Search

Nephrogenic systemic fibrosis, initially called nephrogenic fibrosing dermopathy, has been strongly linked to exposure to gadolinium-based contrast media used in magnetic resonance imaging in patients with renal insufficiency. This review discusses recent advances in our understanding of the pathophysiology and clinical approach to patients with chronic kidney disease who require diagnostic imaging with gadolinium-based contrast media.

The cutaneous microcirculation is organized as two horizontal plexuses. One is situated 1-1.5 mm below the skin surface and the other is at the dermal-subcutaneous junction. Ascending arterioles and descending venules are paired as they connect the two plexuses. From the upper layer, arterial capillaries rise to form the dermal papillary loops that represent the nutritive component of the skin circulation. There are sphincter-like smooth muscle cells at the point where the ascending arterioles divide to form the arteriolar component of the upper horizontal plexus. At the dermal-subcutaneous junction, there are collecting veins with two cusped valves that are oriented to prevent the retrograde flow of blood. Laser Doppler flowmetry has demonstrated vasomotion of red cell flux localized to the sites of ascending arterioles. The simultaneous recording by laser Doppler flowmetry of red cell flux and the concentration of moving red blood cells from individual sites allows one to construct topographic maps of these two values. These two maps, based on initial studies using correlative skin biopsies, can define 1 mm3 volumes of skin that are predominantly arteriolar in composition, venular in composition, or essentially devoid of all microvascular elements. The electron and light microscopic features that define the microvascular segments, when coupled with that ability of laser Doppler flowmetry to define the predominant microvascular segments under the probe, allow one to study both the mechanisms of normal physiologic states and the pathogenetic mechanisms underlying pathologic skin disorders in which the microvasculature plays a predominant role.

The particulate enzyme, endothelial nitric oxide synthase (eNOS), produces nitric oxide to maintain normal vasodilator tone in blood vessels. In this study, we demonstrate that eNOS is a Golgi-associated protein in cultured endothelial cells and intact blood vessels. Using a heterologous expression system in HEK 293 cells, we show that wild-type myristoylated and palmitoylated eNOS, but not mutant, non-acylated eNOS targets to the Golgi. More importantly, HEK 293 cells expressing wild-type eNOS release substantially more NO than cells expressing the mutant, non-acylated enzyme. Thus, eNOS is a novel Golgi-associated protein, and Golgi compartmentalization is necessary for the enzyme to respond to intracellular signals and produce NO.

Previous studies have demonstrated 1) that patterns of inducible endothelial cell expression of endothelial leukocyte adhesion molecule-1 (ELAM-1) and vascular cell adhesion molecule-1 (VCAM-1) in response to cytokines varies both with anatomic position within the dermal microvasculature and with the presence of perivascular inflammatory infiltrates, and 2) that the anatomic architecture of the dermal superficial plexus (SVP) is altered in inflamed lesional but not in univolved skin of psoriatic patients. The present study was designed to evaluate the pattern of cytokine inducibility of ELAM-1 and VCAM-1 in altered dermal microvessels of psoriatic patients. At the light microscope level, preculture biopsies of uninvolved and perilesional skin were indistinguishable by morphology and ELAM-1 and VCAM-1 expression were virtually absent. In contrast, biopsied lesional skin showed elongated capillary loops and increased numbers of T cells compared to uninvolved and perilesional skin. The dermal microvasculature of the SVP of lesional skin contained ELAM-1+ in 29.4% of vessels and VCAM-1+ endothelial cells in 8.7% of vessels. After 24 h of organ culture in medium supplemented with tumor necrosis factor and interleukin-4, ELAM-1+ endothelial cells in the SVP were increased significantly in uninvolved (from mean 0.5% to 27% of vessels), perilesional (from mean 5.5% to 41.8% of vessels), and lesional skin (from mean 29.4% to 45.7% of vessels). VCAM-1 was not inducible on SVP endothelial cells in uninvolved skin but VCAM-1+ endothelial cells were increased significantly in perilesional (from mean 0.7% to 23.7% of vessels) and lesional skin (from mean 8.7% to 41.4% of vessels). In uninvolved and perilesional skin ELAM-1 and VCAM-1 were confined to endothelial cells below the rete. In contrast, endothelial cells of the intrapapillary part of the capillary loop of lesional skin became cytokine responsive, in that ELAM-1 and VCAM-1 could be induced at this site. By immunoelectron microscopy, expression was most intense on the luminal surface of venular endothelial cells and at the interendothelial junctions. In conclusion, we have presented evidence that the cytokine responsiveness of microvascular endothelial cells is altered in psoriasis in a pattern that may explain both the circumscribed nature and the epidermal involvement of the psoriatic plaque.

