Pulmonary Embolism

Jacob T. Gutsche

William C. Hanson III

CASE SUMMARY

A 72-year-old woman with a history of heart failure and diabetes mellitus presents for a hip arthroplasty after fracturing her hip during a fall. The patient undergoes surgical repair without complications and is admitted to a telemetry bed in the hospital for postoperative care. The patient remains in the hospital until postoperative day 5 due to difficulties in titrating the patient’s medications for heart failure and diabetes. On postoperative day 5, the physician caring for the patient is called to emergently evaluate the patient who is exhibiting tachycardia, hypotension, and tachypnea. An ECG shows ST depression in the right-sided leads, and an echocardiogram is emergently performed, which shows severe right heart dilation with new severe tricuspid regurgitation. The patient is presumed to have a high likelihood for a pulmonary embolism (PE) and is immediately anticoagulated with heparin. A stat computed tomography (CT) angiogram confirms the diagnosis, and the patient remains on anticoagulation for 6 months.

What Are the Facts Concerning Pulmonary Embolism?

PE is one of the most dreaded complications in the perioperative period. Many patients are at increased risk for some form of PE by virtue of their procedure or preexisting illnesses. There are multiple types of PE, including venous thromboembolic (blood clot), venous air, fat or debris, and amniotic fluid. While the source of the emboli and timing of symptoms depends on the type of PE, there are many similarities in the morbidity seen secondary to PE. Patients who develop PE typically develop some degree of ventilation/perfusion mismatch and increased pulmonary arterial pressures, which can lead to right heart strain. Because of the high potential for mortality with each of the different types of pulmonary embolic disorders, this diagnosis should be considered by the anesthesiologist confronted with acute pulmonary or cardiovascular collapse. This chapter will define the epidemiology, etiology, diagnoses, and treatment of each of these four types of PE.

What Conditions Predispose to Venous Thromboembolism?

ETIOLOGY

Thromboembolic PE are blood clots or thrombi that develop in the deep veins, typically of the legs or pelvis, and embolize to the pulmonary arterial system. The spectrum of patient presentations ranges from completely asymptomatic to fulminant cardiovascular collapse. The amount of clot and the patient’s physiologic reserve are the primary factors that dictate the severity of the symptoms and signs seen. A large clot burden can obstruct the pulmonary artery and cause right heart strain or failure, as well as severe ventilatory/perfusion mismatch. In addition, PE is often accompanied by the release of vasoactive mediators (thromboxane and serotonin), which also increase pulmonary artery pressures. Intraoperative thromboembolic PE is unusual: While there are multiple case reports of acute massive PE in patients with occult lower extremity deep vein thromboses (DVT) who underwent intraoperative manipulation of the lower extremity,¹ thromboembolic pulmonary emboli are far more likely to present in the postoperative period.

The physiologic factors that increase the risk of developing DVT were originally proposed by Rudolf Virchow in 1860 and include venous stasis, damage to the vessel wall, and hypercoagulability. In the surgical patient, venous stasis occurs upon the induction of general anesthesia because of the vasodilatory effects of the drugs administered and the immobility of extremities under anesthesia. A hypercoagulable state is then induced by the stress response resulting from the surgical intervention.²,³ Finally, patients may experience direct damage to the venous endothelium during surgery. All these conditions can lead to the formation of clot in the calf veins, which propagates proximally to larger veins in the leg (popliteal, femoral, or iliac). Thrombus in these large proximal leg veins are more likely to cause physiologically significant PE than emboli from the smaller calf veins.

DIAGNOSIS

The diagnosis of PE is confirmed by imaging studies in patients with clinical signs and symptoms. The gold standard for diagnoses is a pulmonary angiogram. Because it is an invasive procedure, and its availability is limited in many hospitals—and reasonable alternatives have been developed—the angiogram is not typically employed. Patients are evaluated by history and physical examination. Signs and symptoms aid in formulating a pretest probability of the likelihood of PE. Symptoms include dyspnea, chest pain, cough, and blood-tinged sputum. Signs include fever, tachycardia, tachypnea, and coarse breath sounds. Auscultation may yield a new fourth heart sound or accentuation of the pulmonic component of the second heart sound.

