Henry Ford Hospital Medical Journal

Volume 36 Number 1 Article 9

3-1988

Clinical Manifestations of Acute Pulmonary Embolism: Henry Ford Hospital Experience, A Five-Year Review

Kenneth V. Leeper Jr.

John Popovich Jr.

Deborah Adams

Paul D. Stein

Follow this and additional works at: https://scholarlycommons.henryford.com/hfhmedjournal

Part of the Life Sciences Commons, Medical Specialties Commons, and the Public Health Commons

Recommended Citation Leeper, Kenneth V. Jr.; Popovich, John Jr.; Adams, Deborah; and Stein, Paul D. (1988) "Clinical Manifestations of Acute Pulmonary Embolism: Henry Ford Hospital Experience, A Five-Year Review," Henry Ford Hospital Medical Journal : Vol. 36 : No. 1 , 29-34. Available at: https://scholarlycommons.henryford.com/hfhmedjournal/vol36/iss1/9

This Article is brought to you for free and open access by Henry Ford Health System Scholarly Commons. It has been accepted for inclusion in Henry Ford Hospital Medical Journal by an authorized editor of Henry Ford Health System Scholarly Commons. Clinical Manifestations of Acute Pulmonary Embolism: Henry Ford Hospital Experience, A Five-Year Review

Kenneth V. Leeper, Jr, MD,* John Popovich, Jr, MD,* Deborah Adams, BSN,* and Paul D. Stein, MD^

Clinical findings of 112 patients with angiographically proven pulmonary embolism over a five-year period were analyzed. Recent immobilization, chronic heart disease, deep venous thrombosis, malignancy, and recent surgery were the most frequent predisposing faclors. Only 5% had no identifiable risk factors. Presenting syndromes were circulatory collapse (25%), pulmonary infarction (59%), and uncomplicated pulmonary embolism (18%). Dyspnea and pleuritic chest pain were the predominant symptoms. The combination of dyspnea, pleuritic chest pain, and hemoptysis occurred in 23% ofthe patients. The most frequent signs were tachypnea, tachycardia, and rales. Most ofthe patients demonstrated arterial oxygen tension (PaO^) between 50 and 70 mm Hg, and 9% had a PaO, greater than 80 mm Hg. The mosl common chest x-ray findings were infiltration and consolidalion, pleural effusion, and atelectasis. The most common electrocardiographic abnormalities were nonspecific ST and T-wave changes in 42% of patients. Lung scans were most frequently interpreted as indeterminate probability. A continuing reassessment of the features of pulmonary embolism may assist in selecting patients for confirmatory diagnostic studies. (Henry Ford Hosp Med J 1988; 36:29-34)

ulmonary embolism is common yet frequently remains lism identified by this procedure (Table 1). Charts of these 112 Punrecognized. Consequently, this condition is under­ patients were reviewed for demographic data, symptoms, and diagnosed, especially in patients with a history of cardiac or pul­ signs. Reports ofthe chest radiographs, ventilation perfusion monary disease. The mortality rate from untreated pulmonary lung scans, and pulmonary angiograms were also reviewed. embolism is approximately 30% . while correctly diagnosed and treated patients have a mortality rate of approximately 8% (1). The clinical features and laboratory findings have been charac­ Results teristically described as nonspecific in numerous studies (2-4), Patient population and predisposing factors which is similarly true for the noninvasive diagnostic studies, The patients with angiographically proven pulmonary embo­ including ventilation perfusion scans (5,6). Because of the non- lism included 59 men and 53 women. Their ages ranged from 20 specificity of the noninvasive studies, pulmonary angiography to 89 years. has been used as the "gold standard" in establishing the diag­ nosis of pulmonary embolism (7,8). Some centers, however, have reported that only 30% of patients with suspected pulmo­ Table 1 nary embolism have had pulmonary embolism proven by Acute Pulmonary Embolism angiography (9). Despite the rather grim impression of the nonspecificity of the clinical diagnosis, some investigators have Number of Patients with Suspected Pulmonary shown that the clinical features of pulmonary embolism may be Embolism Undergoing Number of Positive extremely useful in establishing a preliminary diagnosis (10,11), Year Angiography Angiograms thereby eliminating the necessity of performing angiography on 1980 24 12 patients in whom the diagnosis is unlikely. In this investigation, 1981 7S 18 we assess our own experience in an effort to strengthen our diag­ 1982 60 25 1983 65 27 nostic capabilities. 1984 75 30 Total 302 112 Methods A computer search of the hospital medical records and a review of the records in Interventional Radiology were under­ Submitted for publication: September 23. 1987. taken to identify patients who underwent pulmonary angiogra­ Accepted for publication; November 10. 1987. 'Department of Intemal Medicine. Division of Pulmonary and Critical Care Medicine. phy for suspected pulmonary embolism from 1980 through Henry Ford Hospital. 1984. During this five-year period, 302 pulmonary angiograms "i Heart and Va.scular Institute. Cardiovascular Research. Henry Ford Hospital. *Address correspondence to Dr Stein. Heart and Vascular Institute. Cardiova.scular were performed and 112 (37%) patients had pulmonary embo­ Research, Henry Ford Hospital, 2799 W Grand Blvd. Detroit, Ml 48202.

