Providing Contemporary Access to Historical Biospecimen Collections: Development of the NHLBI Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC)

Serum protein responses to Dietary Approaches to Stop Hypertension (DASH) and DASH-Sodium trials and associations with blood pressure changes

OBJECTIVES

The Dietary Approaches to Stop Hypertension (DASH) diet reduces blood pressure, but the mechanisms underlying DASH diet-blood pressure relations are not well understood. Proteomic measures may provide insights into the pathophysiological mechanisms through which the DASH diet reduces blood pressure.

METHODS

The DASH (1994-1996) and DASH-Sodium (1997-1999) trials were multicenter, randomized-controlled feeding trials. Proteomic profiling was conducted in serum collected at the end of the feeding period (DASH, N = 215; DASH-Sodium, N = 390). Multivariable linear regression models were used to identify interactions between 71 DASH diet-related proteins and changes in systolic and diastolic blood pressure. Estimates were meta-analyzed across both trials. Elastic net models were used to identify proteins that predict changes in blood pressure.

RESULTS

Ten significant interactions were identified [systolic blood pressure: seven proteins; diastolic blood pressure: three proteins], which represented nine unique proteins. A high level of renin at the end of the feeding period was associated with greater reductions in diastolic blood pressure in individuals consuming the control than DASH diets. A high level of procollagen c-endopeptidase enhancer 1 (PCOLCE) and collagen triple helix repeat-containing protein 1 (CTHRC1) were associated with greater reductions in systolic blood pressure in individuals consuming the DASH than control diets, and with elevations in systolic blood pressure in individuals consuming the control diets (P for interaction for all tests < 0.05). Elastic net models identified six additional proteins that predicted change in blood pressure.

CONCLUSIONS

Several novel proteins were identified that may provide some insight into the relationship between the DASH diet and blood pressure.

Intracranial Atherosclerotic Disease and Incident Dementia: The ARIC Study (Atherosclerosis Risk in Communities)

BACKGROUND

Studies of the neurovascular contribution to dementia have largely focused on cerebral small vessel disease (CSVD), but the role of intracranial atherosclerotic disease (ICAD) remains unknown in the general population. The objective of this study was to determine the risk of incident dementia from ICAD after adjusting for CSVD and cardiovascular risk factors in a US community-based cohort.

METHODS

We acquired brain magnetic resonance imaging examinations from 2011 through 2013 in 1980 Black and White participants in the ARIC study (Atherosclerosis Risk in Communities), a prospective cohort conducted in 4 US communities. Magnetic resonance imaging examinations included high-resolution vessel wall magnetic resonance imaging and magnetic resonance angiography to identify ICAD. Of these participants, 1590 without dementia, without missing covariates, and with adequate magnetic resonance image quality were followed through 2019 for incident dementia. Associations between ICAD and incident dementia were assessed using Cox proportional hazard ratios adjusted for CSVD (characterized by white matter hyperintensities, lacunar infarctions, and microhemorrhages), APOE4 genotype (apolipoprotein E gene ε4), and cardiovascular risk factors.

RESULTS

The mean age (SD) of study participants was 77.4 (5.2) years. ICAD was detected in 34.6% of participants. After a median follow-up of 5.6 years, 286 participants developed dementia. Compared with participants without ICAD, the fully adjusted hazard ratios (95% CIs) for incident dementia in participants with any ICAD, with ICAD only causing stenosis ≤50%, and with ICAD causing stenosis >50% in ≥1 vessel were 1.57 (1.17-2.11), 1.41 (1.02-1.95), and 1.94 (1.32-2.84), respectively. ICAD was associated with dementia even among participants with low white matter hyperintensities burden, a marker of CSVD.

CONCLUSIONS

ICAD was associated with an increased risk of incident dementia, independent of CSVD, APOE4 genotype, and cardiovascular risk factors. The increased risk of dementia was evident even among participants with low CSVD burden, a group less likely to be affected by vascular dementia, and in participants with ICAD causing only low-grade stenosis. Our results suggest that ICAD may partially mediate the effect that cardiovascular risk factors have on the brain leading to dementia. Both ICAD and CSVD must be considered to understand the vascular contributions to cognitive decline.

Role of ideal cardiovascular health metrics in reducing risk of incident arrhythmias

AIMS

Cardiovascular health (CVH) has been proven to reduce cardiovascular disease burden and mortality, but data are lacking regarding cardiac arrhythmias. The aim of this study was to assess the association between CVH metrics and atrial fibrillation/flutter (AF), ventricular arrhythmias, and bradyarrhythmias.

METHODS AND RESULTS

This study analysed data from the Atherosclerosis Risk in Communities (ARIC) cohort, with participants recruited from four different communities across the United States. Cardiovascular health metrics were scored at baseline (1987-89) following the American Heart Association's recommendations and categorized as poor, intermediate, or ideal. Arrhythmia episodes were diagnosed by International Classification of Diseases (ICD)-9 code. Adjusted associations were estimated using Cox models and event rates and population attributable fractions were calculated by CVH metrics category. The study population consisted of 13 078 participants, with 2548 AF, 1363 ventricular arrhythmias, and 706 bradyarrhythmias occurred. The adjusted hazard ratios (HRs) for ideal (vs. poor) CVH metrics were 0.59 [95% confidence interval (CI): 0.50-0.69] for AF, 0.38 (95% CI: 0.28-0.51) for ventricular arrhythmias, and 0.70 (95% CI: 0.51-0.97) for bradyarrhythmia. The risk of incident arrhythmias decreased steadily as the CVH metrics improved from 0 to 14 scores. The adjusted population attributable fractions were calculated to be 29.9% for AF, 54.4% for ventricular arrhythmias, and 21.9% for bradyarrhythmia, respectively. The association between CVH metrics and incident arrhythmias was also seen in people who remained free of coronary heart disease over the follow-up.

CONCLUSION

Achieving ideal CVH metrics recommendations by AHA in midlife was associated with a lower risk of incident arrhythmias later in life.

Clinical trial data sharing: a cross-sectional study of outcomes associated with two U.S. National Institutes of Health models

The impact and effectiveness of clinical trial data sharing initiatives may differ depending on the data sharing model used. We characterized outcomes associated with models previously used by the U.S. National Institutes of Health (NIH): National Heart, Lung, and Blood Institute's (NHLBI) centralized model and National Cancer Institute's (NCI) decentralized model. We identified trials completed in 2010-2013 that met NIH data sharing criteria and matched studies based on cost and/or size, determining whether trial data were shared, and for those that were, the frequency of secondary internal publications (authored by at least one author from the original research team) and shared data publications (authored by a team external to the original research team). We matched 77 NHLBI-funded trials to 77 NCI-funded trials; among these, 20 NHLBI-sponsored trials (26%) and 4 NCI-sponsored trials (5%) shared data (OR 6.4, 95% CI: 2.1, 19.8). From the 4 NCI-sponsored trials sharing data, we identified 65 secondary internal and 2 shared data publications. From the 20 NHLBI-sponsored trials sharing data, we identified 188 secondary internal and 53 shared data publications. The NHLBI's centralized data sharing model was associated with more trials sharing data and more shared data publications when compared with the NCI's decentralized model.

Metabolomic Profiles Associated With Blood Pressure Reduction in Response to the DASH and DASH-Sodium Dietary Interventions

BACKGROUND

The DASH (Dietary Approaches to Stop Hypertension) diets reduced blood pressure (BP) in the DASH and DASH-Sodium trials, but the underlying mechanisms are unclear. We identified metabolites associated with systolic BP or diastolic BP (DBP) changes induced by dietary interventions (DASH versus control arms) in 2 randomized controlled feeding studies-the DASH and DASH-Sodium trials.

METHODS

Metabolomic profiling was conducted in serum and urine samples collected at the end of diet interventions: DASH (n=219) and DASH-Sodium (n=395). Using multivariable linear regression models, associations were examined between metabolites and change in systolic BP and DBP. Tested for interactions between diet interventions and metabolites were the following comparisons: (1) DASH versus control diets in the DASH trial (serum), (2) DASH high-sodium versus control high-sodium diets in the DASH-Sodium trial (urine), and (3) DASH low-sodium versus control high-sodium diets in the DASH-Sodium trial (urine).

RESULTS

Sixty-five significant interactions were identified (DASH trial [serum], 12; DASH high sodium [urine], 35; DASH low sodium [urine], 18) between metabolites and systolic BP or DBP. In the DASH trial, serum tryptophan betaine was associated with reductions in DBP in participants consuming the DASH diets but not control diets (P interaction, 0.023). In the DASH-Sodium trial, urine levels of N-methylglutamate and proline derivatives (eg, stachydrine, 3-hydroxystachydrine, N-methylproline, and N-methylhydroxyproline) were associated with reductions in systolic BP or DBP in participants consuming the DASH diets but not control diets (P interaction, <0.05 for all tests).

CONCLUSIONS

We identified metabolites that were associated with BP lowering in response to dietary interventions.

REGISTRATION

URL: https://www.

CLINICALTRIALS

gov/ct2/show/NCT03403166; Unique identifier: NCT03403166 (DASH trial). URL: https://www.

CLINICALTRIALS

gov/ct2/show/NCT00000608; Unique identifier: NCT00000608 (DASH-Sodium trial).

