Purpura and Other Hematovascular Disorders

Macrovascular Disruption

Macrovascular disruption most often occurs suddenly as the result of trauma or severe atherosclerotic processes, such as a ruptured abdominal aortic aneurysm. Although management of macrovascular hemorrhage is beyond the scope of this text, reference is made here to macrovascular processes for which the hematologist may be asked to consult and assist.

Causes of macrovascular hemorrhage can be grouped into six broad categories ( Box 10.1 ). The first two, trauma and atherosclerotic processes, are not further discussed in this chapter. The four broad categories that may be called primary vascular disorders of the macrocirculation include disorders of connective tissue, infiltration of macrovascular vessels with amyloid, inflammatory processes involving the macrocirculation, and certain arteriovenous malformations. Designation of a particular disorder as microvascular or macrovascular in origin is imperfect in that some processes may fit into either niche (see Chapter 22 ).

True Disorders of Connective Tissue

Large Vessel Infiltration

Amyloidosis and its hemorrhagic manifestations are generally microvascular in nature but may involve the great vessels of the brain; thus they are discussed here. Amyloidosis is not a single disorder, but a series of disorders, all which have in common the misfolding of any number of proteins, giving rise to β pleated sheets that reside beneath the basement membrane of multiple structures, including blood vessels. All of these proteins have in common the characteristic birefringent apple green color seen on staining of affected tissue with Congo red in polarized light, as well as a typical pattern of microfibrils on electron microscopy. More than 20 types of amyloid have been described, and each abnormality has subtle variations in its clinical manifestation. Vessel wall strength may be reduced; thus simple mechanical stress or sheer force may result in bleeding, which is encountered in 28% of patients with amyloidosis. Purpura about the face ( Fig. 10.1 ) seems especially common in amyloidosis and is discussed in greater detail in the Microvascular Purpura section of this chapter. Of interest, spontaneous rupture of the spleen has been described in amyloidosis and in fact may be the presenting symptom.

A unique subtype of amyloidosis that clearly has hemorrhagic macrovascular ramifications is cerebral amyloid angiopathy (CAA), which has in the past been referred to as congophilic angiopathy. Subtypes of CAA may be sporadic or familial; all have as yet undetermined causes. In CAA, amyloid deposition of the cerebral arteries is characteristic. CAA is extremely common and may be present in nearly one-half of all elderly individuals at autopsy. CAA is one of the leading causes of spontaneous cerebral hemorrhage, accounting for approximately 10% to 15% of all such bleeds (see Chapter 39 ). CAA has medical-legal implications, in that many older patients are administered long-term anticoagulation therapy, which is often blamed for the hemorrhagic event, even though the true underlying cause is frequently CAA. A peculiar and uncertain relationship has been noted between CAA and Alzheimer disease, in that cerebral vessels that contain amyloid are seen in 80% of patients with Alzheimer disease.

Inflammatory Processes

Any one of several inflammatory processes may involve large vessels and subsequently may lead to their rupture and hemorrhage.

Infectious causes have historically included syphilitic aortitis. Although this disorder is rarely seen now, in previous eras, it was the prevalent cause of aortic aneurysm. Tuberculous arteritis can conceivably affect any artery. Mycotic aneurysms may accompany almost any infection but are classically seen in subacute bacterial endocarditis ; unfortunately, they may lead to hemorrhage, particularly in the central nervous system. Human immunodeficiency virus (HIV) infection has been implicated as a cause of inflammatory vasculitis resulting from large vessel damage.

Any of the classic inflammatory rheumatic diseases, such as rheumatoid arthritis, polyarteritis nodosa, relapsing polychondritis, Behçet syndrome, giant cell arteritis, and systemic lupus erythematosus, occasionally involve large vessels characterized by inflammation, weakening, and eventual rupture and sudden hemorrhage. Probably the best understood relation is with rheumatoid vasculitis, which, paradoxically, may occur in older patients with quiescent burned-out arthritic manifestations, yet may perforate large arteries. Kawasaki disease involves an inflammatory process of the arteries that occurs in childhood; in a small number of patients, panvasculitis occurs, with particular reference to coronary vessels, which may lead to aneurysmal dilatation of these vessels.

