Prevalence and clinical correlates of bronchoreversibility in severe emphysema.

Chronic obstructive pulmonary disease (COPD) exhibits airflow obstruction that is not fully reversible. The importance of bronchoreversibility remains controversial. We hypothesised that an emphysematous phenotype of COPD would be associated with decreased bronchoreversibility. 544 patients randomised to the medical arm of the National Emphysema Treatment Trial formed the study group. Participants underwent multiple measurements of bronchoreversibility on a mean of four sessions over 1.91 yrs. They were also characterised by measures of symptoms, quality of life and quantitative measures of emphysema by computed tomography. Mean baseline forced expiratory volume in 1 s (FEV(1)) in this patient population is 24% predicted. 22.2% of patients demonstrated bronchoreversibility on one or more occasions using American Thoracic Society/European Respiratory Society criteria. Few patients (0.37%) had bronchoreversibility on all completed tests. Patients who demonstrated bronchoreversibility were more likely to be male, and have better lung function and less emphysema. 64% of patients demonstrated large (> or =400 mL) changes in forced vital capacity (FVC). In a severe emphysema population, bronchoreversibility as defined by change in FEV(1) is infrequent, varies over time, and is more common in males and those with less severe emphysema. Improvements in FVC, however, were demonstrated in the majority of patients.

Metabolism of 125I-labeled lipoproteins by the isolated rat lung.

The capacity of the isolated perfused rat lung to metabolize the protein moieties of serum lipoproteins was assessed using homologous (rat) and heterologous (human) plasma lipoproteins. The protein and lipid moieties of the plasma lipoproteins were labeled in vivo with Na[125I]. In selected cases the lipoprotein peptides were labeled in vivo with 14C- or 3H-labeled amino acids. Uptake of lipoprotein label during perfusion was monitored by measure of losses in perfusate label and by rises in pulmonary tissue labeling as shown by radioassay and by light and electron microscope radioautography. Lipoprotein degradation was assessed by fractionation of perfusate and lung tissue radioactive material into trichloroacetic acid (TCA)-isoluble, TCA-soluble, and ether-ethanol-soluble fractions. When heparin was included in the perfusion medium, there was selective degradation of the protein portion of very low density lipoprotein (VLDL) in the perfusate and concomitant uptake of radioactive label by the lungs. Low density lipoprotein (LDL)) was neither taken up nor catabolized by the isolated rat lung in the absence or presence of heparin. By light and electron microscopy, the label was localized over the interalveolar septa, predominantly the capillary endothelium. Disappearance of TCA-insoluble radioactivity from the perfusate was associated with the generation of both TCA-soluble iodide and noniodide radioactivity. Greater than 50% of the radioactive label taken up by the lungs was found in the delipidated TCA-insoluble fraction. This study provides in vitro evidence for pulmonary catabolism of VLDL apolipoproteins and uptake of peptide catabolic products of VLDL by the lung.

Traube-Hering waves in the pulmonary circulation of the dog.

Although the pulmonary blood vessels receive an ample supply of nerves, it has been difficult to prove that these nerves operate in the regulation of the pulmonary circulation. In the present study, using the isolated lobe of the lung perfused in situ, waves in pulmonary arterial pressure were induced and were shown, as in the case of Traube-Hering waves in the systemic circulation, to be nervous in origin.

Hemoglobin as a tracer in hemodynamic pulmonary edema.

Stroma-free hemoglobin is an electron-opaque molecule useful as a tracer for the ultrastructural stuty of pulmonary capillary permeability. After this tracer was infused into the isolated pulmonary lobe of the dog, the endothelial junctions of the capillaries, as revealed by electron microscopy, act like distensible pores, thus allowing the tracer to escape when the pulmonary artery pressure was raised above 50 millimeters of mercury.

The effect of anemia on the ventilatory response to transient and steady-state hypoxia.

