integrity in HIV+ individuals on HAART.

Methods: We compared gene expression profiles in gut mucosa and plasma levels of microbial products in male and female HIV+ participants.

Results: A greater proportion of HIV+ women on HAART had sustained levels of GALT CD4+ T-cell activation (37.4% vs. 6.5%) and CD8+ T-cell activation (42.7% vs. 12.52%) than men (p<.05). Comparison of mucosal gene expression showed down-regulation of epithelial regeneration pathways and tight junction protein ZO-1 (2-fold decrease) in women than men on HAART. Indicators of microbial translocation, such as plasma LPS, and bacterial 16S rRNA, were increased in women compared to men on HAART.

Conclusions: These preliminary data suggest that increased microbial translocation may explain why women have sustained CD4+ T-cell activation, despite effective control of viral replication. Studies investigating the effects of HAART in women should focus on the maintenance of mucosal integrity and health.

1 Department of Public Health Sciences, University of California, Davis 2 Medical Investigation of Neurodevelopmental Disorders (M.I.N.D.) Institute, University of California, Davis 3 Department of Pediatrics, University of California, Davis 4 Department of Psychiatry and Behavioral Sciences, University of California, Davis P-50: Maternal Periconceptional Folic Acid Intake and Risk for Autism Spectrum Disorders in the CHARGE Case-control Study

Background and Objective: Periconceptional folate is essential for proper neurodevelopment. Maternal folic acid intake was examined in relation to risk for autism spectrum disorder (ASD) and developmental delay (DD).

Methods: Families enrolled in the CHARGE (Childhood Autism Risks from Genetics and Environment) Study from 2003–2009 were included if their child had a diagnosis of ASD (n=429), DD (n=130) or typical development (TD, n=278) confirmed at the UC Davis M.I.N.D. Institute using standardized clinical assessments. Average daily folic acid was quantified for each mother based on dose, brands, and intake frequency of prenatal vitamins, multivitamins, folic acid-specific vitamins, other supplements, and cereal reported through parental interviews.

Results: Mean folic acid intake was significantly greater for mothers of TD children than for mothers of children with ASD in the first month of pregnancy (P1) (779.0 and 655.0 mcg, respectively, p<0.01). This difference primarily resulted from higher mean folic acid from prenatal vitamins among mothers of TD children. Women reporting average daily intake of 600 mcg folic acid or more during P1 had reduced estimated risk of having children diagnosed with ASD compared to mothers reporting less than 600 mcg (odds ratio=0.62, 95% CI 0.42, 0.92, p=0.02), and risk estimates decreased with increased folic acid (P _d=0.001). Mothers of DD children tended to report lower folic acid intake during the three months before pregnancy relative to mothers of TD children (p=0.08).

Conclusions: Folic acid in the periconceptional period may reduce the risk of ASD. Replication of these findings and investigations of mechanisms involved are warranted.

1 Radiation Oncology, Washington University School of Medicine, St. Louis 2 Molecular Imaging Center, Washington University School of Medicine, St. Louis 3 Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis 4 Radiology, University of Pittsburgh 5 Siteman Cancer Center 6 Developmental Biology, Washington University School of Medicine, St. Louis 7 Cell Biology and Physiology, Washington University School of Medicine, St. Louis P-51: Development of a Bioluminescent Mouse Model of Metastatic Cervical Cancer

Background and Objective: Metastatic cervical cancer develops in a predictable pattern that progresses through lymph node regions before involving the lungs. Previous studies in mice used fluorescence microscopy to monitor this process ex vivo. The objective is to use bioluminescence to track non-invasively mouse models of metastatic human cervical cancer in vivo.

Methods: Lentivirus was used to establish stable expression of EGFP/click beetle green (CBG) in human cervix cancer cell lines ME-180, SiHa and Caski (ATCC). To initiate in vivo growth, 10⁶ cells were mixed with media/matrigel and injected subcutaneously into the flanks of 8–12 week old female SCID mice (Charles River). Flank tumors 1 cm in size were excised and dissected under sterile conditions. Tumors were cut into 2 mm fragments and surgically implanted into the cervices of anaesthetized mice (Cairns and Hill, 2004). Serial bioluminescence and magnetic resonance imaging were used to monitor the development of primary tumors and metastatic disease.

Results: Primary cervical tumors were successfully established from all three lines and grew at predictable rates within the pelvis. MRI data suggests primary tumors extend superiorly within the uterus and abdomen. Increases in local tumor burden were associated with leukocytosis and increased circulating hematopoietic progenitor cells after 2 weeks. Metastatic disease was identified by bioluminescence in lungs at 5 weeks post implant.

Conclusions: Bioluminescence can be used to monitor the development of metastatic cervical cancer in the mouse. This model will be used to test novel imaging strategies and therapeutic interventions for cervical cancer.

1 College of Nursing, Michigan State University 2 Department of Family Medicine, Michigan State University P-53: General Health Perception of Older African-American Breast Cancer Survivors

Background and Objective: Impairment in health status is a major health concern for older African-American breast cancer survivors. The objective of this study was to examine the impact of aging on social function and general health perception of older African-American breast cancer survivors using data from the Medicare Health Outcome Survey. Other variables examined were comorbidity, socioeconomic status, and depressive symptoms.

Methods: General health perception and social functioning was examined in 189 breast cancer survivors using the SF-36 Health Survey from the Medicare Health Outcome Survey Questionnaire. Participants were divided into two age groups: ages 60–70 and 71 and older. Multilinear regression was used to determine predictors of general health perception.

Results: There were no statistically significant differences in general health perception between the two age groups. Social functioning scores were moderately high for each age group. General health perception was positively associated with social functioning, physical functioning, and comorbidity.

Conclusions: Social functioning is a potential target for efforts aimed at the maintenance/improvement of physical health for older African-American breast cancer survivors.

