Nuclear medicine provides several valuable tools for the assessment of the venous vasculature. Most clinicians are familiar with the lung ventilation/perfusion scan (V/Q) for diagnosis of acute pulmonary embolism (PE), but some changes have been made in how this test is used or even interpreted. Small studies using hybrid single-photon emission computed tomography/computed tomography (SPECT/CT) technology with addition of low-dose CT to V/Q SPECT have shown improvement in the diagnostic accuracy of V/Q scintigraphy in the assessment of suspected PE. Other nuclear medicine examinations assessing the venous system itself, such as for acute deep venous thrombosis (DVT), have failed to achieve widespread use.

Acute PE is a common and frequently fatal condition. PE refers to obstruction of the pulmonary artery or one of its branches by thrombus originating elsewhere in the body. Other materials such as tumor, air, or fat are also well-known causes of PE. PE is associated with a mortality rate of approximately 30% without treatment, primarily the result of recurrent embolism. However, accurate diagnosis followed by effective therapy with anticoagulants decreases the mortality rate to between 2% and 8%.

The clinical presentation of PE is variable and nonspecific, which makes its accurate diagnosis difficult. Risk factors include immobilization; surgery within the past 3 months; stroke; history of venous thromboembolism; malignancy; central venous instrumentation within the past 3 months; chronic heart disease; obesity; heavy cigarette smoking (>25 cigarettes per day); deficiency of antithrombin III, protein C, protein S; and factor V Leiden mutation. PE can be classified as acute or chronic. Patients with acute PE typically develop symptoms and signs immediately after obstruction of the pulmonary vessels. In contrast, those with chronic PE tend to develop slowly progressive dyspnea over a period of years because of pulmonary hypertension. In the Prospective Investigation of Pulmonary Embolism Diagnosis II (PIOPED II) study, the most common symptoms associated with PE were dyspnea at rest or with exertion (73%), pleuritic pain (44%), cough (34%), more than two-pillow orthopnea (28%), calf or thigh pain (44%), calf or thigh swelling (41%), and wheezing (21%). The most common signs were tachypnea (54%), tachycardia (24%), rales (18%), decreased breath sounds (17%), an accentuated pulmonic component of the second heart sound (15%), and jugular venous distension (14%). Circulatory collapse was uncommon (8%). Massive PE may be accompanied by acute right ventricular failure. Symptoms or signs of lower extremity DVT were common (47%).¹ Routine laboratory tests, including D-dimer assay, which has high sensitivity and negative predictive value but poor specificity and positive predictive value, are nonspecific. The radiologic diagnostic tests commonly used in the evaluation of patients with suspected PE include chest radiography, radionuclide V/Q scan, CT pulmonary arteriography, and pulmonary angiography. Pulmonary angiography is considered the definitive diagnostic modality or “gold standard” in the diagnosis of acute PE, with a sensitivity of 98% and a specificity of 97%. However, it is infrequently done because it is invasive and expensive compared with the other modalities.

Chest radiographic abnormalities are common in patients with PE. However, they are usually diagnostically unhelpful because of their lack of specificity. In a large, prospective study,² atelectasis or a pulmonary parenchymal abnormality was noted in 69% and 58% of patients with and without PE, respectively; pleural effusion was detected in 47% and 3% of patients with and without PE, respectively; and about 12% of the chest radiographs in patients with pulmonary emboli were interpreted as normal. Various signs described, including Westermark’s sign (attenuation of the pulmonary vessels distal to the embolus), the Fleishner sign (prominence of the proximal pulmonary artery from the embolus), and Hamptons’s hump (pleural-based density resulting from infarction) are rare radiographic findings in patients with acute PE.

