Thromboembolic Disease

The major advance seen in thromboembolic disease in the past year was the use of low-molecular-weight heparin as at-home treatment of uncomplicated deep venous thromboses. In addition, new, previously unappreciated risks for the development of thromboembolic disease were elucidated. These risks include increased levels of homocysteine; previous spontaneous thromboembolism; and, to a much lesser degree, the use of estrogen replacement therapy.

Low-Molecular-Weight Heparin Was Effective for Home Use

Levine M, Gent M, Hirsh J, et al. A comparison of low-molecular-weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep-vein thrombosis. N Engl J Med. 1996; 334:677-81.

Proximal deep venous thrombosis is a common and potentially dangerous clinical problem. The actual incidence of the condition is unknown because many affected patients do not seek medical help. However, the most serious consequence of deep venous thrombosis, pulmonary embolism, causes at least 250 000 hospitalizations and 50 000 deaths per year in the United States.

Hospitalization to initiate treatment with unfractionated sodium heparin, followed by longer-term anticoagulation with warfarin, has been the traditional approach to treating deep venous thrombosis. Optimal duration of heparin therapy is unknown, but clinicians use the drug until warfarin becomes effective, typically for 5 to 6 days.

The use of unfractionated heparin is fraught with problems, however. Because the anticoagulant effects of this formulation are not predictable, blood coagulation must be closely monitored. The drug is typically delivered by continuous intravenous infusion, which also requires close monitoring. In addition, unfractionated heparin can precipitate thrombocytopenia.

Low-molecular-weight heparin is made by chemical or enzymatic hydrolysis of unfractionated heparin. Unlike standard heparin, low-molecular-weight heparin can be given by periodic subcutaneous injection because it has a longer half-life, has more predictable anticoagulant effects, and causes fewer problems with platelet function and counts. As a result, many small studies have suggested that low-molecular-weight heparin, possibly given at home rather than in the hospital, could replace heparin as the first-line treatment of deep venous thrombosis.

Levine and colleagues compared the use of low-molecular-weight heparin administered subcutaneously at home with intravenous standard heparin given in the hospital for treatment of acute proximal deep venous thrombosis. They recruited 500 consecutive patients with this diagnosis. Excluded patients had had two or more previous episodes of deep venous thrombosis or had a history of a bleeding disorder, a history of pulmonary embolism, or a known hypercoagulable state. Cancer was not a specific exclusion criterion unless it made home care difficult. The intervention group (n = 247) received subcutaneous enoxaparin, 1 mg/kg of body weight twice daily. These patients were treated at home or, if hospitalized, were discharged early. The control group (n = 253) was hospitalized and received intravenous standard heparin (bolus dose of 5000 U followed by continuous infusion of 1280 U/h). All patients received warfarin on the second day. The primary outcome events were recurrent venous thromboembolism within 3 months of the original episode and bleeding.

Thromboembolism recurred in 5.3% of the low-molecular-weight heparin group and in 6.7% of the standard heparin group (absolute difference, 1.4 percentage points [95% CI, –3.0 to 5.7 percentage points]. Five patients receiving low-molecular-weight heparin and three patients receiving standard heparin had major bleeding. Mean duration of hospitalization was 1.1 days for the low-molecular-weight heparin group (49% of these patients were not hospitalized) and 6.5 days for the standard heparin group.

Low-molecular-weight heparin administered subcutaneously at home was as safe and effective as standard heparin administered intravenously in the hospital. The use of low-molecular-weight heparin can markedly reduce length of hospitalization, thereby increasing patient convenience and reducing health care costs. This study, along with another report published in the same issue of The New England Journal of Medicine [1], strongly suggests that outpatient treatment of proximal deep venous thrombosis with low-molecular-weight heparin is safe and effective. In addition, recent studies have shown that low-molecular-weight heparin used to treat pulmonary embolism has similar effectiveness [2, 3]. Low-molecular-weight heparin costs four to five times as much as standard heparin, but a cost analysis of treatment for deep venous thrombosis showed that the use of low-molecular-weight heparin would save about $4000 per patient treated [4].

Home treatment of thromboembolic disease is now a reasonable practice, but patients receiving this treatment should have relatively few complicating factors, be cooperative, and have ready access to home care delivery systems. Such an approach is being rapidly adopted across the United States, especially among managed care programs in the western part of the country.

Hyperhomocysteinemia Was Associated with Deep Venous Thrombosis

den Heijer M, Koster T, Blom HJ, et al. Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med. 1996; 334:759-62.

