NHLBI Working Group: The Cardiovascular Consequences of Post-Traumatic Stress Disorder

Background:

Experiencing traumatic events such as violent personal assaults, natural or human-made disasters, serious accidents, or military combat can lead to the development of PTSD, a psychiatric condition characterized by a persistent maladaptive reaction to exposure to severe psychological trauma.  PTSD is associated with an increased risk of CVD events, including coronary heart disease, heart failure, and stroke. It is also associated with CVD risk factors such as hypertension and diabetes, and with accelerated progression of CVD, particularly heart failure.  Evidence for this increased risk includes prospective data showing an association between a PTSD diagnosis or symptoms and CVD events, including coronary artery disease, stroke, and heart failure in cohorts of United States veterans and in a large longitudinal study of U.S. nurses.

Despite these observations, prospective data from well-designed studies with measurement of potential complex confounding factors remain limited. The biological and behavioral mechanisms linking PTSD with CVD risk are not well defined. Mechanisms of stress regulation that are thought to be altered in PTSD and that also may affect the cardiovascular system include the hypothalamic-pituitary-adrenal axis; the thyroid axis; immune regulation; peptide regulation, such as pituitary adenylate cyclase-activating polypeptide; and the renin-angiotensin/vasopressin system; and adrenergic/autonomic nervous system regulation. In addition, possible mechanisms of shared risk across CVD and PTSD include shared genetic and epigenetic pathways, trauma-related factors, and metabolic risk factors. Confounding factors resulting from traumatic events, in addition to PTSD, that can affect CVD risk in those exposed to trauma with PTSD include changes in health risk behaviors (e.g., disrupted sleep patterns, maladaptive dietary changes, tobacco use, substance abuse, physical inactivity, and non-compliance with medical therapy) and other associated comorbidities of PTSD (e.g., depression). There is also a well-described association of PTSD that may occur after a life-threatening CVD event, such as cardiac arrest, or repeated shocks by an internal cardiac defibrillator, highlighting associations of CVD with risk for PTSD.

Discussion:

Participants at the workshop identified numerous major clinical and research gaps and provided recommendations for future scientific directions to illuminate the mechanisms of the relationship of PTSD with CVD, CVD subtypes, and CVD risk factors. A top priority is to establish stronger evidence for causal relationships and genetic/environmental/behavioral modifiers of the relationship of PTSD with CVD/CVD risk factors. Numerous population-based studies and biobanks are now available that may allow more rigorous clinical, translational, and population studies, including studies that incorporate omics (including genomics, proteomics, and metabolomics), critical biomarkers, and high-resolution imaging of the brain and heart.  The most valuable of these for study of PTSD is the U.S. Department of Veterans Affairs (VA) Million Veterans Program (MVP). Other population-based studies and biobanks may require de novo measurements to identify the presence of PTSD.

Questions remain unanswered regarding the basic mechanisms that link PTSD and CVD, including the role of inflammation and immune dysfunction, oxidative stress and mitochondrial abnormalities, autonomic nervous system dysregulation, and endothelial dysfunction. The workshop participants identified the following critical research gaps in PTSD and CVD prevention: 1) the need to create a research network that would facilitate collaboration among investigators and integration of data from both the cardiovascular and psychiatric fields to infer causal relationships from clinical and population studies; 2) the need to identify specific biochemical mechanisms through basic research or human physiological/molecular studies which can inform the discovery of robust preclinical markers of disease; 3) the need to have more refined and evidence-based screening techniques and approaches; and 4) the need to identify therapeutic targets and approaches through translational research ultimately leading to clinical trials.

WORKSHOP RECOMMENDATIONS:

THE FOLLOWING SCIENTIFIC AREA Gaps and Priorities WERE IDENTIFIED:

Clinical/Population Science

Priority: Database/Epidemiological Studies

1. Gap: Lack of prospective, longitudinal measures of both PTSD and CVD (and CVD risk factors) in existing community-based epidemiology cohorts

Recommendations:

• Survey existing cohorts to evaluate suitability of adding trauma/PTSD or CVD outcomes or create a new follow-up study that integrates these elements (e.g.,  explore Department of Defense (DoD) databases where PTSD is well captured,  and identify opportunities for linkages between DoD and VA health databases).
• Add measures of trauma exposure and PTSD symptoms or PTSD diagnosis to ongoing longitudinal studies of CVD outcomes.
• Add measures of CVD and CVD risk factors to ongoing longitudinal studies of trauma/PTSD (especially in adolescents and young adults, which may include DoD databases, but valuable in all ages).
• Explore the development of new studies designed to collect these data with state-of-the-art methodology.

2. Gap: Lack of rigorous measures of trauma exposure and PTSD symptoms in most existing electronic health record (EHR) databases that have data on CVD and CVD risk factors

Recommendations:

• Explore natural language processing, machine learning algorithms, and other artificial intelligence approaches to identify trauma exposure, PTSD phenotypes, and PTSD symptoms in EHR systems (particularly in high yield datasets such as VA health databases) and in studies linked to DoD health databases (e.g., Army Study to Assess Risk and Resilience in Service members (STARRS)/STARRS-Longitudinal Study (LS), and other DoD datasets).