1 Department of Pharmacology & Cancer Biology, Duke University 2 Department of Obstetrics and Gynecology Reproductive Biology and Perinatal Lab, Duke University 3 Department of Chemistry, North Carolina Central University P-54: Smart Therapy for Uterine Fibroids Enabled by the Hyperbranched Polyglycerol Platform

Background and Objectives: Uterine fibroids are diagnosed in one-fourth of all women and contribute more than $2.1 billion annually to health care costs. Despite the high cumulative incidence of fibroids, their etiology is elusive. Hysterectomy is the only treatment to permanently eliminate fibroids. It is obvious but rarely stated that hysterectomies lead to irrevocable loss of fertility. Furthermore, even without hysterectomy, fibroids, particularly subserosal fibroids, can lead to infertility. We have suggested that fibroids develop and grow by a fibrotic process and are not true neoplasms. Clearly, more fundamental knowledge is needed regarding the molecular biology of fibroids. Smart Therapy is needed to enable local delivery of a drug or a combination of drugs that will ablate or inhibit growth of fibroids and prevent hysterectomy. To this end, we have designed a drug delivery system that gels under physiological conditions and degrades over time to release entrapped therapeutic agents. The objectives are to (1) evaluate the degradation properties of the drug delivery system and (2) entrap model drugs in our multifunctional, biocompatible, injectable, thermoresponsive, biodegradable nanocarrier drug delivery system and evaluate drug release.

Methods: Solutions of the drug delivery system in PBS (16.7 wt %) were poured into 2 ml vials and incubated for different periods of time at 37 °C. At predetermined times, samples were quenched in liquid nitrogen and frozen until needed for further studies. The frozen samples were lyophilized and the molecular weights of the resulting products were determined by GPC. A model drug, pirfenidone, was dissolved in the 16.7 wt% drug delivery system solution (0.2 mg/mL in 10 mM PBS, pH 7.4) and incubated at 37 ^oC to create a gel. After 5 min, 3.5 mL of release medium (0.02 wt% NaN3, 0.2 wt% Tween 20, 10 mM PBS) at 37 ^oC was added to the gel. Fresh medium were replaced at designated sampling intervals and subsequently analyzed by optical spectroscopy.

Results: Within the first 16.5 hours, the drug delivery system had lost 95% of its molecular weight. Spectral changes in the proton nuclear magnetic resonance spectroscopy confirmed hydrolytic degradation of delivery system. Pirfenidone release from the delivery system measured by optical spectroscopy showed bimodal kinetics and was consistent with the degradation studies.

Conclusions: Our drug delivery system is biocaptible and has attributes (injectible liquid that solidifies below 35 ^oC before degrading) to provide controlled local delivery of antifibrotic drugs. This localized, non-surgical intervention for fibroids is the first of its type and may prove to be a viable alternative to hysterectomies.

1 University of Minnesota, Minneapolis 2 National University of Singapore, Singapore P-55: Mitochondrial Copy Number is Associated with Breast Cancer Risk

Background and Objective: Mitochondrial DNA (mtDNA) copy number variation is common in breast tumors and reflects the effect of interaction between hereditary and environmental factors. Data on mtDNA copy number and breast cancer risk in prospective cohorts is lacking. We evaluated the association between mtDNA copy number in peripheral blood and breast cancer risk among female participants of the Singapore Chinese Health Study.

Methods: We conducted a nested case-cohort study of 262 breast cancer cases (91 incident cases with pre-diagnostic blood and 171 prevalent cases with post-diagnostic blood) and 324 control women who were free of breast cancer at blood draw. The mtDNA copy number was measured using real time PCR. Analysis of covariance and logistic regression methods were employed to examine the association between mtDNA copy number and breast cancer risk.

Results: mtDNA copy number was significantly lower among incident breast cancer cases than controls (geometric mean: 0.50 vs. 0.58, p=0.03) after adjustment for potential confounders. Among prevalent breast cancer cases, mtDNA copy number increased with decreasing time interval between cancer diagnosis and blood draw: 0.60 in 45 cases with <6 months, 0.58 in 69 cases with 6–12 months, and 0.46 in 57 cases with >12 months (P for trend=0.045). Lower mtDNA copy number was associated with increased breast cancer risk.

Conclusions: Breast cancer diagnosis and/or treatment could increase mtDNA copy number. Prospective studies with a larger study sample size are warranted to confirm the novel inverse association between mtDNA copy number and breast cancer risk.

1 Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago 2 Department of Health Sciences Research, Mayo Clinic 3 Department of Internal Medicine, Mayo Clinic 4 Department of Psychiatry, University of Illinois at Chicago 5 Department of Neurology, Mayo Clinic P-56: Oophorectomy at Younger Age Increases the Risk of Glaucoma

Background and Objective: Glaucoma is the leading cause of blindness worldwide; the majority of persons affected are women. Identifying sex-specific risk factors for glaucoma may support preventive interventions. One such risk factor is early estrogen deficiency, which may increase susceptibility of the optic nerve to glaucoma. The objective is to determine whether bilateral oophorectomy, which causes estrogen deficiency, increases the risk of glaucoma.

Methods: We used the Mayo Clinic Cohort Study of Oophorectomy and Aging, a population-based historical cohort study of all women who underwent oophorectomy from 1950–1987 in Olmsted County, MN. Glaucoma diagnostic codes were searched in the records-linkage system of the Rochester Epidemiology Project. Hazards ratios (HR) were calculated for women with bilateral oophorectomy compared with age-similar referent women without oophorectomy.

Results: Of 1,061 women who underwent bilateral oophorectomy before menopause 155 developed glaucoma, and of 2,361 referent women 279 developed glaucoma. We found no overall increased risk of glaucoma in women who underwent bilateral oophorectomy (HR 1.07, 95% CI 0.88–1.30). However, we found a significantly increased risk of glaucoma in women who underwent bilateral oophorectomy before age 43 years (first tertile of age distribution; HR 1.51, 95% CI 1.11–2.03). In this group, approximately 10% were treated with estrogen to age 50 and treatment did not modify the association (HR 1.47, 95% CI 0.75–2.85).

Conclusions: Our analyses suggest that bilateral oophorectomy at younger age increases the risk of glaucoma and estrogen treatment does not attenuate the risk. Further verification of glaucoma diagnoses and analyses adjusted for potential confounders is warranted.