Henry Ford Hosp Med J—Vol 36, No I, 1988 Pulmonary Fmbolism—Leeper etal 29 Table 2 Table 4 Incidence of Predisposing Factors in Patients with Symptoms of 112 Patients with Pulmonary Embolism Angiographically Proven Pulmonary Embolism Symptom Number Percenl Condition Number Percent Dyspnea 97 87 Recent immobilization 33 29 Pleuritic chest pain 66 5Cough 30 27 Deep venous thrombosis 26 23 Hemoptysis 25 22 Recent surgery 23 21 Calf pain 16 14 Previous pulmonary embolism 13 12 Syncope 14 13 Previous thrombophlebitis 10 9 Chest pain, nonpleuritic 12 1 1 Obesity 8 7 Palpitations 9 8 Trauma 6 4 Estrogen-containing drugs 3 3 No risk factor 7 5

The most frequent predisposing factors included recent im­ Symptoms and physical signs mobilization and chronic heart disease which occurred in 29% The symptoms and physical signs of the 112 patients with pul­ and 28% of the patients with proven acute pulmonary embolism, monary emboli are listed in Tables 4 and 5. Dyspnea, the most respectively. Deep venous thrombosis, documented by invasive frequent symptom, occurred in 87% of the patients. Pleuritic or noninvasive methods, was observed in 23% of patients with chest pain occurred in 59%. An acute onset of either dyspnea or acute pulmonary embolism. Eight patients had positive ve­ pleuritic chest pain occurred in 73% ofthe patients with pul­ nograms, all demonstrating clots above the knee. Thirteen ofthe monary embolism. Cough, hemoptysis, syncope, nonpleuritic 18 patients who had noninvasive studies had clots in the femoral chest pain, calf pain, and palpitations each occurred in less than vein, and five had equivocal studies suggesting femoral vein in­ 30% ofthe patients (Table 3). volvement. Malignancy and recent surgery were predisposing Tachypnea, the most frequent physical sign, occurred in 81% factors in 24% and 21% of patients, respectively. Less frequent ofthe patients. Tachycardia and rales were the next most predisposing factors included a history of pulmonary embolism, frequent physical findingsa t 67%i and 46%, respectively (Table trauma (usually from long bone fractures), or the use of 5). An accentuated pulmonic closure sound, shock, fever (tem­ estrogen-containing dmgs. Obesity as the only identifiable risk perature > 37.8°C), pleural rub, and cyanosis were less com­ factor was present in 7% of the patients (Table 2). Many of the mon, with each noted in less than 20% of the patients (Table 5). patients had multiple risk factors. However, 5% of the patients Either dyspnea or tachypnea (5= 20 breaths/min) occurred in had no recognized risk factor. 96% of patients. The combination of dyspnea and pleuritic chest pain occurred in 47% of patients. However, dyspnea, pleuritic Presenting syndromes of acute pulmonary embolism chest pain, and hemoptysis occurred in 23% (Table 6). The syndrome of circulatory collapse (shock or syncope) oc­ curred in 25% of the patients. The pulmonary infarction syn­ Arterial blood oxygen drome (hemoptysis with or without pleuritic pain or pleuritic Eighty-nine (79%) of the 112 patients with proven pulmonary pain without hemoptysis) occurred in 59% of patients. Uncom­ embolism had arterial blood gases drawn while breathing room plicated pulmonary embolism (dyspnea alone or nonpleuritic air. The distribution of levels of the arterial PO, is shown in the pain alone) occurred in 18% of patients (Table 3). Figure. The most frequent range of arterial PO^ was between 50

