The Very Large Database of Lipids (VLDL)

RATIONALE Closer examination of granular lipid data in a large population offers numerous opportunities to generate new knowledge, ranging from studies examining concordance between commonly used lipid parameters to phenotypic characterization of rare or extreme disorders of lipid metabolism, opening possibilities for better personalizing future treatment of abnormal blood lipids.

STUDY POPULATION The Very Large Database of Lipids (VLDL) currently includes 1,340,614 consecutive adult and 10,294 consecutive pediatric patients. Individuals were clinically referred for lipoprotein cholesterol measurement between 2009 and 2011 using Vertical Auto Profile (VAP) density gradient ultracentrifugation (Birmingham, Alabama, USA).

LIPID MEASUREMENTS The VAP test directly measures cholesterol concentrations of low density lipoprotein, very low density lipoprotein, intermediate density lipoprotein, high density lipoprotein, lipoprotein(a), and their subclasses. Triglycerides in The VLDL are directly measured using the Abbott ARCHITECT C‐8000 system (Abbott Park, Illinois, USA). Lipid distributions in the VLDL closely match those from the population-representative National Health and Nutrition Examination Survey (NHANES) 2007-2008.

STUDY PROCEDURES This study was investigator-initiated. Only de-identified data reach the investigational site. The first data harvest was in 2011. Future harvests are planned. The master database is housed at The Johns Hopkins Hospital in Baltimore, Maryland, and maintained by Drs. Jones and Martin. Only electronic data, and not biospecimens, are sent to Hopkins. The academic investigators have unrestricted access to study data, take responsibility for the accuracy of analyses, and have authority over manuscript preparation and submission.

DATABASE ORGANIZATION To meet the needs of a variety of research questions, data will be organized into 3-year interval datasets (VLDL 2009-2011, VLDL 2012-2015, etc), a summation dataset (first VAP test for each patient), serial lab dataset (patients who have had repeated testing), and ancillary datasets (subsets of patients with coexisting data on other measures such as apolipoproteins, vitamin D, hs-CRP, etc).

INDIVIDUAL STUDIES Individual VLDL studies will typically be based on a priori hypotheses and statistical analysis plans (SAPs) will be peer-reviewed prior to execution. Where the literature does not permit a priori hypothesis generation, studies may be exploratory with the understanding that their findings will be considered hypothesis-generating. The VLDL investigators will periodically register individual studies on clinicaltrials.gov.