Pulmonary embolism (PE) is a common clinical problem characterized by the deposition and embolization of a venous clot. Collectively, deep vein thrombosis (DVT) and PE are referred to as venous thromboembolism (VTE). Patients with PE typically have symptoms related to ventilation-perfusion mismatch and increased pulmonary artery pressures. These abnormalities can lead to hypoxemia and right ventricular strain/failure. Because of the high potential for associated mortality, the diagnosis of PE should be considered by the intensivist confronted with acute pulmonary or cardiovascular failure.
Epidemiology and Natural History
The annual estimated incidence of PE in the United States is 112 per 100,000 adults. The incidence is significantly higher in men, and incidence and mortality increase with advancing age. Mortality rates for PE remain high; data from the International Cooperative Pulmonary Embolism Registry indicate a mortality rate approaching 15% among hemodynamically stable patients and 60% in hemodynamically unstable patients.
Most often, PE arises from DVTs that embolize after 3 to 7 days. In approximately 70% of patients with PE, DVT can be found in the lower limbs. The initial studies on the natural history of VTE were intraoperative assessment during orthopedic surgery. In this setting, DVT of the calf or more proximal venous system was found in approximately 30% of patients. DVT resolved spontaneously after a few days in approximately one third of patients and did not extend in approximately 40%. However, in 25%, the clot evolved into proximal DVT and PE. Major risk factors for the development of VTE are listed in Table 26-1 . Even temporary immobilization for 1 to 2 days will significantly increase the risk of DVT.
|
PE presents with shock or hypotension in 5% to 10% of patients. In some patients without shock, there are signs of right ventricular dysfunction or injury. This abnormality is associated with poorer prognosis.
PE is difficult to diagnose because of the nonspecific clinical presentation or complete lack of symptoms. Among patients with proximal DVT who have lung scans, approximately 50% will have associated, usually clinically asymptomatic, PE.
Pathophysiology
The initial clinical consequences of acute PE are primarily hemodynamic and become apparent when more than 30% to 50% of the pulmonary arterial bed is occluded by thromboemboli. Large or multiple emboli can acutely increase pulmonary vascular resistance. The resultant increased afterload often cannot be overcome by the right ventricle (RV) because a nonpreconditioned, thin-walled RV cannot generate mean pulmonary arterial pressures exceeding approximately 40 mm Hg. Resultant underfilling of the left ventricle (LV) decreases blood pressure and coronary blood flow. The combination of increased RV myocardial workload and a decreased RV coronary perfusion gradient (decreased systemic diastolic pressure – increased intraventricular pressure) contributes to RV ischemia. This ischemia worsens RV dysfunction and may initiate a vicious cycle that can ultimately result in pulseless electrical activity and sudden cardiac death.
In up to one third of patients, right-to-left shunt through a patent foramen ovale may contribute to severe hypoxemia and will also increase the risk for systemic embolization. Ventilation-perfusion mismatch occurs in most cases. Vasoactive mediators such as serotonin released from ischemic lung tissue may exacerbate ventilation perfusion mismatch.
Diagnosis
Evaluating the likelihood of PE in an individual patient on the basis of the clinical presentation is the first and most important step to select an appropriate diagnostic strategy and interpret diagnostic test results.
Clinical Presentation
Suspicion of PE should accompany clinical symptoms such as dyspnea, chest pain, or syncope. These abnormalities are present in more than 90% of patients with PE. The likelihood of PE increases with the number of risk factors present. However, in approximately 30% of cases, PE occurs in the absence of any risk factor. Individual clinical signs and symptoms are not usually helpful because they are neither sensitive nor specific.
Other symptoms include cough and blood-tinged sputum. Signs include fever, tachycardia, tachypnea, cyanosis, and coarse breath sounds. PE is generally associated with hypoxemia. However, up to 20% of patients with PE have a normal arterial oxygen pressure (Pa o 2 [partial pressure of oxygen, arterial) and a normal alveolar-arterial oxygen gradient. Auscultation may reveal a new fourth heart sound or accentuation of the pulmonic component of the second heart sound.
Electrocardiography may reveal new evidence of right ventricular strain, tachycardia, or atrial fibrillation. Electrocardiographic signs of RV strain include inversion of T waves in leads V 1 to V 4 , a QR pattern of the classic S 1 Q 3 T 3 type in the lead V 1 , and an incomplete or complete right bundle-branch block. Electrocardiographic changes are generally associated with the more severe forms of PE, and lack of electrocardiographic changes does not exclude PE.