The effects of anemia upon the ventilatory responses to transient and steady-state hypoxia were studied in unanesthetized goats. Responses to transient hypoxia (inhalation of several breaths of nitrogen) were considered to reflect peripheral chemoreceptor and non-chemoreceptor influences of hypoxia upon ventilatory control. In all goats, severe anemia (hemoglobin 3.1-4.8 g/100ml) markedly heightened the responses to transient hypoxia (from a mean of 0.27 to a mean of 0.75 liter/min/percent fall in SaO2). This phenomenon was substantially reversed by alpha-adrenergic blockade (phenoxybenzamine, 5 mg/kg). In contrast, the ventilatory responses to steady-state hypoxia were unaffected by severe anemia. These data suggest that severe anemia enhances the peripheral chemoreceptor-mediated response to hypoxia through a mechanism involving the alpha-adrenergic system. It also appears that a ventilatory depressant effect of hypoxia which is not mediated by the peripheral chemoreceptors is also enhanced by severe anemia, thereby preventing an increase in the steady-state ventilatory response to hypoxia. Finally, experiments involving variation in oxygen affinity of hemoglobin suggested that O2 tension rather than O2 availability in arterial blood is the major determinant of peripheral chemoreceptor activity.

The application of Starling's law of capillary exchange to the lungs.

The forces governing the movement of water across the pulmonary capillaries were studied in 39 intact, spontaneously breathing dogs. A situation favoring the net movement of water out of the pulmonary capillaries was created by means of partial pulmonary venous obstruction (left atrial balloon catheter) followed by rapid saline hemodilution. A predetermined difference between pulmonary capillary and plasma colloid osmotic pressures was maintained for periods of 1 to 2 hours. Left atrial (P(LA)) and plasma colloid osmotic pressures (pi(pl)) were measured directly. The water content of the lungs was measured serially by an indicator-dilution technique, and at autopsy by drying the lungs. The rate of accumulation of lung water was measured in four groups of animals: in three of the groups, the capillary hydrostatic and colloid osmotic pressures were varied; in the fourth group, the right lymphatic duct was obstructed in addition. The average rate of water accumulation in the lungs varied in a nonlinear way with the level of the capillary hydrostatic-plasma colloid osmotic pressure difference and was unaffected by the level of the capillary hydrostatic pressure. At low levels of P(LA) - pi(pl), water accumulated in the lung at an average rate of 0.09 g per g dry lung per hour per mm Hg pressure difference. At higher levels of P(LA) - pi(pl) the average rate of accumulation was 0.22 g per g per hour per mm Hg DeltaP; in most of the experiments in this group water accumulated in the lungs slowly during the first 30 minutes of the test period and more rapidly as the period was extended. Obstruction of right lymphatic duct outflow did not alter the rate of water accumulation. Based on the control data of the present experiments, the pericapillary pressure in normal lungs is estimated to be of the order of - 9 mm Hg in the normal dog lung. The filtration coefficient for the pulmonary capillaries is estimated to be of the order of one-tenth to one-twentieth of that for canine muscle capillaries. The data of the present study indicate that edema formation in lung tissue cannot be defined solely in terms of intravascular forces, but may be governed to a significant degree by changes in pericapillary forces in the pulmonary interstitium.

Respiratory adjustment to chronic metabolic alkalosis in man.

This study examined the ventilatory adjustment to chronic metabolic alkalosis induced under controlled conditions in normal human volunteers. Metabolic alkalosis induced by buffers (sodium bicarbonate, trishydroxymethylamine methane) or ethacrynic acid was associated with alveolar hypoventilation, as evidenced by a rise in arterial Pco(2), a fall in arterial Po(2), a reduced resting tidal volume, and a diminished ventilatory response to CO(2) inhalation. Alveolar hypoventilation did not occur when metabolic alkalosis was induced in the same subjects by thiazide diuretics or aldosterone despite comparable elevations of the arterial blood pH and bicarbonate concentration.The different ventilatory responses of the two groups could not be ascribed to differences among individuals comprising each group, pharmacological effects of the alkalinizing agents, differences in the composition of the lumber spinal fluid, changes in extracellular fluid volume, or sodium and chloride balance.The differences in ventilatory adjustments were associated with differences in the patterns of hydrogen and potassium ion balance during the induction of alkalosis. Alveolar hypoventilation occurred when hydrogen ions were buffered (sodium bicarbonate, trishydroxymethylamine methane) or when renal hydrogen ion excretion was increased (ethacrynic acid). Alveolar hypoventilation did not occur when induction of similar degrees of extracellular alkalosis was accompanied by marked potassium loss and no demonstrable increase in external hydrogen loss (thiazides and aldosterone).These observations suggest that respiratory depression does not necessarily accompany extracellular alkalosis but depends on the effect of the mode of induction of the alkalosis on the tissues involved in the control of ventilation.

