Cost-Effectiveness Analysis of Transcatheter Arterial Embolization Techniques for the Treatment of Gastrointestinal Bleeding in the United States.

OBJECTIVES: Gastrointestinal (GI) bleeding is a common medical emergency associated with significant mortality. Transcatheter arterial embolization first was introduced by Rosch et al as an alternative to surgery for upper GI bleeding. The clinical success in patients with GI bleeding treated with transcatheter arterial embolization previously has been reported. However, there are no cost-effectiveness analyses reported to date. Here we report cost-effectiveness analysis of N-butyl 2-cyanoacrylate glue (NBCA) and ethylene-vinyl alcohol copolymer (Onyx) versus coil (gold standard) for treatment of GI bleeding from a healthcare payer perspective. METHODS: Fixed-effects modeling with a generalized linear mixed method was used in NBCA and coil intervention arms to determine the pooled probabilities of clinical success and mortality with complications with their confidence intervals, while the Clopper-Pearson model was used for Onyx to determine the same parameters. Models were provided by the "Meta-Analysis with R" software package. A decision tree was built for cost-effectiveness analysis, and Microsoft Excel was used for probabilistic sensitivity analysis. The cost-effective option was determined based on the incremental cost-effectiveness ratio and scatter plots of incremental cost versus incremental quality-adjusted life-years. RESULTS: Comparing scatter plots and incremental cost-effectiveness ratio results, -$1024 and -$1349 per quality-adjusted life-year for Onyx and N-butyl 2-cyanoacrylate glue, respectively, Onyx was the least expensive and most effective intervention. CONCLUSION: Onyx was the dominant strategy regardless of threshold values. Our analyses provide a framework for researchers to predict the target clinical effectiveness for early-stage TAE interventions and guide resource allocation decisions.

Molecular dynamics study of the diffusivity of a hydrophobic drug Cucurbitacin B in pseudo-poly(ethylene oxide-b-caprolactone) micelle environments.

Isobaric-isothermal molecular dynamics simulation was used to study the diffusion of a hydrophobic drug Cucurbitacin B (CuB) in pseudomicelle environments consisting of poly(ethylene oxide-b-caprolactone) (PEO-b-PCL) swollen by various amounts of water. Two PEO-b-PCL configurations, linear and branched, with the same total molecular weight were used. For the branched configuration, the block copolymer contained one linear block of PEO with the same molecular weight as that of the PEO block used in the linear configuration but with one end connecting to three PCL blocks with the same chain length, hereafter denoted PEO-b-3PCL. Regardless of the configuration, the simulation results showed that the diffusivity of CuB was insensitive to the water concentration up to ∼8 wt % while that of water decreased with an increasing water concentration. The diffusivity of CuB (10(-8) cm(2)/s) was 3 orders of magnitude lower than that of water (10(-5) cm(2)/s). This is attributed to the fact that CuB relied on the wiggling motion of the block copolymers to diffuse while water molecules diffused via a hopping mechanism. The rates at which CuB and water diffused into PEO-b-PCL were twice those in PEO-b-3PCL because the chain mobility and the degree of swelling are higher and there are fewer intermolecular hydrogen bonds in the case of PEO-b-PCL. The velocity autocorrelation functions of CuB show that the free volume holes formed by PEO-b-3PCL are more rigid than those formed by PEO-b-PCL, making CuB exhibit higher-frequency collision motion in PEO-b-3PCL than in PEO-b-PCL, and the difference in frequency is insensitive to water concentration.