Laboratory Testing

Laboratory tests (including arterial blood gases) are not typically useful for diagnosing or ruling out PE in the perioperative period. Patients with a normal PACO₂ may still have a PE,⁴ and patients with a high FiO₂/PaO₂ gradient could have one of many diagnoses. D-dimer, a degradation product of cross-linked fibrin, has been studied for the diagnosis of PE. Two common D-dimer tests used are the enzyme-linked immunosorbent assays and the latex agglutination assays. Neither test is helpful in the postoperative period because the D-dimer is very likely to be positive even in the absence of PE.⁵

Imaging Studies

Imaging studies are the modality of choice for diagnosing PE. The [V with dot above]/[Q with dot above] scan is useful for patients with renal insufficiency, since nephrotoxic, intravenous contrast is not used in nuclear medicine scans. An algorithm designed by the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) investigators has simplified the diagnostic approach to these patients.⁶ By assessing the clinical signs and symptoms, the patient is given a pretest probability (low, medium, or high) of having a PE before performing the [V with dot above]/[Q with dot above] scan. The results of a [V with dot above]/[Q with dot above] scan range from a normal to high probability of PE. Combining the pretest probability with the results from a [V with dot above]/[Q with dot above] scan help successfully diagnose or exclude PE.⁷ Previously, patients with a low probability scan (indeterminate) would undergo a pulmonary angiogram for further workup, but now, their chances of having a PE are considered minimal.

Computed Tomography

Helical CT has replaced the ventilation/perfusion scan ([V with dot above]/[Q with dot above] scan) as the study of choice in many hospitals for PE evaluation⁴ because of the speed of the study and the ability to concurrently evaluate potential embolic sources in the legs or pelvis. The results of studies that have evaluated the helical CT have shown sensitivities up to 90% with single detector CT scans.⁸ Multidetector row CT scans have an increased sensitivity for subsegmental PE⁹ but, in postoperative patients, should still be used in conjunction with pretest probability and ultrasonography of the lower extremities.¹⁰

The PIOPED II investigators have recently investigated the use of clinical pretest probability in conjunction with either CT angiogram or combined CT angiogram and CT venography.¹¹ These investigators concluded that the CT angiogram combined with CT venography has a higher sensitivity for venous thromboembolism (VTE). In addition, the pretest probability should still be used in the diagnostic algorithm for PE. The PIOPED II investigators did note that a negative CT angiogram does not rule out a subsegmental PE; studies suggest that it is safe to withhold anticoagulation in patients with low or intermediate pretest probability.⁹,¹² The PIOPED pretest probability is based on the Wells criteria, which uses a scoring system based on signs and symptoms that include DVT, tachycardia, immobilization, and recent surgery among others.⁷ All postsurgical patients would then, by definition, be considered at least intermediate pretest probability, and thus a physician’s global judgement should be used in the postsurgical patient. For diagnosing a PE with a CT scan (see Fig. 13.1), a pretest probability is assessed, and only patients with a high pretest probability and a negative CT scan should have compression ultrasonography of the lower extremities to rule out VTE.

EPIDEMIOLOGY

In addition to the surgical and anesthetic factors, multiple other preexisting conditions place patients at a higher risk for developing a DVT and potential PE in the perioperative period¹³ (see Table 13.1). While the lower extremities are more frequently the cause of pulmonary emboli, patients with upper extremity DVT had a 9% incidence of symptomatic PE in one study.¹⁴ Upper extremity venous thrombosis accounts for about 10% of DVTs and usually occurs as a result of central venous catheters or in patients with cancer. Rarely, DVT can occur spontaneously in the upper extremity, known as Paget-Schroetter syndrome. Clinically, patients present with swelling, erythema, and arm pain affected by the DVT. Ninety percent of pulmonary emboli emanate from extremity (upper and lower) DVT, with the remaining pulmonary emboli arising from pelvic veins, renal veins, inferior vena cava (IVC), and the heart.

Perioperatively, patients are at higher risk for a DVT with prolonged duration of surgery and immobilization,¹⁵ preexisting thrombophilia, and malignancy. One large study used an adminstrative database to estimate the incidence of VTE (within 91 days) in patients undergoing specific types of surgery.¹⁶ This study did not analyze thromboprophylactic practice and excluded patients with a previous diagnosis of VTE. Different thromboprophylactic practices in the groups studied may alter the rate of VTE and change the interpretation of the risk of VTE from a surgery¹⁶ (see Table 13.2).

FIGURE 13.1 Diagnostic algorithm for patients with suspected pulmonary embolism. PE, pulmonary embolism; VTE, venous thromboembolism; CUS, compression ultrasound; CT, computed tomography.

PREVENTION/TREATMENT

The primary mode of preventing postoperative PE is to prevent the formation of DVT. Frequently used therapies include unfractionated heparin, low molecular weight heparin (LMWH), intermittent pneumatic compression (IPC) devices, and elastic stockings. The approach to preventative therapy for postoperative VTE should be based on the patient’s risk for developing VTE¹⁷ as shown in Table 13.3.