1 Vanderbilt Epidemiology Center, Institute of Medicine and Public Health, Department of Obstetrics and Gynecology, Vanderbilt University 2 Center for Human Genetics Research, Vanderbilt University P-57: Risk of Spontaneous Abortions Related to Nonsteroidal Anti-Inflammatory Drug Exposure in the First-Trimester Varies by Race

Background and Objective: Non-steroidal anti-inflammatory drugs (NSAIDs) are the most common medication exposures reported in the first-trimester. Prostaglandins play multiple biologic roles in early pregnancy and their synthesis is inhibited by NSAIDs. The objective was to estimate the association between NSAID exposure during the first-trimester and risk for spontaneous abortion (SAB, <20 weeks).

Methods: Women were enrolled in Right from the Start (2000–2009), a prospective cohort. Data about NSAID exposure was obtained from baseline and first-trimester interviews; miscarriages were self-reported and verified by medical records. We used Cox proportional hazards regression models to estimate association between NSAID exposure and SAB outcome.

Results: Among 4,137 pregnancies, 523 experienced an SAB and 3,614 did not. NSAID exposure was reported by 1,503 (36%) participants. Only maternal age remained in the model as a confounder and race was an effect modifier. NSAID exposure was not associated with SAB risk in unadjusted models (hazard ratio [HR]: 1.03, 95% confidence interval [CI] 0.86–1.22). However, adjusted race-stratified analyses indicated protection from SAB in Blacks (HR: 0.61, 95% CI 0.38–0.99) but not Whites (HR: 1.02, 95% CI 0.84–1.26) or Hispanics (HR: 1.55, 95% CI 0.75–3.21).

Conclusions: NSAID use appears to protect against pregnancy loss for Black but not White women. This finding could reflect confounding factors or a true causal association. It warrants pursuit of evidence about whether known differences in inflammatory pathways or other genetic factors are in play. Further investigation of dose, timing in gestation and indication may help identify biologic mechanisms to explain this disparity in SAB risk.

1 Department of Physiology, Tulane University 2 Department of Medicine, Tulane University 3 Neuroscience Program, Tulane University P-59: Short-Term High Dietary Salt Treatment Decreases Sex-Dependent Difference in Neuronal Activity in the Rat Hypothalamus

Background and Objective: Neuronal activity in the paraventricular nucleus (PVN) of the hypothalamus plays a critical role in the central pathways mediating sympathetic regulation of cardiovascular function. Increased dietary salt triggers signaling pathways that increase oxidative stress and lead to the development and progression of hypertension. We conducted experiments to determine the effects of short term (7 days) high dietary salt (8% NaCl) on neuronal activity in the hypothalamus of male and female rats.

Methods: Expression of c-fos was used as an indirect marker of neuronal activity.

Results: The basal neuronal activity was higher in male hypothalamic tissue compared to age-matched females. High dietary salt treatment increased the neuronal activity in the hypothalamus of both males and females; however the sex-related difference in neuronal activity diminished. TRPV1 (transient receptor potential vanilloid type 1) is shown to play a role in the transduction of osmotic and mechanical stimuli and has multiple effects on the cardiovascular system. Moreover, TRPV1 is widely expressed in the brain. Total TRPV1 expression was higher in the PVN of male rats compared to females and the expression levels are increased after high dietary salt treatment. However, the basal difference between males and females also diminished similarly to the difference in neuronal activity.

Conclusions: The present findings demonstrate that there is a sex-dependent difference in basal neuronal activity and high-salt diet diminishes this difference. Moreover, short-term high-salt diet association with increased neuronal activity in the rat hypothalamus may indicate an early alteration contributing to the development of neurogenic hypertension.

1 Department of Psychiatry, Yale University School of Medicine 2 Yale Child Study Center O-1: Gender Differences in Associations between Stress Response and Substance Use in Adolescents

Background and Objective: Individual differences in adolescents' responses to stress may be a factor in their risk for substance use disorders and links between stress response and substance abuse may be different for girls versus boys. We examined associations between emotional and physiological response to stress and substance use in adolescents by gender.

Methods: We assessed heart rate (HR), blood pressure (BP), self-reported and observed emotion, and salivary cortisol arousal in response to the Trier Social Stress Test-Child version (TSST-C) in 120 (60 girls) low-income inner-city adolescents. Youth's lifetime substance use was determined based on self-report, interview, urine toxicology, and breathalyzer.

Results: Lower diastolic BP response to the stressor was associated with greater odds of being a substance user (Exp[B] = .94, p = .02). Lower observed anxiety responses to the stressor were associated with greater odds of being a substance user for girls but not for boys (Exp[B] = .41, p = .04). In contrast, lower self-reported anxiety responses showed a trend for an association with substance use for boys but not for girls (Exp[B] = .77, p = .06). Heart rate and cortisol responses were not related to substance use for boys or girls.

Conclusions: Findings suggest that dampened physiological (e.g., blood pressure) and emotional responses to stress may be a marker for substance abuse risk in adolescents. The profile of responding that is a marker for risk may differ by gender, highlighting the importance of considering sex differences in understanding substance abuse risk.

Supported by NIH grants: R01-DA-06025 (LM), P50-DA-16556 (RS), and K01-DA-024759(TC).

1 Department of Molecular Microbiology and Microbial Pathogenesis, Washington University School of Medicine, St. Louis 2 Center for Women's Infectious Disease Research, Washington University School of Medicine, St. Louis 3 Department of Biochemistry and Molecular Biophysics, Washington University School of Medicine, St. Louis 4 Department of Laboratory and Genomic Medicine, Washington University School of Medicine, St. Louis 5 Department of Internal Medicine – Infectious Diseases, Washington University School of Medicine, St. Louis O-2: Treatment, Prevention, and Reversal of Antibiotic Resistance in Urinary Tract Infection by Orally Active Mannosides

Background and Objective: Over 15 million women suffer from urinary tract infections (UTI) annually in the U.S. and ∼20–40% have multiple recurrences. This unique high prevalence of chronic/recurrent infections has led to long-term conventional prophylactic therapy which has contributed to an increased antibiotic-resistance of uropathogenic E. coli (UPEC) the causative agent of most UTIs. Elucidation of bacterial pathogenic pathways revealed that the adhesin of type 1 pili, FimH, is an essential virulence factor and thus a novel therapeutic target. We seek to develop FimH antagonists called mannosides to prevent and treat UTI.