Arteriovenous Malformations/Hemangiomas

Recent discoveries have changed our knowledge and understanding of the pathogenesis of this anatomic disorder. Although in the past, most terms used to define these entities as separate diseases were based on epidemiology, clinical presentations, or even treatment, an explosion in our understanding of the basic science of vascular endothelial growth has already begun to change not only our understanding but even our classification of these disorders. Ultimately, no doubt, new treatments will be found as our level of knowledge increases. Accordingly, processes as diverse as Kaposi sarcoma (KS), hereditary hemorrhagic telangiectasia (HHT), and diabetic retinopathy may have more in common than we would otherwise ever have imagined.

Many of these processes have at their core abnormal and sustained proliferation of endothelial cells caused by upregulation of growth-promoting factors or inhibition of growth-promoting apoptosis. Indeed, in HHT, circulating levels of vascular endothelial cell growth factor (VEGF) are consistently higher than normal, and a tendency toward even higher levels has been noted among those patients with HHT who bleed more than other patients with HHT. It is hypothesized that most features of HHT are due to persistence of the activation phase of angiogenesis caused by perturbations in VEGF.

Our understanding of these processes has realigned our thinking regarding hemangiomas, particularly those occurring in children. The International Society for the Study of Vascular Anomalies has produced a new classification scheme that is based on whether the lesion is a tumor (i.e., due to vascular proliferation) or a malformation (i.e., due to structural abnormalities associated with slow endothelial cell turnover). This group carefully dissected what were previously (and probably erroneously) considered to be two manifestations of the same process, namely, localized disseminated intravascular coagulation (DIC) and cavernous hemangioma (previously known as the Kasabach-Merritt syndrome). Students of this disease now argue that what up to this point has been called Kasabach-Merritt syndrome is actually a localized form of coagulation activation that produces thrombocytopenia and arguably DIC; however, most often, it is confined to a unique vascular anomaly, namely, kaposiform hemangioendothelioma (KHE). Most vascular lesions, particularly the common hemangiomas of infancy, do not exhibit the Kasabach-Merritt phenomena, namely, localized DIC and DIC-like changes.

Mulliken and associates defined differences between KHE and otherwise benign infantile hemangioma. Vascular lesions, whether KHE or not, that exhibit the Kasabach-Merritt phenomenon, have recently been reviewed by Kelly. These lesions were initially described by Kasabach and Merritt in 1940 and were believed to be seen in only approximately 1% of cases of hemangioma. Neither the site nor the size of the vascular lesion predicts the syndrome. The Kasabach-Merritt phenomenon is very serious, with a mortality rate of approximately 40%. Patients can bleed into these highly vascular tumors. Not many lesions are biopsied, but the vast majority of patients who have true Kasabach-Merritt phenomena actually have KHE—not benign hemangioma. Therapy includes arterial embolization and general support of the patient with FFP or platelet infusion. Some have advocated the use of glucocorticosteroids, 2 to 3 mg/kg per day. It is hoped that breakthroughs in antiangiogenic sciences may result in therapies based on antiangiogenic strategies.

Microvascular Hemorrhage

Purpura is a general term that describes either small punctate lesions called petechiae or larger lesions called ecchymoses (see Table 10.2 for terminology). Purpura is derived from the Latin term for purple—the color generated by the extravasation of red cells into the skin. Extravasation may result from coagulation disorders, physical trauma, or systemic conditions that lead to alterations in the microvasculature. Purpura, which is due to extravasation of blood from the microcirculation, consisting of the smallest arterioles, capillaries, and postcapillary venules, is thus a disorder of the microcirculation.