Table 3 Table 5 Presenting Syndromes of Acute Pulmonary Physical Signs in 112 Patients with Pulmonary Embolism Embolism in 112 Patients with Angiographically Proven Pulmonary Embolism Physical Signs Number Percent Tachypnea (3= 20/min) 91 81 Syndrome Number Percent Tachycardia (s: lOO/min) 75 67 Circulatory collapse syndrome Rales 52 46 Syncope 15 13 Accentuated P, 18 16 Shock 14 12 Fever> 37.8°C 16 14 Pulmonary infarction syndrome Shock (systolic blood pressure < 100 mm Hg) 15 13 Hemoptysis ± pleuritic pain 25 22 Jugular venous distension 14 12 Pleuritic pain without hemoptysis 4(.) .r/ Fourth heart sound 12 11 Uncomplicated embolism syndrome Third heart sound 9 8 Dyspnea i: II Pleural rub 5 4 Nonpleuritic pain 6 5 Cyanosis 5 4

30 Henry Ford Hosp Med J—Vol 36, No 1, 1988 F*ulmonary Embolism—Leeper etal Table 6 30. 28 Combinations of Dyspnea, Pleuritic Pain, and 26. Hemoptysis in 112 Patients with Angiographically Proven 24. 22. Pulmonary Embolism 20. 18. 16. Symptoms Number Percent 20% 14. Dyspnea only 25 22 12. Dyspnea + pleuritic pain 53 47 10. Dyspnea + pleuritic pain + hemoptysis 26 23 8 . 6 Pleuritic pain 9 8 4. Pleuritic pain -1- hemoptysis 3 3 2. Dyspnea or hemoptysis -1- pleuritic pain 97 87 0. 30-39 40-49 50-59 69-09 70-79 80-89 90-99 PaOi (mmHg) and 59 mm Hg, which occurred in 33% of the patients. An ar­ terial PO, greater than 80 mm Hg was present in 9%, and a PO, less than 40 mm Hg occurred in 6% of the patients. Figure—Ranges of partial pressure of oxygen in arterial blood (PaO,). The numbers in the bars indicate the percentage of Electrocardiographic features patients with the indicated PaO,. Ofthe 112 patients studied, 83 had identifiable electrocar­ diograms at the time of the embolic event. The most common cated in Table 2, recent immobilization, chronic heart disease, abnormality was nonspecific ST segment or T-wave changes, documented deep venous thrombosis, recent surgery, and ma­ which occurred in 57% of patients. A normal electrocardiogram lignancy were the most frequent predisposing factors. All of the was seen in 10% of patients. An S1Q3T3 pattem occurred in 7% patients in our study who had documented deep venous throm­ of patients. A new right bundle branch block was noted in 14 bosis by phleborheography or venography had findings sug­ (17%) of 83 patients, and six of these patients presented with gestive of proximal deep venous thrombosis (involvement of the either shock or syncope. Left axis deviation (^ — 30) and right veins of the thigh). Moser and LeMoine (12) demonstrated that axis deviation (> 90°) occurred in 7% and 10% of the patients. proximal deep venous thrombosis carries a greater risk of embo­ New arrhythmias that seemed to accompany the onset of symp­ lization than distal (calf) venous thrombosis. toms occurred in 12%, with atrial flutter or fibrillation being Chronic heart disease is a major risk factor for the develop­ most common in 10% of the patients. ment of thromboembolic disease, especially in patients ob­ served with both atrial fibrillation and congestive heart failure Chest roentgenographic and lung scan findings (13). Chronic heart disease occurred in 28% of our patients, with The most commonly reported chest x-ray finding was infiltra­ most patients suffering from recurrent congestive heart failure. tion (43%), followed by pleural effusion (37%). Atelectasis was Individuals who weigh 20% more than the standard weight noted in 19%. A normal chest x-ray was present in 18%. Ventila­ for their age, sex, and frame have an increased incidence of tion and perfusion scans were obtained prior to pulmonary an­ thromboembolic disease (14). Adiposity in women was noted to giography in 46 (41%) of 112 patients. Of these scans, 22 (48%) be a factor for significant pulmonary embolism at autopsy (15). were interpreted as indeterminate probability, 19 (41%) as high Obesity was the only identifiable riskfacto r in 7% of the patients probability, and five (11%) as low probability. in our study. The association between malignancy and recurrent throm­ Pulmonary angiography boembolic disease was first described by Trousseau (16). It has Only 112 (37%) of the 302 patients who underwent pulmonary been suggested that recurrent pulmonary embolic events may angiography had demonstrable pulmonary emboli. No an­ presage the diagnosis of an occult neoplasm (17). In a review by giographically related deaths occurred among these 112 patients. Gore et al (17), an increased incidence of cancer of the Ofthe 112 patients with pulmonary embolism, 108 had bilateral breast, gastrointestinal tract, lung, and uterus was observed for angiograms in which the most frequent lobar locations for em­ two years following the diagnosis of deep venous thrombosis boli were the right lower lobe (36%) and the left lower lobe and pulmonary embolism. A shift in patients undergoing pul­ (33%). Right atrial, right ventricular, and pulmonary artery monary angiography for suspected pulmonary embolism, from pressures were not consistently recorded. Ofthe 108 bilateral patients with primary cardiovascular disease to patients with angiograms, 62% had bilateral involvement. One patient had a neoplastic disease, has been reported (18). Malignancy was pulmonary digital subtraction angiogram which demonstrated a noted to be a major riskfacto r (24% of patients) in the develop­ right lower lobe embolus. ment of pulmonary embolism in our review, which was com­ parable to the incidence in patients with chronic heart disease Discussion (28%). Lung, pancreas, gastric, and colonic neoplasms were Pulmonary embolism is not a disease but a complication of the usual tumors associated with pulmonary embolic events in deep venous thrombosis. The factors responsible for the devel­ our patients. opment of deep venous thrombosis are obviously the major The determination of an identifiable predisposing risk factor determinants for subsequent pulmonary embolization. As indi- is important in formulating a clinical diagnosis of pulmonary

