Abstract WP285: Impact of Endovascular Treatment on the Patterns of Interhospital Transfer of Stroke Patients in California

Background: Optimized stroke systems of care can enable equitable access to timely care, including endovascular thrombectomy (EVT). Yet how hospitals are connected in the care of stroke patients is not well-characterized. Given that EVT is only available in specialized centers, stroke systems and patient transfer patterns may have evolved after the 2015 publication of EVT benefit. Primary objective: to map the stroke patient transfer network in California and to determine whether it changed after 2015. Methods: We analyzed California data including every nonfederal hospital admission from pre- (2010-2014) and post-2015 (2016-2017). ICD-9, ICD-10, and DRG codes identified ischemic stroke (IS) hospitalizations. Connections between any 2 hospitals were based on the transfer of > 5 IS patients between them/year. t-tests compared the patient transfer maps pre- vs post-2015 on descriptive network measures: number of hospitals, transfer connections, and patients shared in transfer, and distance traveled in transfer. A hierarchical logistic regression model assessed whether patients were more frequently transferred to EVT-capable hospitals after 2015, adjusting for patient- and hospital-level factors, including a time-by-distance interaction. Results: Among 385,799 IS hospitalizations, 15,522 (4.0%) were transferred. After 2015, patients traveled longer distances in transfer (25.1 vs 28.4 miles, p<0.001) (Figure). The proportion of patients transferred and the number of EVT centers were stable, but, after 2015, patients were more frequently transferred to EVT centers and travelled longer distance to do so (post-2015 OR 3.0, 95% CI 2.5-3.5; distance OR 1.044/mile, 95% CI 1.04-1.05; OR for interaction 1.01, 95% CI 1.003-1.02). Conclusion: The California stroke transfer network significantly changed after the 2015 publication of benefit for EVT, with increased likelihood of transfer to EVT centers and longer distances traveled in transfer.

Validation of transfer map calculation for electrostatic deflectors in the code COSY INFINITY

The code COSY INFINITY uses a beamline coordinate system with a Frenet–Serret frame relative to the reference particle, and calculates differential algebra-valued transfer maps by integrating the ODEs of motion in the respective vector space over a differential algebra (DA). We described and performed computation of the DA transfer map of an electrostatic spherical deflector in a laboratory coordinate system using two conventional methods: (1) by integrating the ODEs of motion using a numerical integrator and (2) by computing analytically and in closed form the properties of the respective elliptical orbits from Kepler theory. We compared the resulting transfer maps with (3) the DA transfer map of COSY INFINITY’s built-in electrostatic spherical deflector element [Formula: see text] and (4) the transfer map of the electrostatic spherical deflector computed using the program GIOS, which uses analytic formulas from a paper1 by Hermann Wollnik regarding second-order aberrations. In addition to the electrostatic spherical deflector, we studied an electrostatic cylindrical deflector, where the Kepler theory is not applicable. We computed the DA transfer map by the ODE integration method (1), and we compared it with the transfer maps by (3) COSY INFINITY’s built-in electrostatic cylindrical deflector element [Formula: see text] and (4) GIOS. The transfer maps of electrostatic spherical and cylindrical deflectors obtained using the direct calculation methods (1) and (2) are in excellent agreement with those computed using (3) COSY INFINITY. On the other hand, we found a significant discrepancy with (4) the program GIOS.

Filtered finiteness of the image of the unstable Hurewicz homomorphism with applications to bordism of immersions

After recent work of Hill, Hopkins and Ravenel on the Kervaire invariant one problem [M. A. Hill, M. J. Hopkins and D. C. Ravenel, On the non-existence of elements of Kervaire invariant one, Ann. Math. (2), 184 (2016), 1–262], as well as Adams’ solution of the Hopf invariant one problem [J. F. Adams, On the non-existence of elements of Hopf invariant one, Ann. Math. (2), 72 (1960), 20–104], an immediate consequence of Curtis conjecture is that the set of spherical classes in H∗Q0S0 is finite. Similarly, Eccles conjecture, when specialized to X=Sn with n> 0, together with Adams’ Hopf invariant one theorem, implies that the set of spherical classes in H∗QSn is finite. We prove a filtered version of the above finiteness properties. We show that if X is an arbitrary CW-complex of finite type such that for some n, HiX≃0 for any i>n, then the image of the composition π∗ΩlΣl+2X→π∗QΣ2X→H∗QΣ2X is finite; the finiteness remains valid if we formally replace X with S−1. As an application, we provide a lower bound on the dimension of the sphere of origin on the potential classes of π∗QSn which are detected by homology. We derive a filtered finiteness property for the image of certain transfer maps ΣdimgBG+→QS0 in homology. As an application to bordism theory, we show that for any codimension k framed immersion f:M↬ℝn+k which extends to an embedding M→ℝd×ℝn+k, if n is very large with respect to d and k then the manifold M as well as its self-intersection manifolds are boundaries. Some results of this paper extend results of Hadi [Spherical classes in some finite loop spaces of spheres. Topol. Appl., 224 (2017), 1–18] and offer corrections to some minor computational mistakes, hence providing corrected upper bounds on the dimension of spherical classes H∗ΩlSn+l. All of our results are obtained at the prime p = 2.

Integrable Derivations and Stable Equivalences of Morita Type

Using that integrable derivations of symmetric algebras can be interpreted in terms of Bockstein homomorphisms in Hochschild cohomology, we show that integrable derivations are invariant under the transfer maps in Hochschild cohomology of symmetric algebras induced by stable equivalences of Morita type. With applications in block theory in mind, we allow complete discrete valuation rings of unequal characteristic.