Factor H preserves alternative complement function during ARDS, linked to improved survival

Background

Effective regulation of complement activation may be crucial to preserving complement function during acute respiratory distress syndrome (ARDS). Factor H is the primary negative regulator of the alternative pathway of complement. We hypothesised that preserved factor H levels are associated with decreased complement activation and reduced mortality during ARDS.

Methods

Total alternative pathway function was measured by serum haemolytic assay (AH50) using available samples from the ARDSnet Lisofylline and Respiratory Management of Acute Lung Injury (LARMA) trial (n=218). Factor B and factor H levels were quantified using ELISA using samples from the ARDSnet LARMA and Statins for Acutely Injured Lungs from Sepsis (SAILS) (n=224) trials. Meta-analyses included previously quantified AH50, factor B and factor H values from an observational registry (Acute Lung Injury Registry and Biospecimen Repository (ALIR)). Complement C3, and complement activation products C3a and Ba plasma levels were measured in SAILS.

Results

AH50 greater than the median was associated with reduced mortality in meta-analysis of LARMA and ALIR (hazard ratio (HR) 0.66, 95% CI 0.45-0.96). In contrast, patients in the lowest AH50 quartile demonstrated relative deficiency of both factor B and factor H. Relative deficiency of factor B (HR 1.99, 95% CI 1.44-2.75) or factor H (HR 1.52, 95% CI 1.09-2.11) was associated with increased mortality in meta-analysis of LARMA, SAILS and ALIR. Relative factor H deficiency was associated with increased factor consumption, as evidenced by lower factor B and C3 levels and Ba:B and C3a:C3 ratios. Higher factor H levels associated with lower inflammatory markers.

Conclusions

Relative factor H deficiency, higher Ba:B and C3a:C3 ratios and lower factor B and C3 levels suggest a subset of ARDS with complement factor exhaustion, impaired alternative pathway function, and increased mortality, that may be amenable to therapeutic targeting.

Identification of Protein Biomarkers of the Dietary Approaches to Stop Hypertension Diet in Randomized Feeding Studies and Validation in an Observational Study

Background The Dietary Approaches to Stop Hypertension (DASH) diet is recommended for cardiovascular disease prevention. We aimed to identify protein biomarkers of the DASH diet using data from 2 randomized feeding studies and validate them in an observational study, the ARIC (Atherosclerosis Risk in Communities) study. Methods and Results Large-scale proteomic profiling was conducted in serum specimens (SomaLogic) collected at the end of 8-week and 4-week DASH diet interventions in multicenter, randomized controlled feeding studies of the DASH trial (N=215) and the DASH-Sodium trial (N=396), respectively. Multivariable linear regression models were used to compare the relative abundance of 7241 proteins between the DASH and control diet interventions. Estimates from the 2 trials were meta-analyzed using fixed-effects models. We validated significant proteins in the ARIC study (N=10 490) using the DASH diet score. At a false discovery rate <0.05, there were 71 proteins that were different between the DASH diet and control diet in the DASH and DASH-Sodium trials. Nineteen proteins were validated in the ARIC study. The 19 proteins collectively improved the prediction of the DASH diet intervention in the feeding studies (range of difference in C statistics, 0.267-0.313; P<0.001 for both tests) and the DASH diet score in the ARIC study (difference in C statistics, 0.017; P<0.001) beyond participant characteristics. Conclusions We identified 19 proteins robustly associated with the DASH diet in 3 studies, which may serve as biomarkers of the DASH diet. These results suggest potential pathways that are impacted by consumption of the DASH diet. Registration URL: https://www.clinicaltrials.gov; Unique identifiers: NCT03403166, NCT00000608.

Systolic Blood Pressure and Cardiovascular Risk in Normotensive Adults

OBJECTIVE

To examine the association of systolic blood pressure (SBP) and cardiovascular risk in normotensive adults.

PATIENTS AND METHODS

This study analyzed data from 7 prospective cohorts between September 29, 1948, and December 31, 2018. Complete information on history of hypertension and baseline blood pressure measurements were required for inclusion. We excluded individuals younger than 18 years old, those with a history of hypertension, and patients with baseline SBP measurements of less than 90 mm Hg or 140 mm Hg or higher. Cox proportional hazards regression and restricted cubic spline models were used to evaluate the hazards of cardiovascular outcomes.

RESULTS

A total of 31,033 participants were included. The mean ± SD age was 45.3±14.8 years, 16,693 of the participants (53.8%) were female, and the mean ± SD SBP was 115.8±11.7. Over a median follow-up of 23.5 years, 7005 cardiovascular events occurred. Compared with those who had SBP levels of 90 to 99 mm Hg, participants with SBP levels of 100 to 109, 110 to 119, 120 to 129, and 130 to 139 mm Hg experienced 23% (hazard ratio [HR], 1.23; 95% CI, 1.07 to 1.42), 53% (HR, 1.53; 95% CI, 1.33 to 1.76), 87% (HR, 1.87; 95% CI, 1.62 to 2.16), and 117% (HR, 2.17; 95% CI, 1.87 to 2.52) increased risks of cardiovascular events, respectively. Compared with follow-up SBP of 90 to 99 mm Hg, the HRs for cardiovascular events were 1.25 (95% CI, 1.02 to 1.54), 1.93 (95% CI, 1.58 to 2.34), 2.55 (95% CI, 2.09 to 3.10), and 3.39 (95% CI, 2.78 to 4.14), respectively, for follow-up SBP levels of 100 to 109, 110 to 119, 120 to 129, and 130 to 139 mm Hg.

CONCLUSION

In adults without hypertension, there is a stepwise increase in risk of cardiovascular events, with increasing SBP starting at levels as low as 90 mm Hg.

Organ-On-A-Chip Database Revealed-Achieving the Human Avatar in Silicon

Organ-on-a-chip (OOC) provides microphysiological conditions on a microfluidic chip, which makes up for the shortcomings of traditional in vitro cellular culture models and animal models. It has broad application prospects in drug development and screening, toxicological mechanism research, and precision medicine. A large amount of data could be generated through its applications, including image data, measurement data from sensors, ~omics data, etc. A database with proper architecture is required to help scholars in this field design experiments, organize inputted data, perform analysis, and promote the future development of novel OOC systems. In this review, we overview existing OOC databases that have been developed, including the BioSystics Analytics Platform (BAP) developed by the University of Pittsburgh, which supports study design as well as data uploading, storage, visualization, analysis, etc., and the organ-on-a-chip database (Ocdb) developed by Southeast University, which has collected a large amount of literature and patents as well as relevant toxicological and pharmaceutical data and provides other major functions. We used examples to overview how the BAP database has contributed to the development and applications of OOC technology in the United States for the MPS consortium and how the Ocdb has supported researchers in the Chinese Organoid and Organs-On-A-Chip society. Lastly, the characteristics, advantages, and limitations of these two databases were discussed.

Dynamic Change of Cardiovascular Health Metrics and Long‐Term Risk of Sudden Cardiac Death: The ARIC Study

Background

The change of cardiovascular health (CVH) status has been associated with risk of cardiovascular disease. However, no studies have explored the change patterns of CVH in relation to risk of sudden cardiac death (SCD). We aim to examine the link between baseline CVH and change of CVH over time with the risk of SCD.

Methods and Results

Analyses were conducted in the prospective cohort ARIC (Atherosclerosis Risk in Communities) study, started in 1987 to 1989. ARIC enrolled 15 792 individuals 45 to 64 years of age from 4 US communities (Forsyth County, North Carolina; Jackson, Mississippi; suburbs of Minneapolis, Minnesota; and Washington County, Maryland). Subjects with 0 to 2, 3 to 4, and 5 to 7 ideal metrics of CVH were categorized as having poor, intermediate, or ideal CVH, respectively. Change in CVH over 6 years between 1987 to 1989 and 1993 to 1995 was considered. The primary study outcome was physician adjudicated SCD. The study population consisted of 15 026 subjects, of whom 12 207 had data about CVH change. Over a median follow‐up of 23.0 years, 583 cases of SCD were recorded. There was a strong inverse association between baseline CVH metrics and time varying CVH metrics with risk of SCD. Compared with subjects with consistently poor CVH, risk of SCD was lower in those changed from poor to intermediate/ideal (hazard ratio [HR], 0.67 [95% CI, 0.48–0.94]), intermediate to poor (HR, 0.73 [95% CI, 0.54–0.99]), intermediate to ideal (HR, 0.49 [95% CI, 0.24–0.99]), ideal to poor/intermediate CVH (HR, 0.23 [95% CI, 0.10–0.52]), or those with consistently intermediate (HR, 0.49 [95% CI, 0.36–0.66]) or consistently ideal CVH (HR, 0.31 [95% CI, 0.13–0.76]). Similar results were also observed for non‐SCD.

Conclusions

Compared with consistently poor CVH, other patterns of change in CVH were associated with lower risk of SCD. These findings highlight the importance of promotion of ideal CVH in the primordial prevention of SCD.