Because of its widespread availability, CT pulmonary angiography (CT-PA) is being increasingly used to diagnose patients with suspected PE.³ In theory, the clot itself can be visualized as a filling defect within the vessel without an invasive procedure (Figure 10-1). In addition to rapid acquisition, the benefits of CT-PA include the ability to detect alternative causes that may explain the patient’s clinical presentation. Many radiologists now feel more comfortable interpreting CT-PA over older V/Q techniques because of differences in levels of experience. In fact, many radiologists and clinicians believe that the results are more definitive than those of a radionuclide V/Q scan. However, this feeling may be misleading because according to the PIOPED II study results, the accuracy of the high-probability V/Q scan (>85%) and that of a low-probability scan (<20%) is comparable to that of the CT-PA but with significantly lower total-body radiation dose. Additionally in CT-PA, emboli are difficult to visualize in more distal branches, and the significance of many defects is actually uncertain. Despite technologic developments in multislice CT cameras and automatic bolus tracking timing systems, the results are frequently inconclusive because of artifacts from motion, slice positioning, or insufficient contrast enhancement technique. It is important to remember that accurate CT-PA requires concomitant pretest clinical probability assessment (Table 10-1). Discordant CT-PA and clinical findings should prompt additional testing. Conventional pulmonary angiography is often required in this setting to definitively exclude PE. The PIOPED II trial concluded that there was increased sensitivity in diagnosis of suspected PE when CT-PA was combined with CT venography to assess the iliac, femoral, and popliteal veins for acute DVT with similar specificity compared with CT-PA alone⁴ The utility of CT venography, however, has proven more limited in widespread clinical use because it remains difficult to obtain technically adequate examinations. Although reported data⁵ suggest that CT-PA is not inferior to V/Q scanning for ruling out PE, the diagnostic accuracy of CT-PA appears to vary widely between institutions. Because this may be caused by differences in the experience of the radiologist interpreting the images on top of the image quality, clinicians should consider their institution’s experience as well as the pretest probability of PE when deciding whether to use CT-PA and whether to pursue other additional diagnostic testing. In practice, when PE is a clinical concern, a reasonable approach in choosing the appropriate diagnostic modality is to begin with the chest radiograph.⁶ If the chest radiograph results are abnormal with opacities, pleural effusion, or atelectasis, CT-PA is an appropriate next step; however, if the chest radiograph results are normal, V/Q scintigraphy is the more prudent examination to follow. Finally, it is important to note that both the PIOPED studies state the importance of clinical assessment to the final diagnosis, and both the modalities CT-PA and V/Q scanning lose considerable value when there is discordance between objective clinical assessment and test results. In fact, the final sentence in PIOPED II states: “Additional testing is necessary when clinical probability is inconsistent with the imaging results.”

The V/Q scan has allowed accurate noninvasive, although indirect, imaging of acute pulmonary emboli for decades. Normally, pulmonary ventilation (V) and perfusion (Q) are matched, with a normal gradient (i.e., the lung apices are less ventilated and perfused compared with the lung bases) (Figures 10-2 and 10-3). In PE, a thrombus occludes the pulmonary artery or one of its branches and reduces distal pulmonary arterial perfusion. The perfusion component of the V/Q scan procedure involves intravenous (IV) injection of tiny radiolabeled particles that will pass from the heart and into the pulmonary arterial circulation. Where there is no obstruction to flow, the particles are trapped in tiny precapillary arterioles distally, and perfusion is detected as homogeneous radioactivity when imaged by the nuclear medicine gamma camera. When a thrombus blocks flow, particles are trapped centrally, and the entire region perfused by the occluded artery appears as an area of “cold,” or diminished, activity. The classic pattern involves large, often wedge-shaped perfusion defects (Figures 10-4 and 10-5).

To achieve accurate results, however, other causes of decreased perfusion must be excluded. Other lung problems, such as chronic obstructive pulmonary disease (COPD) and pneumonia, can cause diminished perfusion as the body automatically shunts blood away from areas that are not normally ventilated. By imaging the patient after he or she breathes a radioactive material (ventilation component of the V/Q scan), areas of abnormal ventilation can be detected and compared with the perfusion portion of the examination. In the case of a non-embolic etiology, a ventilation defect will be seen that matches the perfusion defect. On the other hand, a mismatched defect is seen with acute PE. The perfusion defect occurs in a region of normal ventilation. Ventilation is maintained in this setting because the lung parenchyma usually remains inflated, perfused by the bronchial artery circulation; this is also why pulmonary infarction in PE is uncommon. The hallmark finding of pulmonary embolic disease is a region with reduced pulmonary perfusion while ventilation remains normal, resulting in a V/Q mismatch, the hallmark finding of pulmonary embolic disease.

The V/Q scan remains the modality of choice in instances of renal insufficiency and contrast allergy, which preclude use of CT-PA. In addition, the risks of radiation exposure from the explosion in CT use have raised widespread concerns in the radiology community. It is, therefore, worthwhile to note that CT-PA is associated with significantly greater radiation burden to the body (the absorbed radiation dose is between 8 to 10 mSv), particularly to the radiation-sensitive female breast compared with the V/Q scan (absorbed radiation dose, approximately 2 mSv). Therefore, many physicians advocate the use of the V/Q scan in women, particularly younger patients with lower levels of clinical suspicion for PE.

The exception to this rule is in pregnant women. Although V/Q is considered safe, the consensus evidence points toward a lower fetal radiation dose with shielded CT-PA compared with the V/Q scan.⁸ To reduce exposure in pregnant women, a perfusion-only lung scan can be considered when the chest radiograph is normal and there is no history of chronic lung disease. Additional precautions are needed if a V/Q scan is done in a nursing mother because the technetium-99m (Tc-99m) radiotracer is excreted into breast milk. Breastfeeding should be withheld for 2 days after the examination. Milk can be expressed ahead of time and frozen for the 2 days to allow for the radioactive levels to decay to background.