In Brief

Article Outline

• Copyright

Ever since the landmark clinical observations of Trousseau linking venous thrombosis and malignancy, this association has fascinated generations of physicians and other scientists. This monograph has been created to explore many facets of thrombosis and its relationship to malignancy. Venous thromboembolism is an important medical problem that has generally been underappreciated due to the occult and insidious nature of clinical thrombosis. Patients developing pulmonary emboli often present with sudden death (34%), which makes successful treatment impossible. This problem is uncommon and often masquerades as other conditions such as cardiac events. Furthermore, most fatalities are not subjected to autopsy so the real incidence of this problem may not be apparent. Preventing this disease involves administration of anticoagulants to patients, which can cause bleeding in the surgical patient. Needless to say, this strategy is unpopular among surgeons because often the thrombosis is occult.

Thrombosis prophylaxis has been shown to be very effective in several randomized prospective studies conducted over the past 3 decades. Based on more than 900 articles in the medical literature, consensus guidelines have been developed and revised every 2 years for the past 2 decades. These guidelines are underused for a variety of reasons, including the fact that most thrombosis occurs following discharge and these episodes may be treated by another physician. Large referral centers, for example, are generally unaware of the fate of patients when they return to their local communities. What is particularly troublesome is the fact that two thirds of all fatalities from pulmonary emboli are seen in medical patients. Surgeons in general have more organized thrombosis prophylaxis protocols, although they tend to use nonanticoagulant modalities, which, in some cases, are ineffective based on the level of patient risk.

One of the authors has worked on a risk assessment schema for the past 25 years to provide appropriate thrombosis prophylaxis based on the individual patient’s risk. This risk assessment model is based on retrospective incidence data from clinical trials. It represents in some people’s minds merely a thorough history and physical. Put another way, it parallels the checklist used by pilots before an aircraft is to fly. The risk factors are also weighted depending on the level of risk. For example, a hernia operation in a 35-year-old patient without other risk factors is not the same as a recurrent hernia operation in an elderly obese patient with a past history of cancer or deep vein thrombosis (DVT). Using this system for providing anticoagulants for the surgical patient, the level of risk for developing serious or fatal thrombosis is much greater than is the risk of bleeding. Surgeons must realize that sudden fatalities from thrombosis can occur without prophylaxis and this denies the physician any opportunity to treat the problem. If anticoagulants are used, on the other hand, sudden fatalities from bleeding never occur and, although the bleeding can be a serious problem, there is the opportunity to treat the patient. Individuals with high risk scores or limited mobility may need extended out-of-hospital prophylaxis, which may involve expensive injections. Using a risk model to identify these individuals may help proper selection of those most likely to benefit from the extended treatment. Additionally, national organizations and accreditation agencies are embracing the concept of risk assessment linked to appropriate evidence-based prophylaxis.

Patients with malignancy are at increased risk for developing a thrombosis and patients presenting with thrombosis may harbor a hidden malignancy. Malignancy is a strong risk factor for thrombosis and cancer patients need a greater degree of prophylaxis than do those with benign disease. These individuals have an inherent risk of developing thrombosis because of the hypercoagulable state associated with malignancy. They are also at greater thrombotic risk because of chemotherapy, radiation treatment, central catheters, and prolonged mobility. Thrombosis is considered to be the second most common cause of death in hospitalized patients. One fifth of all thrombotic episodes are thought to occur in those with active cancer. In 1 study of patients with a first episode of thrombosis, malignancy was associated with a 7-fold increase in risk compared with no malignancy. In another study that examined discharge codes, the incidence of DVT was 2% in those with malignancy whereas only a 1% incidence was noted in those with no malignancy. Thrombotic incidence varies with type of cancer, the highest being in those with malignant brain tumors, hematologic malignancies, and several visceral adenocarcinomas. Patients with cancer are also more likely to suffer recurrent thrombotic episodes than are those with benign disease. As many as 34% of patients who suffer a thrombosis have been found to present with a malignancy within 5 years. Unfortunately, in many of these cases, the cancer has already spread before it becomes clinically evident. This was observed in 1 study where 40% of those presenting with a DVT had a metastatic tumor appear within 1 year. Life expectancy is decreased in patients with thrombosis and malignancy, because this combination is associated with a 3-fold increase in mortality versus those with thrombosis and benign disease.