Artificial Intelligence-Based Medical Data Mining

Understanding published unstructured textual data using traditional text mining approaches and tools is becoming a challenging issue due to the rapid increase in electronic open-source publications. The application of data mining techniques in the medical sciences is an emerging trend; however, traditional text-mining approaches are insufficient to cope with the current upsurge in the volume of published data. Therefore, artificial intelligence-based text mining tools are being developed and used to process large volumes of data and to explore the hidden features and correlations in the data. This review provides a clear-cut and insightful understanding of how artificial intelligence-based data-mining technology is being used to analyze medical data. We also describe a standard process of data mining based on CRISP-DM (Cross-Industry Standard Process for Data Mining) and the most common tools/libraries available for each step of medical data mining.

Correlation Analysis of Variables From the Atherosclerosis Risk in Communities Study

The need to test chemicals in a timely and cost-effective manner has driven the development of new alternative methods (NAMs) that utilize in silico and in vitro approaches for toxicity prediction. There is a wealth of existing data from human studies that can aid in understanding the ability of NAMs to support chemical safety assessment. This study aims to streamline the integration of data from existing human cohorts by programmatically identifying related variables within each study. Study variables from the Atherosclerosis Risk in Communities (ARIC) study were clustered based on their correlation within the study. The quality of the clusters was evaluated via a combination of manual review and natural language processing (NLP). We identified 391 clusters including 3,285 variables. Manual review of the clusters containing more than one variable determined that human reviewers considered 95% of the clusters related to some degree. To evaluate potential bias in the human reviewers, clusters were also scored via NLP, which showed a high concordance with the human classification. Clusters were further consolidated into cluster groups using the Louvain community finding algorithm. Manual review of the cluster groups confirmed that clusters within a group were more related than clusters from different groups. Our data-driven approach can facilitate data harmonization and curation efforts by providing human annotators with groups of related variables reflecting the themes present in the data. Reviewing groups of related variables should increase efficiency of the human review, and the number of variables reviewed can be reduced by focusing curator attention on variable groups whose theme is relevant for the topic being studied.

Plasma Proprotein Convertase Subtilisin/kexin Type 9 (PCSK9) in the Acute Respiratory Distress Syndrome

Background

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a serine protease that is a mediator of the immune response to sepsis. PCSK9 is also highly expressed in pneumocytes and pulmonary endothelial cells. We hypothesized that serum PCSK9 levels would be associated with death and ICU outcomes in patients with ARDS.

Methods

Using data and plasma samples from the NIH BioLINCC data repository, we assembled a cohort of 1,577 patients with the acute respiratory distress syndrome (ARDS) enrolled in two previously completed clinical trials, EDEN and SAILS. We measured PCSK9 levels in plasma within 24 h of intubation using commercially available ELISA kits (R&D Systems). We assessed the association of PCSK9 with mortality using Cox proportional hazard models. We also assessed clinical factors associated with PCSK9 level and the association of PCSK9 with the number of days free of mechanical ventilation and days free of ICU care.

Results

In 1,577 ARDS patients, median age was 53 years (IQR 42–65 years) and median APACHE III score 91 (72–111) connoting moderate critical illness. PCSK9 levels were 339.3 ng/mL (IQR 248.0–481.0). In multivariable models, race, cause of ARDS, body mass index, pre-existing liver disease, body temperature, sodium, white blood cell count and platelet count were associated with PCSK9 level. Presence of sepsis, use of vasopressors and ventilator parameters were not associated with PCSK9 level. PCSK9 levels were not associated with in-hospital mortality (HR per IQR 0.96, 95% CI 0.84–1.08, P = 0.47). Higher PCSK9 levels were associated with fewer ICU and ventilator free days.

Conclusions

Plasma PCSK9 is not associated with mortality in ARDS, however higher PCSK9 levels are associated with secondary outcomes of fewer ICU free and ventilator free days. Clinical factors associated with PCSK9 in ARDS are largely unmodifiable. Further research to define the mechanism of this association is warranted.

Cardiovascular risk factors and breast cancer incidence in a large middle-aged cohort study

BACKGROUND

Several studies have demonstrated that cardiovascular risk factors play a role in the etiology of breast cancer. However, the combined effect of cardiovascular risk factors on the risk of breast cancer is still uncertain.

METHODS

Data from the Atherosclerosis Risk in Communities (ARIC) study, a prospective cohort of middle-aged women, were used to investigate the association of individual and combined cardiovascular risk factors with breast cancer. Cox proportional hazards models were applied to calculate the hazard ratio (HR) and 95% confidence intervals (CI).

RESULTS

A total of 7501 women were included. During a mean follow-up of 19.7 years, 576 women were diagnosed with breast cancer. White women and premenopausal status were independently associated with increased risk of breast cancer. Of the individual cardiovascular risk factors, only obesity was independently associated with an increased risk of breast cancer (HR 1.29, 95% CI 1.04-1.61). Compared with women without cardiovascular risk factors, women having three or greater, but not those with fewer than three cardiovascular risk factors, had a significantly higher risk of developing breast cancer (HR 1.27, 95% CI 1.06-1.53). Subgroup analyses indicated that women with three or greater cardiovascular risk factors had higher risk of breast cancer among postmenopausal Black women, but not among premenopausal Black and White women.

CONCLUSIONS

Combinations of cardiovascular risk factors are associated with increased risk of breast cancer in middle-aged women, especially in postmenopausal Black women. Joint interventions to modify cardiovascular risk factors could be used to prevent breast cancer in these higher-risk individuals.

Association Between Temporal Changes in Early Repolarization Pattern With Long-Term Cardiovascular Outcome: A Population-Based Cohort Study

Background

The prognostic value of early repolarization pattern (ERP) remains controversial. We aim to test the hypothesis that temporal changes in ERP are associated with increased risks for sudden cardiac death (SCD) and cardiovascular death.

Methods and Results

A total of 14 679 middle‐aged participants from the prospective, population‐based cohort were included in this analysis, with ERP status recorded at baseline and during 3 follow‐up visits in the ARIC (Atherosclerosis Risk in Communities) study. We related baseline ERP, time‐varying ERP, and temporal changes in ERP to cardiovascular outcomes. Cox models were used to estimate the hazard ratios (HRs) adjusted for possible confounding factors. With a median follow‐up of 22.5 years, there were 5033 deaths, 1239 cardiovascular deaths, and 571 SCDs. Time‐varying ERP was associated with increased risks of SCD (HR, 1.59 [95% CI, 1.25–2.02]), cardiovascular death (HR, 1.70 [95% CI, 1.44–2.00]), and death from any cause (HR, 1.16 [95% CI, 1.05–1.27]). Baseline ERP was also associated with 3 outcomes. Compared with those with consistently normal ECG findings, subjects with new‐onset ERP or consistent ERP experienced increased risks of developing SCD and cardiovascular death. The time‐varying ERP in women, White subjects, and anterior leads and J‐wave amplitudes ≥0.2 mV appeared to indicate poorer cardiovascular outcomes.

Conclusions

Our findings suggest that baseline ERP, time‐varying ERP, new‐onset ERP, and consistent ERP were independent predictors of SCD and cardiovascular death in the middle‐aged biracial population. Repeated measurements of the ERP might improve its use as a risk indicator for SCD.

Plasma Metabolites Associated with a Protein-Rich Dietary Pattern: Results from the OmniHeart Trial

Scope

Lack of biomarkers is a challenge for the accurate assessment of protein intake and interpretation of observational study data. The study aims to identify biomarkers of a protein‐rich dietary pattern.

Methods and Results

The Optimal Macronutrient Intake Trial to Prevent Heart Disease (OmniHeart) trial is a randomized cross‐over feeding study which tested three dietary patterns with varied macronutrient content (carbohydrate‐rich; protein‐rich with about half from plant sources; and unsaturated fat‐rich). In 156 adults, differences in log‐transformed plasma metabolite levels at the end of the protein‐ and carbohydrate‐rich diet periods using paired t‐tests is examined. Partial least‐squares discriminant analysis is used to identify a set of metabolites which are influential in discriminating between the protein‐rich versus carbohydrate‐rich dietary patterns. Of 839 known metabolites, 102 metabolites differ significantly between the protein‐rich and the carbohydrate‐rich dietary patterns after Bonferroni correction, the majority of which are lipids (n = 35), amino acids (n = 27), and xenobiotics (n = 24). Metabolites which are the most influential in discriminating between the protein‐rich and the carbohydrate‐rich dietary patterns represent plant protein intake, food or beverage intake, and preparation methods.

Conclusions

The study identifies many plasma metabolites associated with the protein‐rich dietary pattern. If replicated, these metabolites may be used to assess level of adherence to a similar dietary pattern.

Airflow obstruction, impaired lung function and risk of sudden cardiac death: a prospective cohort study

BACKGROUND

Growing evidence suggests that compromised lung health may be linked to cardiovascular disease. However, little is known about its association with sudden cardiac death (SCD).

OBJECTIVES

We aimed to assess the link between impaired lung function, airflow obstruction and risk of SCD by race and gender in four US communities.

METHODS

A total of 14 708 Atherosclerosis Risk in Communities (ARIC) study participants who underwent spirometry and were asked about lung health (1987-1989) were followed. The main outcome was physician-adjudicated SCD. Fine-Gray proportional subdistribution hazard models with Firth's penalised partial likelihood correction were used to estimate the HRs.