Background: Cutaneous laser Doppler flowmetry enables monitoring of changes in skin perfusion by quantifying the phase shift of laser light induced by moving red blood cells under a fiberoptic probe. It thus can identify the presence of and response to a vasoconstrictive stimulus. However, aspects of the technique must be defined before it can be used with maximum effectiveness. We evaluated the responses of two different laser Doppler outputs, the concentration of moving blood cells (CMBC) and red cell flux (CMBC times cell velocity), and the responses at two sites of probe application, the finger and forearm, during systemic infusions of phenylephrine., Methods and Results: Eight healthy volunteers were monitored with a brachial blood pressure cuff, ECG, and laser Doppler flowmeter probes applied to the palmar surface of the fourth finger and volar forearm of the arm opposite the pressure cuff. After baseline readings were obtained, the subjects received three 10-minute intravenous infusions of phenylephrine at rates of 0.4, 0.8, and 1.6 micrograms.kg-1.min-1. The two parameters, flux and CMBC, trended similarly. Flux and CMBC at the finger declined significantly in response to each infusion (P < .05 using repeated-measures ANOVA with Duncan's multiple range test). In contrast, flux and CMBC of the forearm had highly variable responses, with an overall increase during each infusion (P < .05 for % delta of forearm versus % delta of finger readings during the 0.4 microgram.kg-1.min-1 infusion). Heart rate declined significantly during each infusion, consistent with a baroreceptor-mediated response, even though systolic and diastolic blood pressures each increased by less than 2 mm Hg during the 0.4 microgram.kg-1.min-1 infusion., Conclusions: As expected, laser Doppler readings at the finger decreased during infusion of an alpha 1-agonist. Although, like the digital vessels, forearm vessels have the potential to constrict, the increases in forearm readings suggest that these vessels are highly susceptible to homeostatic responses. The increase in CMBC (a parameter that is sensitive primarily to local changes in vascular caliber) suggested vasodilation of the underlying vessels. The forearm vasodilation and the concomitant decline in heart rate most likely represented vagally mediated baroreceptor activity, which was altered even though blood pressure changed minimally during the 0.4 microgram.kg-1.min-1 infusion. Thus, integrated assessment of skin perfusion at the finger and forearm may provide valuable information about the direct and indirect effects of a vasoactive stimulus. The present application of laser Doppler flowmetry suggests activation of vasodilatory reflexes despite minimal changes in blood pressure.

Dermatosparaxis is a recessively inherited connective-tissue disorder that results from lack of the activity of type I procollagen N-proteinase, the enzyme that removes the amino-terminal propeptides from type I procollagen. Initially identified in cattle more than 20 years ago, the disorder was subsequently characterized in sheep, cats, and dogs. Affected animals have fragile skin, lax joints, and often die prematurely because of sepsis following avulsion of portions of skin. We recently identified two children with soft, lax, and fragile skin, which, when examined by transmission electron microscopy, contained the twisted, ribbon-like collagen fibrils characteristic of dermatosparaxis. Skin extracts from one child contained collagen precursors with amino-terminal extensions. Cultured fibroblasts from both children failed to cleave the amino-terminal propeptides from the pro alpha 1(I) and pro alpha 2(I) chains in type I procollagen molecules. Extracts of normal cells cleaved to collagen, the type I procollagen synthesized by cells from both children, demonstrating that the enzyme, not the substrate, was defective. These findings distinguish dermatosparaxis from Ehlers-Danlos syndrome type VII, which results from substrate mutations that prevent proteolytic processing of type I procollagen molecules.

