Bruce E. Bray

A Randomized Trial of Intracoronary Streptokinase in the Treatment of Acute Myocardial Infarction

Fifty patients with acute myocardial infarction were randomly assigned to receive either intracoronary streptokinase or standard (control) therapy within about three hours after the onset of pain. Coronary perfusion was reestablished in 19 of 24 patients receiving streptokinase. Streptokinase alleviated pain (as indicated by differences in subsequent morphine use). The Killip class was significantly improved after therapy with streptokinase, as were changes in radionuclide ejection fraction between Days 1 and 10 in surviving patients (+3.9 vs. -3.0 per cent, P less than 0.01). The echocardiographic wall-motion index also showed greater improvement after streptokinase treatment (P less than 0.01). Streptokinase therapy was associated with rapid evolution of electrocardiographic changes, which were essentially complete within three hours after therapy, but loss of R waves, ST elevation, and development of Q waves in the convalescent period were greater in the control group (P less than 0.01). The time required to reach peak plasma enzyme concentrations was significantly shorter after streptokinase. The incidence of early and late ventricular arrhythmias was not affected by treatment. We conclude that intracoronary streptokinase appears to have a beneficial effect on the early course of acute myocardial infarction.

Automated extraction of ejection fraction for quality measurement using regular expressions in Unstructured Information Management Architecture (UIMA) for heart failure

OBJECTIVES Left ventricular ejection fraction (EF) is a key component of heart failure quality measures used within the Department of Veteran Affairs (VA). Our goals were to build a natural language processing system to extract the EF from free-text echocardiogram reports to automate measurement reporting and to validate the accuracy of the system using a comparison reference standard developed through human review. This project was a Translational Use Case Project within the VA Consortium for Healthcare Informatics. MATERIALS AND METHODS We created a set of regular expressions and rules to capture the EF using a random sample of 765 echocardiograms from seven VA medical centers. The documents were randomly assigned to two sets: a set of 275 used for training and a second set of 490 used for testing and validation. To establish the reference standard, two independent reviewers annotated all documents in both sets; a third reviewer adjudicated disagreements. RESULTS System test results for document-level classification of EF of <40% had a sensitivity (recall) of 98.41%, a specificity of 100%, a positive predictive value (precision) of 100%, and an F measure of 99.2%. System test results at the concept level had a sensitivity of 88.9% (95% CI 87.7% to 90.0%), a positive predictive value of 95% (95% CI 94.2% to 95.9%), and an F measure of 91.9% (95% CI 91.2% to 92.7%). DISCUSSION An EF value of <40% can be accurately identified in VA echocardiogram reports. CONCLUSIONS An automated information extraction system can be used to accurately extract EF for quality measurement.

An Event Model of Medical Information Representation

Objective : Develop a model for structured and encoded representation of medical information that supports human review, decision support applications, ad hoc queries, statistical analysis, and natural-language processing. Design : A medical information representation model was developed from manual and semiautomated analysis of patient data. The key assumption of the model is that medical information can be represented as a series of linked events. The event representation has two main components. The first component is a frame or template definition that specifies the attributes of the event. The second component is a structured vocabulary, the terms of which are taken as the values of the slots in the event template structure. Individual event instances are linked by specific named relationships. Results : The proposed model was used to represent a chest-radiograph report. Conclusions : The event model of medical information representation provides a mechanism for formal definition of the logical structure of medical data and allows explicit time-oriented and associative relationships between event instances.

Value Driven Outcomes (VDO): a pragmatic, modular, and extensible software framework for understanding and improving health care costs and outcomes

Objective To develop expeditiously a pragmatic, modular, and extensible software framework for understanding and improving healthcare value (costs relative to outcomes). Materials and methods In 2012, a multidisciplinary team was assembled by the leadership of the University of Utah Health Sciences Center and charged with rapidly developing a pragmatic and actionable analytics framework for understanding and enhancing healthcare value. Based on an analysis of relevant prior work, a value analytics framework known as Value Driven Outcomes (VDO) was developed using an agile methodology. Evaluation consisted of measurement against project objectives, including implementation timeliness, system performance, completeness, accuracy, extensibility, adoption, satisfaction, and the ability to support value improvement. Results A modular, extensible framework was developed to allocate clinical care costs to individual patient encounters. For example, labor costs in a hospital unit are allocated to patients based on the hours they spent in the unit; actual medication acquisition costs are allocated to patients based on utilization; and radiology costs are allocated based on the minutes required for study performance. Relevant process and outcome measures are also available. A visualization layer facilitates the identification of value improvement opportunities, such as high-volume, high-cost case types with high variability in costs across providers. Initial implementation was completed within 6 months, and all project objectives were fulfilled. The framework has been improved iteratively and is now a foundational tool for delivering high-value care. Conclusions The framework described can be expeditiously implemented to provide a pragmatic, modular, and extensible approach to understanding and improving healthcare value.