The chest radiograph is usually abnormal, with the most frequently encountered findings (platelike atelectasis, pleural effusion, or elevation of a hemidiaphragm) being nonspecific. However, the chest radiograph is useful in excluding other causes of dyspnea and chest pain.
On the basis of clinical presentation or lack thereof, PE can be divided into three groups: hemodynamically unstable, hemodynamically stable and symptomatic, and asymptomatic and silent with incidental finding.
Hemodynamically Unstable Group
This group includes patients with shock or severe hypotension associated with RV dysfunction and injury. These patients require rapid, specific diagnosis and therapy because of the high mortality risk (short-term mortality >15%).
Any intensive care unit (ICU) patient who is at risk for PE and is hemodynamically unstable should be evaluated for acute right ventricle failure and thrombus in the right ventricle or main pulmonary artery. Acute heart failure is not specific for PE; therefore other causes must be considered. The main therapeutic goal is to rapidly restore flow through the pulmonary circulation.
Hemodynamically Stable and Symptomatic Group
This group of patients can be divided into intermediate- and low-risk subgroups. Intermediate-risk PE is diagnosed when the patient has either RV dysfunction or myocardial injury. Indicators of RV dysfunction include (1) elevated right ventricular pressures and RV dilation, (2) hypokinesis, or (3) pressure overload on echocardiography. Elevations of cardiac troponin T or I indicate RV injury. Initial therapy is aimed at the prevention of further pulmonary thromboembolism.
Asymptomatic and Silent Group with Incidental Finding
Mild, untreated PEs carry a lower immediate mortality than recurrent PEs. Because of the intrinsic fibrinolytic activity of the lung, small PEs usually resolve spontaneously. Withholding anticoagulation treatment in nonmassive PE is an acceptable strategy for patients who have an indeterminate ventilation-perfusion study, negative serial lower extremity venous examination results, adequate cardiopulmonary reserve, and relative-to-absolute contraindications to anticoagulation treatment. The rationale for this approach is based on synthesis of the results of several studies. The optimal management of patients with asymptomatic PE has not been prospectively studied.
Diagnosis
Clinical prediction scores have been widely used, but they do not have the necessary sensitivity and specificity to be used without diagnostic tools. Several modalities are available for confirmation or exclusion of the diagnosis of PE. Laboratory studies, including arterial blood gas measurements, are nonspecific and generally unreliable. Often, but not always, the arterial blood gas will demonstrate hypoxemia and respiratory alkalosis. In one study, the average Pa o 2 in patients with PE was 72 ± 16 mm Hg, as opposed to patients without PE, for whom the Pa o 2 was 70 ± 18 mm Hg. In addition, up to 20% of patients with PE had a Pa o 2 in the normal range, and the alveolar to arterial oxygen gradient was not found to be helpful because there was an average difference of only 2 mm Hg.
Although a negative serum d dimer may be used to rule out PE, the results of this test are often of limited utility in the intensive care population. Patients with malignancy, those who are hospitalized, and pregnant women demonstrate reduced specificity with d -dimer testing. Patients with either low or moderate pretest probability and a negative d dimer have no need to undergo any further testing. However, those with positive tests or high clinical probability will require further investigation because a negative d dimer does not exclude PE in more than 15% of patients with high clinical probability. Furthermore, d dimer is neither sensitive nor specific in the postoperative period.
The use of troponins and brain natriuretic peptide (BNP), often elevated in moderate or large PE, may be useful prognostic tests. In one study, normal levels of BNP had a 100% negative predictive value for hemodynamically stable patients. Elevation of troponins are generally associated with right ventricular dysfunction and ischemia; therefore they are associated with worse outcome.
Diagnostic Tools
Because chest radiography is neither sensitive nor specific, the literature describes two modalities used in the diagnosis of PE: perfusion lung scans (V/Q scans) and computed tomography (CT) pulmonary angiography. The ease and speed of acquiring a CT scan make it the most widely used diagnostic tool for patients with suspected PE.