Henry Ford Hosp Med J—Vol 36, No 1, 1988 Pulmonary Embolism—Leeper et al 31 embolism. Of 215 patients with angiographically proven embo­ pleuritic chest pain who did or did not have pulmonary emboli. lism and no preexisting cardiac or pulmonary disease, Stein et al In an editorial by Martin (22) conceming PaO, data from phase 2 (11) found that predisposing risk factors occurred in 76% of pa­ ofthe UPET study, he contended that a PaO, above 80 mm Hg as tients. In the urokinase pulmonary embolism trial (UPET) (19), being "nonnal" may be misleading without knowledge of the 94% of patients with documented emboli had one or more rec­ PaCO, and the alveolar-arterial gradient to allow for estimation ognized predisposing conditions. Only 5% of the patients in our ofthe contribution of hyperventilation to normoxemia. The Fig­ study had no identifiable risk factors. ure illustrates the varied ranges of partial pressure of oxygen in our patients with acute pulmonary embolism. As noted, most Syndromes patients demonstrated moderate hypoxemia; however, 9% had a Categorizing pulmonary embolism in terms of syndromes PaOj greater than 80 mm Hg, which is in accordance with Bell et helps to clarify the presenting clinical diagnostic pattems. The al's study (3). However, the alveolar-arterial gradients were not syndromes of pulmonary embolism include the circulatory col­ evaluated. lapse syndrome, pulmonary infarction syndrome, and uncom­ Using their inert gas technique, Wagner et al (23) studied the plicated pulmonary embolism syndrome (12). causes of hypoxemia. Recent investigators, using this method, By recognizing the various presenting syndromes of acute demonstrated that during the initial phase of the pulmonary em­ pulmonary embolism, the clinician can estimate the hemo­ bolic event hypoxemia was the result of perfusion of lung units dynamic impact of the embolic event. In the study by Stein et al with a low ventilation/perfusion ratio. Shunt as a major factor of (11), the pulmonary angiographic severity scores were higher in hypoxemia was noted to occur later in the course of the embolic syndromes characterized by circulatory collapse and the uncom­ event (48 hours or more) and only if atelectasis or other factors plicated embolism syndrome. In contrast, the angiographic leading to loss in lung volume were present (24). index of severity was lower in patients with the pulmonary infarction syndrome. Thames et al (20) found that most of the patients with syncope demonstrated 50% or greater obstmction Chest radiographic findings of pulmonary blood flow. A normal chest radiograph was found in 18% of the patients in our series. The presence of a normal chest radiograph was Symptoms and physical findings seen in 24% of 169 patients with angiographically proven Several clinical studies have demonstrated the difficulty of pulmonary embolism in the National Heart, Lung and Blood diagnosing pulmonary embolism based on the history and phys­ Institute (NHLBI) urokinase-streptokinase pulmonary embo­ ical examination (2,3). Nevertheless, some clues have been lism trials (25). found based upon common but nonspecific signs and symp­ In our review, the most frequent radiographic findings toms. Among patients with pulmonary embolism without pre­ were infiltration, pleural effusion, and atelectasis. These existing cardiac or pulmonary disease. Stein et al (11) found that parenchymal signs were more common in patients with the pul­ 96% had either dyspnea, tachypnea, or clinical signs of deep monary infarction syndrome. Other chest radiographic abnor­ venous thrombosis. Our experience paralleled these observa­ malities may include elevation of the hemidiaphragm on the side tions, with either dyspnea or tachypnea occurring in 96% of our ofthe embolus, enlarged pulmonary arteries, a difference in patients. The absence of dyspnea and/or tachypnea would be vascularity of the two sides (Westermark's sign), and car- strong evidence for exclusion of the diagnosis. diomegaly secondary to right ventricular enlargement (25). Data collected from the UPET (19) revealed that the triad of Greenspan et al (26) evaluated whether the chest radiograph pleuritic chest pain, hemoptysis, and thrombophlebitis was alone could be used to diagnose or exclude pulmonary embo­ present in only 4% of patients with acute pulmonary embolus. lism. These investigators found that the chest x-ray can provide The same study also showed that the triad of dyspnea, pleuritic additional information, especially to exclude other disease pro­ chest pain, and hemoptysis was observed in only 28% of patients cesses (rib fracture or pneumothorax). On the other hand. Stein with pulmonary embolism. From our review, the combination et al (25) found frequent radiographic signs that may give clues of dyspnea, pleuritic chest pain, and hemoptysis was seen in to the diagnosis. In particular, an elevated hemidiaphragm was 23% of patients. Stein et al (11) analyzed a subset of patients found in 26% of patients with pulmonary embolism. In our from the UPET study who had no history of previous car­ study, the chest x-ray report of an elevated hemidiaphragm diopulmonary disease. They found, even among patients with was rare (only 5%), but this may have represented a lack of de­ no cardiac or pulmonary disease, that the combination of dys­ scription of this finding. It would be important to consider a pnea, pleuritic chest pain, and hemoptysis occurred in only 22% normal chest x-ray, in the clinical sening of acute dyspnea with of patients with acute pulmonary embolism. hypoxemia and no evidence of bronchospasm, to be strongly suggestive of acute pulmonary embolism. Arterial blood gases Hypoxemia is a consistent and important clinical feature of Electrocardiogram acute pulmonary embolism and is present in most patients with In patients without prior cardiopulmonary disease, elec­ pulmonary embolism (21). Bell etal (3) reported that the average trocardiographic abnormalities were present in 87% of 90 pa­ PaO, was 60 mm Hg and that 10%> of the patients had a PaO, tients with angiographically documented emboli (27). The most greaterthan 80 mm Hg. McNeil et al (21) showed that no signifi­ common abnormalities were nonspecific T-wave changes in cant difference occurred in the PaO, of young patients with 42% of patients, with RST-segment abnormalities present in

