The Diagnosis and Treatment of Acute Pulmonary Embolism
; ; ;
Background: Pulmonary embolism (PE) is a cardiovascular emergency with high morbidity and mortality.
Methods: Review of relevant literature retrieved by a selective Medline search, including current guidelines.
Results: Hemodynamically unstable patients are considered to have high-risk PE, whereas hemodynamically stable patients are considered to have non-high-risk PE. After classification into one of these two risk groups, patients undergo further diagnostic evaluation for PE according to the appropriate risk-adapted algorithm. Patients who are in cardiogenic shock or have persistent arterial hypotension (high-risk PE) should undergo multidetector computed tomography (MDCT) or echocardiography at once, so that a PE, if present, can be treated immediately by thrombolysis. For hemodynamically stable patients with non-high-risk PE the proper diagnostic strategy is determined by the clinical probability of PE, which can be calculated with the aid of validated scoring systems and is based on both MDCT and D-dimer levels. For further risk stratification in hemodynamically stable patients, tests are performed to detect right ventricular dysfunction or myocardial injury, either of which indicates intermediate-risk PE. In addition to specific therapy, patients with high-risk PE, patients at high risk for hemorrhage and these with severe renal insufficiency should be anticoagulated with unfractionated heparin. All other patients should be treated with low-molecular-weight heparin or fondaparinux. Thereafter, long-term oral anticoagulation with vitamin K antagonists is recommended.
Conclusion: Modern algorithms have considerably simplified the diagnosis and treatment of acute PE. It would be desirable for these algorithms to be rapidly implemented in routine practice, because speedy diagnosis and immediate treatment can lower the morbidity and mortality associated with PE.
Despite many medical advances, acute pulmonary embolism (PE) remains a cardiovascular emergency with high morbidity and mortality. With clinically suspected PE, rapid and targeted treatment is essential because speedy diagnosis and immediate therapy can lower the morbidity and mortality associated with PE (1). However, the non-specific clinical presentation and the variety of suggested diagnostic algorithms, some of which are complex, can impede speedy and certain diagnosis (2). In light of this, the authors carried out a selective Medline literature review for this review article, taking into consideration a recent comprehensive review of the guidelines issued by the European Society for Cardiology (ESC) (3), the official comments from the German Cardiac Society (4) and the German interdisciplinary S2 guidelines (5). The authors would like to inform the reader of clear diagnostic procedures—dependent on the hemodynamic status of the patient—which have been simplified compared to previous review articles. The authors would also like to suggest risk-adapted, evidence-based therapeutic strategies that conform to these guidelines.
Definition, epidemiology and clinical issues
A pulmonary artery embolism is defined as a partial or complete occlusion of a pulmonary arterial branch (6). Approximately 70% of cases are caused by pelvic or leg thromboses (e1, e2, 3). Deep vein thrombosis and PE are different presentations of the same underlying pathophysiological event, venous thromboembolism (VTE) (1). Precise figures for the incidence of PE are not available. The annual incidence of diagnosed VTE is 150 to 200 cases per 100 000 population (7). As well as an unknown number of clinically silent embolisms, the non-specific clinical presentation hampers diagnosis, meaning the actual disease frequency is underestimated. The diagnosis is made within the lifetime only in about 30% of cases with confirmation via post mortem examination (8). Estimates for Germany begin at more than 350 000 cases annually. In the acute phase the mortality rate is 7% to 11%, meaning that up to 40 000 patients probably die as a result of a PE each year in Germany (2–4, e3).
Many patient-related and situational factors may contribute to the development of a VTE (Table 1 gif ppt) (1, 9). As well as those patients that have to undergo major surgical procedures, non-surgical patients are also at risk. Awareness of the risk factors is essential if individualized and risk-adapted prophylaxis is to be implemented (10). In clinical practice, however, PE also occurs in about 20% of cases in patients without recognizable risk factors (11).