A probe holder for laser Doppler velocimetry was designed to allow a narrow (1.24-mm) probe to be moved in increments of 1.04 mm, approximately the width of the probe window itself, so that four contiguous 1-mm2 spots in a 2.29 x 2.29 mm area could be sampled. An area as large as 2.48 x 2.48 cm can be measured with this device. The flux was fed into a computer by an analog/digital board, for visualization as a wave-form on the monitor, and for analysis by fast Fourier transforms and power spectrum analysis. Each spot sampled was given a unique x,y coordinate and the mean amplitude of the flux was designated as the z coordinate. With the aid of software, the values of these three coordinates were mathematically processed to produce contour graphs with shading that represented a map of the arteriolar circulation in the skin. This methodology allows for accurate movement and placement of the probe on the skin and the ability to relocate a spot within 1 mm of its original location after intervals of hours to days. Video images of the superficial plexus at the sites where the maps were made confirm the topographic appearance of these maps. Data generated by these maps raise the intriguing possibility that microcirculatory vessels may be engaged in a division of labor--some involved primarily with tissue perfusion, others with thermoregulation.

We studied 10 cutaneous telangiectatic lesions of hereditary hemorrhagic telangiectasia (HHT), ranging in size from pinpoint to 2 mm, by light and electron microscopy. Four representative lesions were reconstructed by computer from serial 1- or 2-mm plastic embedded sections. The earliest clinically detectable lesion of HHT is a focal dilatation of postcapillary venules, which continue to enlarge and eventually connect with dilated arterioles through capillaries. As the vascular lesion increases in size, the capillary segments disappear and a direct arterio-venous communication is formed. This entire sequence of morphologic events is associated with a perivascular mononuclear cell infiltrate in which the majority of cells are lymphocytes and the minority are monocytes/macrophages by ultrastructure. Comparison of these findings with the telangiectatic mats of scleroderma and cherry angiomas revealed that the former, previously shown to be composed of dilated postcapillary venules, are also associated with perivascular infiltrates, but the latter, which are produced by capillary loop aneurysms, are not.

Laser Doppler velocimetry (LDV) was performed on the chest, back, and abdomen of four healthy volunteers. As the probe was moved over distances of 2-6 mm, the red-cell flux varied by 100%, but was associated with three distinctive wave patterns. Correlative skin biopsies showed that a high flux, pulsatile pattern superimposed on vasomotor activity was found when the probe was directly over an ascending elastic arteriole with its immediate branches; low flux, pulsatile flow with minimal or no vasomotor activity was found when the probe was off center relative to the ascending arteriole and its branches; and a low flux, non-pulsatile pattern occurred when the probe window was situated between ascending arterioles over an area in the upper horizontal plexus composed primarily of capillaries and post-capillary venules.

The microvessels in the buttock skin of 15 patients with long-standing juvenile diabetes were studied both by electron microscopy and three-dimensional (3D) computer reconstruction of a prototypical diabetic postcapillary venule. Endothelial cell gaps were found in postcapillary venules and capillaries, but only in association with an increased deposition of basement membrane-like material in the vascular wall. In parallel with the increased amounts of deposited basement membrane-like material, the space between pericytes and endothelial cells was wider and the cytoplasmic processes that formed the contact points between them were longer and thinner than normal. Pericytes, devoid of any cytoplasmic contacts with the underlying endothelial cells, were observed as isolated cells within the outer third of the vascular wall in markedly thickened vessels. These observations offer an explanation for the known increased vascular permeability of diabetic vessels, and suggest a possible explanation for the development of diabetic retinopathy with aneurysm formation.