The hematologist may be consulted to evaluate patients with purpura because the differential diagnosis of disorders that may result in purpura includes many hematologic processes. Because purpuric disorders require careful consideration of hematologic and dermatologic causes, the hematologist should be familiar with hematologic purpura, as well as its imitators. One may wish to consult with a dermatologist for many patients. The differential diagnosis of disorders that lead to purpuric lesions encompasses a considerable range of processes, from mild chronic dermatologic disorders to rapidly progressive, life-threatening illnesses such as meningococcemia with DIC.

HHT (also known as OWR) is not truly a purpuric disorder because blood is not extravasated. Discussion of HHT syndrome traditionally is placed among discussions of nonthrombocytopenic purpura. HHT is thoroughly discussed in Chapter 11 .

Historical Perspective

Because of their ready visibility, purpuric lesions have been described throughout history. Victims of the bubonic plague, which circled the globe in the Middle Ages, killing untold millions of people, often were purpuric (hence Black Death), which led to the rapid recognition of the affliction and banishment by others. Typhus is claimed to have killed more soldiers throughout history than all battles combined. Scurvy, particularly the prevention of scurvy, which was appreciated and practiced by the British Navy, was highly instrumental in England's defeat of Napoleon's navy at Trafalgar because the French Navy did not practice scurvy prevention. Before the adoption of antiscorbutic measures by the British Navy, 1500 patients with scurvy were admitted to the main naval hospital in England each year. After antiscorbutic policies were initiated, scurvy was essentially eliminated from the British Navy, and extant records reveal only two cases in the 5-year period following Lord Nelson's destruction of the French Navy in 1805. This story has been engagingly recounted in the medical literature.

Although purpura had been known for centuries, scientific interest in these lesions was established in the 18th century, when Werlhof, then serving as court physician to King George II of England, accurately described what is now called acute idiopathic thrombocytopenic purpura (ITP). At the very beginning of the 19th century, Willan reviewed purpura and proposed five categories in the first rudimentary attempt to explain purpura as a rightfully circumscribed area of clinical science. The five subtypes of purpura that he described consisted of the following: (1) purpura simplex, (2) purpura hemorrhagica, (3) purpura urticans, (4) purpura senilis, and (5) purpura contagiosa. More likely than not, the five subtypes of purpura that he explained 200 years ago would now be regarded as, respectively, purpura simplex, acute ITP, Henoch-Schönlein purpura (HSP), senile purpura, and meningococcemia, as well as other acute bacterial infections related to purpura. Approximately 100 years later, at the turn of the 19th century, purpura was reviewed by Austin Flint in his textbook. Scientists' understanding of purpura had not substantially advanced since the time of Willan's review. Flint clearly separated clinical purpura from hemophilia and traumatic hemorrhage. He preferred the term purpura rheumatica to purpura urticans and clearly described what we now refer to as HSP. He did not like Willan's term, purpura hemorrhagica , because he recognized that all purpura was indeed hemorrhagic. He preferred to revert to the older European name for ITP, namely, morbus maculosus Werlhofii , or Werlhof disease. Although he did not use a separate descriptive category for scurvy, he correctly described the two primary differences between ITP and scurvy—gum swelling and bleeding characteristic of advanced scurvy but not of ITP—and that ITP did not respond to dietetic manipulation, as did scurvy. William Osler in a major textbook of medicine in that era (competing with that of Flint) gave a similar account of purpura. He again clearly separated hemophilia from the purpuric disorders. Revealing his strength in observational medicine, he was the first to accurately describe petechiae, differentiating them from ecchymoses. He described cases of purpura following iodine administration, which were probably the result of iodine-induced cutaneous vasculitis (CV). He preferred the term cachectic purpura to senile purpura . Osler also clearly distinguished HHT (now one of the several diseases bearing his name, OWR syndrome) from hemophilia, while recognizing that both were hereditary hemorrhagic diatheses.