With acute PE there is a mechanical obstruction of the pulmonary circulation system (1). The hemodynamic consequences are determined by the size of the embolism, any pre-existing cardiopulmonary diseases, and the intensity of pulmonary vasoconstriction. With hemodynamically significant PE, the sudden increase in pulmonary arterial pressure can cause acute right ventricular dysfunction and lead to the interventricular septum deviating to the left with a fall in the left ventricular preload (1). There is a danger of a subsequent reduction in coronary perfusion and cardiac output with cardiogenic shock and myocardial ischemia (2, 6). In most cases death due to acute PE can be traced back to acute right heart failure.
Suspicion of acute PE is raised by symptoms such as sudden onset dyspnea and tachypnea, chest pain, hemoptysis or syncope but these symptoms are neither sensitive nor specific due to the variety of possible differential diagnoses (1). Additional examinations such as chest x-rays, ECG or blood gas analysis are also unsuitable to confirm or exclude suspected PE with sufficient certainty but they do help with differential diagnosis (12).
Initial risk stratification and diagnostic strategies
In contrast to earlier recommendations, the current guidelines recommend a practical procedure if acute PE is suspected (3). Initially the hemodynamic stability of the patient (shock, persistent arterial hypotension) should be assessed to enable evaluation of the likely risk of patient death due to PE during the acute phase in the hospital or within 30 days (Figure 1 gif ppt). The advantage of this simplified division is that the diagnostic and therapeutic strategies can be adapted to the urgency of the situation and that no invasive hemodynamic parameters are required (2). Previous classification systems (such as the classification of severity according to Grosser) should no longer be used (3). The current ESC guidelines suggest risk-adapted management and therefore recommend two different diagnostic algorithms for patients with suspected high-risk versus non-high-risk PE (3).
Suspected high-risk PE
Suspected high-risk PE (hemodynamically unstable patient) is an acute life-threatening medical emergency (2). The clinical probability of PE is (almost) always high in this situation (12). The suggested algorithm (Figure 2 gif ppt) recommends using multidetector computed tomography (MDCT) with imaging of the pulmonary arteries to confirm PE (3). Bedside emergency echocardiography should only be carried out if there is significant hemodynamic instability (transport into CT not possible) to immediately determine if there is an indication for—potentially life-saving—thrombolysis. In light of the urgency and the non-invasiveness, transthoracic echocardiography should initially be carried out (6). The following echocardiographic parameters are indicative of pulmonary embolism (e4):
- Abnormal right ventricular wall motion
- Right ventricular dilatation
- Paradoxical septal motion
- Tricuspid valve insufficiency
- Increased pulmonary arterial pressure
- Inferior vena cava congestion
- Dilated pulmonary artery
Thrombi can occasionally be documented in the right heart transthoracically while transesophageal echocardiography allows direct imaging of thrombi in the pulmonary arteries (3). Taking into consideration availability, the individual ability of the examiner, and the extent of the hemodynamic instability, a transesophageal echocardiogram can also be done. In case of a negative MDCT or echocardiogram, other causes of the hemodynamic instability must be sought.
This algorithm is based on the consensus of the ESC Committee for Practice Guidelines (evidence level C).
Suspected non-high-risk PE
To select a suitable diagnostic strategy for hemodynamically stable patients prior to laboratory tests or imaging, the clinical probability of PE is estimated using a simple and validated scoring system (1, 2). The Wells score (Table 2 gif ppt) (13) allows standardized classification of patients on the basis of the clinical probability of PE (low, intermediate, or high). The result should be documented (12).
In order to prevent patients being placed unnecessarily on potentially hazardous long-term anticoagulation, diagnostic certainty has priority with suspected non-high-risk PE. In principle, many diagnostic procedures or combinations of examinations are suitable to reliably confirm or exclude PE (2, 3). Taking current diagnostic and management studies (14–16) and clinical reality into consideration, the ESC Committee for Practice Guidelines has decided to recommend a uniform diagnostic algorithm (Figure 3 gif ppt) on the basis of highly sensitive (enzyme linked immunosorbent assay [ELISA]) D-dimer determination and MDCT (3).
