Portal de Periódicos da CAPES

Fatal Pulmonary Embolism: Old Pitfalls, New Challenges

1995; Elsevier BV; Volume: 70; Issue: 5 Linguagem: Inglês

10.4065/70.5.501

1942-5546

Kenneth Μ. Moser,

Atrial Fibrillation Management and Outcomes

Several fundamental misconceptions still inhibit the diagnosis of pulmonary embolism, as emphasized by Morgenthaler and Ryu in the current issue of Mayo Clinic Proceedings (pages 417 to 424). Such misconceptions are critical when pulmonary embolism goes unrecognized and untreated and eventuates in death. Even more disquieting is the fact that fatal pulmonary emboli constitute only a small portion—perhaps 10%—f all emboli. Thus, although the frequency of undiagnosed and untreated fatal pulmonary emboli in this report is striking, extrapolation of these figures to the much larger population that must have experienced nonfatal emboli should concern all physicians. What are the miscon­ ceptions involved, and what might be done about them? Deep Venous Thrombosis.— The first misconception is that pulmonary embolism is “the problem.” In fact, as has long been recognized, the actual problem is deep venous thrombosis (DVT), almost exclusively arising in the deep veins of the legs.1Havig O Source of pulmonary emboli.Acta Chir Stand Suppl. 1977; 478: 42-47Google Scholar, 2Moser KM Fedullo PF Venous thromboembolism: three simple decisions (parts I and 2).Chest. 1983; 83 (256–260): 117-121Crossref PubMed Scopus (13) Google Scholar, 3Sevitt S Gallagher N Venous thrombosis and pulmonary embolism: a clinico-pathological study in injured and bumed patients.Br J Surg. 1961; 48: 475-489Crossref PubMed Scopus (474) Google Scholar When DVT is absent, pulmonary embolism does not occur. Furthermore, if pulmonary embolism has already occurred and DVT is absent, embolism will not recur (unless other, rare sites of venous thrombosis exist). The failure to recognize the importance of DVT is reflected in the data presented by Morgenthaler and Ryu. In only 6 of their 92 patients with fatal pulmonary embolism, impedance plethysmography, duplex ultrasonography, or venography had been done in an effort to detect lower extremity thrombosis. Nonetheless, 88% of the 92 patients who died of pulmonary embolism had risk factors4Milbank Mem Fund Q. 1972; 50: 1-261Google Scholar, 5Prevention of venous thrombosis and pulmonary embolism [Consensus Conference].JAMA. 1986; 256: 744-749Crossref PubMed Scopus (179) Google Scholar for venous thrombosis (and, therefore, pulmonary embolism). Also reflecting the lack of understanding of the criticality of DVT is the fact that no type of prophylaxis had been used in approximately half of the 92 patients. Other investigators have reported an even poorer rate of prophylaxis in “at-risk” patients.6Anderson Jr, FA Wheeler HB Goldberg RJ Hosmer DW Forcier A Patwardhan NA Physician practices in the prevention of venous thromboembolism.Ann Intern Med. 1991; 115: 591-595Crossref PubMed Scopus (205) Google Scholar If the frequency of pulmonary embolism, including fatal embolism, is to be decreased, physician behavior must change. A National Institutes of Health consensus conference emphasized this need in 1986.5Prevention of venous thrombosis and pulmonary embolism [Consensus Conference].JAMA. 1986; 256: 744-749Crossref PubMed Scopus (179) Google Scholar That conference recommended that the intensity of prophylaxis be matched to the magnitude of risk by using one or a combination of the three available preventive modalities: intermittent venous compression devices, heparin, or warfarin sodium. In the rare circumstance in which prophylaxis cannot be applied, at-risk patients should be monitored with impedance plethysmography7Hull RD Hirsh J Carter C Jay RM Ockelford PA Buller HE et al.Diagnostic efficacy of impedance plethysmography for clinically suspected deep-vein thrombosis: a randomized trial.Ann Intern Med. 1985; 102: 21-28Crossref PubMed Scopus (282) Google Scholar or duplex ultrasonography.8Heijboer H Cogo A Buller HR Prandoni P ten Cate JW Detection of deep vein thrombosis with impedance plethysmography and realtime compression ultrasonography in hospitalized patients.Arch Intern Med. 1992; 152: 1901-1903Crossref PubMed Scopus (62) Google Scholar Finally, if neither prophylaxis nor monitoring is possible (for example, in patients with extensive lower extremity trauma), insertion of an inferior vena caval filter can be considered,9Kantcr B Moser KM The Greenfield vena caval filter.Chest. 1988; 93: 170-175Crossref PubMed Scopus (47) Google Scholar Manifestations of Pulmonary Embolism.—The second misconception involves the symptoms and signs of and pulmonary embolism. Investigators have long known that only 50% (or less) of patients with DVT—even extensive DVT -have any symptoms or signs. That fact is why DVT—its complication, pulmonary embolism-is such a treacherous disorder.10Moser KM Venous thromboembolism.Am Rev Respir Dis. 1990; 141: 235-249Crossref PubMed Scopus (332) Google Scholar Waiting for such signs and symptoms (for example, heat, redness, pain, edema, or positive Homans' sign) can be a fatal mistake. Massive, fatal pulmonary embolism can be the first symptom of extensive DVT. Only prophylaxis can avoid such an event. Equally unacceptable is the approach of requiring “classic” symptoms of pulmonary embolism before that diagnosis is considered. Morgenthaler and Ryu note that the classic symptoms of pulmonary embolism were frequently absent in patients who died of embolism. Neither the term “classic” nor “usual” should be applied if pulmonary embolism is to be diagnosed promptly. Investigators have repetitively demonstrated the high frequency (40 to 60%) of asymptomatic pulmonary embolism in patients with proximal DVT (popliteal vein and above);11Huisman MV Builer HR ten Cate JW van Royen EA Vrecken J Kerstcn MJ et al.Unexpected high prevalence of silent pulmonary embolism in patients with deep venous thrombosis.Chest. 