Voronoi Polyhedra Analysis of Optimized Arterial Tree Models

Topological and metric properties of Voronoi polyhedra (VP) generated by the distal end points of terminal segments in arterial tree models grown by the method of constrained constructive optimization (CCO) are analyzed with the aim to characterize the spatial distribution of their supply sites relative to randomly distributed points as a reference model. The distributions of the number N _f of Voronoi cell faces, cell volume V, surface area S, area A of individual cell faces, and asphericity parameter of the CCO models are all significantly different from the ones of random points, whereas the distributions of V, S, and are also significantly different among CCO models optimized for minimum intravascular volume and minimum segment length (p < 0.0001). The distributions of N _f , V, and S of the CCO models are reasonably well approximated by two-parameter gamma distributions. We study scaling of intravascular blood volume and arterial cross-sectional area with the volume of supplied tissue, the latter being represented by the VP of the respective terminal segments. We observe scaling exponents from 1.20 ± 0.007 to 1.08 ± 0.005 for intravascular blood volume and 0.77 ± 0.01 for arterial cross-sectional area. Setting terminal flows proportional to the associated VP volumes during tree construction yields a relative dispersion of terminal flows of 37% and a coefficient of skewness of 1.12. © 2003 Biomedical Engineering Society. PAC2003: 8719Uv, 8710+e, 4720Ky, 0260Pn, 0230Oz     

Generation of an Anatomically Based Three-Dimensional Model of the Conducting Airways

An anatomically accurate model of the conducting airways is essential for adequately simulating gas mixing, particle deposition, heat and water transfer, and fluid distribution. We have extended a two-dimensional tree-growing algorithm to three dimensions for generation of a host-shape dependent three-dimensional conducting airway model. Terminal branches in the model are both length limited and volume-supplied limited. A limit is imposed on the maximum possible branch angle between a daughter and parent branch. Comparison of the resulting model with morphometric data shows that the algorithm produces branching and length ratios, path lengths, numbers of branches, and branching angles very close to those from the experimental data. The correlation between statistics from the generated model and those from morphometric studies suggests that the conducting airway structure can be described adequately using a supply and demand algorithm. The resulting model is a computational mesh that can be used for simulating transport phenomena. © 2000 Biomedical Engineering Society. PAC00: 8719Uv, 8710+e     

The Effects of Sodium Nitrite on Neocortical Neuron Activity during Performance of Defensive and Inhibitory Conditioned Reflexes

Administration of the NO-generating substance sodium nitrite to conscious non-immobilized rabbits at a dose of 11 mg/kg (s.c.) decreased the intensity and duration of the short-latency modality-specific components of the responses of neurons in the visual cortex (to flashes of light) and sensorimotor cortex (to pain reinforcement). Decreases in neuron activation in the visual cortex in response to flashes of light occurred independently of their biological significance, i.e., as the signal for a defensive reflex and on the background of conditioned inhibition. The long-latency activatory components of the response of sensorimotor and visual cortex neurons to pain reinforcement, the inhibitory pause in the responses of visual cortex neurons to flashes of light, and the disinhibitory effect of pain reinforcement showed smaller changes after sodium nitrite. The results obtained here support the suggestion that different neuromediators are involved in transmitting modality-specific and modality-non-specific influences to neocortical neurons during learning, and provide the basis for suggesting that sodium nitrite has a neurotropic action when given systemically.     

Effect of alpha adrenergic receptor stimulation on integrated systemic venous bed

Summary In 29 mongrel dogs, venous return was completely drained from the caval veins to an oxygenator and returned to the femoral arteries with a roller pump at a constant perfusion rate. Blood pressures in both caval veins were kept constant by an automatic control system. The oxygenator weight was recorded and showed changes of the intracorporeal blood volume, i. e. of systemic venous blood volume since allowance was made for parallel changes in arterial blood volume.Xylometazolin (5–20 g/kg) and norepinephrine (2–8 g/kg), both infused within 3 min, decreased systemic venous blood volume by up to 7 ml/kg. After denervation of arterial pressoreceptors, nearly similar results were obtained. After beta receptor blockade with prindolol (600 g/kg), the effects on venous system were unchanged. After alpha receptor blockade with phentolamin (2500 g/kg), all effects were abolished. Phentolamin as such increased venous volume by 3 ml/kg.It is concluded that stimulation of venous alpha receptors effects predominantly a constriction of the systemic venous bed and hereby must increase venous return and cardiac output.     