The fibrin fragment D-dimer is produced by degradation of crosslinked fibrin by plasmin (1). The negative predictive value is extremely high, meaning that PE is unlikely with a D-dimer antigen level below a test-specific threshold (12, e5). Modern ELISA test kits have a sensitivity of >95% and a specificity of about 40% (3). Further diagnostics and anticoagulation can be omitted for patients with a low or intermediate clinical probability if a highly sensitive assay yields a negative result (15). A positive result, however, only indicates the necessity of further (imaging) diagnostics. Age, pregnancy and a range of pathological conditions often lead to fibrin formation, which may lead to a non-specific increase in the D-dimer antigen level, and the predictive value of a positive result with regard to the presence of PE is still further reduced (5). Immediately postoperatively the D-dimer antigen level is likewise regularly elevated above the normal value, meaning that exclusion of PE using D-dimer determination is difficult (e5, e6). With a high clinical probability of PE D-dimer determination is not recommended, because a negative test result is only expected for a few patients (2).
MDCT reveals the extent of the PE and also allows possible differential diagnoses. It has replaced ventilation-perfusion scintigraphy and pulmonary angiography as the gold standard. The PIOPED II study documented a sensitivity of 83% and a specificity of 96% (14). The negative predictive value of a negative MDCT is significantly dependent on the clinical probability ascertained with the Wells score (low: 96%, intermediate: 89%, high: 60%). Pelvic or leg CT venography can in principle be included in the same examination in order to determine the presence and extent of a pelvic or leg thrombosis. In light of the additional radiation exposure, the ESC recommends compression ultrasonography of the lower extremities as an additional examination to increase the level of diagnostic certainty in case of doubt (Figure 3) (3).
Further risk stratification
For non-high-risk PE (no hemodynamic instability) the ESC recommends additional early risk stratification (evidence level B) (Table 3 gif ppt). Patients with right ventricular (RV) dysfunction and/or myocardial injury have an intermediate PE-related early mortality risk, as the mortality rate in studies was 3% to 15% (3, 17, 18, e7). Patients without RV dysfunction or myocardial injury have the best prognosis in studies (mortality rate <2%) and are therefore classified in the low risk category (4).
Because of its non-invasiveness and rapid availability even in an emergency, transthoracic echocardiography is of value to assess right ventricular function (6). In a current meta-analysis the risk of death due to PE is elevated by a factor of 2.5 with evidence of RV dysfunction (17). On the other hand, normal echocardiographic findings indicate a very good prognosis, as in prospective randomized studies the PE-related early mortality was about 1% for this constellation (e8, 19). That no uniform criteria for verification of RV dysfunction have been established is problematic (2, 17).
Cardiac biomarkers are a sensible addition to echocardiography for further risk stratification of hemodynamically stable patients with acute PE (20, 21). In a recent meta-analysis, elevation of cardiac troponin I or T indicated increased mortality and risk of complications (18). Normal troponin levels on the other hand indicate a very good prognosis in the acute phase of PE (2). B-type (brain) natriuretic peptide (BNP) and N-terminal fragment of BNP (NT-proBNP) are synthesized after myocardial stress and reveal the neurohumoral activation that takes place with ventricular dysfunction (2). Negative test results have a high predictive value for a good prognosis (22); however, the specificity is low and prospectively validated cutoffs are not yet available (e9). In the early phase of myocardial ischemia there is already detectable elevation of the cytoplasmic protein h-FABP (heart-type fatty acid binding protein). Study results indicate that h-FABP determination would allow better prognostic assessment in future (e9).
Risk-adapted therapeutic strategies with acute PE
Apart from hemodynamic stabilization and reversal of hypoxemia, the therapeutic goals for acute PE are—depending on the severity—prevention of appositional thrombus growth, restoration of pulmonary blood flow, and prevention of recurrences (8). If there is no contraindication, parenteral anticoagulation is therefore obligatory. The options available include unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), or fondaparinux (3). Where suspicion of an acute PE is high (high or intermediate clinical probability), initial anticoagulation—with consideration of the bleeding risk—must be initiated before a definitive diagnosis is available (evidence level C) (2).