Numerous inherited and acquired factors predispose some patients to thromboembolic disease (Table 1) [5, 6]. Cancer is believed to be the most frequent risk factor. Most patients with deep venous thrombosis do not have cancer, however, and experts recommend that physicians do not search for cancer when it is not clinically obvious. Another recently discovered and common risk factor is the factor V Leiden mutation. This single-point mutation in the gene coding for coagulation factor V results in a form of factor Va that resists degradation by activated protein C and leads to a relative hypercoagulable state. The mutation is found in about 5% of persons in the United States [7].

Another possible risk factor for thromboembolic disease is mild hyperhomocysteinemia, a known risk factor for the development of atherosclerosis and vascular disease. It has long been known that classic homocysteinuria is associated with deep venous thrombosis.

In a case–control study, den Heijer and colleagues sought to determine whether high plasma homocysteine levels are a risk factor for deep venous thrombosis in the general population. They recruited 269 patients with a first episode of deep venous thrombosis and 269 healthy controls. Total plasma homocysteine levels were measured, and hyperhomocysteinemia was defined as a homocysteine level greater than the 95th percentile in the control group (18.5 µmol/L). Other independent risk factors for thrombosis, including known hypercoagulable states, use of oral contraceptives, and pregnancy, were assessed.

Of the 269 patients with deep venous thrombosis, 10% had plasma homocysteine levels greater than the 95th percentile of the control group (odds ratio, 2.5 [95% CI, 1.2 to 5.2]). The association between homocysteinemia and venous thrombosis was stronger among women and older patients. The association was highest in patients with a homocysteine level of at least 22 µmol/L. The exclusion of patients who had hypercoagulable states, were pregnant, or used oral contraceptives did not affect risk estimates.

High plasma homocysteine levels are a risk factor for deep venous thrombosis. The strength of this association is similar to that reported for hyperhomocysteinemia and arterial vascular disease. Assays for homocysteine should be part of the laboratory evaluation when high-risk patients are tested for a hypercoagulable state (Table 2). Although this has not yet been demonstrated by controlled clinical trials, treatment of hyperhomocysteinemia with folic acid or other vitamins may reduce the risk for subsequent vascular disease, both venous and arterial.

Why hyperhomocysteinemia is associated with vascular disease is unclear. However, homocysteine added to cultured endothelial cells makes the cells prothrombotic. Hyperhomocysteinemia can be hereditary or acquired. Hereditary causes are linked to defects in either the enzyme cystathionine ß-synthase or the enzyme methylenetetrahydrofolate reductase, commonly called MTHFR. In the U.S. population, 50% of persons are heterozygous for the thermolabile form of MTHFR, 10% are homozygous, and 40% are "normal." Homozygotes are more sensitive to folic acid deficiency and, with a healthy diet, usually have normal homocysteine levels. If they become folate deficient, however, homocysteine levels can increase. Acquired causes of hyperhomocysteinemia include folic acid deficiency and vitamin B₁₂ deficiency. To determine the presence of vitamin B₁₂ deficiency, the methylmalonic acid level should be measured; this test is more sensitive and specific than measurement of the vitamin B₁₂ level. Treatment for folic acid deficiency consists of 5 mg of folic acid per day for 2 weeks, followed by 2 mg/d indefinitely. The cost of 1000 one-mg tablets of folic acid is about $12. Physicians should remeasure the homocystine level about 8 weeks after treatment begins. If the level has not decreased to normal, B complex vitamins may be added. Because folate therapy can mask the hematologic manifestation of pernicious anemia, a yearly assay of methylmalonic acid might be prudent.

It is important to note, however, that no studies have yet shown that treating hyperhomocysteinemia actually reduces the risk for complications of vascular disease.

Absolute Risk for Deep Venous Thrombosis Associated with Estrogen Was Small

Daly E, Vessey MP, Hawkins MM, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet. 1996; 348:977-80.

The list of benefits of postmenopausal hormone replacement therapy-such as protection against osteoporosis, cardiovascular disease, and possibly dementia of the Alzheimer type-is growing [8, 9]. However, an association between use of oral contraceptives that contain estrogen and deep venous thrombosis has been well demonstrated [10]. In a hospital-based case–control study, Daly and colleagues investigated whether current use of post-menopausal hormone replacement therapy is associated with venous thromboembolism.

All 45- to 64-year-old women who were admitted to hospitals in Oxford, England, were screened for a suspected diagnosis of venous thromboembolism during a 1-year period. Idiopathic venous thromboembolism was found in 101 patients, and these patients were compared with 178 hospital controls. Information on medical and gynecologic history, use of oral contraceptives, use of hormone replacement therapy, use of other drugs, lifestyle, and socioeconomic characteristics was obtained. Women were classified as current hormone replacement therapy users if they had received such treatment in the month before admission.