Comparison of four methods for testing high-level aminoglycoside resistance in enterococci

In a prospective study the prevalence of high-level aminoglycoside resistance (MIC 2,000 µg/ml) among 62 clinically significant enterococci was investigated. A total of 10⁵ organisms were inoculated a) onto a plate containing 2,000 µg/ml of gentamicin or streptomycin; b) into a microtube for dilution MIC determinations for gentamicin, amikacin, tobramycin and streptomycin; and c) into a single tube containing 500 µg/ml of gentamicin, amikacin, tobramycin or streptomycin in supplemented Mueller-Hinton broth. In addition, tubes containing 500 µg/ml of gentamicin, amikacin, tobramycin or streptomycin were inoculated with five enterococcal colonies (crude method). For 45 of the 62 isolates, MICs of gentamicin, amikacin and tobramycin were 500 µg/ml, while 17 (27 %) showed high-level resistance. The MICs of streptomycin were 500 µg/ml for 42 of 62 isolates, and 2,000 µg/ml for 20 (32.3 %). For 8 of the 17 (47 %) isolates showing high-level gentamicin resistance, MICs of streptomycin were 500 µg/ml. There was complete agreement between the results of the plate method, the microtube dilution MIC and the tube inoculated with 10⁵ CFU, but the crude method gave discordant results for two isolates. It is concluded that a tube containing 500 µg/ml of aminoglycoside is a simple, accurate and inexpensive method for determining high-level aminoglycoside resistance.     

Structural Model of the Muscle Spindle

A model of the muscle spindle was developed based on its anatomical structure. The model contains three intrafusal fibers (bag1, bag2, and chain), two efferents (dynamic efferent to the bag1 fiber and static efferent to bag2 and chain fibers), and two afferents [primary (Ia) and secondary (II)]. As in the real muscle spindle, the spindle model, under the modulation of efferents, responds to the extrafusal muscle fiber length. Both outputs (Ia and II afferents) of the model were compared extensively with published data, under both sinusoidal stretch (with different stretch amplitudes and frequencies) and ramp and hold stretch (with different stretch amplitudes and velocities) in three different fusimotor activation conditions (dynamic stimulation, static stimulation, and without stimulation). Model Ia afferent responses fit the published data well with active gamma input, but less well in the passive state. Model II afferent responses also fit the published data, although less quantitative data were available for comparison. The model correctly predicted the fractional power dependence of the primary and secondary ending responses on stretch velocity. The current model provides a powerful tool for simulation studies of neuromusculoskeletal systems, and demonstrates the feasibility of using a structural approach to model complex neurophysiological systems. © 2002 Biomedical Engineering Society. PAC02: 8719Ff, 8719La, 8719St     

Integrative Models for Understanding the Structural Basis of Regional Mechanical Dysfunction in Ischemic Myocardium

Myocardial ischemia and many other cardiac pathologies are associated with regional ventricular dysfunction. Since the distributions of stress and material properties cannot be measured directly in intact myocardium, understanding how regional alterations in myocardial strain or segment function are related to underlying cellular dysfunction must be deduced from theoretical models. Here, we describe how anatomically detailed, three-dimensional computational models can be used in conjunction with experimental or clinical studies to elucidate the structural basis of regional dysfunction in acutely ischemic and ischemic-reperfused (stunned) myocardium in vivo. Integrative experimental and computational analysis shows that: (1) in acutely ischemic myocardium, the transition from abnormal systolic strain in the ischemic region to normal shortening in adjacent, normally perfused tissue is governed primarily by systolic blood pressure and regional fiber orientation rather than the geometry of the perfusion boundary; and (2) in stunned myocardium, the degree of reperfusion injury to the contractile apparatus may be uniform across the wall thickness despite observations that the extent of ischemia and the impairment of regional strain during reperfusion are both significantly greater in the subendocardium. © 2000 Biomedical Engineering Society. PAC00: 8719Hh, 8719Uv, 8719Ff, 8719Rr, 8710+e     

Oxidation of 14C-β-hydroxybutyrate administered by ventriculo-cisternal perfusion in the dog

Summary A ventriculo-cisternal perfusion system (from the 3rd ventricle to cisterna magna) was employed to study the oxidation of Na-DL-3-hydroxybuty-rate-3-¹⁴C when added to the perfusion fluid. ¹⁴CO₂ appeared in the perfusate and in the blood taken from both the confluens sinuum and a peripheral artery, in the course of the perfusion. Specific activity of CO₂ in the perfusate usually exceeded that of the venous blood, which in turn was always higher than that of arterial blood. The calculated rate of oxidation of OHB to CO₂ was about 20 mmole/min, or 5–6% of the perfused OHB. It is postulated that OHB supplied locally in CSF may serve some supplemental physiological role as an oxidizable substrate.Supported by Grant HE 03130 from the National Heart Institute.     