The three-dimensional relationships between smooth muscle cells and endothelial cells and between pericytes and endothelial cells in four segments of the microcirculation were analyzed by computer reconstructions from serial electron micrographs. In elastic-containing arterioles, the smooth muscle cells formed an inner longitudinal layer above and parallel to the elastica and an outer spiral layer. In the terminal arterioles the two layers of smooth muscle cells and elastica were replaced by a single smooth muscle cell that completely encircled the endothelial cell tube. The pericytes in the post-capillary venules completely encircled and gripped the endothelium through multiple contact points from their lateral processes. In the large venules the pericytes only partially encircled the endothelial cell tube and were more randomly placed.

The blotchy mouse has an X chromosome mutation affecting crosslinking of collagen and elastin, which results in aneurysmal dilatation of the aorta. The age of onset, patterns of distribution, and histologic features of these lesions have not been characterized in detail in previous studies. Male normal and blotchy mice 1 to 8 months of age were killed and latex was injected into the left ventricles to facilitate exposure, examination, histologic sampling, and photography of the aorta. Aneurysms were not detected in any normal animals but the affected animals had a progressive increase in the incidence of aneurysms with age, reaching 100% by 6 months. Most aneurysms occurred in the ascending aorta, with some also present in the descending thoracic and abdominal segments. Some animals had multiple aneurysms. Histologically the blotchy mice aortas exhibited disrupted elastic lamellae and thickening of the interlamellar spaces. These spaces contained conspicuously pleomorphic smooth muscle cells, confirmed by electron microscopy. These changes occurred as early as 21 days, when there was no gross evidence of aneurysmal development. Aortic aneurysms develop in blotchy mice in a consistent fashion, with characteristic gross and histologic changes. These animals provide a practical model for further studies of aneurysmal disease, including possible therapeutic interventions to prevent aneurysm development.

Eight types of telangiectases were studied by light and electron microscopy and by 3-dimensional reconstruction from photomicrographs. Five were macular: mat telangiectasia of scleroderma, generalized essential telangiectasia, nevus flammeus, and 2 macular types not previously described. Three were papular: cherry angioma, angiokeratoma (Fabry), and angiokeratoma (Fordyce). The macular telangiectases were produced by dilatation of postcapillary venules of the upper horizontal plexus. There was no evidence of neovascularization or vascular malformation. The walls of the dilated venules were thickened by the peripheral deposition of basement membrane-like material admixed with reticulin fibers. The ultrastructure and configuration of the papular telangiectases were different. The cherry angioma was produced by spherical and tubular dilatations of capillary loops in dermal papillae. Each abnormally dilated loop was connected to the neighboring loop or loops by tortuous vascular channels. The vessels in the upper horizontal plexus were not involved. Ultrastructurally, the cherry angiomas were composed of both venous capillaries and postcapillary venules whose walls were thickened in a manner identical to that observed in the macular telangiectases. The angiokeratomas of Fabry and Fordyce were also produced by vascular abnormalities predominantly involving the dermal papillae. Ultrastructurally these vessels were similar to the small collecting veins which are normally found at the dermal-subcutaneous interface. Thus, the papular telangiectases also arose by alterations of the existing microvasculature rather than by proliferation of new vessels with random anastomoses. Reconstruction of the upper horizontal plexus from normal skin showed an undulating network of arterioles and their accompanying postcapillary venules. A 3-layered plexus arranged as venules, arterioles, and venules was not found.