Among patients with venous thromboembolism, 42.7% were current users of hormone replacement therapy compared with 24.7% of controls (adjusted odds ratio, 3.5 [CI, 1.8 to 7.0]). No association was found with past use of hormone replacement therapy. The risk seemed to be highest among current users who had recently begun receiving therapy (P = 0.001). Higher mean body mass index, varicose veins, and superficial thrombophlebitis were more common in patients with thromboembolism than in controls. Fewer patients with thromboembolism were from higher socioeconomic groups.

This report, along with three others published in the same issue of The Lancet [11-13], suggests that women receiving hormone replacement therapy have an increased risk for venous thromboembolism. However, the absolute risk remains small: Only one excess case of thrombosis will be observed in 5000 women receiving hormone replacement therapy for 1 year. This small increased risk for thromboembolism is outweighed by the benefits of treatment. However, it remains unknown whether hormone replacement therapy confers higher risks in women with hypercoagulable states or a history of venous thromboembolism.

Risk for Recurrent Spontaneous Deep Venous Thrombosis Was High

Prandoni P, Lensing AW, Cogo A, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996; 125:1-7.

Deep venous thrombosis can result in pulmonary embolism, venous ulceration, chronic leg pain, and intractable edema. In general, most physicians treat deep venous thrombosis with warfarin therapy for 3 to 6 months; this therapy reduces the short-term complication rate to about 5%. What remains unknown is the long-term clinical course of patients treated for deep venous thrombosis.

Prandoni and colleagues conducted a cohort study to track the clinical course of deep venous thrombosis in patients with a mean age of 63 years. They recruited 355 consecutive patients with a first episode of symptomatic deep venous thrombosis. Patients were admitted to the hospital and were treated with intravenous standard heparin or subcutaneous low-molecular-weight heparin. Warfarin therapy was started on days 5 to 7 and was continued for 3 months to maintain an international normalized ratio of 2.0 to 3.0. All patients also wore elastic graduated compression stockings. Patients were seen at 3 and 6 months and then every 6 months for as long as 8 years.

The cumulative incidence of recurrent venous thromboembolism was 17.5% after 2 years, 24.6% after 5 years, and 30.3% after 8 years. The presence of cancer and hypercoagulable states increased the risk for recurrent venous thromboembolism (hazard ratios of 1.72 and 1.44, respectively). Patients whose deep venous thrombosis was associated with surgery or recent trauma had a decreased risk for recurrent thromboembolism (hazard ratios of 0.36 and 0.51, respectively). The cumulative incidence of the post-thrombotic syndrome (criteria scores were based on pain, cramps, heaviness, pruritus, paresthesia, pretibial edema, skin induration, hyperpigmentation, new venous ectasia, redness, calf tenderness, and ulceration) was 22.8% after 2 years, 28% after 5 years, and 29.1% after 8 years. The development of ipsilateral recurrent deep venous thrombosis was strongly associated with the post-thrombotic syndrome (hazard ratio, 6.4). One of every five recurrent thromboembolic episodes was pulmonary embolism; this event was fatal in half of the affected patients. Forty percent of the recurrent thrombosis episodes occurred in the previously asymptomatic leg.

This study indicates that the risk for recurrent venous thromboembolism is high, particularly after a spontaneous or unprovoked episode of deep venous thrombosis. The high incidence of recurrent venous thromboembolism after cessation of anticoagulant therapy suggests that prolongation of antithrombotic treatment should be considered in patients who are at relatively high risk. Development of the post-thrombotic syndrome should also prompt the clinician to consider long-term anticoagulant therapy.

A recently reported clinical trial [14] underscores these findings. Patients with a second episode of deep venous thrombosis were randomly assigned to receive warfarin for 6 months or indefinitely. Of patients treated for 6 months, 20.7% had recurrent thromboembolic events and 2.7% had major bleeding complications. In those given indefinite therapy, the rate of thromboembolism recurrence was 2.6% and the bleeding rate was 8.6%. Mortality rates did not differ between the two groups. About six patients would need to be treated indefinitely to prevent one episode of recurrent thromboembolism, and 17 patients would have to be treated to cause one major bleeding episode.