RESULTS

Over a median follow-up of 25.4 years, 706 (4.8%) subjects experienced SCD. The incidence of SCD was inversely associated with FEV₁ in each of the four race and gender groups and across all smoking status categories. After adjusting for multiple measured confounders, HRs of SCD comparing the lowest with the highest quintile of FEV₁ were 2.62 (95% CI 1.62 to 4.26) for white males, 1.80 (95% CI 1.03 to 3.15) for white females, 2.07 (95% CI 1.05 to 4.11) for black males and 2.62 (95% CI 1.21 to 5.65) for black females. The above associations were consistently observed among the never smokers. Moderate to very severe airflow obstruction was associated with increased risk of SCD. Addition of FEV₁ significantly improved the predictive power for SCD.

CONCLUSIONS

Impaired lung function and airflow obstruction were associated with increased risk of SCD in general population. Additional research to elucidate the underlying mechanisms is warranted.

The National Heart, Lung, and Blood Institute data: analyzing published articles that used BioLINCC open access data

Background: Data sharing is now a mandatory prerequisite for several major funders and journals, where researchers are obligated to deposit the data resulting from their studies in an openly accessible repository. Biomedical open data are now widely available in almost all disciplines, where researchers can freely access and reuse these data in new studies. We aim to study the BioLINCC datasets, number of publications that used BioLINCC open access data, and the citations received by these publications.

Methods: As of July 2019, there was a total of 194 datasets stored in BioLINCC repository and accessible through their portal. We requested the full list of publications that used these datasets from BioLINCC, and we also performed a supplementary PubMed search for other publications. We used Web of Science (WoS) to analyze the characteristics of publications and the citations they received, where WoS database index high quality articles.

Results: 1,086 published articles used data from BioLINCC repository for 79 (40.72%) datasets, where 115 (59.28%) datasets did not have any publications associated with it. Of the total publications, 987 (90.88%) articles were WoS indexed. The number of publications has steadily increased since 2002 and peaked in 2018 with a total number of 138 publications on that year. The 987 open data publications (i.e., secondary publications) received a total of 34,181 citations up to 1 ^st October 2019. The average citation per item for the open data publications was 34.63. The total number of citations received by open data publications per year has increased from only 2 citations in 2002, peaking in 2018 with 2361 citations.

Conclusion: Majority of BioLINCC datasets were not used in secondary publications. Despite that, the datasets used for secondary publications yielded publications in WoS indexed journals and are receiving an increasing number of citations.

Data mining in clinical big data: the frequently used databases, steps, and methodological models

Many high quality studies have emerged from public databases, such as Surveillance, Epidemiology, and End Results (SEER), National Health and Nutrition Examination Survey (NHANES), The Cancer Genome Atlas (TCGA), and Medical Information Mart for Intensive Care (MIMIC); however, these data are often characterized by a high degree of dimensional heterogeneity, timeliness, scarcity, irregularity, and other characteristics, resulting in the value of these data not being fully utilized. Data-mining technology has been a frontier field in medical research, as it demonstrates excellent performance in evaluating patient risks and assisting clinical decision-making in building disease-prediction models. Therefore, data mining has unique advantages in clinical big-data research, especially in large-scale medical public databases. This article introduced the main medical public database and described the steps, tasks, and models of data mining in simple language. Additionally, we described data-mining methods along with their practical applications. The goal of this work was to aid clinical researchers in gaining a clear and intuitive understanding of the application of data-mining technology on clinical big-data in order to promote the production of research results that are beneficial to doctors and patients.

A Low-Sodium DASH Dietary Pattern Affects Serum Markers of Inflammation and Mineral Metabolism in Adults with Elevated Blood Pressure

BACKGROUND

The blood pressure-lowering effects of the Dietary Approaches to Stop Hypertension (DASH) dietary pattern and reduced sodium intake are well established. The effects on other biomarkers related to vascular health are of interest and might assist in explaining the effects of the DASH diet and sodium reduction.

OBJECTIVES

We hypothesized that a low-sodium DASH diet improves (lowers) biomarkers of inflammation [C-reactive protein (CRP) and soluble urokinase plasminogen activator receptor (suPAR)] and mineral metabolism [phosphorus and fibroblast growth factor-23 (FGF23)].

METHODS

We conducted a secondary analysis of the DASH-Sodium trial using frozen serum samples. This controlled feeding study randomly assigned 412 adults (≥22 y) with elevated blood pressure (120-159/80-95 mmHg) to consume either a DASH diet or control diet. Within each arm, participants received 3 sodium levels [low (1150 mg), intermediate (2300 mg), high (3450 mg)] in random sequence, each for 30 d. To maximize contrast, samples collected at the end of the low-sodium DASH (n = 198) and high-sodium control (n = 194) diets were compared. Between-diet differences in serum CRP, suPAR, phosphorus, and FGF23 concentrations were assessed using linear regression adjusted for age, sex, race, income, education, smoking status, and BMI.

RESULTS

CRP concentrations did not differ between groups (P = 0.83), but suPAR was higher after the low-sodium DASH diet than the high-sodium control [geometric mean 2470 pg/mL (95% CI: 2380, 2560 pg/mL), compared with 2290 pg/mL (95% CI: 2210, 2380 pg/mL); P = 0.006]. Phosphorus was higher after the low-sodium DASH diet [geometric mean 3.50 mg/dL (95% CI: 3.43, 3.57 mg/dL)] compared with the high-sodium control diet [geometric mean 3.39 mg/dL (95% CI: 3.33, 3.46 mg/dL); P = 0.04]. FGF23 was also higher after the low-sodium DASH diet [geometric mean 35.3 pg/mL (95% CI: 33.3, 37.3 pg/mL) compared with 28.2 pg/mL (95% CI: 26.6, 29.8 pg/mL); P < 0.001].

CONCLUSIONS

Contrary to our hypothesis, biomarkers of inflammation and mineral metabolism were increased or unchanged by a low-sodium DASH diet compared with a high-sodium control diet in adults with elevated blood pressure.

Longitudinal change in lung function and subsequent risks of cardiovascular events: evidence from four prospective cohort studies

BACKGROUND

Lung function is constantly changing over the life course. Although the relation of cross-sectional lung function measure and adverse outcomes has been reported, data on longitudinal change and subsequent cardiovascular (CV) events risks are scarce. Therefore, this study is to determine the association of longitudinal change in lung function and subsequent cardiovascular risks.

METHODS

This study analyzed the data from four prospective cohorts. Subjects with at least two lung function tests were included. We calculated the rate of forced respiratory volume in 1 s (FEV1) and forced vital capacity (FVC) decline for each subject and categorized them into quartiles. The primary outcome was CV events, defined as a composite of coronary heart disease (CHD), chronic heart failure (CHF), stroke, and any CV death. Cox proportional hazards regression and restricted cubic spline models were applied.

RESULTS

The final sample comprised 12,899 participants (mean age 48.58 years; 43.61% male). Following an average of 14.79 (10.69) years, 3950 CV events occurred. Compared with the highest FEV1 quartile (Q4), the multivariable HRs for the lowest (Q1), 2nd (Q2), and 3rd quartiles (Q3) were 1.33 (95%CI 1.19, 1.49), 1.30 (1.16, 1.46), and 1.07 (0.95, 1.21), respectively. Likewise, compared with the reference quartile (Q4), the group that experienced a faster decline in FVC had higher HRs for CV events (1.06 [95%CI 0.94-1.20] for Q3, 1.15 [1.02-1.30] for Q2, and 1.28 [1.14-1.44] for Q1). The association remained robust across a series of sensitivity analyses and nearly all subgroups but was more evident in subjects < 60 years.

CONCLUSIONS

We observed a monotonic increase in risks of CV events with a faster decline in FEV1 and FVC. These findings emphasize the value of periodic evaluation of lung function and open new opportunities for disease prevention.

Body mass index trajectories during mid to late life and risks of mortality and cardiovascular outcomes: Results from four prospective cohorts

BACKGROUND

Our understanding of the weight-outcome association mainly comes from single-time body mass index (BMI) measurement. However, data on long-term trajectories of within-person changes in BMI on diverse study outcomes are sparse. Therefore, this study is to determine the associations of individual BMI trajectories and cardiovascular outcomes.

METHODS

The present analysis was based on data from 4 large prospective cohorts and restricted to participants aged ≥45 years with at least two BMI measurements. Hazard ratios (HR) and 95% confidence intervals(95%CI) for each outcome according to different BMI trajectories were calculated in Cox regression models.

FINDINGS

The final sample comprised 29,311 individuals (mean age 58.31 years, and 77.31% were white), with a median 4 BMI measurements used in this study. During a median follow-up of 21.16 years, there were a total of 10,192 major adverse cardiovascular events (MACE) and 11,589 deaths. A U-shaped relation was seen with all study outcomes. Compared with maintaining stable weight, the multivariate adjusted HR for MACE were 1.53 (95%CI 1.40-1.66), 1.26 (95%CI 1.16-1.37) and 1.08 (95%CI 1.02-1.15) respectively for rapid, moderate and slow weight loss; 1.01 (95%CI 0.95-1.07), 1.13 (95%CI 1.05-1.21) and 1.29 (95%CI 1.20-1.40) respectively for slow, moderate and rapid weight gain. Identical patterns of association were observed for all other outcomes. The development of BMI differed markedly between the outcome-free individuals and those who went on to experience adverse events, generally beginning to diverge 10 years before the occurrence of the events.