Reflex effects of aerosolized histamine on phrenic nerve activity.

Studies were conducted in anesthetized, paralyzed dogs on the effect of aerosolized histamine on phrenic nerve activity. The paralyzed dogs were ventilated in phase with their recorded phrenic nerve activity at a constant inspiratory flow-rate, using a cycle-triggered ventilator. Phrenic nerve activity was measured before and during administration of aerosolized histamine while the inspiratory flow-rate and arterial blood gases were kept constant. In addition, before and after histamine, phrenic nerve activity was recorded for single bursts during which the ventilator was switched off. The effects of histamine on respiratory resistance were prevented by prior administration of isoproterenol and atropine. Although no changes occurred in respiratory resistance, histamine increased the instantaneous magnitude of phrenic nerve activity. The effect was evident early in the inspiratory period and was found even when the lungs were not inflated. Inflation of the lungs excited phrenic nerve activity; this effect increased after histamine. All of these actions of histamine were abolished by vagotomy. We conclude that histamine increased phrenic nerve activity during inspiration by a vagal reflex.

The effects of abnormal sympathetic nervous function upon the ventilatory response to hypoxia.

THE VENTILATORY RESPONSE TO HYPOXIA WAS STUDIED IN TWO GROUPS OF SUBJECTS WITH ABNORMAL SYMPATHETIC NERVOUS CONTROL: (a) human subjects with familial dysautonomia (Riley-Day syndrome), and (b) unanesthetized goats treated with an alpha-adrenergic blocking agent (phenoxybenzamine). The ventilatory response to hypoxia was evaluated in two ways: (a) from the slope of the relationship between ventilation and alveolar P(Co2) ([unk]V(E)-P(ACo2) slope) during the rebreathing of hypoxic and hyperoxic gases, and (b) from the change in ventilation produced when hypoxia was abruptly relieved. The ventilatory and circulatory responses of the unanesthetized, phenoxybenzamine-treated goats were qualitatively similar to those of dysautonomic patients. In contrast to the sustained stimulation of ventilation produced by hypoxia in normal subjects, hypoxia either did not change, or decreased, the [unk]V(E)-P(ACo2) slope of dysautonomic patients and phenoxybenzamine-treated goats; CO(2)-free hypoxia produced a fleeting hyperventilation, which was followed by apnea when hypoxia was abruptly relieved. Unlike normal subjects, the dysautonomic patients and phenoxybenzamine-treated goats became hypotensive while hypoxic. The results indicate that peripheral chemoreceptor reflex responses to hypoxia are preserved in subjects in whom sympathetic nervous responses are impaired. However, the central nervous depression of ventilation by hypoxia is enhanced simultaneously. The inordinate central depression is attributed to the inability of the dysautonomic subjects and goats to maintain systemic blood pressure and, consequently, cerebral blood flow during hypoxia, thereby aggrevating central nervous hypoxia.

Autoradiographic distribution of thyrotropin-releasing hormone receptors in the African lungfish Protopterus annectens.

We used quantitative autoradiography to examine the distribution of thyrotropin-releasing hormone (TRH) receptors in the central nervous system (CNS) of the African lungfish Protopterus annectens. We found that the distribution of TRH receptors throughout the CNS of the lungfish was heterogeneous with the highest concentrations (500-800 fmol/mg protein) in the olfactory bulb and telencephalon, moderately high concentrations (200-500 fmol/mg protein) in the diencephalon, and moderate (50-200 fmol/mg protein) to low (less than 50 fmol/mg protein) concentrations in the brainstem and spinal cord. Except for the motor nuclei of the cranial nerves and spinal cord, TRH receptors were concentrated in the acellular regions. In the telencephalon and diencephalon, the receptor density was inversely related to cellular density. These results provide a neuroanatomic and neuropharmacologic basis for further investigations of TRH in the African lungfish.

Contractile mechanics and interaction of the right and left ventricles.