In summary, the data suggest that if a patient has a reversible risk factor, such as hip fracture, only 6 to 12 weeks of therapy is needed. The recurrence rate in this group is low. In patients with idiopathic or spontaneous thrombosis, therapy should last for at least 6 months, perhaps longer. For patients with a second spontaneous thromboembolic event, long-term warfarin therapy is warranted. Warfarin should also be given indefinitely to many patients with thrombosis and a known hypercoagulable state.

Finally, the question remains as to who should be assessed for hypercoagulability. Reasons for such study include the following: assessment of the need for anticoagulation, treatment of specific disorders, family studies, surgical prophylaxis, and pregnancy or estrogen therapy. The cost of a workup for hypercoagulability is estimated to be about $600.

Diagnosis and Treatment of Anemia

Two common types of anemia received attention in 1996. Pernicious anemia was found to be more common than generally believed. Because modern dietary supplementation may mask the anemic effects of this condition, neurologic disease may be the first manifestation. Iron deficiency was found to be highly associated with poor scholastic achievement in urban young women; this may be a factor in the slow advancement of children living in poverty.

Pernicious Anemia Was Common in Elderly Women

Carmel R. Prevalence of undiagnosed pernicious anemia in the elderly. Arch Intern Med. 1996; 156:1097-100.

Pernicious anemia is the most common cause of cobalamin (vitamin B₁₂) deficiency. It has historically been noted in up to 1% of elderly persons living in northern Europe, but it is increasingly recognized in young black and Hispanic women. The true prevalence in the United States has been unknown. The illness is invariably associated with atrophic gastritis, which in turn is related to an increased risk for gastric carcinoma. If left untreated, pernicious anemia can lead to complex neurologic symptoms, including peripheral neuropathies, paraesthesias, balance and gait disturbances, and such neuropsychiatric disorders as dementia. These neurologic deficits can be permanent if left untreated for more than 6 months. In addition, patients with pernicious anemia are prone to develop other autoimmune diseases, including IgA deficiency, and polyglandular endocrine insufficiency.

Carmel estimated the prevalence of undiagnosed and untreated pernicious anemia in elderly persons. He recruited 729 persons 60 years of age or older by random solicitation at two adult education centers, a seniors social club, an apartment complex for free-living elderly persons, and a Department of Veterans Affairs outpatient clinic. The group comprised 41% white persons, 25% African-American persons, 20% Hispanic persons, and 14% Asian-American persons. Fifty-nine percent of participants were men. About 35% of participants were 60 to 69 years of age; 17% were older than 80 years of age. Blood samples for measurement of cobalamin and anti-intrinsic factor antibody levels were obtained from each participant. All persons found to have a low serum cobalamin level or to be positive for anti-intrinsic factor antibody were invited for further studies, including a complete blood count, measurement of fasting gastrin levels, and a Schilling test.

Table 3 lists the study findings. Seventeen participants had pernicious anemia. Three of these had previously received a diagnosis but had been treated inadequately. Eight of the 17 participants were shown to have pernicious anemia by an abnormal result on the Schilling test, and 9 (who declined follow-up testing) were shown to have pernicious anemia on the basis of low cobalamin levels (or abnormal methylmalonic acid or homocysteine levels) combined with positivity for anti-intrinsic factor antibody. Eight of the 9 also had elevated gastrin levels. Most of the 17 patients had only minimal clinical manifestations of pernicious anemia.

Undiagnosed subclinical pernicious anemia is common in elderly persons, especially elderly white and black women. Extrapolation of these results indicates that about 800 000 elderly persons in the United States have undiagnosed and untreated pernicious anemia. One reason that the high prevalence of pernicious anemia is important is because of the widespread use of folate supplementation to prevent thrombotic disorders or neural tube defects. Folate may mask the hematologic manifestations of pernicious anemia and therefore produce neurologic disease without the warning of megaloblastic anemia. The gold standard of diagnosis remains the Schilling test. However, the test for intrinsic factor has a sensitivity of about 70% and a specificity of 99%. Thus, if the test result is positive in a typical older white or black woman, the patient has an almost 100% chance of having the disease. The wholesale cost of 1000 µg of vitamin B₁₂ is about $0.50. However, the injection costs up to $20; this charge, coupled with the expense of the necessary office visits, would represent significant cost and inconvenience if 800 000 persons were to receive monthly injections. If the indirect association of pernicious anemia with venous thromboembolic disease (through hyperhomocysteinemia) is shown to be true, diagnosis and treatment may prove to be well worth the costs.

Iron Deficiency Was Associated with Low Scholastic Achievement

Bruner AB, Joffe A, Duggan AK, et al. Randomised study of cognitive effects of iron supplementation in non-anaemic iron-deficient adolescent girls. Lancet. 1996; 348:992-6.