Methods: Mannose derivatives were rationally designed using high-resolution structures of mannose bound FimH and synthesized. The resulting compounds were tested for potency in FimH-dependent hemagluttination and biofilm inhibition assays. The efficacy of orally delivered mannosides in treating and/or preventing acute and chronic cystitis was evaluated in a murine model. Specifically, we determined the ability of mannosides to prevent UPEC colonization, invasion, and formation of recalcitrant intracellular bacterial communities within the bladder epithelium. Mass-spectrometry was utilized to determine bioavailability.

Results: We identified potent orally active mannosides that demonstrated fast-acting efficacy in the treatment and prevention of chronic UTI and potentiated first-line empiric UTI therapy of TMP-SMZ, reversing clinical resistance.

Conclusions: This work shows for the first time that an orally administered mannoside is a potent fast-acting therapeutic for the treatment and prevention of UTI in vivo. Use of mannoside FimH inhibitors promises to provide substantial benefit to women suffering from recurrent and chronic UTI.

1 Department of Medicine, Pennsylvania State University College of Medicine 2 Department of Public Health Sciences, Pennsylvania State University College of Medicine 3 Department of Microbiology and Immunology, Pennsylvania State University College of Medicine 4 Department of Biology and Entomology, Pennsylvania State University Eberly College of Science 5 Department of Pathology, Pennsylvania State University College of Medicine O-3: Behavioral, Clinical, and Biological Factors Associated with Anal Human Papillomavirus (HPV) Infection among HIV-infected Women

Background and Objective: HIV-infected women are at high risk for HPV-associated cancers due to immunosuppression and co-infection with HPV, but anal HPV infection is understudied. The study's objective is to estimate the prevalence of anal HPV infection and to assess factors associated with anal HPV infection among HIV-infected women.

Methods: HIV-infected women were recruited from five HIV/AIDS outpatient clinics. A behavioral survey was self-administered by the participants to collect information on socio-demographics, HPV-related knowledge, sexual history, and high-risk behaviors. Oral, vaginal, and anal swabs were collected by the clinicians and were tested for the presence of 37 type-specific HPV DNA using a PCR-based assay. Participants' sera were tested for antibody responses against 14 types of HPV virus-like particles (VLPs) by ELISA. Clinical information was obtained from the medical records.

Results: A total of 60 HIV-infected women were enrolled; they were mainly older than 40 years, non-Hispanic white, had low knowledge about HPV, and had been infected with HIV for more than 10 years. None had received HPV vaccination. High-risk sexual behaviors and substance use were commonly reported. Overall, HPV DNA was detected among 57% of women, and the prevalences of oral, vaginal, and anal HPV infections were 8%, 37% and 42%, respectively. About 18% had multiple-site infection (≥2 anatomical sites) or multiple-type infection (≥2 HPV types). Anal HPV infections with HPV16/18/6/11 were not common (9%). While most women (80%) had positive antibody responses to multiple HPV VLPs, they were either not currently infected or infected with different HPV types.

Conclusions: Anal HPV infection, especially with non-vaccine covered types, was common among HIV-infected women. Antibodies induced from natural infection may have protective effects against HPV re-infection.

1 Department of Psychiatry, Vanderbilt University 2 Department of Preventive Medicine, Vanderbilt University 3 Department of Biostatistics, Vanderbilt University 4 Department of Pediatrics, Vanderbilt University O-4: Increasing First Trimester Use of Opioid Analgesics

Background and Objective: Use of opioid analgesics has increased dramatically in recent years, but little research examines whether use of these medications in pregnancy has increased. Given that opioids may pose a risk for the developing fetus, information on early fetal exposure is urgently needed. The objective is to quantify whether the prevalence of prescribed opioid analgesics during the first trimester among pregnant women enrolled in Tennessee Medicaid increased from 1995 to 2009.

Methods: This is a retrospective cohort study of 268,987 pregnancies from 1995 to 2009 among Tennessee Medicaid-insured women. Pregnancies were identified from birth and fetal death certificates, and linked to previously validated computerized pharmacy records to determine medication use. Poisson regression was used to estimate trends over time, rate ratios, and 95% confidence intervals.

Results: During the study period, 14.5% of pregnancies filled at least one prescription for an opioid analgesic during the first trimester. Use increased from 8.6% of pregnancies in 1995 to 20.1% of pregnancies in 2009. After adjusting for maternal characteristics, the prevalence of any first trimester use in 2009 was 2.3 times greater than in 1995 (95% C.I. = 2.1 to 2.5). Older, white, non-Hispanic mothers with less than a high school education and one or more prior pregnancies were more likely to have filled at least one prescription for an opioid analgesic during the first trimester.

Conclusions: First trimester use of opioid analgesics increased from 1995 to 2009 among Tennessee Medicaid-insured pregnant women. Potential consequences of early fetal exposure to these medications warrant careful examination.

1 Department of Medicine, University of California, San Francisco 2 Keck School of Medicine, University of Southern California 3 Department of Genetics, Harvard Medical School 4 Department of Epidemiology, Harvard School of Public Health 5 Cancer Prevention Institute of California 6 University of Hawaii Cancer Center 7 Instituto Nacional de Salud Publica, Cuernavaca, Mexico O-5: Admixture Mapping Identifies Regions on 6q25 and 11p15 Associated with Breast Cancer in U.S. Latinas

Background and Objective: U.S. Latinas have lower breast cancer incidence compared to non-Hispanic Whites. Higher European genetic ancestry among Latinas is associated with increased risk of breast cancer. This suggests that there could be a genetic component to the observed difference in breast cancer incidence between populations. We used an admixture mapping approach to detect particular regions along the genome with strong departures in ancestry between Latina women with breast cancer and healthy controls which would indicate the presence of genetic risk variants within those regions.