Therapeutic strategies with high-risk PE
As well as general circulatory support and therapeutic anticoagulation, hemodynamically unstable patients with confirmed PE require immediate thrombolysis to relieve the right ventricle (evidence level A) (3). The following active substances and dosage regimens are recommended in the literature (5, 12):
- Alteplase (rtPA): 10 mg IV bolus over 1 to 2 minutes followed by 90 mg over 2 hours (with body weight <65 kg maximal 1.5 mg/kg)
- Urokinase: 3 million IU over 2 hours
- Streptokinase: 1.5 million IU over 2 hours.
In Germany, reteplase and tenecteplase are not approved for acute PE.
International guidelines for cardiopulmonary resuscitation recommend the administration of a thrombolytic agent for suspected pulmonary embolism even during resuscitation (23, e10). Resuscitation should then be continued for 60 to 90 minutes if stabilization is not achieved. With an absolute contraindication to thrombolysis (e11) or failure of thrombolytic therapy, surgical or interventional restoration of blood flow can be used depending on logistics and expertise on site (evidence level C) (2).
In addition to a specific therapy, supportive measures are also of central importance in the treatment of PE patients with hemodynamic instability, because acute right heart failure is the primary cause of death. In the case of hypotension, noradrenaline (norepinephrine) is the catecholamine of first choice. For normotensive patients with low cardiac output, dobutamine is used and for hypotensive patients with cardiogenic shock adrenaline (epinephrine) is used (3). Selective reduction of pulmonary arterial pressure using inhalative nitric oxide or prostacyclin aerosol in smaller clinical trials led to an improvement in the ventilation-perfusion ratio with improvement in oxygenation, reduction in pulmonary arterial pressure, and an increase in cardiac output (6).
Therapeutic strategies with non-high-risk PE
Normotensive patients (non-high-risk PE) receive therapeutic anticoagulation using LMWH or fondaparinux in a weight-adjusted dose (evidence level A) (5). UFH is preferable for patients with a very high risk of bleeding or severe renal insufficiency (target aPTT is 1.5 to 2.5 times normal value) (evidence level C) (2). Routine thrombolytic therapy is not recommended (evidence level B).
Patients with normal arterial blood pressure at the time of diagnosis but with evidence of RV dysfunction and/or myocardial injury (Table 3) have an intermediate risk of PE-related early mortality (18). The therapeutic consequences of assignment to this category, apart from the necessity of hospitalization and initial intensive care monitoring, have not yet been definitively clarified. A large European multicenter study (Pulmonary Embolism International Thrombolysis Study: EudraCT No.: 2006–005328–18) is currently investigating the significance of thrombolysis in PE patients with intermediate risk (12).
Hemodynamically stable patients with no evidence of RV dysfunction or myocardial injury (low risk) have a mortality rate of <2% with effective anticoagulation (e7). There is, therefore, no indication for thrombolysis or mechanical restoration of flow (evidence level B) (3, 12). Selected patients may be suitable for early discharge or treatment on an outpatient basis (e7). Precise criteria for outpatient treatment of PE are, however, not currently available (3).
Secondary prophylaxis and prevention of recurrence
The ESC recommendations regarding the duration of oral anticoagulation agree for the most part with the current guidelines of the American College of Chest Physicians (3, e12). For hemodynamically stable patients the administration of vitamin K antagonists (VKAs) should be initiated as early as the first or second day. Heparin or fondaparinux therapy is continued in conjunction for at least five days (evidence level A) and only stopped when the international normalized ratio (INR) is in the therapeutic range (2.0 to 3.0) on two consecutive days (evidence level C) (12). For PE secondary to reversible risk factors, treatment with a VKA for three months is recommended (evidence level A). In light of the lasting increase in the risk of recurrence after the initial occurrence of an ‘idiopathic’ (unprovoked) PE, we recommend continuing treatment with a VKA for at least three months (evidence level A). Provided anticoagulation is stable and the risk of bleeding is low, indefinite continuation of this therapy should be considered (evidence level B). Patients with PE and malignancies should be treated with LMWH for the first three to six months (evidence level B) and they should subsequently receive lifetime anticoagulation with VKA or LMWH or until the cancer is ‘cured’ (evidence level C).