Specific patterns of neuron arrangement and of synaptic articulation in the medial geniculate body

Summary Golgi and electron microscopic analysis of the known cellular layers in concentric shells of the ventro-lateral portion of the medial geniculate body revealed a flat grid of high density neuropil filling the space between the geniculocortical relay cells, forming essentially a single cell layer in each lamina. The skeleton of this neuropil grid is made up by the interdigitating dendritic tufts of the geniculocortical relay cells, joined together by a rich system of desmosomoid adhesion plaques. The holes of the skeleton are filled in by the multilobed dendritic appendages of Golgi type II interneurons and the grape-like terminals of the inferior collicular specific afferents. Additional axon terminals of other sources — terminals of descending corticogenicular fibers, axons of the Golgi type II interneurons and terminals of the initial collaterals of the geniculocortical relay cells — contribute only to a very insignificant fraction of neuropil volume. The Golgi type II interneurons are oriented in perpendicular direction to the cell layers so that they may bridge with their dendrites several successive layers.Although the general expression synaptic glomeruli used in other relay nuclei for this type of specific synaptic arrangement is hardly applicable to this grid-like neuropil, the essential synaptic articulation pattern of all thalamic relay nuclei is well maintained. The specific inferior collicular afferents are presynaptic to both relay cell dendrites and to the multilobed dendritic appendages of Golgi type II cells, which in turn are presynaptic to the same dendritic regions of the relay cells receiving the bulk of the specific afferents.     

Mechanical environment,donor age,and presence of endothelium interact to modulate porcine artery viability ex vivo

Though ex vivo culture of arteries is a widely used model of native arteries and is closely aligned with efforts to generate tissue-engineered arteries, the effects of culture conditions on artery viability are poorly characterized. To investigate factors regulating long-term viability of cultured arteries, carotid arteries from neonatal and adolescent pigs were perfused for up to 27 days with steady laminar flow ranging from 2% to 200% of physiological flow rates. Arteries from neonatal animals (2 weeks old, 5 kg) were susceptible to spontaneous progressive endothelial denudation followed by deterioration of the vessel wall that spread from luminal to abluminal regions. Subphysiological levels of flow and pressure abrogated this deterioration. Arteries harvested from adolescent (6 months old, 100 kg) animals maintained viability and retained structure for at least 9 days as assessed by normal histology, presence of intact endothelium, normal mitochondrial activity, and low levels of cell death and proliferation, unless the vessels were subjected to superphysiological levels of flow or the endothelium was intentionally denuded. Adolescent arteries perfused at subphysiological, but not physiological, flow rates maintained viability and normal structure for at least 27 days. These data indicate that under the appropriate conditions, arteries may be cultured long term but careful attention to the viability is merited. © 2002 Biomedical Engineering Society. PAC2002: 8780Rb, 8719Tt, 8719Uv, 8768+z     

The extracellular space of rat intestine “in vivo”

Summary The extracellular space (ECS) of rat jejunum, ileum and colon were determined in in vivo condition both by means of a continuous perfusion technique and after ligature of the renal pedicles. As a marker, intravenously injected¹⁴C-polyethylenglycol (¹⁴C-PEG) was used.The data indicate that the ECS of rats whose renal pedicles were ligated, increases throughout the experiment, due most probably to a hypertonic expansion of the extracellular space. On the contrary, in the experiments with a continuous perfusion the ECS remains constant throughout the experiment.Total tissue water does not change in the two types of experiments whereas the intracellular water decreases only if the renal pedicles are ligated.     

The effects of intraportl infusions of glucagon on the hepatic arterial and portal venous vascular beds of the dog: Inhibition of hepatic arterial vasoconstrictor responses to noradrenaline

The sympathetically-innervated hepatic arterial and portal venous vascular beds of the dog were perfused simultaneously in situ. Glucagon was infused into the hepatic portal vein (1–10 g/min); it caused increases in hepatic portal vascular resistance and tended to reduce the hepatic arterial vascular resistance. Extrahepatic effects of intraportal infusions of glucagon included increases in superior mesenteric blood flow and heart rate and falls in systemic arterial pressure.A test dose of noradrenaline (10 g) injected into either the hepatic artery or the portal vein caused both hepatic arterial and portal venous vasoconstriction. The hepatic arterial constrictor responses to noradrenaline were antagonized intraportal infusions of glucagon. In contrast, intraportal glucagon did not antagonize the portal constrictor responses to intraarterial or intraportal noradrenaline.Elevated portal blood glucagon concentrations may protect the hepatic arterial blood flow from vasoconstriction due to elevated systemic levels of vasoactive substances including catecholamines.     