32 Henry Ford Hosp Med J—Vol 36. No 1. 1988 Pulmonary Embolism—Leeper et c 41%. We also found nonspecific ST segment and T-wave Pulmonary angiography changes to be the most common abnormality. Left axis deviation Pulmonary angiography continues to be the "gold standard" test and right axis deviation occurred at equal frequency (7%) (27). for the diagnosis of pulmonary embolism (7,8). In the UPET study Signs of cor pulmonale (S.QjT,, right bundle branch block, P- (19), in which over 800 pulmonary angiograms were pjerformed, pulmonale, or right axis deviation) were noted in 26% of pa­ the morbidity was less than 1% and the mortality was less than tients with submassive pulmonary embolism and in 32% with 0.01%. A review of 1,350 angiograms performed at Duke Medical massive pulmonary embolism (27). Although we did not sepa­ Center over an 11 -year period revealed only three deaths (34). There rate our patient group into massive or submassive categories, were no angiographically related deaths in our series. we noted that in six of 14 with syncope or shock a new right bun­ A negative pulmonary angiogram, despite even a strong clinical dle branch block was present. In the UPET study (19), 23% of suspicion, excludes the diagnosis of pulmonary embolism. Among patients with submassive pulmonary embolism had normal 167 patients with clinically suspected pulmonary embolism who electrocardiograms, whereas in our study 10% of patients had had negative pulmonary angiograms and who were followed for six normal electrocardiograms. Although no specific electrocar­ months, none had a clinical event suggestive of pulmonary embo­ diographic features are diagnosric of pulmonary embolism, lism (35). Those who died and had postmortem examinations did these nonspecific ECG abnormalities in the proper clinical set­ not demonstrate pulmonary emboli. ting support the diagnosis. Arrhythmias occurred in 12% of our patients, most of which Summary were atrial fibrillation or atrial flutter. This finding occurred We retrospectively reviewed the clinical features, laboratory more frequently in our study than in others, which we cannot data, chest radiographic reports, and lung scans of patients with explain. In 90 patients with pulmonary emboli without previous angiographically proven pulmonary embolism to reassess the cardiopulmonary disease, no arrhythmia occurred (27). Of the features of pulmonary embolism which may assist the clinician 23 patients with acute pulmonary embolism and no heart disease in selecting patients for confirmatory diagnostic studies. We re­ in Szucs et al's study (28), only 4% had atrial flutter or at­ alize evaluation of the clinical features of pulmonary embolism rial fibrillation. Even in patients who had previous cardiac or only in those patients with angiographic documentation creates pulmon.i.ry disease, only 3% had premature atrial contractions, a degree of bias when determining these clinical characteristics. 