As a result of the development of advanced multislice spiral CT technology as well as increased availability and acceptance of pulmonary CT angiography, the role of ventilation/perfusion lung scanning in the diagnosis of pulmonary emboli (PE) must be redefined (5,6).

Several schemes for interpretation of ventilation-perfusion scintigram have been developed over the years. These included the Bilo, original PIOPED, and the revised PIOPED (Table 2.1) criteria. Currently, another set of categories has been suggested by the PIOPED II study. PIOPED II is a prospective multicenter study designed to objectively evaluate the new scan categories to include a very low probability category and also to evaluate the efficacy of CT in the diagnosis of PE.

In PIOPED II, the revised PIOPED criteria for high probability have been maintained, the intermediate and low probabilities have been revised, and a very low probability criterion has been established (7,8).

The very low probability criteria in PIOPED II are as follows:*

A very low probability interpretation, according to these authors, has a positive predictive value (PPV) of 8.2%. However, a very low probability scan combined with a low clinical probability has a PPV of 3.1%, and when these data were applied only to a female patient aged <40 years, it was only 2%. Based on these findings, the authors conclude the combination of a low clinical probability and very low probability interpretation reliably excludes acute pulmonary embolus and can be used in patients whose computed tomography angiography (CTA) may be disadvantageous (women of reproductive age, compromised renal function, allergic risk).

Regardless of what criteria are being used, the classification of lung scans as normal, high probability, or very low probability is easier and the results are more reliable. However, distinguishing between low probability and intermediate probability may be difficult and some degree of uncertainty is unavoidable. In addition, an interobserver variability as much as 30% is not uncommon. Therefore adhering to the strict diagnostic criteria of lung scan interpretation may not be practical for all patients all the time. It appears that using the criteria combined with interpretative experience (gestalt), clinical evaluation, and consideration of the total situation is the best approach for the diagnosis of PE. The combination of clinical assessment and V/Q interpretation improves the chances of making the correct diagnosis when compared with either scan or clinical assessment alone (Table 2.2). It is important to note that the scintigraphic findings and clinical impression of PE are more commonly concordant than discordant.

The basic concept in formulating a rational approach to the diagnosis of PE is to observe all the clinical data—pretest probability factors and stratification of the patient with respect to underlying cardiopulmonary disease as well as ancillary scintigraphic finding and radiographs that either favor or disfavor the presence of embolic disease (9).

History: A healthy 36-year-old male with a family history of thrombophilia (not known at the time) presents with chest pain, dyspnea, and apprehension. His chest x-ray is normal. He receives a V/Q scan for evaluation of probable pulmonary emboli (Figs. 2.1 A, B).

Findings: There are multiple large, wedge-shaped, pleural-based, mismatched defects in both lung fields. In addition, there is generalized reduced perfusion of the right lung as compared with the left. The lung scan findings indicate a high probability of acute pulmonary emboli, which were also visualized on CT angiography (Fig. 2.1 C) done on the same day. Multiple filling defects within the lumen of contrast-filled pulmonary arteries are seen on both sides. A follow-up perfusion scan (Fig. 2.1 D) after completion of anticoagulation therapy shows complete resolution of the perfusion abnormalities.

Discussion: Multiple large, segmental, mismatched perfusion defects are important characteristics for pulmonary emboli which, in patients with high clinical probability, have a positive predictive value of 96%.

Other causes of V/Q mismatch are previous PE, bronchogenic carcinoma, previous radiation therapy to the lung, pneumonia, tuberculosis, collagen vascular disease, intravenous drug abuse, tumor emboli, etc.; however, most of these can be excluded by clinical and other diagnostic tests.

Perfusion defects due to acute pulmonary emboli gradually regress with time and following therapy; the majority resolve within 3 months. Resolution is quicker in patients with no previous cardiovascular disease (3 to 7 days).

Obtaining a perfusion lung scan at the completion of anticoagulant therapy and at discharge is of clinical value in future management of the patient. This scan will serve as a baseline for the diagnosis of recurrent PE, which is seen as the development of a new defect.