Voltage clamp analysis in isolated cardiac fibres as performed with two different perfusion chambres for double sucrose gap

Summary Voltage clamp experiments were performed on isolated frog atrial trabeculae disposed in 2 different perfusion chambres for double sucrose gap. In one of the perfusion chambres, a liquid partition system (L.P.S.) was used; in the other, a vaseline sealing (V.P.S.) method was used to separate the various fluid compartments. From the linear electrical properties, the elements of an apparent equivalent electrical circuit were calculated. The apparent nodal capacity was significantly larger in fibres disposed in the liquid partition system than those disposed in the vaseline partition system. The apparent error factor resulting from the presence of the series resistance was larger in the L.P.S. than in the V.P.S. The apparent error factor becomes relatively large when the membrane conductance increases, such as during the flow of the inward current.A rapid desactivation of the peak inward current was found on bringing the command potential back to the resting potential in the V.P.S. This was not found in the L.P.S., indicating better voltage control with the V.P.S. than with the L.P.S.Transmembrane microelectrode recordings during voltage clamp experiment in the V.P.S. indicated satisfactory voltage control during the flow of the peak inward current.Adequate voltage control is lost when notches or irregularities appear on the current traces.     

The Stress–Strain Behavior of Coronary Stent Struts is Size Dependent

Coronary stents are used to re-establish the vascular lumen and flow conditions within the coronary arteries; the typical thickness of a stent strut is 100 m, and average grain sizes of approximately 25 m exist in stainless steel stents. The purpose of this study is to investigate the effect of strut size on the stress strain behavior of 316 L stainless steel. Other materials have shown a size dependence at the micron size scale; however, at present there are no studies that show a material property size dependence in coronary stents. Electropolished stainless steel stent struts within the size range of 60–500 m were tensile tested. The results showed that within the size range of coronary stent struts a size dependent stress–strain relationship is required to describe the material. Finite element models of the final phase of fracture, i.e., void growth models, explained partially the reason for this size effect. This study demonstrated that a size based stress–strain relationship must be used to describe the tensile behavior material of 316 L stainless steel at the size scale of coronary stent struts. © 2003 Biomedical Engineering Society. PAC2003: 8719Rr, 8780Rb, 8719Uv     

Effect of Blood Flow on Thermal Equilibration and Venous Rewarming

In this study we have explored the feasibility of using an isolated rat limb as an animal model for studying countercurrent arterial thermal equilibration and venous rewarming in muscle tissue. Unlike in vivo experiments in which animal models have been used for studying thermoregulation or temperature response in tissue under various physiological conditions, isolated organ or tissue provides for better control and more accurate measurement of the blood perfusion rate. It has been shown that the induced perfusion rate in the rat limb can vary from 3 ml/(min 100 g) at normal physiological conditions to 25 ml/(min 100g) during hyperemic conditions. Temperature distributions along the countercurrent arteries and veins have been measured using fine thermocouple wires. We observed a 25%–78% thermal equilibration along the femoral artery and its branches in intermediate size vessels between 700 and 300 m diameter. This equilibration depends strongly on the local perfusion rate. In comparison, local perfusion rate plays a minor role in determining the overall venous rewarming in the rat hind limb. Approximately 70%–80% of the heat leaving the artery is recaptured by the countercurrent vein. This agrees well with our previous theoretical and experimental results, which show a dramatic shift in thermal equilibration between the supply artery and vein tissue cylinder and the secondary vessel tissue cylinder as the flow rate changes. © 2003 Biomedical Engineering Society. PAC2003: 8719Uv, 8719Pp     

Certain morphological alterations in isolated portions of the intestine after their denervation

Summary Isolated intestinal segments were prepared in dogs. One segment was made by the usual Thiry method. The other was additionally denervated by one of two methods and then, either left within the abdominal cavity, or else transplanted subcutaneously.The morphological and physiological alterations produced were observed over a period of several years.The central nervous system exerts a powerful trophic effect upon the intestine. The denervated segments are restored to central nervous system control by ingrowth of new nerve fibers. In the segment that had been left within the abdomen, these fibers come from nerves which normally innervate the digestive organs. The transplanted segment receives nerve fibers from the subcutaneous net. It is an experimental fact that the intra-abdominal segments tend to return to normal to a much greater extent than do the segments transplanted under the skin.Presented by Active Member Acad. Med. Sci. USSR V. N. Chernigovskii     

