Shortening of oblique left ventricular (LV) fibers results in torsion. A unique relationship between volume and torsion is therefore expected, and the effects of load and contractility on torsion should be predictable. However, volume-independent behavior of torsion has been observed, and the effects of load on this deformation remain controversial. We used magnetic resonance imaging (MRI) with tagging to study the relationships between load and contractility, and torsion. In ten isolated, blood-perfused canine hearts, ejection was controlled by a servopump: end-diastolic volume (EDV) was controlled by manipulating preload parameters and end-systolic volume (ESV) by manipulating afterload using a three-element windkessel model. MRI was obtained at baseline, two levels of preload alteration, two levels of afterload alteration, and dobutamine infusion. An increase in EDV resulted in an increase in torsion at constant ESV (preload effect), whereas an increase in ESV resulted in a decrease in torsion at constant EDV (afterload effect). Dobutamine infusion increased torsion in association with an increase in LV peak-systolic pressure (PSP), even at identical EDV and ESV. Multiple regression showed correlation of torsion with preload (EDV), afterload (ESV), and contractility (PSP; r = 0.67). Furthermore, there was a close linear relationship between torsion and stroke volume (SV) and ejection fraction (EF) during load alteration, but torsion during dobutamine infusion was greater than expected for the extent of ejection. Preload and afterload influence torsion through their effects on SV and EF, and there is an additional direct inotropic effect on torsion that is independent of changes in volume but rather is force dependent. There is therefore potential for the torsion-volume relation to provide a load-independent measure of contractility that could be measured noninvasively.

Histochemical staining for cytochrome oxidase (CO) and axonal transport of horseradish peroxidase (HRP) were used to investigate thalamocortical connections in the vibrissa-barrel system of adult rats. CO staining revealed that the medial division of the ventrobasal thalamic nucleus (VBm) consists of intensely stained rod-like configurations, containing thalamocortical projection neurons and intervening neuropil, separated by lighter-stained septa. CO-dark rods span the thickness of VBm, are arranged in a pattern of rows and arcs that resembles the distribution of vibrissae on the mystacial pad, and are similar to the cytoarchitectonic structures termed "barreloids" in the mouse thalamus. Based upon the dimensions of CO-dark structures and the numerical density of neurons in VBm we estimated that a barreloid in the rat may contain 250-300 neurons. HRP injections into lamina IV of the somatosensory cortex led to retrograde labeling of neurons within one or more barreloids. When injections were centered within the CO-dark hollows of cortical barrels about 95% of retrogradely labeled neurons were located in the barreloid that is isomorphic to the injected barrel; up to 5% of labeled neurons were located within a single adjacent barreloid. Barrel hollow injections that also included a barrel side yielded a larger proportion of labeled neurons in non-isomorphic barreloids. Interestingly, such extra-barreloid labeling was topologically consistent in that HRP-labeled neurons were distributed among barreloids that corresponded to cortical barrels nearest the injected barrel side. Injections into the septa between barrels similarly resulted in labeling within barreloids that corresponded to cortical barrels flanking the septal injection site. Following lamina IV injections the density of labeled neurons tended to be highest in the ventrolateral one-half to two-thirds of VBm. Retrograde labeling of neurons in the dorsomedial one-third to one-half of VBm was more often observed after HRP injections at the lamina V/VI border. Thus, barreloid neurons may be heterogeneous with respect to their laminar pattern of terminations within the somatosensory cortex. Some HRP injections in the cortex resulted in orthograde labeling of corticothalamic axons in the barreloids. When observed, labeled corticothalamic axons arborized principally within the barreloid isomorphic to the injected barrel column. Indeed, terminal labeling was densest in the vicinity of neurons retrogradely labeled by the same injection.

No abstract

No abstract