That thrombocytopenia was causally related to the petechiae of ITP was recognized by the end of the 19th century. The existence of nonthrombocytopenic purpura received a large boost in credibility when Wolbach in 1919 described that purpura in Rocky Mountain spotted fever (RMSF) was associated with the infestation of rickettsiae in the walls of microcirculatory vessels. In 1942 Wintrobe opined that although thrombocytopenia was characteristic in ITP, in many purpuric diseases, platelet counts were normal; he therefore concluded that “some obscure change in the capillary endothelium” must account for at least some cases of nonthrombocytopenic purpura. In 1948 Haden and colleagues showed that cutaneous hemorrhage could occur with a normal platelet count and a normal clotting system. They promulgated the concept of “increased capillary fragility,” a term that by virtue of its accuracy, has continued to be used up to now. In 1952, Spaet, and in 1953, Ackroyd clearly, convincingly, and permanently established vascular damage as the primary cause of nonthrombocytopenic purpura. Study of the microcirculation and endothelial biology in particular became possible with electron microscopy, endothelial cell culture, and other modern biomedical techniques.

Microvascular Structure–Function Interrelations

Study of the fine structure of the microcirculation was pioneered by Majno. The microcirculation is defined as terminal arterioles, capillaries, and postcapillary venules. Fig. 10.2 depicts a normal capillary, which will be used as the model of the basic microvascular structural unit. The primary anatomic difference between capillaries and arterioles, as well as postcapillary venules, is that both of the latter have an investiture of smooth muscle cells that control blood flow. The capillary, by anatomic definition, lacks smooth muscle investiture because its function is not to regulate traffic; rather, as the business end of the circulation, it is the site of unimpeded gas, fluid, and nutrient exchange. Accordingly, its structure is simple. The primary structural unit is a series of two to five endothelial cells that are joined by tight junctions. Endothelial cells are now known to be extremely influential, providing a great deal of secretory activity on both luminal and abluminal surfaces and assuming, under appropriate stimulation, a neutral, antithrombotic, or prothrombotic stance. The subendothelial basement membrane is immediately beneath the closed circle of endothelial cells. This ill-defined material probably affords some degree of structural integrity but chiefly is highly procoagulant in the event that blood makes contact with it by virtue of a breach in the endothelial membrane. Liberally scattered both longitudinally and circumferentially around capillaries are collagen bundles, which offer resilience against mechanical stresses. Around capillaries are pericytes, whose function is probably largely one of support. Accordingly, the four chief functioning members of the structure of the capillary circulation that keep the circulatory system closed are (1) the endothelium; (2) the subendothelial basement membrane; (3) collagen; and (4) pericytes.

The histologic hallmark of purpura is extravasation of red blood cells from the microcirculation. Fig. 10.3 shows light microscopic examination of a biopsy specimen of a single petechia in which red cells have extravasated from a nearby capillary. In Fig. 10.4 , extravasated red cells abound, with at least one phagocytized by a tissue macrophage. Iron from the heme of red blood cells remains in the skin, causing hemosiderin deposition characteristic of the extremities of patients who have had long-standing purpura.

Pathophysiologic Categories of Purpura

Box 10.2 presents a classification of purpura and hematovascular lesions.

Purpura Not Associated With Known Microvascular Pathology

Frequently, purpura is identified in which no known anatomic aberration of the microcirculation occurs. These conditions are truly purpuric in that if one chooses to biopsy such lesions, extravasated red cells are seen.

Mechanical Purpura

The mechanical strength of the capillary unit is finite; thus any pressure that exceeds that limit could well lead to extravasation of red cells. A vacuum of 200 mm Hg applied to normal skin will extravasate red cells. That the human mouth can generate this much suction is exemplified by lesions about the neck known as “hickeys.” Suction cupping can result in a circular type of purpura. Petechiae on the face and neck may result from increased venous pressure after vomiting or seizures or even from prolonged hanging upside down by one's feet to alleviate back pain. Purpura on the palms and soles of the feet may result from leisure activities such as weight-lifting or from traumatic blows from one's avocation or occupation. Formation of petechiae may be seen following choking, asphyxiation, seizure, barotrauma, and electrocution and even a few may be seen on normal infants. The chief diagnostic criterion is the history of activities prior to the appearance of purpura.