INTERPRETATION

Our findings may signal an underlying high-risk population and inspire future studies on weight management.

FUNDING

National Natural Science Foundation of China, Guangdong Natural Science Foundation.

Urine Metabolites Associated with the Dietary Approaches to Stop Hypertension (DASH) Diet: Results from the DASH-Sodium Trial

SCOPE

Serum metabolomic markers of the Dietary Approaches to Stop Hypertension (DASH) diet are previously reported. In an independent study, the similarity of urine metabolomic markers are investigated.

METHODS AND RESULTS

In the DASH-Sodium trial, participants are randomly assigned to the DASH diet or control diet, and received three sodium interventions (high, intermediate, low) within each randomized diet group in random order for 30 days each. Urine samples are collected at the end of each intervention period and analyzed for 938 metabolites. Two comparisons are conducted: 1) DASH-high sodium (n = 199) versus control-high sodium (n = 193), and 2) DASH-low sodium (n = 196) versus control-high sodium. Significant metabolites identified using multivariable linear regression are compared and the top 10 influential metabolites identified using partial least-squares discriminant analysis to the results from the DASH trial. Nine out of 10 predictive metabolites of the DASH-high sodium and DASH-low sodium diets are identical. Most candidate biomarkers from the DASH trial replicated. N-methylproline, chiro-inositol, stachydrine, and theobromine replicated as influential metabolites of DASH diets.

CONCLUSIONS

Candidate biomarkers of the DASH diet identified in serum replicated in urine. Replicated influential metabolites are likely to be objective biomarkers of the DASH diet.

Association of early repolarization pattern with cardiovascular outcomes in middle-aged population: A cohort study

BACKGROUND

Large cohort studies provide conflicting evidence regarding the prognostic value of early repolarization pattern (ERP) in the general population, complicated by the complex or heterogeneous definitions of ERP applied in different studies.

HYPOTHESIS

We hypothesized that ERP was associated with increased cardiovascular risk with the definition of ERP recommended by the expert consensus statements.

METHODS

A total of 13673 middle-aged subjects from the prospective, population-based Atherosclerosis Risk in Communities (ARIC) study were included in this analysis. Cox models were used to estimate the hazard ratios (HRs) adjusted for possible confounding factors. ERP was defined as ST-segment elevation ≥0.1mV at the end of the QRS or J wave on the QRS downstroke in two or more contiguous leads.

RESULTS

Compared with those without ERP, subjects with ERP had a significantly increased risk of developing sudden cardiac death (SCD) (HR, 1.48; 95% CI, 1.08-2.04) and death from coronary heart disease (CHD) (HR, 1.45; 95% CI, 1.10-1.92) after a median follow-up of 20.1 years. ERP was significantly predictive of SCD in females, whites, younger people, and subjects with relatively low cardiovascular risk. ERP with ST-segment elevation appeared to indicate poor cardiovascular outcomes. ERP was associated with an absolute risk increase of 93.3 additional SCDs per 100 000 person-years.

CONCLUSIONS

Our findings suggest that ERP was an independent predictor of SCD and CHD death in the middle-aged biracial population.

Pulmonary and systemic hemodynamics are associated with myocardial injury in the acute respiratory distress syndrome

Background

Whether right and left heart hemodynamics are associated with myocardial injury in the acute respiratory distress syndrome (ARDS) is not known.

Methods

We performed a retrospective cohort study of subjects who had right heart catheterization within the ALVEOLI trial and Fluid and Catheter Treatment Trial. Myocardial injury was assessed using a highly sensitive troponin assay (hsTn; Abbot ARCHITECT). Hemodynamic variables included right atrial pressure, pulmonary artery wedge pressure, cardiac index and stroke volume, pulmonary vascular resistance, pulmonary arterial compliance, and pulmonary effective arterial elastance. We performed linear, logistic, and Cox regression to determine the association of hemodynamic variables with myocardial injury and to determine if hemodynamics mediated the association between myocardial injury and death.

Results

Among 252 ARDS patients, median day 0 troponin was 65.4 (13.8–397.8) ng/L. Lower cardiac index (β −0.23 SE 0.10; P < 0.001) and stroke volume (β −0.26 SE 0.005; P < 0.001), higher pulmonary vascular resistance (β 0.22 SE 0.11; P < 0.001), lower pulmonary arterial compliance (β −0.24 SE 0.06; P < 0.001), and higher arterial elastance (β 0.27 SE 0.43; P < 0.001) were associated with greater myocardial injury in univariable and adjusted models. Changes in stroke volume, cardiac index, pulmonary arterial compliance, pulmonary vascular resistance, and arterial elastance were all associated with progressive myocardial injury over three days. hsTn levels were associated with mortality; however, the association was attenuated after adjustment for each of stroke volume, pulmonary vascular resistance, pulmonary arterial compliance, and arterial elastance.

Conclusion

Pulmonary vascular hemodynamics are associated with myocardial injury in ARDS, while filling pressures are not. Pulmonary vascular disease may represent a treatable contributor to myocardial injury in ARDS.

Association between resting painless ST-segment depression with sudden cardiac death in middle-aged population: A prospective cohort study

BACKGOUND

Silent electrocardiographic ST change predicts future coronary heart disease (CHD) incidence and mortality, but the prognostic significance of painless ST-segment depression (STD) with respect to sudden cardiac death (SCD) in subjects without apparent CHD remain unclear. This study sought to test the association between non-ischemic resting STD and risk of SCD in the general population.

METHODS

A total of 14,935 middle-aged subjects from the prospective, population-based Atherosclerosis Risk in Communities (ARIC) study were included in this analysis. Cox models were used to estimate the hazard ratios (HRs) adjusted for possible confounding factors. STD was defined as ST-segment depression of ≥0.05 mV in two or more contiguous leads.

RESULTS

A total of 626 sudden cardiac death occurred during the mean follow-up of 20.4 years. Compared with those without STD, subjects with resting painless STD of at least 0.05 mV had a significantly increased risk of developing SCD (adjusted HR, 1.45; 95% CI, 1.20 to 1.76), and those with STD ≥ 0.1 mV had even higher risk of SCD (adjusted HR, 1.90; 95% CI, 1.25 to 2.88). Significant interactions were present between gender and STD (P = .03), and between race and STD (P = .01). STD was significantly predictive of SCD in males (adjusted HR, 1.57; 95% CI, 1.22-2.01) and in whites (adjusted HR, 1.65; 95% CI, 1.27-2.14). STD in lateral leads and global leads were strong predictors of SCD.

CONCLUSIONS

Resting painless STD was an independent predictor of SCD in the middle-aged population without previously diagnosed CHD.

Brief introduction of medical database and data mining technology in big data era

Data mining technology can search for potentially valuable knowledge from a large amount of data, mainly divided into data preparation and data mining, and expression and analysis of results. It is a mature information processing technology and applies database technology. Database technology is a software science that researches manages, and applies databases. The data in the database are processed and analyzed by studying the underlying theory and implementation methods of the structure, storage, design, management, and application of the database. We have introduced several databases and data mining techniques to help a wide range of clinical researchers better understand and apply database technology.

Racial Differences in Sudden Cardiac Death

BACKGROUND

Blacks have a higher incidence of out-of-hospital sudden cardiac death (SCD) in comparison with whites. However, the racial differences in the cumulative risk of SCD and the reasons for these differences have not been assessed in large-scale community-based cohorts. The objective of this study is to compare the lifetime cumulative risk of SCD among blacks and whites, and to evaluate the risk factors that may explain racial differences in SCD risk in the general population.

METHODS

This is a cohort study of 3832 blacks and 11 237 whites participating in the Atherosclerosis Risk in Communities Study (ARIC). Race was self-reported. SCD was defined as a sudden pulseless condition from a cardiac cause in a previously stable individual, and SCD cases were adjudicated by an expert committee. Cumulative incidence was computed using competing risk models. Potential mediators included demographic and socioeconomic factors, cardiovascular risk factors, presence of coronary heart disease, and electrocardiographic parameters as time-varying factors.

RESULTS

The mean (SD) age was 53.6 (5.8) years for blacks and 54.4 (5.7) years for whites. During 27.4 years of follow-up, 215 blacks and 332 whites experienced SCD. The lifetime cumulative incidence of SCD at age 85 years was 9.6, 6.6, 6.5, and 2.3% for black men, black women, white men, and white women, respectively. The sex-adjusted hazard ratio for SCD comparing blacks with whites was 2.12 (95% CI, 1.79-2.51). The association was attenuated but still statistically significant in fully adjusted models (hazard ratio, 1.38; 95% CI, 1.11-1.71). In mediation analysis, known factors explained 65.3% (95% CI 37.9-92.8%) of the excess risk of SCD in blacks in comparison with whites. The single most important factor explaining this difference was income (50.5%), followed by education (19.1%), hypertension (22.1%), and diabetes mellitus (19.6%). Racial differences were evident in both genders but stronger in women than in men.

CONCLUSIONS

Blacks had a much higher risk for SCD in comparison with whites, particularly among women. Income, education, and traditional risk factors explained ≈65% of the race difference in SCD. The high burden of SCD and the racial-gender disparities observed in our study represent a major public health and clinical problem.