Left ventricular mass increases during cardiac allograft vascular rejection

OBJECTIVES This study evaluated whether left ventricular mass increases during cellular or vascular (humoral) cardiac allograft rejection. BACKGROUND An increase in left ventricular mass during cellular cardiac allograft rejection has been described by other investigators, although controversy has existed over the validity of these findings. Left ventricular mass changes have not been evaluated in the setting of vascular (humoral) cardiac allograft rejection. METHODS To determine the effect of allograft rejection on left ventricular mass, we retrospectively reviewed endomyocardial biopsy results and corresponding echocardiograms in 41 cardiac transplant recipients undergoing treatment for allograft rejection. Left ventricular mass was assessed by two-dimensional echocardiography using the method of Schiller. Maintenance immunosuppression included cyclosporine in all patients. RESULTS Although significant changes in left ventricular wall thickness, mass and dimensions were not observed in patients experiencing moderate or severe cellular allograft rejection (International Society for Heart and Lung Transplantation grades III and IV, n = 27), marked changes were noted in patients with vascular (humoral) rejection (n = 14). Patients with vascular rejection demonstrated an echocardiographic mean (+/- SEM) increase in left ventricular wall mass (from 109 +/- 17 to 151 +/- 17 g), and left ventricular wall thickness (from 1.3 +/- 0.1 to 1.6 +/- 0.1 cm) during the rejection episode. Additionally, vascular rejection was associated with a trend toward an increase in left ventricular systolic dimension (from 2.6 +/- 0.1 to 3.0 +/- 0.2 cm) and a decrease in left ventricular fractional shortening and increased incidence of hemodynamic compromise with rejection (50% for vascular vs. 11% for cellular rejection). CONCLUSIONS Left ventricular mass increases during episodes of vascular (humoral) rejection, but there is no significant change in left ventricular mass during cellular cardiac allograft rejection.

Comparative effects of diltiazem, propranolol, and placebo on exercise performance using radionuclide ventriculography in patients with symptomatic coronary artery disease: Results of a double-blind, randomized, crossover study

The effects of oral diltiazem (120 mg), propranolol (100 mg), and placebo on exercise performance and left ventricular function were compared before and during symptom-limited supine bicycle exercise by means of multigated radionuclide ventriculography in 12 patients with documented, symptomatic coronary artery disease; a double-blind, randomized crossover protocol was used. Diltiazem increased ejection fraction (EF) at submaximal exercise (+7.0 absolute percentage points, p less than 0.02) and maximal exercise (+8.1 percentage points, p less than 0.01). Exercise EF was increased by 13.6 percentage points (p less than 0.02) in patients with decreased ventricular function (resting EF less than 50%). Propranolol had no effect on exercise EF at any stage, even when patients with EF less than 50% were excluded. The increase in total exercise time was significant after diltiazem (+27%, p less than 0.01) but not after propranolol (+16%, p = NS). As expected, propranolol decreased both resting (-9 bpm, p less than 0.01) and exercise heart rates (-27 bpm, p less than 0.001), whereas diltiazem had no significant effect. Propranolol decreased resting diastolic blood pressure (-8 mm Hg, p less than 0.02), exercise systolic (-27 mm Hg, p less than 0.001) and diastolic (-9 mm Hg, p less than 0.01) blood pressures, and rest (p less than 0.01) and exercise (p less than 0.001) double product. Diltiazem decreased resting systolic blood pressure (-9 mm Hg, p less than 0.01) and both resting (-8 mm Hg, p less than 0.001) and exercise (-9 mm Hg, p less than 0.01) diastolic blood pressures.(ABSTRACT TRUNCATED AT 250 WORDS)

Countering imbalanced datasets to improve adverse drug event predictive models in labor and delivery

BACKGROUND The IOM report, Preventing Medication Errors, emphasizes the overall lack of knowledge of the incidence of adverse drug events (ADE). Operating rooms, emergency departments and intensive care units are known to have a higher incidence of ADE. Labor and delivery (L&D) is an emergency care unit that could have an increased risk of ADE, where reported rates remain low and under-reporting is suspected. Risk factor identification with electronic pattern recognition techniques could improve ADE detection rates. OBJECTIVE The objective of the present study is to apply Synthetic Minority Over Sampling Technique (SMOTE) as an enhanced sampling method in a sparse dataset to generate prediction models to identify ADE in women admitted for labor and delivery based on patient risk factors and comorbidities. RESULTS By creating synthetic cases with the SMOTE algorithm and using a 10-fold cross-validation technique, we demonstrated improved performance of the Naïve Bayes and the decision tree algorithms. The true positive rate (TPR) of 0.32 in the raw dataset increased to 0.67 in the 800% over-sampled dataset. CONCLUSION Enhanced performance from classification algorithms can be attained with the use of synthetic minority class oversampling techniques in sparse clinical datasets. Predictive models created in this manner can be used to develop evidence based ADE monitoring systems.