Clinical and laboratory observations have strongly suggested that leukocytoclastic angitis is an immune complex disease. Since immune complexes can be visualized as electron-dense deposits by electron microscopy (EM), this method was used in conjunction with direct immunofluorescence (IF) to determine whether complexes could be demonstrated in spontaneous lesions, and in uninvolved skin in which the vessels were made permeable by the local injection of histamine. Histamine-induced wheals were produced in the uninvolved skin of patients with active angitis. In the resulting wheal, EM studies revealed electron-dense deposits characteristic of immune complexes in postcapillary venules and direct IF studies demonstrated complement and immunoglobulins in the vessel walls. Neutrophils in varying stages of disintegration were present thereby reproducing the histopathologic changes of spontaneous lesions. EM and IF studies of nonmanipulated uninvolved skin also revealed electron-dense deposits and immune reactants in the vessel walls. Neutrophils were not present, however. This observation indicates that immune complexes are deposited in vessels before tissue damage ensues. Study of spontaneous lesions older than 24 hr revealed only fibrin by EM and no immune reactants by direct IF. In spontaneous lesions less than 24 hr old, electron-dense deposits and fibrin were seen by EM, and complement and immunoglobulins by IF. Histamine-induced wheals should be a useful device to investigate patients with disorders that have an immune complex pathogenesis.

Electron microscopy was used to define the ultrastructure of the capillary loops in the dermal papillae of normal, eczematous, and psoriatic skin. Capillary loops were reconstructed from 1-mum plastic-embedded sections. Ultrathin sections were taken at 4- to 6-mum intervals for correlation with the thick sections. The capillary loops of normal forearm and elbow skin could be divided into two segments: an intrapapillary and extrapapillary portion. The intrapapillary portion had the ultrastructural characteristics of an arterial capillary--homogeneous-appearing basement membrane without bridged fenestrations. The ascending limb in the extrapapillary portion was also an arterial capillary and the descending limb in the same portion had venous characteristics--multilayered basement membrane. The intrapapillary arterial loop developed venous characteristics abruptly after the vessel left the dermal papilla proper. The capillary loops in acute nummular eczema were identical in ultrastructure to those of normal skin. Bridged fenestrations were found in one capillary loop in eczema but not in normal skin. By contrast, the intrapapillary capillary loops in psoriasis were venous capillaries which were characterized by bridge fenestrations and multilayered basement membrane. The ultrastructural features of the capillary loops and those of the microcirculatory segments in the horizontal dermal plexus should provide sufficient criteria for one to evaluate the nature of vascular abnormalities that occur in the upper dermis.

In a previous paper we showed that the microvessels in a psoriatic plaque as studied by electron microscopy returned to normal before the labeling index of the basal cells did during successful therapy with PUVA or the Goeckerman treatment. In this paper we studied the same parameters in 4 additional psoriatic patients: 2 received oral methotrexate and 2 were treated with a topical steroid under plastic wrap occlusion. The labeling index of the basal cells returned to normal in 3 and near normal in 1. The histologic features of the psoriatic epidermis became normal except for mild to moderate acanthosis, but the capillary loops in the dermal papillae retained their venous capillary ultrastructure and showed no signs of reversion to a normal arterial capillary configuration. The lack of response of the dermal capillaries to the topical steroid and oral methotrexate during the initial clinical improvement raises the possibility that the clinical relapses in psoriasis which may promptly follow discontinuation of topical steroid therapy and oral methotrexate may be related to an inability of these drugs to restore the microvasculature to normal in such situations.

A computer graphics system for constructing 3-dimensional models of tissues from serial sections has been designed around a Motorola 68000 CPU. The system consists of a Dual Systems 16-bit microcomputer running the UNIX operating system. A Scion Corporation MicroAngelo color graphics system drives the color monitor. Three-dimensional imaging is enhanced by hidden line removal and area filling by color. A computer-drawn model of a cherry angioma is compared with a previously constructed physical model of the same angioma to illustrate how the system operates.