Quantifying Importance of Major Risk Factors for Coronary Heart Disease

Supplemental Digital Content is available in the text.

Background:

To optimize preventive strategies for coronary heart disease (CHD), it is essential to understand and appropriately quantify the contribution of its key risk factors. Our objective was to compare the associations of key modifiable CHD risk factors—specifically lipids, systolic blood pressure (SBP), diabetes mellitus, and smoking—with incident CHD events based on their prognostic performance, attributable risk fractions, and treatment benefits, overall and by age.

Methods:

Pooled participant-level data from 4 observational cohort studies sponsored by the National Heart, Lung, and Blood Institute were used to create a cohort of 22 626 individuals aged 45 to 84 years who were initially free of cardiovascular disease. Individuals were followed for 10 years from baseline evaluation for incident CHD. Proportional hazards regression was used to estimate metrics of prognostic model performance (likelihood ratio, C index, net reclassification, discrimination slope), hazard ratios, and population attributable fractions for SBP, non–high-density lipoprotein cholesterol (non–HDL-C), diabetes mellitus, and smoking. Expected absolute risk reductions for antihypertensive and lipid-lowering treatment were assessed.

Results:

Age, sex, and race capture 63% to 80% of the prognostic performance of cardiovascular risk models. In contrast, adding either SBP, non–HDL-C, diabetes mellitus, or smoking to a model with other risk factors increases the C index by only 0.004 to 0.013. However, primordial prevention could have a substantial effect as demonstrated by population attributable fractions of 28% for SBP≥130 mm Hg and 17% for non–HDL-C≥130 mg/dL. Similarly, lowering the SBP of all individuals to <130 mm Hg or lowering low-density lipoprotein cholesterol by 30% would be expected to lower a baseline 10-year CHD risk of 10.7% to 7.0 and 8.0, respectively (absolute risk reductions: 3.7% and 2.7%, respectively). Prognostic performance decreases with age (C indices for age groups 45–54, 55–64, 65–74, 75–84 are 0.75, 0.72, 0.66, and 0.62, respectively), whereas absolute risk reductions increase (SBP: 1.1%, 2.3%, 5.4%, 10.3%, respectively; non–HDL-C: 1.1%, 2.0%, 3.7%, 5.9%, respectively).

Conclusions:

Although individual modifiable CHD risk factors contribute only modestly to prognostic performance, our models indicate that eliminating or controlling these individual factors would lead to substantial reductions in total population CHD events. Metrics used to judge importance of risk factors should be tailored to the research objectives.

Bringing radiomics into a multi-omics framework for a comprehensive genotype-phenotype characterization of oncological diseases

Genomic and radiomic data integration, namely radiogenomics, can provide meaningful knowledge in cancer diagnosis, prognosis and treatment. Despite several data structures based on multi-layer architecture proposed to combine multi-omic biological information, none of these has been designed and assessed to include radiomic data as well. To meet this need, we propose to use the MultiAssayExperiment (MAE), an R package that provides data structures and methods for manipulating and integrating multi-assay experiments, as a suitable tool to manage radiogenomic experiment data. To this aim, we first examine the role of radiogenomics in cancer phenotype definition, then the current state of radiogenomics data integration in public repository and, finally, challenges and limitations of including radiomics in MAE, designing an extended framework and showing its application on a case study from the TCGA-TCIA archives. Radiomic and genomic data from 91 patients have been successfully integrated in a single MAE object, demonstrating the suitability of the MAE data structure as container of radiogenomic data.

Development of a Biobank from a Legacy Collection in Universitas Gadjah Mada, Indonesia: Proposed Approach for Centralized Biobank Development in Low-Resource Institutions

Introduction: The establishment of a biobank requires specific expertise along with relatively expensive infrastructure and appropriate technology. This causes certain challenges in biobank implementation for research in low-middle-income countries. Biobank development with established specimens and data collection (legacy collection) was an approach used in the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada. This approach aimed to identify the resources available at present, while providing nontechnical information for further development of a centralized biobank. Materials and Methods: Retrospective modeling was done in 2015 by recruiting existing specimen collections and their associated data. The steps were as follows: (1) informing research stakeholders through discussion with experts and stakeholders; (2) identifying specimen collections to be used; (3) determining the system, infrastructure, and consumables needed; (4) determining inclusion criteria; (5) building an in-house database system; (6) organizing data and physical specimen collections; and (7) validating data and physical sample arrangement. All technical procedures were built into standard operating procedures. Results: The model included specimens from one -80°C freezer. The associated data included demographic, clinical diagnosis, and physical sample information. Samples came from six studies, collected between 2001 and 2014. A web-based database was built based on the MySQL programming system. Information on biospecimens from a total of 4196 subjects collected in 11,358 vials was entered into the database, following physical rearrangement of vials in the -80°C freezer with one-dimensional barcodes taped to vials, boxes, and racks. A validation test was done for data concordance between the database and physical arrangement in the -80°C freezer, showing no discrepancies. Conclusion: This report demonstrated current technical and nontechnical insights to further develop a centralized biobank for health research at an academic institution in Indonesia.

Pre-Vent: the prematurity-related ventilatory control study

BACKGROUND

The increasing incidence of bronchopulmonary dysplasia in premature babies may be due in part to immature ventilatory control, contributing to hypoxemia. The latter responds to ventilation and/or oxygen therapy, treatments associated with adverse sequelae. This is an overview of the Prematurity-Related Ventilatory Control Study which aims to analyze the under-utilized cardiorespiratory continuous waveform monitoring data to delineate mechanisms of immature ventilatory control in preterm infants and identify predictive markers.

METHODS

Continuous ECG, heart rate, respiratory, and oxygen saturation data will be collected throughout the NICU stay in 500 infants < 29 wks gestation across 5 centers. Mild permissive hypercapnia, and hyperoxia and/or hypoxia assessments will be conducted in a subcohort of infants along with inpatient questionnaires, urine, serum, and DNA samples.

RESULTS

Primary outcomes will be respiratory status at 40 wks and quantitative measures of immature breathing plotted on a standard curve for infants matched at 36-37 wks. Physiologic and/or biologic determinants will be collected to enhance the predictive model linking ventilatory control to outcomes.

CONCLUSIONS

By incorporating bedside monitoring variables along with biomarkers that predict respiratory outcomes we aim to elucidate individualized cardiopulmonary phenotypes and mechanisms of ventilatory control contributing to adverse respiratory outcomes in premature infants.

Discovering Pediatric Asthma Phenotypes on the Basis of Response to Controller Medication Using Machine Learning

RATIONALE

Pediatric asthma has variable underlying inflammation and symptom control. Approaches to addressing this heterogeneity, such as clustering methods to find phenotypes and predict outcomes, have been investigated. However, clustering based on the relationship between treatment and clinical outcome has not been performed, and machine learning approaches for long-term outcome prediction in pediatric asthma have not been studied in depth.

OBJECTIVES

Our objectives were to use our novel machine learning algorithm, predictor pursuit (PP), to discover pediatric asthma phenotypes on the basis of asthma control in response to controller medications, to predict longitudinal asthma control among children with asthma, and to identify features associated with asthma control within each discovered pediatric phenotype.

METHODS

We applied PP to the Childhood Asthma Management Program study data (n = 1,019) to discover phenotypes on the basis of asthma control between assigned controller therapy groups (budesonide vs. nedocromil). We confirmed PP's ability to discover phenotypes using the Asthma Clinical Research Network/Childhood Asthma Research and Education network data. We next predicted children's asthma control over time and compared PP's performance with that of traditional prediction methods. Last, we identified clinical features most correlated with asthma control in the discovered phenotypes.

RESULTS

Four phenotypes were discovered in both datasets: allergic not obese (A⁺/O^-), obese not allergic (A^-/O⁺), allergic and obese (A⁺/O⁺), and not allergic not obese (A^-/O^-). Of the children with well-controlled asthma in the Childhood Asthma Management Program dataset, we found more nonobese children treated with budesonide than with nedocromil (P = 0.015) and more obese children treated with nedocromil than with budesonide (P = 0.008). Within the obese group, more A⁺/O⁺ children's asthma was well controlled with nedocromil than with budesonide (P = 0.022) or with placebo (P = 0.011). The PP algorithm performed significantly better (P < 0.001) than traditional machine learning algorithms for both short- and long-term asthma control prediction. Asthma control and bronchodilator response were the features most predictive of short-term asthma control, regardless of type of controller medication or phenotype. Bronchodilator response and serum eosinophils were the most predictive features of asthma control, regardless of type of controller medication or phenotype.

CONCLUSIONS

Advanced statistical machine learning approaches can be powerful tools for discovery of phenotypes based on treatment response and can aid in asthma control prediction in complex medical conditions such as asthma.

Trimethylamine N-oxide and incident atherosclerotic events in high-risk individuals with diabetes: an ACCORD trial post hoc analysis

INTRODUCTION

Type 2 diabetes mellitus (T2D) confers high atherosclerotic cardiovascular disease (ASCVD) risk. The metabolite trimethylamine N-oxide (TMAO) derived via gut flora has been linked to excess ASCVD.

