Implementation of a clinical decision support alert to prevent use of intravenous ß-blockers and nondihydropyridine calcium channel blockers in hospitalized patients with acute decompensated heart failure.

PURPOSE: Intravenous (IV) ß-blockers (BBs) and nondihydropyridine calcium channel blockers (NDCCBs) are harmful in patients with acute decompensated heart failure (ADHF), but they are commonly used for rate control in atrial fibrillation (AF). This study evaluated the implementation of a clinical decision support (CDS) alert in the electronic health record (EHR) to prevent the use of these agents for AF in patients with ADHF, as well as results from the alert's continuous quality improvement. METHODS: This was a single-center, retrospective, quasi-experimental pre/post analysis of hospitalized adult patients with an ejection fraction of less than 40% documented during their encounter. Groups corresponding to encounters before and after introduction of the alert were compared, and the first version of the alert was compared to its second version that was refined by iterative design. RESULTS: For all patient hospital encounters, the rate of IV BB and NDCCB orders decreased in the period after alert implementation from 16.2% to 12% (P < 0.001). The alert's override rate decreased from 83.8% for the first version to 70.1% after iterative design (P = 0.015). CONCLUSION: This study demonstrates that a CDS alert can be used in the EHR to reduce the use of potentially harmful IV BBs and NDCCBs in patients with ADHF for rate control. User compliance with the alert was improved by applying human factors design principles and iterative design during continuous quality improvement.

Identifying the prevalence of clinically actionable drug-gene interactions in a health system biorepository to guide pharmacogenetics implementation services.

Understanding patterns of drug-gene interactions (DGIs) is important for advancing the clinical implementation of pharmacogenetics (PGx) into routine practice. Prior studies have estimated the prevalence of DGIs, but few have confirmed DGIs in patients with known genotypes and prescriptions, nor have they evaluated clinician characteristics associated with DGI-prescribing. This retrospective chart review assessed prevalence of DGI, defined as a medication prescription in a patient with a PGx phenotype that has a clinical practice guideline recommendation to adjust therapy or monitor drug response, for patients enrolled in a research genetic biorepository linked to electronic health records (EHRs). The prevalence of prescriptions for medications with pharmacogenetic (PGx) guidelines, proportion of prescriptions with DGI, location of DGI prescription, and clinical service of the prescriber were evaluated descriptively. Seventy-five percent (57,058/75,337) of patients had a prescription for a medication with a PGx guideline. Up to 60% (n = 26,067/43,647) of patients had at least one DGI when considering recommendations to adjust or monitor therapy based on genotype. The majority (61%) of DGIs occurred in outpatient prescriptions. Proton pump inhibitors were the most common DGI medication for 11 of 12 clinical services. Almost 25% of patients (n = 10,706/43,647) had more than one unique DGI, and, among this group of patients, 61% had a DGI with more than one gene. These findings can inform future clinical implementation by identifying key stakeholders for initial DGI prescriptions, helping to inform workflows. The high prevalence of multigene interactions identified also support the use of panel PGx testing as an implementation strategy.

Phosphorylation status of fetuin-A is critical for inhibition of insulin action and is correlated with obesity and insulin resistance.

Fetuin-A (Fet-A), a hepatokine associated with insulin resistance, obesity, and incident type 2 diabetes, is shown to exist in both phosphorylated and dephosphorylated forms in circulation. However, studies on fetuin-A phosphorylation status in insulin-resistant conditions and its functional significance are limited. We demonstrate that serum phosphofetuin-A (Ser312) levels were significantly elevated in high-fat diet-induced obese mice, insulin-resistant Zucker diabetic fatty rats, and in individuals with obesity who are insulin resistant. Unlike serum total fetuin-A, serum phosphofetuin-A was associated with body weight, insulin, and markers of insulin resistance. To characterize potential mechanisms, fetuin-A was purified from Hep3B human hepatoma cells. Hep3B Fet-A was phosphorylated (Ser312) and inhibited insulin-stimulated glucose uptake and glycogen synthesis in L6GLUT4 myoblasts. Furthermore, single (Ser312Ala) and double (Ser312Ala + Ser120Ala) phosphorylation-defective Fet-A mutants were without effect on glucose uptake and glycogen synthesis in L6GLUT4 myoblasts. Together, our studies demonstrate that phosphorylation status of Fet-A (Ser312) is associated with obesity and insulin resistance and raise the possibility that Fet-A phosphorylation may play a role in regulation of insulin action.

Ultrasensitive protein detection and imaging: comparison of Lumitein™, ProteoSilver™, SYPRO(®) Ruby, and Coomassie (®) Brilliant Blue gel stains.

Following sodium dodecyl sulfate polyacrylamide gel electrophoresis, proteins can be visualized by various methods of detection and imaging. Traditional methods of protein gel detection and imaging have been improved and expanded through technological advancement. Today, the detection of proteins, resolved on gels, can be accomplished with a variety of stains with various sensitivities. Digital cameras used in the imaging of protein gels are not only more sensitive than their film precursors, but they can be used in combination with imaging software that offers a host of useful applications. Here we describe the UVP BioImaging System in combination with LabWorks Image and Acquisition software to provide a comparison of four different protein gel stains: Lumitein™, ProteoSilver™, SYPRO(®) Ruby, and Coomassie(®) Brilliant Blue. We demonstrate that the detection sensitivity limit appears to be between 100 and 500 ng/protein band of protein with Coomassie(®) Brilliant Blue, 10-50 ng/protein band with Lumitein™ and SYPRO(®) Ruby, and as little as 5 ng/protein band with the ProteoSilver™ stain.