This paper describes the ultrastructure of the microcirculatory vessels in the mid- and lower dermis and subcutaneous fat. Reconstruction of vessel walls, tracing out the courses of individual vessels, and survey examination of vessels were carried out by various combinations of routine and serial, ultrathin, and 1-micron sections. Intracellular myofilamentous bundles associated with extracellular filaments were a characteristic feature of the endothelial cells in arterioles possessing an internal elastic lamina, but were only rarely seen in endothelial cells of venules. The ultrastructural features of these bundles and filaments suggested both contractile and anchoring functions. The elastic lamina of the arterioles was shown to be composed of individual elastic fibers oriented in the long axis of the blood vessel under the endothelium, rather than being a continuous sheet as in larger arteries and arterioles. The capillaries in the fat had walls of normal thickness (0.1-0.3 micron) similar to capillaries in other organs, in contrast to the dermal capillaries whose walls are 2-3 micron thick. The arterioles, capillaries, and venules in the fat were frequently devoid of veil cells in contrast to those in the dermis. Venous capillaries with bridged fenestrations were found in close proximity to eccrine sweat glands and hair bulbs in the forearm, trunk, and buttock skin.

We studied by light and electron microscopy the elastic fibers in he sun exposed and sun protected skin of normal and psoriatic individuals of different ages in order to separate the changes of actinic damage from those of chronological aging. The sun exposed skin showed 2 types of elastic fiber abnormalities-one related to actinic damage and the other to chronological aging. The sun protected buttock skin showed only the latter. From ages 30 to 70, a minority of the elastic fibers exhibited abnormalities that appeared to represent a process of fiber disintegration. After age 70, the majority of elastic fibers showed these abnormalities. These abnormalities were present without accompanying inflammatory cells. Also, there was morphological evidence of continuing synthesis of elastic fibers during the lifetime of these subjects, except that from ages 50-93, the fibers appeared to be loosely, rather than compactly, assembled. Incubation of dermal slices from buttock skin of young adults with porcine pancreatic elastase and bovine chymotrypsin produced elastic fiber degradation that closely simulated the changes that were observed in aged sun protected skin. We propose that one of the features of cutaneous aging is a slow, spontaneous, progressive degradative process inherent in the elastic fiber that can be enzymatically accelerated from decades to hours by elastase and chymotrypsin.

Electron microscopy was used to define the ultrastructure of the various segments of the human cutaneous microcirculation in normal forearem skin. The organization of the vessels in the horizontal plexus of the papillary dermis was reconstructed from 1-mum plasticembedded sections. Ultrathin sections were taken at 10- to 20-mum intervals over a distance of 450 mum. Arterioles were followed through the capillary bed to their venous connections. Terminal arterioles, arterial and venous capillaries, and postcapillary venules were identified on the basis of size, cellular composition of their walls, and their relationship to the other segments of the microvascular bed. The arterial segments were characterized by a homogeneous basement membrane and the venous segments by a multilaminated basement membrane. The elastic lamina in the arterioles was a discontinuous layer which gradually disappeared from the arteriolar wall to form an external sheath just before the arteriole connected with the arterial capillary segment. The vascular walls varied from 1 to 5 mum in all of the segments of the microvascular bed. Criteria are proposed for identifying the various segments of the microcirculation so that their roles in dermatoses and vascular malformations can be evaluated.

Valve-containing collecting veins 70-120 micron in diameter were found at the interface of the dermis and fat. Two cusped valves with sinuses were found at most points where 25- to 50-micron venules entered these larger veins as well as in the lumens of the larger veins unrelated to branch points. The walls of these valve-containing veins in their narrowest portions were composed of smooth muscle cells, collagen fibers, and homogeneous-appearing basement membrane material. As the vessel became wider, elastic fibers first appeared just inside the most peripheral layer of smooth muscle cells, and gradually occupied the spaces between all the layers of smooth muscle cells. However the subendothelial zone was frequently devoid of elastic fibers. The valves were always pointed in the direction of the larger vessel as would be expected in a collecting vein. Recent experiments by other workers have demonstrated that the capillary blood flow in human nail-fold capillaries is pulsatile, thereby providing physiologic reasons for the presence of valve-containing veins at the dermis-fat interface.