RESEARCH DESIGN AND METHODS

We analyzed data, biospecimens, and major adverse cardiovascular events (MACEs) from the prospective multicenter randomized Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial to assess its value in 330 high-risk individuals with T2D without evident atherosclerotic disease at enrollment.

RESULTS

Incident cardiovascular events occurred in 165 cases; 165 controls matched by age, sex, and treatment arm experienced no incident events during follow-up. Cases and controls (mean age 64.5 years) had similar mean glycated hemoglobin (HbA1c) (8.2%) and mean 10-year ASCVD risk (23.5%); groups also had similar use of statins and antihypertensive medications at baseline and follow-up. Baseline plasma TMAO levels did not differ between groups after adjusting for ASCVD risk score, HbA1c, and estimated glomerular filtration rate, nor did TMAO distinguish patients suffering incident MACE from those who remained event-free.

CONCLUSIONS

TMAO's prognostic value for incident ASCVD events may be blunted when applied to individuals with T2D with poor glycemic control and high baseline ASCVD risk. These results behoove further translational investigations of unique mechanisms of ASCVD risk in T2D.

Progressive myocardial injury is associated with mortality in the acute respiratory distress syndrome

PURPOSE

Myocardial injury connotes worse prognosis in the Acute Respiratory Distress Syndrome (ARDS), however the prognostic connotation of changes in cardiac troponin (cTn) levels in ARDS patients is not known.

METHODS

We performed a study of 908 ARDS patients enrolled in two previously completed ARDS Network trials. We obtained plasma samples via the NIH BIOLINCC repository and measured cTn using the ARCHITECT STAT high sensitivity troponin-I assay (Abbott Laboratories) at trial day 0 and 3. We constructed Cox proportional hazard models to determine the association between 60-day mortality and quintiles of percentage change in high-sensitivity troponin (ΔhsTnI).

RESULTS

The median percent change in hsTnI (%ΔhsTnI) from day 0 to day 3 was -58.2% (IQR -79.0 to 0%). After multivariable adjustment, participants with a 32.1% or greater increase in hsTnI between day 0 and day 3 (highest quintile) had a 2.27 fold increased risk for mortality (95% CI 1.29 - 3.99, p = 0.002) as well as fewer ventilator-free and ICU-free days compared to the lowest quintile.

CONCLUSION

Progressive myocardial injury in ARDS patients is associated with worse outcome, independent of severity of critical illness. Investigation of the mechanisms underlying this relationship is warranted to guide possible strategies to mitigate myocardial injury in ARDS.

Overview and experience of the YODA Project with clinical trial data sharing after 5 years

The Yale University Open Data Access (YODA) Project has facilitated access to clinical trial data since 2013. The purpose of this article is to provide an overview of the Project, describe key decisions that were made when establishing data sharing policies, and suggest how our experience and the experiences of our first two data generator partners, Medtronic, Inc. and Johnson & Johnson, can be used to enhance other ongoing or future initiatives.

Serum untargeted metabolomic profile of the Dietary Approaches to Stop Hypertension (DASH) dietary pattern

Background

The Dietary Approaches to Stop Hypertension (DASH) dietary pattern is recommended for cardiovascular disease risk reduction. Assessment of dietary intake has been limited to subjective measures and a few biomarkers from 24-h urine collections.

Objective

The aim of the study was to use metabolomics to identify serum compounds that are associated with adherence to the DASH dietary pattern.

Design

We conducted untargeted metabolomic profiling in serum specimens collected at the end of 8 wk following the DASH diet (n = 110), the fruit and vegetables diet (n = 111), or a control diet (n = 108) in a multicenter, randomized clinical feeding study (n = 329). Multivariable linear regression was used to determine the associations between the randomized diets and individual log-transformed metabolites after adjustment for age, sex, race, education, body mass index, and hypertension. Partial least-squares discriminant analysis (PLS-DA) was used to identify a panel of compounds that discriminated between the dietary patterns. The area under the curve (C statistic) was calculated as the cumulative ability to distinguish between dietary patterns. We accounted for multiple comparisons with the use of the Bonferroni method (0.05 of 818 metabolites = 6.11 × 10-5).

Results

Serum concentrations of 44 known metabolites differed significantly between participants randomly assigned to the DASH diet compared with both the control diet and the fruit and vegetables diet, which included an amino acid, 2 cofactors and vitamins (n = 2), and lipids (n = 41). With the use of PLS-DA, component 1 explained 29.4% of the variance and component 2 explained 12.6% of the variance. The 10 most influential metabolites for discriminating between the DASH and control dietary patterns were N-methylproline, stachydrine, tryptophan betaine, theobromine, 7-methylurate, chiro-inositol, 3-methylxanthine, methyl glucopyranoside, β-cryptoxanthin, and 7-methylxanthine (C statistic = 0.986).

Conclusions

An untargeted metabolomic platform identified a broad array of serum metabolites that differed between the DASH diet and 2 other dietary patterns. This newly identified metabolite panel may be used to assess adherence to the DASH dietary pattern. This trial was registered at http://www.clinicaltrials.gov as NCT03403166.

Patient consent to publication and data sharing in industry and NIH-funded clinical trials

BACKGROUND

Participants are recruited into clinical trials under the assumption that the research will contribute to medical knowledge. Therefore, non-publication trials-and, more recently, lack of data sharing-are widely considered to violate the trust of trial participants. Existing practices regarding patient consent to publication and data sharing have not been evaluated. Analyzing informed consent forms (ICFs), we studied what trial participants were told regarding investigators' intention to contribute to medical knowledge, publish trial results, and share de-identified trial data.

METHODS

We obtained 98 ICFs of industry-funded pre-marketing trials for all (17) antibiotics approved by the European Medicines Agency and 46 ICFs of publicly funded trials from the National Heart, Lung and Blood Institute Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC) data repository. Three authors independently reviewed ICFs to identify and extract what was stated or implied regarding: (1) publication of results; (2) sharing de-identified data; (3) data ownership; (4) confidentiality of identifiable data; and (5) whether the trial will produce knowledge that offers public benefit. Consensus was obtained from the two reviewers with the greatest overall agreement on all five measures. Disagreements were resolved through discussion among all authors.

RESULTS

Four (3%) trials indicated a commitment to publish trial results; 140 (97%) did not commit to publishing trial results; six (4%) indicated a commitment to share de-identified data with third party researchers. Commitments to share were more common in publicly funded trials than industry-funded trials (7% vs 3%). A total of 103 (72%) ICFs indicated the trials will or may produce knowledge that offers public benefits, while 131 (91%) ICFs left unstated who "owned" trial data; of those with statements, the sponsor always claimed ownership. Patient confidentiality was guaranteed in 137 (95%) trials.

CONCLUSIONS

Our results suggest that consent forms rarely disclose investigators' intentions regarding the sharing of de-identified data or publication of trial results.

DASH, the data and specimen hub of the National Institute of Child Health and Human Development

The benefits of data sharing are well-established and an increasing number of policies require that data be shared upon publication of the main study findings. As data sharing becomes the new norm, there is a heightened need for additional resources to drive efficient data reuse. This article describes the development and implementation of the Data and Specimen Hub (DASH) by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) to promote data sharing from NICHD-funded studies and enable researchers to comply with NIH data sharing policies. DASH’s flexible architecture is designed to archive diverse data types and formats from NICHD’s broad scientific portfolio in a manner that promotes FAIR data sharing principles. Performance of DASH over two years since launch is promising: the number of available studies and data requests are growing; three manuscripts have been published from data reanalysis, all within two years of access. Critical success factors included NICHD leadership commitment, stakeholder engagement and close coordination between the governance body and technical team.

Prevalence and Prognostic Association of Circulating Troponin in the Acute Respiratory Distress Syndrome

OBJECTIVE

Circulating cardiac troponin has been associated with adverse prognosis in the acute respiratory distress syndrome in small and single-center studies; however, comprehensive studies of myocardial injury in acute respiratory distress syndrome using modern high-sensitivity troponin assays, which can detect troponin at much lower circulating concentrations, have not been performed.

DESIGN

We performed a prospective cohort study.

SETTING

We included patients enrolled in previously completed trials of acute respiratory distress syndrome.

PATIENTS

One thousand fifty-seven acute respiratory distress syndrome patients were included.

INTERVENTIONS

To determine the association of circulating high-sensitivity troponin I (Abbott ARCHITECT), with acute respiratory distress syndrome outcomes, we measured high-sensitivity troponin I within 24 hours of intubation. The primary outcome was 60-day mortality.

MEASUREMENTS AND MAIN RESULTS

Detectable high-sensitivity troponin I was present in 94% of patients; 38% of patients had detectable levels below the 99th percentile of a healthy reference population (26 ng/L), whereas 56% of patients had levels above the 99th percentile cut point. After multivariable adjustment, age, cause of acute respiratory distress syndrome, temperature, heart rate, vasopressor use, Sequential Organ Failure Assessment score, creatinine, and PCO2 were associated with higher high-sensitivity troponin I concentration. After adjustment for age, sex, and randomized trial assignment, the hazard ratio for 60-day mortality comparing the fifth to the first quintiles of high-sensitivity troponin I was 1.61 (95% CI, 1.11-2.32; p trend = 0.003). Adjusting for Sequential Organ Failure Assessment score suggested that this association was not independent of disease severity (hazard ratio, 0.95; 95% CI, 0.64-1.39; p = 0.93).