Methods: Genome-wide association data from 1,461 U.S. Latinas with breast cancer and 1,280 U.S. Latina controls were used to estimate and test locus specific genetic ancestry for association with breast cancer risk in case-control analyses using logistic regression models.

Results: We identified two regions where genetic ancestry was significantly associated with breast cancer risk: 6q25 (p = 8.2 × 10⁻⁷) and 11p15 (p = 2.9 × 10⁻⁶). In both regions the risk of breast cancer increased with higher European ancestry. The peak of the 6q25 signal includes the estrogen receptor 1 gene (ESR1) and 5' region, which has been associated with breast cancer in previous studies. The 11p15 signal is within a region not previously implicated in breast cancer. Combined analysis of the SNPs genotyped and imputed in these regions found no single variant that explained the admixture signals.

Conclusions: Identification of the underlying biologically relevant variants responsible for the admixture mapping signals will contribute to the improvement of predictive models for breast cancer risk assessment both in Latinas and non-Hispanic Whites, as well as to a better understanding of breast cancer etiology.

1 Department of Physical Medicine and Rehabilitation, Northwestern University Feinberg School of Medicine 2 Division of Occupational Science and Occupational Therapy, University of Southern California O-6: The Association of Pelvic Girdle Pain and Urinary Incontinence in Pregnancy

Background and Objectives: Pelvic girdle pain (PGP) and urinary incontinence (UI) are common problems during pregnancy. The musculoskeletal changes, particularly of the pelvic floor musculature (PFM), that occur during this time may be an underlying risk factor in both these conditions. The study has two objectives: (1) to investigate the association between UI and PGP in pregnant women in the second trimester of pregnancy and (2) to determine the association between UI and PFM strength by PGP status.

Methods: Fifty-two pregnant women, 26 with and 26 without PGP, were recruited in the second trimester. Women were defined as having PGP if they reported pain intensity of 3/10 or higher on self-report and at least 2/4 positive physical examination tests. Following international guidelines, women were classified as having incontinence if they score 1 or higher on the International Consultation on Incontinence Questionnaire-Short Form (ICIQ). Vaginal PFM examination assessing for strength using the Modified Oxford scale was also performed on 48 women.

Results: There was not a statistically significant association between PGP status and continence status (Fishers Exact Test (FET) p < 0.25), in the second trimester. Women without PGP and strong PFM never reported incontinence. Women with PGP and/or weak PFM were equally likely to report incontinence.

Conclusion: There is no association with PGP and incontinence or PFM weakness and incontinence in these second trimester pregnant women. Further research of this possible link in the third trimester and postpartum is warranted.

1 Department of Radiation Oncology, Duke University 2 Department of Molecular Genetics and Microbiology, Duke University O-7: Gene Expression Response of Luminal and Basal Breast Cancer Cell Lines to Ionizing Radiation

Background and Objective: There are significant differences in how breast tumors respond to ionizing radiation. A growing body of data suggests that breast tumor biology (luminal, basal) contributes to the probability of locoregional control in women with breast cancer. We hypothesized that this may be, in part, due to molecular differences in radiation response among biologically diverse breast tumors exposed to ionizing radiation.

Methods: Six commercially available human breast cancer cell lines known to display gene expression patterns of distinct breast subtypes (luminal—ZR751, T47D, MCF7; basal—SUM159, SUM149, MDA-MB-231) were selected for further study. Cell lines were maintained in appropriate media and grown in triplicate to account for individual sample variability. After reaching confluence, RNA was harvested before and 24 hours after a single dose of 5Gy. Clonogenic cell survival assays were performed to assess the varying degree of radiosensitivity and compare with gene expression to identify markers and determinants of phenotypic responses.

Results: Resistance to ionizing radiation was significantly associated, but not in perfect agreement with, the basal type breast cells. Similarly, pre- and post-radiation gene expression mostly tracked with basal versus luminal cell types, but the distinct patterns of gene expression seen between the two cell types were tightly associated with varying degrees of radiosensitivity.

Conclusion: These data are the first to link patterns of gene activation to radiation response. Our long-term goal is to further characterize radiation gene expression response to identify biomarkers of radiation resistance and novel approaches to enhance radiosensitivity.

1 Neuroendocrine Unit, Massachusetts General Hospital and Harvard Medical School 2 Department of Psychiatry, Massachusetts General Hospital and Harvard Medical School O-10: Hypoleptinemia in Anorexia Nervosa and Hypothalamic Amenorrhea is Associated with Symptoms of Depression, Independent of Body Mass Index

Background and Objective: The fat-derived hormone leptin is anorexigenic, and levels are decreased in low-weight and high in obese women. Animal data suggest that leptin may have anxiolytic and anti-depressant properties. Depression and anxiety are common in anorexia nervosa (AN), an illness associated with low weight, body fat, and leptin levels. Similarly, there is an increased prevalence of anxiety and depression in normal-weight women with hypothalamic amenorrhea (HA), a disorder also characterized by hypoleptinemia. It is unknown whether low leptin levels contribute to the development of mood disorders in these populations. We performed a cross-sectional study to investigate the relationship between leptin levels and symptoms of anxiety and depression in AN and HA.

Methods: We studied 64 women: 15 AN, 12 normal-weight HA, 20 normal-weight healthy (HC), and 17 overweight or obese (OB) healthy controls. We measured fasting serum leptin levels and administered Hamilton Rating Scales for Anxiety (HAM-A) and Depression (HAM-D).

Results: HAM-D scores were higher in AN than other groups, higher in HA than OB, and lowest in HC. HAM-A scores were higher in AN than OB and HC, higher in HA than OB, and lowest in HC. Leptin was lowest in AN, higher in HC than HA, and highest in OB. Leptin levels were inversely associated with HAM-D scores (r = −0.43, p = 0.0004); this relationship remained significant after controlling for BMI. Leptin levels were inversely associated with HAM-A scores (r = −0.34, p = 0.006); this relationship was not significant after controlling for BMI.