3. Gap: Lack of data on racial, ethnic, and sex differences in PTSD-CVD risk association

Recommendations:

• Identify CVD cohorts that focus on underrepresented groups (e.g., Jackson Heart Study, MVP, Million Hearts Study, and Strong Heart Study) for further studies on relationships of PTSD and CVD.
• Explore the development of new studies directly addressing this question.

Priority: Experimental/Interventional Studies

1. Gap:  Lack of comprehensive, robust, longitudinal approaches, including multivariable analysis and systems biology approaches, to study biological mechanisms linking PTSD and CVD

Recommendation:

• Identify CVD and PTSD datasets/cohorts that have longitudinal, multivariable (omics and other) data available; if not available, develop such cohorts/datasets.

2. Gap:  Lack of comprehensive evaluation of multiple health behaviors as mediators/moderators of PTSD-CVD association

Recommendation:

• Identify CVD and PTSD datasets/cohorts that have deep data on multiple health behaviors (e.g., MVP and STARRS/STARRS-LS) that could be utilized for further studies of relationships; if not available, develop such cohorts/datasets.

3. Gap: Lack of interventional experimental studies to enhance the evidence for or against causal mechanism(s) in the PTSD-CVD association

Recommendations:

• Add PTSD symptom diagnostic measures to studies targeting potential CVD risk mediators (e.g., anti-inflammatory trials and renin-angiotensin receptor inhibitor trials). Consider as candidate trials a range of ongoing randomized controlled trials in CVD, some of which are beginning to enroll patients with evidence of inflammatory states.
• Identify or develop stress probes suitable for measurement of cardiovascular stress reactivity and other physiological responses to stress as they relate to PTSD, to be implemented in clinical and population studies and randomized interventional trials. 

4. Gap: Lack of clinical studies assessing how intervening on PTSD risk or symptoms impacts CVD risk factors or CVD outcomes

Recommendation:

• Add measures of CVD risk factors and outcomes to PTSD treatment trials. Reasonable cardiovascular risk measures expected to change in the timeframe of a clinical trial would include heart rate variability (HRV), endothelial function (flow-mediated dilation), metabolic risk factors, and other biomarkers of CVD risk that could be combined with psychophysiological measures/outcome.

Priority: Genetic and Genomic Studies

1. Gap: Lack of large-scale datasets with deep genetic data and both PTSD and CVD (and CVD risk factor) measures for genetic and genomics studies

Recommendation:

• Capitalize Leverage large existing consortia (e.g., MVP, UK Biobank, Psychiatric Genomics Consortium, STARRS/STARRS-LS, and DoD Serum Repository) to include PTSD and CVD phenotypes in data collection.
• Explore whole-genome sequencing databases that capture both PTSD and CVD to identify other candidate linkages of PTSD and CVD (e.g., STARRS/STARRS-LS and MVP).
• Further leverage other whole-genome sequencing databases, including those that may capture either mental health disorders or CVD, such as NHLBI’s Trans-Omics for Precision Medicine (TOPMed) program, the National Human Genome Research Institute’s Consortium for Common Disease Genetics (CCDG), or the Whole Genome Sequencing for Psychiatric Disorders (WGSPD) Consortium, to dissect candidate genomic regions for associations between PTSD and CVD.

2. Gap: Lack of population cohort studies and biobanks to harness genetics and genomics to understand shared risks of PTSD and CVD, to identify components and subtypes of PTSD with increased CVD risk, to identify possible gene targets for interventions to reduce CVD risk in PTSD, and to enhance causal inference

Recommendations:

• Use methods such as Mendelian randomization to test the causal relationships between PTSD/PTSD symptom clusters and potential biological mechanisms (i.e., inflammation, autonomic nervous system, and HRV) and CVD.
• Identify genetic and/or epigenetic profiles of PTSD that would also be candidates for CVD prevention.

3. Gap: Lack of polygenic risk scores from genome-wide association studies (GWAS) of PTSD and CVD to inform potential causal PTSD-CVD associations and identify potential therapeutic targets

Recommendation:

• Explore genetic studies to identify polygenic risk scores and their role in the PTSD-CVD association, as well as their possible utility in therapeutic clinical trials.

Basic and Translational Science:

Priorities:  Atherosclerosis, Circulating Factors, Brain-Heart Communications

1. Gap:  Lack of studies of basic mechanisms, including immune function, autonomic dysregulation and neurohormonal factors, linking PTSD and CVD in human studies and animal models

Recommendations:

• Determine the mechanistic role of autonomic dysregulation in the link between PTSD and CVD using multidisciplinary approaches that may also address the roles of immune activation/inflammation, neuropeptide, and hypothalamic-pituitary-adrenal axis dysregulation.
• Conduct basic mechanistic experimental studies using appropriate cardiovascular and PTSD-based animal models simultaneously (e.g., paradigms that focus on long-term effects of stress, not acute stress).