Electron microscopy was used to define the ultrastructure of the capillary loops in the dermal papillae of psoriatic lesions. Pustular psoriasis of von Zumbusch and psoriasis vulgaris were studied before and after treatment with the Goeckerman regimen. Capillary loops were reconstructed from 1-mum plastic-embedded sections. Ultrathin sections were taken at intervals for correlation with the 1-mum sections. There were no ultrastructural differences between the capillary loops in psoriasis vulgaris and pustular psoriasis. The intrapapillary portion of the loop was predominantly a venous capillary. Four basic ultrastructural loop patterns were recognized which can serve as markers in studying the responses of psoriatic loops in various experimental situations. Following 3 weeks of Goeckerman therapy, the morphology of psoriatic capillary loops changed from venous capillaries to arterial capillaries which are found in the papillae of normal skin. This transformation was observed to begin 48 to 72 hr after the initiation of therapy.

We examined by electron microscopy the calf and buttock skin of 15 juvenile diabetics, 9-36 years old, to determine whether microangiopathy was present. We found in 14 of 15 buttock specimens, and in all 15 calf skin specimens, that the vascular walls were thickened by the peripheral deposition of a belt of basement membrane-like material admixed with variable amounts of collagen (reticulin) fibrils. In addition there appeared to be an increased amount of unidentified material, most likely basement membrane, deposited within the vascular wall itself. We also found age-related elastic fiber abnormalities in the dermis and in the arterioles which were identical to those observed in normal, nondiabetic aged individuals 50-93 years old. We suggest that the vascular and elastic fiber abnormalities in diabetic skin may represent a manifestation of accelerated aging rather than being related to the various underlying metabolic derangements in diabetes.

We studied by light and electron microscopy the microcirculatory vessels in the sun exposed and sun protected skin of normal and psoriatic individuals in order to separate the features of actinic damage from those of chronological aging. In actinically damaged skin, the vascular walls of postcapillary venules and of arterial and venous capillaries were thickened by the peripheral addition of a layer of basement membrane-like material. The veil cells which were intimately related to these layers often had dilated cisternae of rough endoplasmic reticulum containing electron dense material. In 3 of 8 individuals, 70, 70 and 72 yr old, the buttock skin showed mold vascular wall thickening. In 5 other patients, 59-88 yr old the vessels of the buttock skin were normal. In 4 individuals 80-93 yr old, the vessels were abnormally thin (0.5-1.0 micrometer). The veil cells were either absent or decreased in number in these specimens. We propose that (1) the veil cell is responsible for the synthesis and maintenance of the peripheral portion of the vascular wall of the dermal microcirculatory vessels; (2) the veil cell is stimulated to produce excessive basement membrane-like material in response to UV light, factors associated with diabetes mellitus, and possibly to factors associated with the early phase of chronological aging; and (3) with progressive aging there is a decrease in the number and synthetic activity of veil cells which correlates with the appearance of abnormally thin walled vessels.

Endothelial cell gaps in psoriatic vessels and histamine-induced gaps in forearm skin of normal controls were reconstructed in 3 dimensions by a computer graphics system. The gaps in psoriatic vessels were present within the cell, at the intercellular junction, or concurrently at both sites. Histamine-induced gaps were found at the intercellular junction or at both intracellular and intercellular locations. The gaps were linear to oval and often contained cytoplasmic processes from one of the endothelial cells, suggesting that gap formation represents a cellular injury rather than a purely physiologic reversible phenomenon.

To provide a morphologic basis for a better understanding of the "aging" of human skin, the authors studied the three-dimensional arrangement and chronological alterations of the fibrous components of the connective tissue using rats aged 2 weeks to 24 months with a new technique for scanning electron microscopy. These studies showed that with postnatal growth there was a dynamic rearrangement of the collagen and elastic fibers: an ordered arrangement of mature collagen bundles was attained by producing a distortion of the elastic fiber meshwork of relatively straight fibers. During adulthood, there was a subsequent tortuosity of the distorted elastic fibers coupled with an incomplete rebuilding of the elastic fiber network, laid down in a form to interlock with the collagen bundles. These changes provide a model for explaining manifestations of aged skin, such as laxity, sagging, and wrinkling. The tortuously fixed elastic fibers imply that they have been stretched and have lost their original elasticity and ability to restitute short and straight. Interlocking of both collagen and elastic fibers should disturb the two independent fibrous systems, as would normally be the case, and thus decrease tissue compliance.