The heart and lungs, together with hemoglobin, provide for the transport of oxygen from the atmosphere to the metabolizing tissue. The oxygenation of blood and the circulation of oxygenated blood are precisely synchronized so that the heart and lungs constitute an integrated cardiopulmonary unit. The functional integration of the heart and lungs is fostered by their anatomic arrangement and mechanical interaction. The cardiopulmonary unit consists of the right and left ventricles (two in-series pumps composed of cardiac muscle), which are mechanically coupled by the lungs. The factors that control cardiac muscle shortening (fiber length, afterload and myocardial contractile state) also regulate the pumping behavior of each ventricle. Because the ventricles are aligned in series a perturbation in the mechanical events of one ventricle will influence the behavior of the other ventricle. The interventricular septum and pericardium further promote the mechanical interplay between ventricles. Intrathoracic pressure (the pressure that surrounds the cardiopulmonary unit) creates an additional interaction between the ventricles as well as the heart and lungs.

Etiology and pathogenesis of primary pulmonary hypertension: a perspective.

In recent years, considerable advances have been made in treating primary pulmonary hypertension (PPH). These have provided a series of therapeutic options, ranging from the oral administration of calcium channel blockers to the continuous infusion of prostacyclin and/or lung transplantation. These therapeutic advances have highlighted the need for the better understanding of etiology and pathogenesis. Among the key uncertainties, the following are defined as leading uncertainties: (1) the nature of the initiating lesion; (2) the shared pathogenetic mechanisms that culminate in the pathologic lesions of PPH; (3) the molecular genetic bases for familial PPH and for susceptibility to PPH; (4) understanding of the obliterative-proliferative occlusive process in the small muscular pulmonary arteries; and (5) redefinition of "primary" and "secondary," ie, a revised nomenclature of pulmonary hypertension. A revised classification based on etiology is presented.

Endothelium: a distributed organ of diverse capabilities.

A historical survey of the growth of ideas about the circulation of the blood and about the endothelium is provided as a background to the specialized papers that follow. Over the years, in the course of clarifying its structure and function, the status of the endothelium has gradually evolved from that of an inert semipermeable barrier that determines the exchange of substances between blood vessels and tissue spaces to that of a specialized organ that is unusual in its dispersion throughout the body as the lining of blood vessels. This unique arrangement has enabled it to play different roles in different parts of the vascular tree and in different organs and to manifest distinctive responses to injury according to its structure and location. As a corollary, because its responsiveness is strongly influenced by its ambience and neighboring tissues, it seems inevitable that extrapolations from function in vitro to function in vivo require considerable prudence.

Circulatory adaptation to bimodal respiration in the dipnoan lungfish.

In the dipnoan lungfish, Protopterus aethiopicus, P. annectens, and Lepidosiren paradoxa, the ductus is a short powerful muscular vascular trunk forming a channel for communication between the systemic and pulmonary circulations. In structure, the dipnoan ductus is very similar to the ductus arteriosus (Botalli) in the mammal. Innervation is abundant, consisting of myelinated and nonmyelinated nerve fibers issuing, at least in part, from the vagus. Neurons are present in the adventitia, and numerous nerve profiles, filled with small agranular vesicles, are closely associated with the myocytes, suggesting strong cholinergic control. Perfusion of the ductus in vitro using hypoxic saline causes it to dilate; conversely it is constricted by alpha-agonists. Dopamine and prostaglandin E2 are potent dilators, whereas the beta-agonist, isoproterenol, and acetylcholine are less powerful. A vasomotor segment has been identified on the pulmonary artery (PAVS) close to its junction with the ductus. Its location and structure are similar to the corresponding segment in amphibians and reptiles. It is innervated by endings filled with small clear vesicles. Granular vesicle cells are also present within the adventitia. The PAVS is constricted by acetylcholine. As in amphibians, alpha-agonists and hypoxic saline are without vasomotor effects. Based on the anatomic and physiological observations, a concept of cyclic perfusion of the gas exchangers in Dipnoi is proposed. During the alternation between air breathing (emersion) and apneic phases (immersion), the pattern of the circulation in the lungfish oscillates between that of a tetrapod and a fish.

Effects of increased ventilation on lung lymph flow in unanesthetized sheep.