Systematic use of venous filters to prevent PE recurrence is not recommended (evidence level B) (3, 24). Venous filters may, however, be indicated in exceptional cases if therapeutic anticoagulation is absolutely contraindicated or a PE has recurred despite adequate anticoagulation (evidence level C) (2, 25). The venous filter should be removed as soon as possible in order to avoid secondary vena cava thromboses and thromboembolisms.
Conflict of interest statement
Prof. Böttiger is Chairman of the European Resuscitation Council (ERC).
The remaining authors declare that no conflict of interest exists according to the guidelines of the International Committee of Medical Journal Editors.
Manuscript received on 27 July 2009, revised version accepted on 16 November 2009.
Translated from the original German by language & letters.
Dr. med. Alexander Schellhaaß
Klinik für Anaesthesiologie, Universitätsklinikum Heidelberg
Im Neuenheimer Feld 110
69120 Heidelberg, Germany
@For eReferences please refer to:
N Engl J Med 2008; 359: 2804–13. MEDLINE
Chest 2002; 122: 1440–56. MEDLINE
Klinik für Anästhesiologie und operative Intensivmedizin, Klinikum Stuttgart, Katharinenhospital, Prof. Dr. med. Walther
Department of Cardiology, University General Hospital Alexandroupolis, Griechenland: Prof. Dr. med. Konstantinides
Klinik für Anästhesiologie und Operative Intensivmedizin, Universitätsklinikum Köln: Prof. Dr. med. Böttiger
|1.||Tapson VF: Acute pulmonary embolism. N Engl J Med 2008; 358: 1037–52. MEDLINE|
|2.||Konstantinides SV: Acute pulmonary embolism revisited: thromboembolic venous disease. Heart 2008; 94: 795–802. MEDLINE|
|3.||Torbicki A, Perrier A, Konstantinides S, et al.: Guidelines on the diagnosis and management of acute pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). Eur Heart J 2008; 29: 2276–315. MEDLINE|
|4.||Konstantinides S, Janssens U, Mayer E, Hasenfuß G: Kommentar zu den ESC-Leitlinien „Guidelines on Diagnosis and Management of Acute Pulmonary Embolism“. Kardiologe 2009; 3: 272–82.|
|5.||Interdisziplinäre S2-Leitlinie: Lungenembolie (LE). Vasa 2005; 34: 15–24.|
|6.||Walther A, Böttiger BW: Lungenembolie. Wien Med Wochenschr 2008; 158: 610–4. MEDLINE|
|7.||Goldhaber SZ: Pulmonary embolism. Lancet 2004; 363: 1295–305. MEDLINE|
|8.||Walther A, Böttiger BW: Lungenembolie. Notfallmedizin up2date 2007; 2: 57–70.|
|9.||Anderson FA Jr., Spencer FA: Risk factors for venous thromboembolism. Circulation 2003; 107: 9–16. MEDLINE|
|10.||Schulman S: Current strategies in prophylaxis and treatment of venous thromboembolism. Ann Med 2008; 40: 352–9. MEDLINE|
|11.||Goldhaber SZ, Visani L, De Rosa M: Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet 1999; 353: 1386–9. MEDLINE|
|12.||Konstantinides S: Clinical practice. Acute pulmonary embolism. |
N Engl J Med 2008; 359: 2804–13. MEDLINE
|13.||Wells PS, Anderson DR, Rodger M, et al.: Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000; 83: 416–20. MEDLINE|
|14.||Stein PD, Fowler SE, Goodman LR, et al.: Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006; 354: 2317–27.|
|15.||van Belle A, Buller HR, Huisman MV, et al.: Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA 2006; 295: 172–9. MEDLINE|
|16.||Perrier A, Roy PM, Sanchez O, et al.: Multidetector-row computed tomography in suspected pulmonary embolism. N Engl J Med 2005; 352: 1760–8. MEDLINE|
|17.||Sanchez O, Trinquart L, Colombet I, et al.: Prognostic value of right ventricular dysfunction in patients with haemodynamically stable pulmonary embolism: a systematic review. Eur Heart J 2008; 29: 1569–77. MEDLINE|