V/Q scans have been used to detect the presence of perfusion defects within the patient’s pulmonary circulation. The major advantage of V/Q scans is the avoidance of nephrotoxic radiographic contrast. In the PIOPED (Prospective Investigation of Pulmonary Embolism Diagnosis) study, 755 patients underwent V/Q scans and selective pulmonary angiography within 24 hours of the symptoms that suggested PE. Thirty-three percent of the patients had angiographic evidence of PE. Almost all patients with PE (98%) had abnormal V/Q scan findings. Thus, V/Q scans are highly sensitive for acute PE. However, although PE was documented by angiography in 88%, only 41% of the patients with PE had a high-probability scan. Most patients with PE (75%) had an intermediate- or low-probability scan. Thus, specificity was low. In postoperative patients with significant atelectasis, consolidation, or PE, the negative predictive value is low. The V/Q scan is the study of choice for pregnant patients to avoid unnecessary radiation exposure.
High-resolution multidetector computed tomography (MDCT) has replaced the V/Q scan as the study of choice to diagnose PE. CT scanning is widely available, can be performed rapidly, and provides clear anatomic and pathologic lung images (so that the clinician often obtains a diagnosis despite negative results from an angiographic examination) and the ability to concurrently evaluate potential embolic sources in the legs or pelvis. Four-slice MDCT scans have an increased sensitivity for subsegmental PE. In two studies comprising approximately 100 patients, sensitivities for the detection of PE with four-slice CT angiography were reported to be 96% and 100%, with respective specificities of 98% and 89%. The combination of arterial-phase and venous-phase CT angiography appears more sensitive (90%) and specific (96%) than the arterial phase alone. Postoperative patients with high clinical suspicion of PE and a negative MDCT scan should undergo lower extremity ultrasonography. Patients with impaired renal function should undergo hydration before administration of contrast and preferably receive nonionic contrast. Alternatively, these patients may undergo pulmonary scintigraphy, venous ultrasound, or magnetic resonance imaging.
However, if hemodynamic instability is present, echocardiography should be performed to evaluate right ventricular function. Right ventricular dysfunction is associated with increased mortality, especially in patients with hemodynamic instability. During the diagnostic and treatment phase, echocardiography may assist clinical decision-making, although only 30% to 40% of patients have any echocardiographic abnormalities. Evidence of right ventricular failure, such as severe hypokinesis, dilatation, or the McConnell sign (severe hypokinesis of the free wall with preserved apical function), may prompt an immediate surgical or catheter-based thrombectomy. Increased tricuspid regurgitation, chamber dilatation, and septal shift are suggestive of volume-pressure overload. If transesophageal echo is performed, then emboli may be seen in the main pulmonary arteries.
Treatment
The immediate priority is stabilization of the patient who is compromised by hemodynamic or respiratory instability. In some cases, severe hypoxemia and respiratory failure may require supplemental oxygen or mechanical ventilation.
Without treatment, mortality from hemodynamically unstable PE approaches 30%. In treated patients, the overall mortality decreases to 15%. The treatment of PE in the postoperative patient is complicated by the inherent potential for bleeding with therapeutic anticoagulation (TAC) and thrombolytics.
For acute PE, the options for treatment include TAC, inferior vena cava (IVC) filter placement to prevent continued embolization from the lower extremities, clot thrombolysis, and surgical or catheter embolectomy. Hemodynamically stable patients diagnosed with PE should receive TAC with intravenous unfractionated heparin (UFH) or subcutaneous low-molecular-weight heparin (LMWH). The risk of major bleeding from initiation of TAC is less than 3%. Meta-analyses have shown that LMWH treatment, when adjusted to body weight, is at least as effective and safe as dose-adjusted UFH. However, in postoperative and critically ill patients and in patients in whom epidural catheters have been placed, the shorter half-life and reversibility of intravenous UFH provides a safety buffer over LMWH. Furthermore, LMWH should be avoided in patients with a creatinine clearance less than 30 mL/min because of renal excretion. Therefore, despite the absence of randomized prospective trials, when there is a risk for clinically significant bleeding, UFH may be safer. In patients with moderate clinical suspicion, TAC should be started if the diagnostic evaluation is expected to exceed 4 hours; if suspicion is low, then TAC should be started if evaluation delay is greater than 24 hours. As described previously, heparin should be adjusted to the goal activated partial thromboplastin time (aPTT), and anti–factor Xa levels should be checked if the patient requires large doses of UFH without achieving therapeutic aPTT. Treatment duration with anticoagulation is often a minimum of 3 months ranging to indefinite depending on the risk factors. Patients who have had PE are at high risk of recurrence, especially those with hypercoagulable states such as malignancy or inherited thrombophilic disorders such as protein C and S deficiency. Traditionally, patients with VTE have been transitioned to vitamin K antagonists, but newer anticoagulants, such as direct thrombin inhibitors and factor Xa inhibitors, are currently under investigation for use in long-term anticoagulation. The safety of these agents in the immediate postoperative period is unclear. The advantage of these agents includes a stable dosing regimen with reliable anticoagulation.