Controversy exists whether the initiating events in psoriasis are primarily epidermal or dermal (vascular). To study this point, serial biopsies from 6 patients were taken from the periphery of individual plaques before, and at 1 to 3 day intervals during, Goeckerman and PUVA treatments. Part of the biopsy was studied by electron microscopy to determine the fine structure of the capillary loops and part was incubated with tritiated thymidine to determine the labeling index (LI) of the basal cells. Normal appearing buttock skin of the 11 other psoriatic patients not under treatment was studied by identical methods. In 4 of the 6 treated patients, the capillary loops began to return toward normal 3 to 8 days before the LI began to decrease. Two patients did not show a return toward normal of either capillaries or LI during the period of the experiment. The LI was elevated in the normal appearing buttock skin of 6 of 11 untreated psoriatics. In 4 of the 6, the loops were normal arterial capillaries. We did not observe abnormal (venous) capillaries associated with a normal LI in the other 5 untreated patients. These data support the concept that the initiating factors in psoriasis are in the epidermis, but epidermal hyperplasia cannot occur without vascular proliferation. Understanding the factors responsible for shortening the capillary loops during epidermal normalization and for inhibition of capillary growth in the presence of an increased LI could lead to other ways of controlling psoriasis.

The cutaneous microvasculature is organized into upper and lower horizontal plexuses with the dermal capillary loops arising from the upper plexus. The arteriolar and venular sides of the microvasculature can be identified by the ultrastructure of the mural basement membrane material. Collecting venules present in the lower dermis contain valves. Periadventitial cells (veil cells) are present around all microvessels. Their size and number appear to correlate with the quantity of mural basement membrane material found in cutaneous vessels in diabetes, actinic damage, and chronological aging. The contractile cells of the vascular wall surround the endothelial cell tube in a manner suggesting specific functions. The smooth muscle cells in the arteriolar segment form a sleeve, whereas each pericyte in the postcapillary venular simultaneously makes many contacts with several underlying endothelial cells. The common telangiectases can be explained by abnormalities in this organization and ultrastructure rather than by neovascularization or random anastomoses. The macular telangiectases seen in scleroderma, generalized essential telangiectasia, and nevus flammeus are produced by dilatation of the postcapillary venules of the upper horizontal plexus. Cherry angiomas are produced by spherical and tubular dilatations of capillary loops in dermal papillae with tortuous cross-connections between individual loops. Angiokeratomas of Fabry and Fordyce have the ultrastructure of collecting venules that contain valves, and appear to represent the ectopic development or placement of small valve-containing collecting veins. The cutaneous lesions of hereditary hemorrhagic telangiectasia represent arteriovenous communications.

The veil cells around normal, diabetic, and aged vessels were reconstructed in 3 dimensions by a computer graphics system from 120-140 serial ultrathin sections. The normal vessel was surrounded by a single layer of veil cells which had a wrinkled and pleated surface. The diabetic vessels were surrounded by 3-6 layers of cellular material produced by increased numbers of veil cells and their associated cytoplasmic sheets. The veil cells around aged vessels appeared to have the same length as young and diabetic veil cells but were underdeveloped in their lateral extensions so that they did not cover the vessel circumferentially as well as did the normal veil cells. Preliminary data suggest that young, diabetic, and aged veil cells have the same metabolic activity per unit area of cytoplasm and are of the same length. The abnormal thickness or thinness of the vascular wall of dermal microvessels appears to be related to the degree of development and numbers of veil cells around the vessels rather than any change in their basic metabolic activity from normal.