5% had atrial fibrillation, and 9% had premature ventricular The clinical features of pulmonary embolism have been dem­ contractions (19). onstrated in the literature as being nonspecific. However, the following clinical parameters should be sought in selecting Lung scans patients with suspected pulmonary embolism for pulmonary The clinical utility of ventilation perfusion imaging has come angiography: 1) appropriate predisposing factors; 2) an acute, under considerable controversy recently, especially in regard unexplained onset of dyspnea, tachypnea, or tachycardia; 3) ar­ to focusing on the sensitivity and specificity of this nonin­ terial blood gases demonstrating hypoxemia or a widening of vasive test (29). Several retrospective studies have compared the alveolar-arterial oxygen gradient; 4) an electrocardiogram scintigraphic findings with pulmonary angiographic findings, showing nonspecific ST segment and T-wave changes; 5) a nor­ resulting in different criteria for the interpretation of ventilation mal chest radiograph in the presence of unexplained dyspnea or perfusion scans in patients with pulmonary embolism (30,31). showing an elevated hemidiaphragm or parenchymal abnor­ These studies showed that a scan pattem characterized as multi­ malities; and 6) abnormal ventilation perfusion scans, which, ple mismatched perfusion defects, involving greater than 75% whether of low or indeterminate probability, in the presence of of the bronchopulmonary segment, is associated with a high the above findings warrant angiographic pursuit. probability of pulmonary embolism (90%). In a lung scan that demonstrates perfusion defects, matching with radiographic and Acknowledgment ventilation abnormalities, the likelihood of pulmonary embo­ We thank Ida Bomm for her assistance in the preparation of lism is indeterminate (25% to 35%). A scan pattem which shows this manuscript. that all perfusion defects are less than 25% of the segment and are matched with ventilation abnormalities but not to chest References radiographic abnormalities has a low likelihood of pulmonary 1. Dalen JE. Alpert JS. Natural history of pulmonary embolism. Prog Car­ embolism (less than 10%). diovasc Dis 1975:17:259-70. This scheme has been challenged by a prospective Canadian 2. Goodall RJR, Greenfield LJ. Clinical correlations in the diagnosis of pul­ study (29), which suggests that low probability scan patterns monary embolism. Ann Surg 1980:191:219-23. may be associated with a higher incidence of venous throm- 3. Bell WR, Simon TL, DeMets DL. The clinical features of submassive and massive pulmonary emboli. Am J Med 1977;62:355-60. boemboli than previously reported. Nevertheless, a completely 4. Wenger NK, Stein PD. Willis PW III. Massive acute pulmonary embo­ normal ventilation perfusion scan virtually eliminates the diag­ lism: The deceivingly nonspecific manifestations. JAMA 1972:220:843-4. nosis of pulmonary embolism (32,33). 5. Tow DE. Simon AL. Comparison of lung scanning and pulmonary Our institution is currently involved with a multicentered angiography in the detection and follow-up of pulmonary embolism: NHLBI study of the prospective investigation of pulmonary The urokina.se-pulmonary embolism trial experience. Prog Cardiovasc Dis 1975;17:239-45. embolism diagnosis (PIOPED), which assesses the efficacy of 6. Bell WR, Simon TL, A comparative analysis of perfusion scans with pul­ ventilation perfusion scanning. monary angiograms. Am Heart J 1976;92:700-6.