Most patients with high-probability lung scans and a high clinical likelihood of PE require treatment and need no further diagnostic test to confirm diagnosis. Utilization of CT angiography in the workup of patients suspicious for PE is increasing. However, in patients with a high probability for PE, the discriminatory power of V/Q and CT angiography is similar and concordant positive results have been reported in 86% of interpretations.

In evaluating patients with PE, attention should be given to the risk factors for the development of DVT and PE, the most notable of which are use of estrogen, history of previous DVT or PE, and presence of concurrent DVT. Genetic disorder of thrombus formation (thrombophilia) should be kept in mind when PE occurs in young patients with no other predisposing factors. These patients are characteristically younger than the average patient with thrombosis.

Diagnosis: Acute pulmonary embolism with complete resolution.

History: A 55-year-old male who had undergone renal transplantation and now has declining renal function was flown from a long distance to the Medical Center for further evaluation and possible renal biopsy.

Shortly following his admission, the patient developed a sudden onset of right-sided chest pain and dyspnea. Cardiac enzyme, ECG, and chest radiographs were not contributory and the patient received a V/Q scan for the evaluation of probable pulmonary emboli.

Findings: V/Q lung scans (Figs. 2.2 A, B) show multiple large and moderate sized wedge-shaped, pleural-based, mismatched defects in both lungs.

Immediately following the lung scan, a lower extremity venous Doppler examination (Fig. 2.2 C) was performed. The duplex study shows large segmental venous thromboses involving the midportion of the superficial femoral vein and extending through the left popliteal vein.

The patient was managed appropriately for thromboembolic disease. A lung scan at 3-month follow-up (Fig. 2.2 D) shows persistence of multiple defects with no significant interval change.

Discussion: In the proper clinical context and in a patient with no history of previous pulmonary emboli and a clear chest radiograph, the presence of multiple large mismatched segmental defects represents a high-probability study with a positive predictive value of 96%.

In the diagnosis of PE, regardless of the scan category, attention should be paid to the clinical presentation as well as risk factors contributing to the formation of DVT and subsequent PE. Genetic factors, use of oral contraceptives, pregnancy, puerperium, malignancy, immobilization, trauma, recent surgery, central line placement, and long airplane flight (traveler’s thrombosis) are among the most important risk factors to consider.

Coexistence of PE and DVT is not unusual. The majority of PE arise from the proximal vein of the lower extremity. About one-third of patients who have PE also have detectable DVT, but more than half of the patients with PE do not demonstrate thrombosis, suggesting that the thrombus has already migrated to the lung.

Silent emboli are also not uncommon and suspected perfusion defects consistent with PE may be seen in up to 30% of patients with a positive venous duplex scan. Although these patients normally receive anticoagulant therapy, obtaining a baseline perfusion lung scan would be helpful for evaluation of future episodes of PE.

Perfusion defects resulting from pulmonary emboli may begin to resolve within 3 to 4 days. Resolution of perfusion defect is quicker in patients with no previous cardiopulmonary disease but slower when a large defect occurs in a patient with preexisting cardiopulmonary disease. A defect not resolving by 2 to 3 months may persist for many years, and some of these patients may develop chronic changes of pulmonary emboli. It is to be noted that the most common cause of a false-positive high-probability interpretation of a lung scan is the history of PE in the past.

Pulmonary emboli are responsible for a substantial number of patients suffering from a secondary pulmonary arterial hypertension. The process is initiated by single or recurrent episodes of PE. The unresolved thrombosis proceeds to the organization stage and fibrosis, which results in stenosis or complete obstruction of the pulmonary artery; this is demonstrable by CT or conventional pulmonary angiography. On lung scan, these patients would have segmental defects in both lung fields that remain unchanged over time. These patients may benefit from the surgical procedure of pulmonary endarterectomy. It is to be noted that although CT angiography appears to be
the first-line imaging procedure for the evaluation of patients suspected of having PE, V/Q pulmonary scintigraphy has a higher sensitivity and specificity for distinguishing pulmonary arterial hypertension of chronic thromboembolic disease from other causes of pulmonary arterial hypertension (idiopathic, emphysema, pulmonary fibrosis, pulmonary venous occlusive disease, sarcoidosis, and congenital heart disease such as atrial septal defect). A normal scan practically excludes chronic PE as the cause of the patient’s pulmonary arterial hypertension. The patient with a chronic inflammatory disorder of the lungs and those with splenectomy are at especially high risk for developing chronic embolic pulmonary arterial hypertension.