Conclusions: Leptin may mediate symptoms of depression independent of BMI.

1 Department of Preventive Medicine and Community Health, University of Texas Medical Branch 2 Sealy Center on Aging, University of Texas Medical Branch 3 Department of Obstetrics and Gynecology, University of Texas Medical Branch O-14: Depressive Symptoms, Gender, and Cardiovascular Disease in Older Adults

Background and Objectives: Depressive symptoms and cardiovascular disease influence life expectancy and quality of life in older adults and often co-occur. The direction of association between these conditions and the role of gender in these relationships is unclear. The objectives are to investigate (1) the relationship between depressive symptoms and incident cardiovascular disease, (2) the relationship between cardiovascular disease and incident depressive symptoms, and (3) whether these relationships are stronger in women than in men.

Methods: The Health and Retirement Study is a longitudinal study of adults in the U.S., born in 1931–1941. Respondents provided information on heart disease and depressive symptoms in 2002, 2004, 2006, and 2008. Information on other health and social characteristics was collected in 1992 and 2002. We used discrete-time hazard models to examine relationships in both directions between heart disease and depressive symptoms.

Results: In 2002, 10,344 individuals did not have heart disease, and 9,721 did not have depressive symptoms. In fully adjusted gender-stratified models, depressive symptoms were related to increased risk of incident heart disease only in women (OR = 1.21; 95% CI = 1.03, 1.42, p value for interaction between gender and depressive symptoms = 0.04). Heart disease in 2002 was associated with a moderate increase in risk of incident depressive symptoms in fully adjusted models (OR = 1.07; 95% CI = 0.99, 1.15). Gender did not modify this relationship.

Conclusions: Depressive symptoms are a risk factor for cardiovascular disease in women. Cardiovascular disease is related to subsequent depressive symptoms in men and women.

1 University of California, Davis School of Medicine 2 Boston University School of Medicine 3 New England Baptist Hospital 4 University of California, San Francisco 5 Northwestern University 6 University of Iowa 7 University of Alabama, Birmingham O-15: Compartment Involvement Patterns in Knee Osteoarthritis: The MOST Study

Background and Objective: The knee joint is divided into the medial and lateral compartments, with a large majority of individuals with knee osteoarthritis (OA) having evidence of loss of articular cartilage in the medial compartment. Malalignment is associated with progression of knee OA. Few studies have examined the prevalence and epidemiology of lateral compartment involvement. The objective was to compare compartmental patterns of knee joint space narrowing (JSN) in women and men and assess whether these are related to knee alignment.

Methods: The NIH-funded Multicenter Osteoarthritis (MOST) Study is an observational study of persons aged 50 to 79 years with either symptomatic knee OA or at high risk of knee OA. JSN was defined as present or absent for each tibiofemoral compartment. Mechanical axis angle was measured using full-limb films. We compared the proportion of JSN in lateral compartments between women and men using a logistic regression model adjusting for covariates, and used generalized estimating equations to account for correlation between two knees within a person. We added in knee alignment as a further adjustment.

Results: Of 5,202 knees (2,652 participants), 1,532 (29.5%) had medial JSN, and 425 (8.2%) had lateral JSN. Lateral JSN was more prevalent in women than in men (prevalence odds ratio [pOR] = 1.9, 95% CI 1.5–2.4). After adjusting for malalignment, the gender association became insignificant (pOR = 1.1, 95% CI 0.9–1.5).

Conclusions: Women are more likely to have lateral JSN than men. Valgus malalignment may contribute to the higher prevalence in women.

1 School of Criminal Justice, Michigan State University 2 Psychology Department, Michigan State University 3 School of Social Work, Michigan State University O-16: Post-Separation Abuse of Children whose Mothers Divorce Intimate Partner Abusing Fathers

Background and Objective: Perpetrators of intimate partner violence often do not limit their abuse to intimate partners: approximately 30 to 60 percent of children of abused mothers are also abused by their fathers. Mothers' efforts to gain safety for themselves and their children are complicated by court-ordered custody and parenting time decrees that require mothers and children to keep abusive fathers in their lives. This allows perpetrators opportunities to continue their abuse. This qualitative study examined whether custody and parenting time arrangements were used by fathers to facilitate abuse of their children and whether mothers detected negative mental health consequences due to the abuse.

Methods: Semi-structured interviews were conducted with 19 mothers who had experienced intimate partner violence by their ex-husbands and who had received custody determinations for shared children from family court. Qualitative data analyses were conducted using Analytic Induction.

Results: Ten of the nineteen (53%) mothers reported that their children (n = 20) had experienced either physical or emotional abuse or neglect by their fathers since the initial separation. For 17 of these children (85%), mental health symptoms following father-interactions were often severe and resulted in the children requiring counseling (n = 17), engagement in self-harm (n = 3), and even mental health hospitalizations (n = 2, with one child experiencing three hospitalizations).

Conclusions: Women's claims of intimate partner violence must be taken into consideration by courts when deciding custody and parenting time arrangements, not just for the safety of the mothers, but for the safety and mental well-being of the children as well.

1 Department of Pediatrics, University of Miami Miller School of Medicine 2 Department of Epidemiology, Michigan State University School of Human Medicine 1: Internalizing Symptoms and Marijuana Involvement in African-American Youth: Sex/Gender Influences

Background and Objective: Sex/gender differences in adolescent marijuana involvement have been understudied in African-American youth. The objective of the study was to examine male-female variation in marijuana involvement among African-American adolescents and to estimate the degree to which internalizing symptoms might predict marijuana smoking by mid-adolescence.

Methods: The sample was drawn from the longitudinal Miami Prenatal Cocaine Study (MPCS) cohort. Four hundred and seven African-American adolescents completed at least one exam at 12, 14–15, and/or 16–17 years, with marijuana use assessed by self-report and toxicology. Self-reported anxiety and depressive symptoms were measured at 12 years. Odds ratio estimates (OR) are from the generalized linear model with generalized estimating equations.