The presence of central venous catheters was found in 1 study to be the highest predictive factor for those with upper extremity DVT. Catheter-related thrombosis is a serious problem that is subclinical many times until serious embolization occurs. When it is discovered, there is always a dilemma regarding removing the catheter and losing access, or just treating the thrombosis. Chemotherapy is another risk factor for thrombosis and the risk is magnified when using certain additional drug combinations such as steroids or erythropoietic drugs.

The importance of risk assessment in cancer patients cannot be overemphasized, including continuing prophylaxis for as long as the patient is at risk. This may entail long-term administration in selected individuals. Several risk assessment models are available and can be used to guide thrombosis prophylaxis regimens. It is important to use anticoagulant prophylaxis and the administration of 1 of several low molecular weight heparins (LMWHs) following discharge has been shown to lower the incidence of thrombosis in cancer patients following abdominal surgical operations. A 50% reduction in recurrent thrombosis in cancer patients at 1 year was observed when treatment with LMWH for a 6-month period was pursued instead of using only a short period of LMWH treatment and 6 months of oral anticoagulation. Consensus guidelines currently recommend against using oral anticoagulants for the first 6 months for treatment of thrombosis in cancer patients.

Even more exciting are the apparent antineoplastic effects of anticoagulants and inhibitors of angiogenesis. Several studies have shown an improved survival in cancer patients treated with LMWH compared with other anticoagulants or placebo in patients without thrombosis. The combination of prospects for increased survival and reduced chance of recurrent thrombosis have caused some to recommend very long-term LMWH prophylaxis in those with cancer, especially when the disease is still active. Further studies are under way to confirm and refine these observations, and for most clinicians, more data will be necessary to establish long-term prophylaxis as a widespread approach.

Another exciting concept regarding the pathophysiology of thrombosis is the relationship between inflammation and thrombosis. The contributory roles of microparticles (MPs), P-selectin, and tissue factor (TF) to vascular inflammation and thrombosis have been a topic of research for 1 of the authors for at least 20 years. This story begins with Virchow, who indicated that cancers tended to occur at sites of chronic inflammation. It has been observed that the development of cancers from inflammation may be driven by inflammatory cells as well as mediators, including cytokines, chemokines, and enzymes, which collectively establish an inflammatory microenvironment. Inflammation involving lymphocytes, plasma cells, macrophages, and other inflammatory cells generate a great amount of growth factors, cytokines, reactive oxygen, and nitrogen species. These may not only cause DNA damage that leads to tumor growth but they also promote vascular inflammation that sets the stage for thrombosis during the natural history of tumor progression. The selectins represent transmembrane molecules expressed on the surface of leukocytes and endothelial cells. They are important in leukocyte adhesion at the site of inflammation and injury. Microparticles, thought of as cell dust at 1 time because they are shed from a variety of circulating cells, are a storage pool of biologically important effectors. High levels of endothelial MPs have been reported in hypercoagulable cancer states. MPs have also been seen incorporated into the growing thrombus in association with TF, which is known to be the prime cellular initiator of coagulation.

P-Selectin is stored in the platelets and endothelial cells and, when released, facilitates adhesion of leukocytes to activated platelets, a central event in thrombus formation. P-Selectin inhibition has been found to be as effective as LMWH in promoting thrombus resolution and in preventing reocclusion, all without the risk of bleeding. P-Selectin inhibition also appears to promote fibrinolysis and to decrease vein wall fibrosis. These observations are based on work conducted in primates involving major experimental venous thrombosis. We are all eagerly awaiting clinical trials to see if these experimental findings can be duplicated in humans. If this were to be true, one would have a way to prevent thrombosis without bleeding complications. Furthermore, this type of drug may also inhibit the progression of cancer and improve survival without the complications associated with the current anticoagulants.