2. Gap: Lack of human tissues (e.g., brain and heart) to study basic mechanisms of CVD and PTSD linkage

Recommendation:

• Develop and link or use currently available biobank data (e.g., National PTSD Brain Bank of VA, which presently does not collect non-brain tissue, but potentially could) to describe and characterize the underlying biology of atherosclerosis as it relates to PTSD, to facilitate collaborative mechanistic studies (e.g., shared tissue samples), and to address potential roles of other various organ dysfunction (e.g., heart, adrenals, kidney, and vasculature) that may play an important role in CVD.

3. Gap: Lack of clear characterization of the specific abnormalities of brain circuitry, including brain signatures/function/activity (e.g., function in the amygdala) in PTSD that may be contributing to CVD

Recommendations:

• Develop additional mechanistic human studies, with support by relevant animal models, that could provide simultaneous imaging and neurophysiological measures of relevant cardiovascular and brain function.
• Utilize circuit- and cell-specific tools to examine a causal role of PTSD-associated circuitry to cardiovascular dysfunction and pathology in appropriate animal models.

4. Gap: Identify the role of bioenergetics (e.g., oxidative stress) and markers of mitochondrial health in PTSD and subsequent studies of association with CVD

Recommendation:

• Include measures of bioenergetics in human cohorts and clinical trials.

5. Gap: Identify whether genes associated with CVD modulate PTSD-associated mechanisms

• Develop studies to examine the functional role of CVD-associated genes in enduring effects of trauma in appropriate animal and tissue models.

Publication Plans:

A white paper outlining the recommendations that arose from the workshop is in preparation.

NHLBI Organizing Team:

Lisa Schwartz Longacre, PhD, Division of Cardiovascular Sciences (DCVS)/Heart Failure and Arrhythmia Branch (HFAB)

Rebecca Campo, PhD, DCVS/Clinical Applications and Prevention Branch (CAPB)

Narasimhan Danthi, PhD, DCVS/Advanced Technologies and Surgery Branch

George Sopko, MD, DCVS/HFAB

Catherine Stoney, PhD, FABMR, DCVS/CAPB

Michael Twery, PhD, Division of Lung Diseases, National Center on Sleep Disorders

Federal and Non-Federal Partners:

Terri Gleason, PhD, US Department of Veterans Affairs

Mario Rinaudo, MD, US Department of Veterans Affairs

Ian M. Kronish, MD, MPH, American Heart Association

Andrew Midzak, PhD, US Army Medical Research and Materiel Command

Additional NIH Institutes and Centers Represented:

Center for Scientific Review

Eunice Kennedy Shriver National Institute of Child Health and Human Development

National Cancer Institute

National Center for Advancing Translational Sciences

National Center for Complementary and Integrative Health

National Institute of Allergy and Infectious Diseases

National Institute of Arthritis and Musculoskeletal and Skin Diseases

National Institute on Drug Abuse

National Institute of Mental Health

National Institute of Nursing Research

Workshop Chairs:

• Christopher J. O’Donnell, MD, MPH, VA Boston Healthcare System and Harvard Medical School
• Murray B. Stein, MD, MPH, University of California San Diego and VA San Diego Healthcare System

Workshop Members:

• Beth Cohen, MD, MAS, University of California, San Francisco and VA Medical Center San Francisco
• Zahi A. Fayad, PhD, Icahn School of Medicine at Mount Sinai
• Charles F. Gillespie, MD, PhD, Emory University
• Joel Gelernter, MD, Yale University and VA Connecticut Healthcare System
• Jessica M. Gill, PhD, RN, FAAN, National Institute of Nursing Research
• Marti Jett, PhD, ST, US Army Medical Research and Materiel Command
• Jeff Kibler, PhD, Nova Southeastern University
• Ian M. Kronish, MD, MPH, Columbia University
• Howard Kruth, MD, Division of Intramural Research, NHLBI
• Israel Liberzon, MD, Texas A&M University Health Science Center
• Paul Marvar, PhD, The George Washington University
• Nehal Mehta, MD, MSCE, Division of Intramural Research, NHLBI
• Thomas A. Mellman, MD, Howard University
• Deborah Murdock, PhD, Children’s Hospital of Philadelphia
• Jeanie Park, MD, MS, Emory University School of Medicine, Atlanta VA Health Care System
• Renato Polimanti, PhD, Yale University School of Medicine and VA Connecticut Healthcare System
• Jordan W. Smoller, MD, ScD, Massachusetts General Hospital and Harvard Medical School
• Victoria Risbrough PhD, University of California, San Diego and VA San Diego Healthcare System
• Robert J. Ursano, MD, Uniformed Services University of the Health Sciences
• Viola Vaccarino, MD, PhD, Emory University and Atlanta VA Medical Center
• Richard S. Vander Heide, MD, PhD, Louisiana State University
• Clyde W. Yancy, MD, MSc, Northwestern University