Results: Estimated across assessments, males were more likely to initiate marijuana use by mid-adolescence (OR = 4.4; p < 0.05). At 16–17-years, males were also more likely to exhibit greater marijuana involvement than females as indicated by repeated lifetime use and current use (p < 0.01). Males reported greater anxiety symptoms, and females reported more depressive symptoms at age 12 years (p < 0.01). However, neither symptoms of depression (p > 0.3) nor anxiety (p > 0.1) at age 12 predicted marijuana use through 16–17 years. Anxiety symptoms were related to marijuana use cross-sectionally at age 12 in females (p = 0.002) but not in males (p = 0.577). Prenatal cocaine exposure was not associated with marijuana smoking.

Conclusions: Within an urban African-American cohort, there was a male excess in marijuana involvement by mid-adolescence. Internalizing symptoms did not predict initiation of marijuana through mid-adolescence. Anxiety was linked with marijuana use during early adolescence only, with a stronger association for girls than boys.

Acknowledgments: P50DA024594; R01DA006556; K01DA16720.

1 Division of Occupational Therapy, Medical University of South Carolina 2 Division of Physical Therapy, College of Health Professions, Medical University of South Carolina 3 Department of Pediatrics-Neonatology, Medical University of South Carolina 2: Early Gender-Related Motor Skills Differences in Female and Male Preterm Infants

Background and Objective: Due to improvements in neonatal care, more extremely preterm infants now survive and these infants are at greater risk for neurodevelopmental disorders, placing caregiving burdens on mothers who must cope with the children's developmental uncertainties and possible lifelong impairments. The mental health of mothers of children with disabilities may be improved by early identification and intervention strategies that decrease the severity of developmental delays. As early screening is incompletely done at best, we designed this pilot trial to determine very early quantifiable abnormal motor and muscle patterns in preterm infants, and to compare these to brain imaging results by gender.

Methods: Infant motor tests, motion kinematics, and muscle ultrasound were performed on 10 female and male preterm infants at term, 6 weeks, and 12 weeks corrected gestational age to establish preliminary normative data on differences in motor skill acquisition between high- and low-risk preterm infants. Infants who scored at-risk on tests of infant motor performance at term had MRS/DTI for measures of brain injury.

Results: Kinematics can detect early atypical head movements of high-risk preterm infants in the first 3 months of life. Additionally, muscle ultrasound can detect abnormalities of the gastrocnemius muscle architecture at 6 weeks providing further support that muscles are affected at earlier time periods than previously thought.

Conclusion: Early identification of abnormal motor and muscle patterns in preterm high-risk infants will facilitate targeted intervention during early development, prior to the onset of significant clinical deficits associated with cerebral palsy.

1 Department of Psychiatry, Harvard Medical School 2 Division of Women's Health, Brigham and Women's Hospital 3 Department of Brain and Cognitive Engineering, Korea University 4 Department of Medicine, Harvard Medical School and Massachusetts General Hospital 5 Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology 6 Athinoula Martinos Center for Biomedical Imaging, Massachusetts Institute of Technology 7 Department of Behavioral Medicine, Columbia University Medical Center 8 Department of Medicine, Harvard Medical School 4: Hypoactivation of the Stress Response Circuitry in Depression Associated with Loss of Parasympathetic Control of the Heart and Dysfunction in the HPG Axis: A Combined Analysis of fMRI and Heart Rate Variability

Background and Objective: The comorbidity of major depressive disorder (MDD) and cardiovascular disease (CVD) represents the fourth leading cause of morbidity and mortality worldwide, and women have higher rates. Thus, understanding the co-occurrence of these disorders will have major public health significance, particularly for women. Brain regions implicated in mood dysregulation also are implicated in the neural control of the heart through the autonomic nervous system (ANS). Additionally, MDD in women has been linked to dysregulation of the hypothalamic-pituitary-gonadal (HPG) axis. In this functional magnetic resonance imaging (fMRI) study, we investigated whether abnormal hypoactivation in fronto-limbic stress response circuitry in women with MDD, demonstrated in our recent work, is associated with dysregulation of the ANS and disruption of the HPG axis.

Methods: Ten women with MDD in remission made comparable to 10 healthy controls were scanned during the late follicular/midcycle phase of the menstrual cycle on a 3T GE magnet using a visual stress challenge in tandem with acquisition of heart rate, from which we calculated the high frequency (HF) component of the R-R interval variability (RRV), a common measure of parasympathetic ANS. Baseline gonadal hormones included estradiol and progesterone.

Results: Using SPM2 with HF-RRV as a regressor in general linear modeling of brain activity of the stress response, we demonstrated that hypoactivations of anterior hypothalamus, hippocampus, anterior cingulate cortex (ACC), and orbitofrontal cortex were significantly associated with lower parasympathetic control of the heart in MDD women. MDD women had higher progesterone and lower estradiol than healthy controls. Inclusion of estradiol and progesterone levels as covariates attenuated the effect of HF-RRV on brain activity differences between groups in the hippocampus, ACC, and OFC, suggesting that gonadal hormone deficits accounted, in part, for the effect of HF-RRV on brain activity deficits in MDD. Thus, findings provide evidence that hypoactivity of the stress response circuitry in MDD is significantly associated with dysregulation of the ANS, specifically parasympathetic cardiac control, and with disruption of the HPG axis in women.

Conclusions: Findings have implications for understanding the relationship between mood, arousal, heart regulation, and hormonal status, and may provide insights into the co-morbidity of MDD and CVD in women.

1 Center for Bioinformatics and Genomics, Tulane University 2 University of Missouri-Kansas City 5: Genome-wide Gene Expression Study of Peripheral Blood Monocytes for Osteoporosis

Background and Objective: Osteoporosis is a disease of bone fragility, with an increased risk of low-trauma fracture. The disease has significant impact to the health of the elderly, especially old women. Bone mineral density (BMD) is a key parameter to characterize osteoporosis and genetics plays an important role in BMD variation. An important mechanism for osteoporosis is an excess of bone resorption by osteoclasts over bone formation by osteoblasts. Peripheral blood monocytes (PBMs) are precursors of osteoclasts and play an essential role in osteoclastogenesis by producing cytokines, such as TNF and IL-1. The objective of the study was to identify genes important to osteoporosis through analysis of PBMs.