To determine the effect of an increase in spontaneous minute ventilation on lung fluid balance, we added external dead space to the breathing circuit of six tracheostomized, unanesthetized, spontaneously breathing sheep in which lung lymph fistulas had been created surgically. The addition of 120-180 ml of dead space caused minute ventilation to increase by 50-100% (secondary to increases in both tidal volume and frequency), without changing pulmonary arterial pressure, pulmonary capillary wedge pressure, cardiac output, or arterial blood gas tensions. The increase in spontaneous ventilation was associated with an average increase of 27% in lung lymph flow (P less than 0.05) and an average reduction of 11% in the lymph-to-plasma concentration ratio (L/P) for total protein (P less than 0.05). Lymph flow and L/P for total protein approached stable values after 2-3 h of hyperpnea, and the increase in lymph flow persisted for at least 18 h of dead-space breathing. Removal of dead space was associated with a rapid return (within 45 min) of lymph flow to base-line levels. These results suggest that hyperpnea increases the pulmonary transvascular filtration rate. Since no changes in vascular pressures or cardiac output were observed, this increase in transvascular filtration is most likely due to a fall in interstitial fluid pressure.

Chromatographic demonstration of reversible changes in endothelial permeability.

This report describes a new in vitro method for measuring the diffusional permeability of an endothelial monolayer and its use in investigating the modulation of permeability by various agents, e.g., isoproterenol, propranolol, dibutyryl adenosine 3',5'-cyclic monophosphate (cAMP), and cytochalasin D. To determine permeability, tracers of different molecular weights were applied simultaneously on a chromatography column containing confluent endothelial cells cultured on porous microcarrier beads. The Sangren-Sheppard model was used to determine the permeability of the endothelial monolayer from the tracer elution profiles. For six radiolabeled tracers the mean (+/- SD) permeabilities (cm/s x 10(-5)) in order of increasing tracer molecular weight were [3H]water, 82.0 +/- 28.8; [14C]urea, 49.5 +/- 9.5; [14C]mannitol, 13.3 +/- 4.7; [14C]-sucrose, 14.1 +/- 2.5; [3H]polyethylene glycol (900 mol wt), 4.80 +/- 1.61; and [3H]polyethylene glycol (4,000 mol wt), 1.97 +/- 1.01. These permeabilities deviate less from in vivo values than those obtained in other in vitro systems and are 10 times higher than in vivo estimates. The values were reproducible for up to the 4 h tested. Modulation of endothelial monolayer permeability was studied in a separate series of experiments. The beta-adrenergic agonist isoproterenol (10(-6) M) decreased the permeability to mannitol by 36% and to polyethylene glycol (900 mol wt) by 49%; in both instances the decrease in permeability was reversed by propranolol. Propranolol alone had no effect. Dibutyryl cAMP (10(-3) M) decreased the permeability to mannitol by 40% and to polyethylene glycol by 47%; permeability returned to base line when dibutyryl cAMP was removed. Cytochalasin D (1 microgram/ml) increased permeability by 350% for mannitol and 380% for polyethylene glycol; the permeability change was reversed after removal of cytochalasin D. The results indicate that cell-column chromatography is a powerful method that can be used to characterize the permeability of endothelial monolayers and to investigate permeability changes produced by various agents.

Adaptation to reflex effects of prolonged lung inflation.

Adaptation to the reflex effects of sustained changes in lung volume on inspiratory duration (TI), expiratory duration (TE), and the phrenic neurogram was examined. Test inflations in gallamine-paralyzed dogs anesthetized with pentobarbital sodium were made during a 6-min trial while the animal was not ventilated: 2 min at functional residual capacity (FRC), 2 min at elevated airway pressure, and 2 min back at FRC. The dogs were hyperoxygenated and arterial PCO2 was kept constant by an infusion of tris (hydroxymethyl) aminomethane. The maintained inflations produced minimal changes in TI. On return to FRC, TI was prolonged in proportion to the magnitude of the prior inflation. In contrast, inflation produced marked prolongation of TE, which then adapted back toward preinflation values. On return to FRC, TE shortened initially to values below control. This shortening increased with greater prior lung inflations. The times to reestablish steady-state values upon return to FRC differed for TI (14.8 +/- 4.6 s) and TE (33.8 +/- 12.7 s). The magnitude of the phrenic neurogram at a fixed time from onset of inspiration and its slope were unchanged with inflation. These results indicate that respiratory phase durations are influenced not only by pulmonary afferent input within each respiratory cycle but also by prior vagal afferent activity that engages central processes with long, although different, time constants. Afferent input to the slow central process controlling TI is not gated to only one phase of the respiratory cycle.

Permeability characteristics of cultured endothelial cell monolayers.