Approximately 20% of the world's population has iron deficiency of various causes. Worldwide, 10% to 40% of adolescent girls are iron deficient; in the United States, this Figure is about 14%. However, only about one third are anemic. Persons treated for iron deficiency anemia report improved memory, mood, attention, and energy before hemoglobin levels increase. Pica, too, disappears before hemoglobin levels begin to increase. Many theories of brain chemistry have been put forth to explain these observations; until now, however, no study had truly shown that iron supplementation is helpful in patients without anemia.

In a clinical trial, Bruner and colleagues assessed the effects of iron supplementation on cognitive function in adolescent girls with nonanemic iron deficiency. They recruited 716 girls from four Baltimore, Maryland, high schools and screened them for nonanemic iron deficiency. This condition was defined as a ferritin level less than 12 µg/L and a normal hemoglobin level. Ninety-eight girls had nonanemic iron deficiency; 81 of these were enrolled in the trial. The participants were randomly assigned to receive oral ferrous sulfate, 650 mg twice daily, or placebo for 8 weeks. Outcomes were results of four tests of attention and memory that were done before and after the intervention.

Of the 78 girls who completed the study, 5 developed anemia and were excluded from the analysis. Ethnic distribution, mean age, serum ferritin, hemoglobin level, and cognitive test scores of the two groups did not differ at baseline. Ferritin levels improved in the treatment group (27.3 µg/L compared with 12.1 µg/L in the control group). Girls who received iron performed better on the tests of verbal learning and memory than did girls in the control group (P < 0.02). The only side effect reported was darker stools.

This study suggests that measurement of ferritin levels should be considered as a screening tool in asymptomatic young women and probably in young men in underprivileged groups. A caveat to the use of measurement of ferritin levels as a screening test is that levels can be elevated, and therefore falsely reassuring, in persons with chronic inflammatory disorders of many types, in patients with abnormal liver function test results, or both. Iron supplementation should be given to persons with low ferritin levels. Whether such treatment results in long-term improvement in academic achievement remains to be shown. In addition, it would be unwise to generalize these finding to all boys or to adults because the predilection of these groups for iron deficiency is different from that of adolescent girls. Finally, when treating iron deficiency, the clinician must be careful not to overtreat and create the problems associated with iron overload.

Adjunctive Therapy in Multiple Myeloma

Pamidronate Decreased Bone Complications in Multiple Myeloma

Berenson JR, Lichtenstein A, Porter L, et al. Efficacy of pamidronate in reducing skeletal events in patients with advanced multiple myeloma. Myeloma Aredia Study Group. N Engl J Med. 1996; 334:488-93.

Multiple myeloma accounts for 1% of all malignant conditions and 10% of hematologic malignant conditions. One of its most devastating complications is skeletal disease, which occurs in 80% of patients with multiple myeloma and is associated with extreme bone pain. This bone disease is due to increased bone resorption; it has been thought that the bisphosphonate drugs, which inhibit resorption, may be effective in treating the skeletal complications of multiple myeloma.

Early trials of bisphosphonates were relatively unimpressive, but they were done with first-generation drugs: etidronate and clodronate. Pamidronate is a second-generation bisphosphonate. Berenson and colleagues tested the efficacy of pamidronate for reducing skeletal events in patients with advanced multiple myeloma.

In a randomized trial, 392 patients with stage III multiple myeloma and at least one lytic lesion received either placebo or pamidronate (90 mg) as a 4-hour intravenous infusion given every 4 weeks for nine cycles in addition to antimyeloma therapy. The patients were stratified according to whether they were receiving first-line or second-line antimyeloma chemotherapy at study entry. Outcomes were pathologic fracture, irradiation of bone or surgery on bone, spinal cord compression, hypercalcemia (defined by presence of symptoms or a serum calcium level 12 mg/dL), bone pain, use of analgesic drugs, performance status, and quality of life. These variables were assessed monthly.

The results (Table 4) are adequately robust to allow one to conclude that monthly infusions of pamidronate provide significant protection against skeletal complications and improve the quality of life of patients with stage III multiple myeloma. Subsequent studies have shown similar results in patients with breast cancer that had metastasized to bone [15, 16]. Whether the use of the oral bisphosphonate alendronate will prove equally effective remains to be determined.

Dr. Feinstein: Department of Medicine, University of Southern California School of Medicine, 1355 San Pablo, Suite 139, Los Angeles, CA 90033.

Dr. Roberts (Series Editor): Madrona Medical Group, 4370 Cordata Parkway, Bellingham, WA 98226-8075.