Patients who cannot undergo anticoagulation (such as those with intracranial bleeding) commonly have an IVC filter placed as soon as possible to prevent further embolization. Again, although this approach is logical, IVC filters have not been shown to increase overall survival.
After the success of thrombolytics in the management of acute myocardial infarction, thrombolysis has been proposed as therapy for massive PE. Commonly used thrombolytic agents include tissue plasminogen activator, streptokinase, and urokinase. Alternative thrombolytic agents include lanoteplase, tenecteplase, and reteplase. These agents all convert plasminogen to plasmin, which in turn breaks down fibrin and promotes clot lysis. A recent meta-analysis comprising 16 randomized trials including 2115 patients reported a lower mortality in patients treated with thrombolytics (2.2% vs 3.9%). Unfortunately, major bleeding rates (9.2% vs 3.4%) and intracranial hemorrhage (1.5% vs 0.2%) were significantly higher in patients receiving thrombolytic therapy when compared with TAC. Unfortunately, this meta-analysis pooled trials with different thrombolytic agents and dosing regimens, making it difficult to conclude which agent or dose should be used. Almost half of the patients in the meta-analysis came from a large multicenter trial (PEITHO [Pulmonary Embolism Thrombolysis]) comparing thrombolytics and heparin with placebo and heparin for intermediate-risk PE in normotensive patients with evidence of RV dysfunction. Despite a reduction in 7-day mortality in the thrombolytics group, the difference in 30-day mortality did not reach statistical significance. Furthermore, the incidence of intracranial and major hemorrhage (11.5% vs 2.4%) was significantly higher in patients receiving thrombolysis. Until further evidence emerges, thrombolytics should be reserved for hemodynamically unstable patients who will not tolerate thrombectomy. Moreover, thrombolysis cannot be recommended for patients with recent major surgery, intracranial lesions, or traumatic injury. Relative contraindications include recent major bleeding, pregnancy, and uncontrolled hypertension.
Pulmonary embolectomy has been performed in patients with massive PE, in those who are hemodynamically unstable despite heparin and fluid resuscitation, and in poor candidates for thrombolysis. Patients with life-threatening PEs may be placed on extracorporeal membrane oxygenation for stabilization and taken to the operating room for open thrombus extraction. No prospective clinical trials have evaluated outcomes from surgical embolectomy. All available data consist of case report and case series. The largest series of pulmonary embolectomies at one institution was reported by Meyer and colleagues in Paris in 1991. During a 20-year period from 1968 to 1998, 96 (3%) of 3000 patients with confirmed PE underwent pulmonary embolectomy under cardiopulmonary bypass. The overall hospital mortality rate was 37.5%. A recent series comparing surgical thrombectomy to thrombolytics reported an early mortality rate of 3.6% in 28 surgical patients, but patients undergoing surgical embolectomy after failed thrombolysis had a mortality rate of 27%. In general, embolectomy is considered a therapy of last resort and should not be considered for most patients with PE.
Several catheter-based embolectomy techniques are available and can be categorized as thrombus fragmentation with pigtail or balloon catheter, rheolytic embolectomy with a hydrodynamic catheter, suction embolectomy with an aspiration catheter, and rotational thrombectomy. None of these techniques have been compared in randomized controlled trials (RCTs) to surgical thrombectomy or thrombolytics. Catheter-based embolectomy should be considered in patients failing thrombolysis as an alternative to surgical thrombectomy as dictated by experience and available expertise.
Related posts:
Should Fever Be Treated?
Should Exacerbations of COPD Be Managed in the Intensive Care Unit?
Are Anti-inflammatory Therapies in ARDS Effective?
Has Outcome in Sepsis Improved? What Has Worked? What Has Not Worked?
What Is the Role of Autonomic Dysfunction in Critical Illness?
How Should Acid-Base Disorders Be Diagnosed and Managed?
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