Efficiency of CYP2C9 Genetic Test Representation for Automated Pharmacogenetic Decision Support

OBJECTIVES We investigated the suitability of representing discrete genetic test results in the electronic health record (EHR) as individual single nucleotide polymorphisms (SNPs) and as alleles, using the CYP2C9 gene and its polymorphic states, as part of a pilot study. The purpose of our investigation was to determine the appropriate level of data abstraction when reporting genetic test results in the EHR that would allow meaningful interpretation and clinical decision support based on current knowledge, while retaining sufficient information in order to enable reinterpretation of the results in the context of future discoveries. METHODS Based on the SNP & allele models, we designed two separate lab panels within the laboratory information system, one containing SNPs and the other containing alleles, built separate rules in the clinical decision support system based on each model, and evaluated the performance of these rules in an EHR simulation environment using real-world scenarios. RESULTS Although decision-support rules based on allele model required significantly less computational time than rules based on SNP model, no difference was observed on the total time taken to chart medication orders between rules based on these two models. CONCLUSIONS Both, SNP- and allele-based models, can be used effectively for representing genetic test results in the EHR without impacting clinical decision support systems. While storing and reporting genetic test results as alleles allow for the construction of simpler decision-support rules, and make it easier to present these results to clinicians, SNP-based model can retain a greater amount of information that could be useful for future reinterpretation.

Controlled terminology for clinically-relevant indexing and selective retrieval of biomedical images

Existing clinical nomenclatures do not provide comprehensive, detailed coverage for multispecialty biomedical imaging. To address clinical needs in this area, the College of American Pathologists (CAP), secretariat of the Systematized Nomenclature of Human and Veterinary Medicine (SNOMED), has entered into partnership with the DICOM (Digital Imaging and Communications in Medicine) Standards Committee, the American College of Radiology, the American Dental Association, the American Academy of Ophthalmology, the American Society for Gastrointestinal Endoscopy, the American Academy of Neurology, the American Veterinary Medical Association, and other professional specialty organizations to develop the controlled terminology that is needed for diagnostic imaging applications. Terminology development is coordinated with ongoing development and maintenance of the DICOM Standard. SNOMED content is being enhanced in two general areas: 1) imaging procedure descriptions and 2) diagnostic observations. The SNOMED DICOM Microglossary (SDM) has been developed to provide context-dependent value sets (SDM Context Groups) for DICOM codedentry data elements and semantic content specifications (SDM Templates) for reports and other structures composed of multiple data elements. The capability of storing explicitlylabeled coded descriptors from the SDM in DICOM images and reports improves the potential for selective retrieval of images and related information. A pilot test of distributed multispecialty terminology development using a World Wide Web (WWW) application was performed in 1997, demonstrating the feasibility of large-scale distributed development of SDM

Long-term follow-up after intracoronary streptokinase for myocardial infarction: A randomized, controlled study

Intracoronary streptokinase (SK) may have beneficial effects on the in-hospital course of acute myocardial infarction (MI), but long-term outcome is unknown. We evaluated the outpatient course of 50 MI patients, randomly treated with either SK (n = 24) or standard therapy (n = 26), who presented within 2.7 +/- 0.7 hours of symptoms. Coronary reperfusion occurred in 19 (79%) SK patients. Survivors were followed for a mean of 18.7 months (range 11 to 28.5); information was current in 48 patients (96%). Both groups received antiplatelet therapy for 3 months. A total of five deaths occurred in the control group and two in the SK group, including one posthospital death in each. Nonfatal MIs totaled five in control patients and three in SK patients, including five posthospital MIs (three control, one SK). Differences in major events (death or nonfatal MI) favoring SK did not quite reach statistical significance (10 control vs 5 SK). Bypass surgery was performed in seven SK and four control patients (NS). Angina occurred in more control (15) than SK (six) patients (p less than 0.01), and more control patients used long-acting nitrates (14 control, three SK; p less than 0.01). Palpitations were noted by nine control and one SK patient (p less than 0.01), and documented late arrhythmias were present in four control patients and no SK survivors (p less than 0.05). Symptoms suggestive of heart failure were present in seven control and one SK patient (p less than 0.01); two control patients were hospitalized for failure. Use of beta blockers, calcium channel blockers, and other cardiac medications did not differ.(ABSTRACT TRUNCATED AT 250 WORDS)