|18.||Becattini C, Vedovati MC, Agnelli G: Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation 2007; 116: 427–33. MEDLINE|
|19.||Grifoni S, Olivotto I, Cecchini P, et al.: Short-term clinical outcome of patients with acute pulmonary embolism, normal blood pressure, and echocardiographic right ventricular dysfunction. Circulation 2000; 101: 2817–22. MEDLINE|
|20.||Tulevski II, ten Wolde M, van Veldhuisen DJ, et al.: Combined utility of brain natriuretic peptide and cardiac troponin T may improve rapid triage and risk stratification in normotensive patients with pulmonary embolism. Int J Cardiol 2007; 116: 161–6. MEDLINE|
|21.||Vuilleumier N, Le Gal G, Verschuren F, et al.: Cardiac biomarkers for risk stratification in non-massive pulmonary embolism: a multicenter prospective study. J Thromb Haemost 2009; 7: 391–8. MEDLINE|
|22.||Klok FA, Mos IC, Huisman MV: Brain-type natriuretic peptide levels in the prediction of adverse outcome in patients with pulmonary embolism: a systematic review and meta-analysis. Am J Respir Crit Care Med 2008; 178: 425–30. MEDLINE|
|23.||Nolan JP, Deakin CD, Soar J, Böttiger BW, Smith G: European Resuscitation Council guidelines for resuscitation 2005. Section 4. Adult advanced life support. Resuscitation 2005; 67(Suppl 1): 39–86. MEDLINE|
|24.||The PREPIC study group: Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study. Circulation 2005; 112: 416–22. MEDLINE|
|25.||Hoppe H: Optional vena cava filters: indications, management, and results [Optionale Vena-cava-Filter: Indikationen, Management und Ergebnisse]. Dtsch Arztebl Int 2009; 106(24): 395–402. VOLLTEXT|
|e1.||Dalen JE: Pulmonary embolism: what have we learned since Virchow? Natural history, pathophysiology, and diagnosis. |
Chest 2002; 122: 1440–56. MEDLINE
|e2.||Kearon C: Natural history of venous thromboembolism. Circulation 2003; 107: I22–30. MEDLINE|
|e3.||Moerchel C, Kroeger K: Prophylaxe tiefer Bein- und Beckenvenenthrombose. Dtsch Arztebl 2007; 104: A 2886–93. VOLLTEXT|
|e4.||Dörffler-Melly J, Amann-Vesti B: Diagnostik und Management der akuten Lungenembolie. Herz 2007; 32: 35–41. MEDLINE|
|e5.||Dempfle CE: Bestimmung des D-Dimer-Antigens in der klinischen Routine. Dtsch Arztebl 2005; 102(7): A 428–32. VOLLTEXT|
|e6.||Dindo D, Breitenstein S, Hahnloser D, et al.: Kinetics of D-dimer after general surgery. Blood Coagul Fibrinolysis 2009; 20: 347–52. MEDLINE|
|e7.||Aujesky D, Roy PM, Le Manach CP, et al.: Validation of a model to predict adverse outcomes in patients with pulmonary embolism. Eur Heart J 2006; 27: 476–81. MEDLINE|
|e8.||Kasper W, Konstantinides S, Geibel A, Tiede N, Krause T, Just H: Prognostic significance of right ventricular afterload stress detected by echocardiography in patients with clinically suspected pulmonary embolism. Heart 1997: 346–9. MEDLINE|
|e9.||Puls M, Dellas C, Lankeit M, et al.: Heart-type fatty acid-binding protein permits early risk stratification of pulmonary embolism. Eur Heart J 2007; 28: 224–9. MEDLINE|
|e10.||2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005; 112: IV1–203. MEDLINE|
|e11.||Van de Werf F, Ardissino D, Betriu A, et al.: Management of acute myocardial infarction in patients presenting with ST-segment elevation. The Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J 2003; 24: 28–66. MEDLINE|
|e12.||Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ: Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th ed.). Chest 2008; 133: 454–545. MEDLINE|