Henry Ford Hosp Med J—Vol 36. No 1. 1988 Pulmonary Embolism—Leeper etal 33 7. Dalen JE, Brooks HL, Johnson LW, Meister SG, Szucs MM, Dexter L, ventilation-perfusion ratios in nonnal subjects breathing air and 100% O,. J Clin Pulmonary angiography in acute pulmonary embolism: Indications, techniques, Invest 1974;54:54-68. and results in 367 patients, AmHeartJ 1971;81:175-85, 24. Huet Y, Lemaire F. Brun-Buis.son C, et al. Hypoxemia in acute pulmo­ 8. Menzoian JO. Williams LF. Is pulmonary angiography essential for the nary embolism. Chest 1985;88:829-36. diagnosis of acute pulmonary embolism? Am J Surg 1979;137:543-8. 25. Stein PD, Willis PW III, DeMets DL, Greenspan RH. Plain chest roent­ 9. Ormond RS. Gale HH, Drake EH, Gahagan T. Pulmonary angiography genogram in patients with acute pulmonary embolism and no pre-existing car­ and pulmonary embolism. Radiology 1966:86:658-62. diac or pulmonary dfsease. Am J Noninvasive Cardiol 1987;1:171-6. 10. Stein PD. Willis PW III, Dalen JE. Importance of clinical a.ssessment in 26. Greenspan RH, Ravin CE. Polansky SM, McLoud TC. Accuracy of selecting patients for pulmonary arteriography. Am J Cardiol 1979;43:669-71. the chest radiograph in diagnosis of pulmonary embolism. Invest Radiol 11. Stein PD, Willis PW III, DeMets DL. History and physical examination in 1982:17:539-43. acute pulmonary embolism in patients without preexisting cardiac or pulmonary 27. Stein PD, Dalen JE, Mclntyre KM, Sasahara AA, Wenger NK, Willis disease. AmJCardiol 1981:47:218-23. PW III. The electrocardiogram in acute pulmonary embolism. Prog Cardiovasc 12. Moser KM, LeMoine JR. Is embolic risk conditioned by location of deep Dis 1975:17:247-57. venous thrombosis? Ann Intern Med 1981:94:439-44. 28. Szucs MM Jr, Brooks HL, Grossman W, et al. Diagnostic sensitivity 13. Coon WW. Risk factors in pulmonary embolism. Surg Gynecol Obstet of laboratory findings in acute pulmonary embolism. Ann Intern Med 1976:143:385-90. 1971;74:161-6. 14. Coon WW, Coller FA. Clinicopathologic correlation in thromboem­ 29. Hull RD, Hirsh J. Carter CJ, et al. Diagnostic value of ventilation-per­ bolism. Surg Gynecol Obstet 1959;109:259-69. fusion lung scanning in patients with suspecled pulmonary embolism. Chest 15. GoldhaberSZ, Savage DD, Garrison RJ, etal. Risk factors for pulmonary 1985;88:819-28. embolism: The Framingham study. Am J Med 1983:74:1023-8. 30. Biello DR, Mattar AG, McKnight RC, Siegel BA. Ventilation-perfusion 16. Trousseau A. Phlegmasia albadolens. Clinique Medicaie de I'Hotel Dieu studies in suspected pulmonary embolism. AJR 1979;133:1033-7. de Paris. 2nd ed. Paris: Bailliere, 1865;3. 31. Alderson PO, Rujanavech N, Seeker-Walker RH, McKnight RC. The role 17. Gore JM, Appelbaum JS, Greene HL, Dexter L, Dalen JE. Occult cancer of "'Xe ventilation studies in the scintigraphic detection of pulmonary embo­ in patients with acute pulmonary embolism. Ann Intem Med 1982:96:556-60. lism. Radiology 1976;120:633-40. 18. Marsh JD, Glynn M. Torman HA. Pulmonary angiography: Application 32. Stein PD. Low-dose heparin for prevention of pulmonary embolism and in a new spectrum of patients. Am J Med 1983;75:763-70. significance of normal lung scan: Resulls of a survey of three ACCP scientific 19. The urokinase pulmonary embolism trial: A national cooperative study. sessions. Cardiopulm Med 1982:21:12. Circulation I973;47(suppl 2):1-I08. 33. Kipper MS, Moser KM, Kortman KE. Ashbum WL. Long term follow- 20. Thames MD, Alpert JS, Dalen JE. Syncope in patients with pulmonary up of patients with suspected pulmonary embolism and a normal lung scan: Per­ embolism. JAMA 1977;238:2509-11. fusion scans in embolic suspects. Chest 1982:82:411-5. 21. McNeil BJ, Hessel SJ, Branch WT, Bjork L, Adelstein SJ, Measures of 34. Mills SR, Jackson DC, Older RA, Heaston DK. Moore AV. The inci­ clinical efficacy, II. The value ofthe lung scan in the evaluation of young patients dence, etiologies, and avoidance of complications of pulmonary angiography in with pleuritic chest pain. J Nucl Med 1976;17:163-9. a large series. Radiology 1980;136:295-9. 22. Martin L. Arterial blood oxygen tension in pulmonary embolism (edi­ 35. Novelline RA, Baltarowich OH. Athanasoulis CA, et al. The clinical torial). Am J Med 1977:63:168. course of patients with suspected pulmonary embolism and a negative pulmo­ 23. Wagner PD, Laravuso RB, Uhl RR, West JB. Continuous distributions of nary arteriogram. Radiology 1978:126:561-7.

34 Henry Ford Hosp Med J—Vol 36, No 1, 1988 Pulinonary Embolism—Leeper et ai