1989; 95: 498-502Crossref PubMed Scopus (325) Google Scholar, 12Moser KM Fedullo PF LittleJohn JK Crawford R Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis.JAMA. 1994; 271: 223-225Crossref PubMed Scopus (368) Google Scholar thus, many patients have no symptoms-classic or otherwise. The only reasonably common symptoms are dyspnea of sudden onset and tachycardia. Although these findings suggest the possible diagnosis of pulmonary embolism, they clearly are nonspecific.10Moser KM Venous thromboembolism.Am Rev Respir Dis. 1990; 141: 235-249Crossref PubMed Scopus (332) Google Scholar In actuality, hemoptysis and pleuritic chest pain are not symptoms or signs of pulmonary embolism. They are indicative of pulmonary infarction, and infarction occurs infrequently as a result of pulmonary embolism (in perhaps 10% of all embolic events). Furthermore, when pulmonary infarction (or congestive atelectasis) does occur, it does not express itself for many hours after the embolic event—a situation that precludes early diagnosis.10Moser KM Venous thromboembolism.Am Rev Respir Dis. 1990; 141: 235-249Crossref PubMed Scopus (332) Google Scholar Thus, hemoptysis and pleuritic chest pain are neither classic nor usual. Even when present, these findings are nonspecific, and their absence does not rule out pulmonary embolism. One manifestation of pulmonary embolism noted by Morgenthaler and Ryu deserves special comment—namely, syncope. In patients encountered in emergency departments, syncopal episodes obviously are associated with numerous diagnostic possibilities, among which pulmonary embolism should be listed. Syncope is indicative of massive pulmo­ nary embolism. If death does not immediately ensue, some distal movement of thrombus, thrombolysis, or thrombus fragmentation may allow sufficient recovery for the patient to awaken. Indeed, cursory examination may show a patient who appears to be “normal,” and various neurologic disorders (for example, transient ischemic attack or seizure) or transient arrhythmic episodes often are appropriately considered. If pulmonary embolism is the cause, however, careful examination may disclose an accentuated pulmonary valve closure sound or a right ventricular tap (or both findings). Other studies—for example, electrocardiography, echocardiography, or ventilation-perfusion lung scanning-if done, will lead to the correct diagnosis. The first step is to recognize that syncope may be the initial manifestation of extensive pulmonary embolism. Pathologic Findings.—Another interesting observation by Morgenthaler and Ryu is the presence, at autopsy, of both organizing and acute embolic material in 50% of their patients. This finding relates to another long-emphasized feature of lethal pulmonary embolism-that is, an initial nonlethal embolic event is often converted to a lethal one by an embolic recurrence from the uncontrolled site of venous thrombosis. “Control” of venous thrombosis can be achieved in two ways: (1) by administration of anticoagulant therapy to prevent growth of fresh thrombus (with the assumption that fibrinolysis and organization will resolve the residual thrombus) or (2) by insertion of an inferior vena caval filter. Insertion of such a filter, however, does not mean that anticoagulant therapy should be withheld (unless such therapy is contraindicated). Extension or recurrence of venous thrombosis is not prevented by insertion of a filter. Indeed, detachment of fresh venous thrombi poses the potential hazard of obstructing the filter and then growing through it. Therefore, whenever possible, anticoagulation should be continued after insertion of an inferior vena caval filter. The pathologic finding of organized and fresh embolic material also relays another message about the need to recognize pulmonary embolism and institute therapy promptly. In situ growth of pulmonary emboli can occur and, again, may convert mild or moderate embolic obstruction to more severe, potentially fatal obstruction. (In patients with extensive, chronic embolic obstruction, fresh proximal thrombus often is noted.13Moser KM Auger WR Fedullo PF Chronic major-vessel thromboembolic pulmonary hypertension.Circulation. 1990; 81: 1735-1743Crossref PubMed Scopus (417) Google Scholar) Anticoagulant therapy early after pulmonary embolism can prevent such in situ extension. Recommendations.—Some reports suggest that death due to pulmonary embolism may not be a major problem because many such deaths occur in patients with other serious medical problems.14Carson JL Kelley MA Duff A Weg JG Fulkerson WJ Palevsky HI The clinical course of pulmonary embolism.N Engl J Med. 1992; 326: 1240-1245Crossref PubMed Scopus (975) Google Scholar The current report also emphasizes the frequency of comorbid conditions and attempts to classify patients into various prognostic categories. Indeed, some patients at risk for venous thromboembolism are dying of other disorders (for example, disseminated malignant disease), and those responsible for the patient's care (in conjunction with the patient and family) must make a philosophic-medical decision about the approach to prevention, diagnosis, and management of venous thromboembolism. Nevertheless, considerable caution should be exercised in adopting a less-than-aggressive approach to venous thromboembolism in most at-risk patients. Many patients who die of pulmonary embolism, or suffer substantial disability from venous thrombosis or embolism, have nonlethal (or no severe) comorbid conditions. Therefore, the overall approach to the problem should be aggressive: apply prophylaxis to all patients at risk; use the available noninvasive and, when necessary, invasive techniques for the diagnosis of venous thrombosis and pulmonary embolism; and promptly treat those with proximal venous thrombosis and embolism. Any tendency to complacency under the guise of “the patient was going to die anyway” is unacceptable and, in fact, hazardous. Except in rare instances, only the fully informed patient should make decisions about prevention or treatment of potentially disabling or lethal complications such as venous thromboembolism.