TABLE 13.1 Risk Factors for Deep Venous Thrombosis

Advancing age

Obesity

Previous venous thromboembolism^a

Trauma^a

Neoplasm^a

Respiratory failure

Infection

Inflammatory bowel disease

Antiphospholipid syndrome^a

Dyslipoproteinemia

Nephrotic syndrome

Paroxysmal nocturnal hemoglobinuria

Myeloproliferative diseases

Behcets syndrome

Varicose veins

Superficial vein thrombosis

Congenital venous malformation

Long distance travel

Prolonged bed rest

Immobilization

Limb paresis

Pregnancy

Oral contraceptives

Hormone replacement therapy

Heparin-induced thrombocytopenia^a

Chemotherapy

Tamoxifen

Thalidomide

Antipsychotics

Central venous catheter^a

Vena cava filter

Intravenous drug abuse

^aSignifies major risk factor.

Data from: Kyrle PA, Eichinger S. Deep vein thrombosis. Lancet 2005;365:1163.

Antiplatelets

Aspirin and other antiplatelet agents are not as effective as fractionated or unfractionated heparin⁸ for DVT prophylaxis. There is a risk of significant bleeding with aspirin, and recent studies supporting its use are limited by sample size and methods of screening for DVT.¹⁸

Compression Devices

In addition to anticoagulants, pneumatic compression devices and elastic stockings may be used to prevent DVT. These are especially useful in the patient who is considered a bleeding risk and in whom anticoagulants are contraindicated. Pneumatic compression devices enhance venous blood flow in the lower extremities and reduce levels of plasminogen activator inhibitor-1.¹⁹ Contraindications to pneumatic compression devices include a diagnosis of DVT in the extremity to be compressed. Several studies support the use of pneumatic compression devices in the general surgical patient.²⁰,²¹,²²,²³

TABLE 13.2 Surgical Procedure Related to Risk of Venous Thromboembolism

Surgical Procedure	Incidence Vte (%)
NEUROSURGERY
Excision/destruction/biopsy brain	2.3
Spinal cord surgery	0.5
HEAD AND NECK
Sinus surgery	0.2
Thyroid or parathyroid surgery	0.1
CARDIAC OR THORACIC
Coronary artery bypass grafting	1.1
Valve replacement	0.5
VASCULAR
Abdominal aortic surgery	1.7
Head and neck endartectomy	0.2
GASTROINTESTINAL
Splenectomy	1.6
Excision of small bowel	1.5
Exploratory laparotomy	0.7
UROLOGIC
Nephrectomy	0.4
GYNECOLOGIC SURGERY
Total abdominal hysterectomy	0.3
ORTHOPEDIC
Total hip arthroplasty	2.4
Total knee arthroplasty	1.9
Shoulder arthroplasty	0.5
VTE, venous thromboembolism.
Data from: White RH, Zhou H, Romano PS. Incidence of symptomatic venous thromboembolism after different elective or urgent surgical procedures. Thromb.Haemost. 2003;90:446.

Regional Anesthesia

Regional anesthesia has been studied as a means for lowering postoperative VTE risk. Whereas epidural anesthesia does lower the incidence of intraoperative VTE,²⁴,²⁵ epidural analgesia does not appear to decrease this risk. When considering the choice of epidural or spinal anesthesia, the patient’s coagulation status must be evaluated. Patients who are anticoagulated when a neuraxial (spinal or epidural) procedure is initiated are at risk for perispinal hematoma. This rare but serious complication of neuraxial procedures is preventable by a working knowledge of when to discontinue and restart anticoagulants²⁶ (see Table 13.4).

TABLE 13.3 Preventive Therapy Related to Risk of Developing Postoperative Venous Thromboembolism

Level of Risk^a		Prevention Therapy
LOW		Early mobilization
	Minor surgery
	Age <40 y
	No risk factors
MODERATE		UFH q2h, LMWH, IPC
	Minor surgery (patient with risk factors)
	Age 40-60 y (patient with no risk factors)
	Major surgery (age <40, no risk factors)
HIGH		UFH q8h, LMWH, IPC
	Nonmajor surgery (age >60 or risk factors)
	Major surgery (age >40 or risk factors)
HIGHEST		LMWH, adjust heparin dose
	Major surgery (age >40 and major risk factors^b)	LMWH, adjust heparin dose
^aRisk factors are as shown in Table 13.1.^bMajor risk factors; prior VTE, malignancy, hip or knee replacement, major trauma, hip fracture surgery, spinal cord injury, diagnosis of thrombophilia.
UFH, unfractionated heparin; LMWH, low molecular weight heparin; IPC, intermittent pneumatic compression devices. Data from: Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest. 2001;119:132S.