Three-layer volume conductor model and software package for applications in surface electromyography

On comparing multichannel surface electromyographic measurements of the m. biceps brachii with simulations performed with a previously developed two-layer volume conduction model, we found substantial discrepancies. To incorporate an apparent distorting effect of the skin tissue, the model was extended to three layers. This new model describes the potential resulting from an eccentric bioelectric source in a finite, cylindrical, and anisotropic volume conductor consisting of three layers, representing muscle, subcutaneous fat, and skin tissue. This contribution presents the governing mathematical equations of the three-layer volume conductor model as well as our approach to their solution. A comparison of various models shows that the three-layer model best describes measured potential distributions. Furthermore, we present the ANVOLCON (analytical volume conductor) software package. This package has been developed to facilitate the use of the model for scientific and educational purposes and is freely available from http://www.mbfys.kun.nl/knf © 2002 Biomedical Engineering Society. PAC2002: 8719Nn, 8719Ff, 8710+e, 8780-y     

Usefulness of low-priming-volume cardiopulmonary bypass circuits and dilutional ultrafiltration in neonatal open-heart surgery

In neonate open-heart surgery, cardiopulmonary bypass (CPB) with extreme hemodilution induces an increased capillary permeability and accumulation of extravascular fluid, resulting in organ dysfunction. We evaluated the effects of a reduced priming volume for CPB and dilutional ultrafiltration (DUF) during neonatal open-heart surgery. Nineteen consecutive neonates with complete transposition of the great arteries who underwent an arterial switch operation were retrospectively assigned into two groups: the high-priming-volume circuit group (group A, n = 9) and the low-priming-volume circuit group (group B, n = 10). Patients in group B underwent surgery with a miniaturized CPB circuit and using the DUF technique. The priming volume of group B was nearly two-thirds that of group A. The water balance value after CPB and surgery was significantly lower in group B (–126 ± 118ml, –116 ± 116ml) than in group A (88 ± 218ml, 83 ± 165ml). Systolic blood pressure just after CPB was higher in group B (67.9 ± 9.1mmHg) than in group A (55.4 ± 10.3mmHg). Postoperative ventilatory support was shorter in group B (45 ± 19h) than in group A (68 ± 27h). In neonatal cardiac surgery, low-priming-volume CPB circuits and DUF improve the water balance during surgery and may attenuate any inflammatory reaction, which would help preserve postoperative organ function.     

Volumetric assessment of the capillary filtration coefficient in the cat small intestine

The capillary filtration coefficient of the cat small intestine was measured in four ways; the conventional measurement was compared with the zero time extrapolation technique, and with two methods designed to assess the end of the blood volume shift due to venous pressure elevation. If the capillary filtration coefficient (CFC) was determined using small elevations of venous pressure (10 mm Hg or less), there was good agreement between the techniques. Larger elevations of venous pressure resulted in apparent overestimates of CFC measured by the conventional technique, probably because of increased smearing of the blood volume shift into the component of the response measured as the CFC. In general, larger elevations of venous pressure gave smaller CFC's, and using venous pressure elevations of 10 mm Hg or less, the CFC was greater if the final elevated venous pressure was 10 mm Hg than if it was 20 or 30 mm Hg. Changes in CFC due to noradrenaline and isoprenaline (about 15 ng/ml arterial blood concentration) were in agreement when small venous pressure elevations were used and CFC measured in three different ways.     

Structure of the intramural nerves of the rat bladder

The bladder of adult female rats receives 16,000 axons (i.e., is the target of that many ganglion neurons) of which at least half are sensory. In nerves containing between 40 and 1200 axons cross-sectional area is proportional to number of axons; >99% of axons are unmyelinated. A capsule forms a seal around nerves and ends abruptly where nerves, after branching, contain 10 axons. A single blood vessel is present in many of the large nerves but never in nerves of <600 axons. The number of glial cells was estimated through the number of their nuclei. There is a glial nucleus profile every 76 axonal profiles. Each glial cell is associated with many axons and collectively covers 1,000 m of axonal length. In all nerves a few axonal profiles contain large clusters of vesicles independent of microtubules. The axons do not branch; they alter their relative position along the nerve; they vary in size along their length; none has a circular profile. All the axons are fully wrapped by glial cells and never contact each other. The volume of axons is larger than that of glial cells (55%–45%), while the surface of glial cell is twice as extensive as that of axons; there are 2.27 m² of axolemma and 4.60 m² of glial cell membrane per gram of nerve. Of the mitochondria of a nerve 3/4 are in axons and 1/4 in glial cells.