Factitious Purpura

When patients with purpura provide a history that is vague or not credible, one should suspect factitious purpura. A variety of suction devices have been applied surreptitiously to every imaginable part of the body to produce purpura for whatever gain the patient may seek. Fig. 10.5 shows a pattern of purpura that clearly suggests that the patient raked reachable parts of his body with a gardening implement. Factitious purpura is more common than one may at first appreciate but can be suspected when the patient has been seen repeatedly by multiple physicians with no hemostatic or underlying medical conditions found. This purpura tends to be well circumscribed and is found chiefly in areas that can be readily reached by the patient. The number and appearance of these purpuric lesions may come and go without an easy explanation; long “remissions” may occur.

Psychogenic Purpura

This is an unusual form of purpura that is increasingly regarded as factitious in origin. It is indisputable that patients with this type of purpura harbor major deep emotional disturbances and a great deal of unresolved emotional conflict. Lesions of this type of purpura usually begin with bruises that are heralded after a variable lead time by a feeling of warmth, stinging, or swelling. Later, an ecchymotic lesion might appear. These bruises can vary from small to rather large. They have many of the features of factitious purpura in that they are often linear, have well-demarcated edges and chiefly occur in areas that can be reached by the patient ( Fig. 10.6 ). The relationship of this disorder to autoerythrocyte sensitization as promulgated by Gardner and Diamond is uncertain.

Jafferany and Bhattacharya recently reviewed this syndrome. It is clearly linked to highly emotional and psychological stressful events that are alleged to trigger the appearance of these skin lesions. Of interest, the repetitive nature of this syndrome, absence of clear diagnostic value of traditional hemostatic tests, skin biopsies, red blood cell injection tests, and overall natural history of the syndrome caused these authors to suggest ordering fewer such tests, especially if they have already been performed. Because family history and lack of hemorrhage with surgery are typical, such studies can be minimized; they proffer that the strongest diagnostic feature of this disorder is the psychological profile.

Purpura Simplex

This vascular phenomenon is a nonpathologic normal process. Many people have small bruises that are associated with trauma of daily living. Women bruise or report bruising more frequently than men. Fig. 10.7 shows an ecchymotic area on the lateral surface of the thigh. It is striking how frequently these bruises appear approximately 30 inches above the floor—the height of most furniture, cabinets, and tables around the house or at work. Purpura simplex may result from being pinched; such lesions are also referred to as “devil's pinches” if an individual has no recall of being pinched. Because this purpura is not of pathologic origin, further evaluation is not necessary. Such patients may safely undergo surgery or invasive procedures.

Bruises and Hematomas

Bruises (including purpura simplex) are not palpable (i.e., not true hematomas) but are flat within the surface of the skin. Bruises result from trauma but of course can be exacerbated by platelet or coagulation defects to become larger bruises or even true hematomas. Simple bruises have been somewhat arbitrarily defined as smaller than 3 cm in diameter and not palpable; they usually number no more than four to six over the body. “Normal bruising” has been quantified in healthy infants, of whom 13% may have up to four bruises up to 10-mm maximum diameter. Such bruises tend to occur over bony prominences and increase in frequency as the child's mobility increases. Bruises not confined to bony prominences or in unusual places (soles or palms) may raise questions of abuse. If bruises are larger and more numerous, consideration may be given to a hemostatic defect, especially if the masses are palpable (i.e., true hematomas). Fig. 10.8 shows a large hematoma of the shin after an athletic incident that served as the diagnostic event for a teenager with heretofore undiagnosed mild hemophilia A with 7% of normal factor VIII activity. Large ecchymotic areas with hematoma formation ( Fig. 10.9 ) provide the typical presentation of factor VIII inhibitors, as discussed in Chapter 5 .

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here