The purpose of this study was to characterize the permeability characteristics of an in vitro endothelial cell monolayer system and relate this information to available in vivo data. We cultured bovine fetal aortic endothelial cells on fibronectin-coated polycarbonate filters and confirmed that our system was similar to others in the literature with regard to morphological appearance, transendothelial electrical resistance, and the permeability coefficient for albumin. We then compared our system with in vivo endothelium by studying the movement of neutral and negatively charged radiolabeled dextran tracers across the monolayer and by using electron microscopy to follow the pathways taken by native ferritin. There were a number of differences. The permeability of our monolayer was 10-100 times greater than seen in intact endothelium, there was no evidence of "restricted" diffusion or charge selectivity, and ferritin was able to move freely into the subendothelial space. The reason for these differences appeared to be small (0.5-2.0 micron) gaps between 5 and 10% of the endothelial cells. Although the current use of cultured endothelial cells on porous supports may provide useful information about the interaction of macromolecules with the endothelium, there appear to be differences in the transendothelial permeability characteristics of these models and in vivo blood vessels.

Passage of uncharged dextrans from blood to lung lymph in awake sheep.

To examine how molecular size alone influences the passage of macromolecules from the pulmonary microcirculation into lymph collected from the caudal mediastinal lymph node of the sheep, we infused polydisperse uncharged [3H]dextrans intravenously at a constant rate over a period of 7.5 h in nine awake sheep with lung lymph fistulas. Lymph and plasma were collected during hours 5.5-7.5 of the infusions, and the [3H]dextrans were separated by molecular sieve chromatography into fractions that ranged from 1.6 to 8.4 nm in effective molecular (Stokes-Einstein) radius. Lymph-to-plasma (L/P) ratios for [3H]dextrans were near 1.0 at 1.6-nm radius, decreased with increasing molecular size, and approached zero at radii above 5.0 nm. We confirmed that these L/P ratios represented steady-state values by extending the duration of the infusion to approximately 30 h in two of the nine sheep and finding that the L/P ratios remained unchanged. These results were consistent with molecular sieving through a homoporous membrane with cylindrical pores of 5.0-nm radius. We also found that the L/P ratio for albumin [0.76 +/- 0.13 (SE)] in five of the same sheep was much higher than that for the [3H]dextran fraction of the same effective molecular radius [0.11 +/- 0.02 (SE)]. These results suggest that the movement of macromolecules from the pulmonary microcirculation into pulmonary lymph collected from the caudal mediastinal node of the sheep is influenced by both molecular size and molecular charge and that, compared with uncharged dextrans, the steady-state passage of anionic endogenous proteins from plasma to lymph is enhanced.

Effect of 100% O2 on passage of uncharged dextrans from blood to lung lymph.

Since charge as well as size may influence the passage of plasma proteins from blood to lung lymph, we used uncharged dextrans as tracers to study the effects of hyperoxic lung injury on the molecular sieving properties of the pulmonary microcirculation in unanesthetized sheep. Polydisperse [3H]dextran was infused intravenously into five sheep before and after the animals breathed 100% O2 until lymph flow increased threefold (66-84 h). Lymph-to-plasma concentration ratios (L/P) were determined for [3H]dextran fractions of graded molecular sizes (1.6-8.4 nm effective radius) from samples obtained during the infusions. Before hyperoxia the blood-lymph barrier was highly restrictive to transport of [3H]dextrans above 5.0 nm in radius; steady-state L/P for these molecules averaged 0.03 or less. After the sheep breathed 100% O2, [3H]dextrans as large as 8.4 nm radius appeared in the lymph. Posthyperoxia, the L/P were significantly increased relative to prehyperoxia base-line values for every [3H]dextran fraction larger than 2.0 nm radius (P less than 0.05). In contrast, neither the L/P for albumin or total protein changed significantly. At autopsy, electron microscopy showed widespread damage to the endothelium of the alveolar capillaries with infrequent gaps between endothelial cells. In two control sheep, inhalation of compressed air for 96 h had no effect on lymph flow or L/P for the [3H]dextrans. We conclude that O2 poisoning reduced the selective sieving of uncharged dextrans across the blood-lymph barrier of the lungs and allowed larger dextrans to enter the lymph. These larger molecules may have leaked from the pulmonary microcirculation via disruptions in the continuity of the endothelial lining.