CONCLUSIONS

Circulating troponin is detectable in over 90% of patients with acute respiratory distress syndrome and is associated with degree of critical illness. The magnitude of myocardial injury correlated with mortality.

Resource implications of preparing individual participant data from a clinical trial to share with external researchers

BACKGROUND

Demands are increasingly being made for clinical trialists to actively share individual participant data (IPD) collected from clinical trials using responsible methods that protect the confidentiality and privacy of clinical trial participants. Clinical trialists, particularly those receiving public funding, are often concerned about the additional time and money that data-sharing activities will require, but few published empirical data are available to help inform these decisions. We sought to evaluate the activity and resources required to prepare anonymised IPD from a clinical trial in anticipation of a future data-sharing request.

METHODS

Data from two UK publicly funded clinical trials were used for this exercise: 2437 participants with epilepsy recruited from 90 hospital outpatient clinics in the SANAD trial and 146 children with neuro-developmental problems recruited from 18 hospitals in the MENDS trial. We calculated the time and resources required to prepare each anonymised dataset and assemble a data pack ready for sharing.

RESULTS

The older SANAD trial (published 2007) required 50 hours of staff time with a total estimated associated cost of £3185 whilst the more recently completed MENDS trial (published 2012) required 39.5 hours of staff time with total estimated associated cost of £2540.

CONCLUSIONS

Clinical trial researchers, funders and sponsors should consider appropriate resourcing and allow reasonable time for preparing IPD ready for subsequent sharing. This process would be most efficient if prospectively built into the standard operational design and conduct of a clinical trial. Further empirical examples exploring the resource requirements in other settings is recommended.

TRIAL REGISTRATION

SANAD: International Standard Randomised Controlled Trials Registry: ISRCTN38354748 . Registered on 25 April 2003.

MENDS

EU Clinical Trials Register Eudract 2006-004025-28 . Registered on 16 May 2007. International Standard Randomised Controlled Trials Registry: ISRCTN05534585 /MREC 07/MRE08/43. Registered on 26 January 2007.

Providing researchers with online access to NHLBI biospecimen collections: The results of the first six years of the NHLBI BioLINCC program

The National Heart, Lung, and Blood Institute (NHLBI), within the United States' National Institutes of Health (NIH), established the Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC) in 2008 to develop the infrastructure needed to link the contents of the NHLBI Biorepository and the NHLBI Data Repository, and to promote the utilization of these scientific resources by the broader research community. Program utilization metrics were developed to measure the impact of BioLINCC on Biorepository access by researchers, including visibility, program efficiency, user characteristics, scientific impact, and research types. Input data elements were defined and are continually populated as requests move through the process of initiation through fulfillment and publication. This paper reviews the elements of the tracking metrics which were developed for BioLINCC and reports the results for the first six on-line years of the program.

Pulmonary Arterial Compliance in Acute Respiratory Distress Syndrome: Clinical Determinants and Association With Outcome From the Fluid and Catheter Treatment Trial Cohort

OBJECTIVES

Pulmonary vascular dysfunction is associated with adverse prognosis in patients with the acute respiratory distress syndrome; however, the prognostic impact of pulmonary arterial compliance in acute respiratory distress syndrome is not established.

DESIGN, SETTING, PATIENTS

We performed a retrospective analysis of 363 subjects with acute respiratory distress syndrome who had complete baseline right heart catheterization data from the Fluid and Catheter Treatment Trial to test whether pulmonary arterial compliance at baseline and over the course of treatment predicted mortality.

MAIN RESULTS

Baseline pulmonary arterial compliance (hazard ratio, 1.18 per interquartile range of 1/pulmonary arterial compliance; 95% CI, 1.02-1.37; p = 0.03) and pulmonary vascular resistance (hazard ratio, 1.28 per interquartile range; 95% CI, 1.07-1.53; p = 0.006) both modestly predicted 60-day mortality. Baseline pulmonary arterial compliance remained predictive of mortality when pulmonary vascular resistance was in the normal range (p = 0.02). Between day 0 and day 3, pulmonary arterial compliance increased in acute respiratory distress syndrome survivors and remained unchanged in nonsurvivors, whereas pulmonary vascular resistance did not change in either group. The resistance-compliance product (resistance-compliance time) increased in survivors compared with nonsurvivors, suggesting improvements in right ventricular load.

CONCLUSIONS

Baseline measures of pulmonary arterial compliance and pulmonary vascular resistance predict mortality in acute respiratory distress syndrome, and pulmonary arterial compliance remains predictive even when pulmonary vascular resistance is normal. Pulmonary arterial compliance and right ventricular load improve over time in acute respiratory distress syndrome survivors. Future studies should assess the impact of right ventricular protective acute respiratory distress syndrome treatment on right ventricular afterload and outcome.

Use of the National Heart, Lung, and Blood Institute Data Repository

BACKGROUND

Research on data sharing from clinical trials has focused on elucidating perceptions, barriers, and attitudes among trialists and study participants with respect to sharing data. However, little information exists regarding utilization or associated publication of articles once clinical trial data have been widely shared.

METHODS

We analyzed administrative records of investigator requests for data access, linked publications, and bibliometrics to describe the use of the National Heart, Lung, and Blood Institute data repository.

RESULTS

From January 2000 through May 2016, a total of 370 investigators requested data from 1 or more clinical trials. Requests for trial data have been increasing, with 195 investigators (53%) initiating requests during the last 4.4 years of the study period. The predominant reason for requesting data was post hoc secondary analysis of new questions (72%), followed by analytic or statistical approaches to clinical trials (9%) and meta-analyses or pooled study research (7%). Of 172 requests with online project descriptions, only 2 requests were initiated for reanalysis of primary-outcome findings. Data from 88 of 100 available clinical trials were requested at least once, and the median time from repository availability to first request was 235 days. A total of 277 articles were published on the basis of data from 47 trials. Citation metrics from 224 articles indicated that half of the publications have cumulative citations that rank in the top 34% normalized for subject category and year of publication.

CONCLUSIONS

Demand for trial data for secondary analysis has been increasing. Requesting data for the a priori purpose of reanalysis or verification of original findings was rare.

Data sharing in clinical trials: An experience with two large cancer screening trials

Paul Pinsky of the US National Cancer Institute and colleagues describe the implementation and outcomes of web-based data sharing from the PLCO and NLST cancer screening trials.

Efficiently Maintaining a National Resource of Historical and Contemporary Biological Collections: The NHLBI Biorepository Model

INTRODUCTION

Reducing costs by improving storage efficiency has been a focus of the National Heart, Lung, and Blood Institute (NHLBI) Biologic Specimen Repository (Biorepository) and Biologic Specimen and Data Repositories Information Coordinating Center (BioLINCC) programs for several years.

METHODS

Study specimen profiles were compiled using the BioLINCC collection catalog. Cost assessments and calculations on the return on investments to consolidate or reduce a collection, were developed and implemented.

RESULTS

Over the course of 8 months, the NHLBI Biorepository evaluated 35 collections that consisted of 1.8 million biospecimens. A total of 23 collections were selected for consolidation, with a total of 1.2 million specimens located in 21,355 storage boxes. The consolidation resulted in a savings of 4055 boxes of various sizes and 10.2 mechanical freezers (∼275 cubic feet) worth of space.

CONCLUSION

As storage costs in a biorepository increase over time, the development and use of information technology tools to assess the potential advantage and feasiblity of vial consolidation can reduce maintenance expenses.

Data sharing through an NIH central database repository: a cross-sectional survey of BioLINCC users

OBJECTIVE

To characterise experiences using clinical research data shared through the National Institutes of Health (NIH)'s Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC) clinical research data repository, along with data recipients' perceptions of the value, importance and challenges with using BioLINCC data.

DESIGN AND SETTING

Cross-sectional web-based survey.

PARTICIPANTS

All investigators who requested and received access to clinical research data from BioLINCC between 2007 and 2014.

MAIN OUTCOME MEASURES

Reasons for BioLINCC data request, research project plans, interactions with original study investigators, BioLINCC experience and other project details.

RESULTS

There were 536 investigators who requested and received access to clinical research data from BioLINCC between 2007 and 2014. Of 441 potential respondents, 195 completed the survey (response rate=44%); 89% (n=174) requested data for an independent study, 17% (n=33) for pilot/preliminary analysis. Commonly cited reasons for requesting data through BioLINCC were feasibility of collecting data of similar size and scope (n=122) and insufficient financial resources for primary data collection (n=76). For 95% of respondents (n=186), a primary research objective was to complete new research, as opposed to replicate prior analyses. Prior to requesting data from BioLINCC, 18% (n=36) of respondents had contacted the original study investigators to obtain data, whereas 24% (n=47) had done so to request collaboration. Nearly all (n=176; 90%) respondents found the data to be suitable for their proposed project; among those who found the data unsuitable (n=19; 10%), cited reasons were data too complicated to use (n=5) and data poorly organised (n=5). Half (n=98) of respondents had completed their proposed projects, of which 67% (n=66) have been published.

CONCLUSIONS

Investigators were primarily using clinical research data from BioLINCC for independent research, making use of data that would otherwise have not been feasible to collect.