Methods: A genome-wide expression profiling analysis of PBMs in 69 middle-aged Caucasian females (age 47–53) with extremely high (n = 41) or low (n = 28) hip BMD was performed using Affymetrix Human Exon 1.0 ST arrays.

Results: We identified a number of genes showing significant evidence of differential expression between high vs. low BMD subjects. Among the genes, some were known to be important factors for osteoclast differentiation (e.g, CTNNBIP1, p = 2.89E-4, fold-change = 1.45 in low- vs. high-BMD subjects) and some were previously associated with BMD variation (e.g., C1orf61, p = 3.59E-5, fold-change = 0.61, and NAMPT, p = 6.13E-4, fold-change = 1.58). Interestingly, we also identified some novel candidate genes for BMD variation, such as CCDC62 (p = 2.95E-3, fold-change = 0.83), which was known to have an estrogen receptor regulation function and may act on monocyte-mediated postmenopausal bone loss through that function.

Conclusions: Our findings, together with previous evidence, suggested interesting candidate genes for BMD regulation and risk of osteoporosis.

1 Center for Bioinformatics and Genomics, Tulane University School of Public Health and Tropical Medicine 2 College of Life Sciences, Hunan Normal University, China 3 Department of Genetics, Boys Town National Research Hospital 4 Center of Systematic Biomedical Research, University of Shanghai for Science and Technology, China 5 College of Life Sciences and Bioengineering, Beijing Jiaotong University, China 6 School of Medicine, University of Missouri-Kansas City 6: Peripheral Blood Monocyte-Expressed Anxa2 Gene is Involved in Pathogenesis of Osteoporosis in Humans

Background and Objective: Low bone mineral density (BMD) is a risk factor of osteoporosis and has strong genetic determination. Peripheral blood monocytes (PBMs) could access to bone resorption surfaces and differentiate into osteoclasts to resorb bone. The objective of this study is to identify osteoporosis susceptibility genes and characterize their functions.

Methods: Three independent samples (Samples I, II, and III) containing Caucasians with extremely low vs. high hip BMD (N = 28, 80, 44, respectively) and one cohort of 1,000 unrelated Caucasians (Sample IV) were used. Proteomics analysis of PBMs was conducted in Sample I to discover proteins/genes functionally relevant to BMD. The identified proteins/genes were validated at mRNA level in PBMs from Sample II. Finally, the validated genes were tested for association with BMD in Samples III and IV.

Results: In proteomics analysis of PBMs, Anxa2 was found to be upregulated 2.0-fold in subjects with low BMD, which was verified by Western blotting, and replicated at mRNA level. Three SNPs in Anxa2 were found to be associated with BMD variation (p < 0.05). The above integrative evidence strongly supports that Anxa2 is involved in the pathogenesis of osteoporosis in humans. Furthermore, monocyte migration across endothelial barrier in vitro was elevated up to 4.9-fold by Anxa2 stimulation, suggesting that it may stimulate more PBMs to migrate from blood to bone resorption surfaces to differentiate into osteoclasts in vivo, hence leading to increased bone resorption and decreased BMD.

Conclusions: This study identified a novel osteoporosis susceptibility gene, and suggested a novel pathophysiological mechanism, mediated by Anxa2, for osteoporosis in humans.

1 Pelvic Floor Research Group 2 Biomechanics Research Laboratory, Department of Mechanical Engineering 3 Department of Obstetrics and Gynecology, University of Michigan 7: Normal Ranges and Shape Analysis in 3-D Vaginal Morphology

Background and Objective: Pelvic organ prolapse is associated with changes in vaginal size, shape, and position. The objective of the study was to develop a technique to quantify vaginal dimension and report preliminary findings on their relationship to body size.

Methods: MRI of 10 subjects of 10^th percentile (age: 48.5 ± 6.7), 10 of 50^th percentile (age: 48.3 ± 6.7), and ten of 90^th percentile (age: 48.1 ± 7.3) height were selected from the control group of an ongoing case and control study. Supine, multi-planar MR imaging was performed. ImageJ v1.44 was used to establish the vaginal location and contour in both axial and middle sagittal scans. Width was assessed at five equally spaced locations. A MATLAB™ program was used for data analysis.

Results: Vaginal widths at the five equally spaced locations from bottom to top vary with the 10^th, 50^th, and 90^th percentile height. The corresponding p-values are 0.134, 0.077, 0.174, 0.702, and 0.789. Anterior vaginal lengths of the subjects with 10^th, 50^th, and 90^th percentile height are 68.7 ± 10.3 (length unit: mm), 67.6 ± 10.5, and 71.8 ± 9.8. The corresponding p-value is 0.504. Posterior vaginal lengths of the subjects with 10^th, 50^th, and 90^th percentile height are 122.1 ± 16.6, 118.9 ± 12.3, and 122.3 ± 10.0. The corresponding p-value is 0.968. Cervix lengths of the subjects with 10^th, 50^th, and 90^th percentile height are 36.4 ± 7.5, 34.0 ± 7.0, and 30.8 ± 5.7. The corresponding p-value is 0.070. Average middle sagittal vaginal shapes of the three groups vary with each other and have overlaps over each other.

Conclusions: Vaginal dimension has large variation between individuals but it is not primarily explained by body height.

Acknowledgments: We gratefully acknowledge support from the Eunice Kennedy Shriver National Institute of Child Health and Human Development Grants R01 HD 38665 and the Office of Research on Women's Health SCOR on Sex and Gender Factors Affecting Women's Health P50 HD044406.

Disclosures: Jiajia Luo: Research Support from American Medical Systems, and Kimberly-Clark Corporation. James Ashton-Miller: Research Support from American Medical Systems, and Kimberly-Clark Corporation; Consultant for American Medical Systems. John DeLancey: Research Support from American Medical Systems, and Kimberly-Clark Corporation; Consultant for American Medical Systems and Johnson & Johnson.