Safety of Computed Tomographic Angiography in the Evaluation of Patients With Acute Stroke: A Single-Center Experience.

BACKGROUND AND PURPOSE: Noncontrasted head computed tomography (NCHCT) has long been the standard of care for acute stroke imaging. New guidelines recommending advanced vascular imaging to identify eligible patients for endovascular therapy have renewed safety concerns on the use of contrast in the emergent setting without laboratory confirmation of renal function. METHODS: We compared computed tomographic angiography (CTA) versus NCHCT alone during acute stroke evaluation with focus on renal safety and timeliness of therapy delivery. We reviewed data on all emergency department patients for whom the Acute Stroke Intervention Team was activated between December 2013 and September 2014. Primary outcomes included acute kidney injury and change in serum creatinine from presentation to 24 to 48 hours (Δ serum creatinine [Cr]). We assessed therapy delay using door-to-CT and door-to-needle times. RESULTS: Of 289 patients requiring Acute Stroke Intervention Team activation, 157 received CTA and 132 NCHCT only. There was no difference between groups in mean Cr at 24 to 48 hours (1.06 CTA; 1.40 NCHCT; P=0.059), ΔCr (-0.07 CTA, -0.11 NCHCT, P=0.489), or rates of acute kidney injury (5 CTA, 7 NCHCT, P=0.422). There was no significant difference in mean intravenous tissue plasminogen activator treatment times (68.11 minutes CTA, 81.36 minutes NCHCT; P=0.577). In the 157 patients who underwent CTA, 16 (10.2%) vascular anomalies and 55 (35.0%) high-grade stenoses or occlusions were identified. CONCLUSIONS: CTA acquisition during acute stroke evaluation was safe with regards to renal function and did not delay appropriate therapy delivery. Acute CTA acquisition offers additional clinical value in rapid identification of vascular abnormalities.

The role of alcohol use in emergency department episodes.

This study investigates the association between alcohol use and emergency-department (ED) utilization in the United States using nationally representative data from the 2001 National Health Interview Survey (N = 33,326). Estimates from our probit models indicate that among men, current drinkers are less likely to have visited the ED in the past year than former drinkers. Among women, lifetime abstainers are less likely than current drinkers to have had an ED episode. Finally, frequency of binge drinking significantly increases the likelihood of ED visits for men. The results suggest that focusing solely on problem drinking provides a limited perspective.

Biclustering models for structured microarray data.

Microarrays have become a standard tool for investigating gene function and more complex microarray experiments are increasingly being conducted. For example, an experiment may involve samples from several groups or may investigate changes in gene expression over time for several subjects, leading to large three-way data sets. In response to this increase in data complexity, we propose some extensions to the plaid model, a biclustering method developed for the analysis of gene expression data. This model-based method lends itself to the incorporation of any additional structure such as external grouping or repeated measures. We describe how the extended models may be fitted and illustrate their use on real data.