Figures 2.2 E and 2.2 F shows a chest radiograph and perfusion scan of a 60-year-old patient with pulmonary arterial hypertension secondary to PE. On perfusion scan (Fig. 2.2 F), in addition to the presence of a segmental perfusion defect, there is generalized reduced perfusion of the right lung as compared with the left, correlating with increased lucency and reduced vascularity of the right lung seen on the radiograph (Fig. 2.2 E); this suggests a previous embolus in the right pulmonary artery. Reduced perfusion is more pronounced peripherally, consistent with diminished peripheral flow due to attenuation and pruning of the peripheral vascular branches, a known feature of pulmonary arterial hypertension.

A patient with primary pulmonary arterial hypertension may present with clinical and radiographic findings similar to those of secondary pulmonary artery hypertension; however, the perfusion lung scan in the former is normal; therefore a perfusion lung scan can be used to differentiate the two entities.

Diagnosis: Pulmonary emboli coexistent with venous thrombosis of lower extremity; unresolved perfusion defects.

History: A 63-year-old alcoholic with hepatic cirrhosis and multifocal atrial tachycardia was referred for V/Q scintigraphy to rule out PE (Figs. 2.3 A, B).

Findings: There is a large, well-defined, wedge-shaped perfusion defect involving the right upper lobe best seen in the lateral projection (Fig. 2.3 B), in the area in which there is some ventilation (Fig. 2.3 A). The radiographic abnormality (Fig. 2.3 C) is smaller than the perfusion abnormality.

Discussion: A perfusion defect substantially larger than the chest x-ray opacity with preservation of some ventilation is one of the criteria for a high-probability scan.

Given this patient’s history of alcohol use and esophageal varices, he is considered to be at high risk for anticoagulation. The PE diagnosis was confirmed by angiography (Fig. 2.3 D) and a Greenfield filter was inserted.

A perfusion defect in an area of pulmonary parenchymal opacity does not mean an intermediate probability for PE in every instance. The scan can be classified as high probability when the criteria are met. In this case, the presence of some
ventilation supports the diagnosis of PE. The segmental nature and wedge-shaped defect extending to the pleural surface is also a characteristic pattern for PE and should not be weighed lightly where the diagnosis of PE is concerned.

In patients with a high-probability scan and a relatively high clinical likelihood of PE, there should be no need for further diagnostic tests; however, in patients undergoing more aggressive therapy, such as inferior vena cava filtration, thrombolysis, or embolectomy, pulmonary angiography is indicated and justified.

Diagnosis: High probability of acute PE.

History: A 62-year-old man presented to the emergency room with dyspnea, chest pain and hypoxemia. Acute myocardial infarction was excluded and an emergency V/Q scan was performed (Figs. 2.4 A and 2.4 B). An admission chest x-ray was taken (Fig. 2.4 C).

Findings: The chest x-ray reveals enlargement of the right pulmonary artery but no parenchymal infiltrates. The lung scan shows diffuse heterogeneous distribution of radioaerosol and a large mismatched perfusion defect involving virtually the entire right lung.

Discussion: A scintigraphic finding of near absence of right lung perfusion in an adequately ventilated lung implies a high probability for PE. Other causes of a near absence of unilateral lung perfusion defect include the following: bronchogenic carcinoma causing compression of the pulmonary artery; congenital heart disease and after shunt procedure; fibrosing mediastinitis usually due to histoplasmosis; postradiation fibrosis; unilateral hyperlucent lung (Swyer-James syndrome); pulmonary arterial hypoplasia; giant bullae; mediastinal hematoma resulting from aortic dissection (valve replacement); Marfan’s syndrome; and several other uncommon conditions. Most of these entities can be excluded by history and radiographic findings.

In addition to the clinical presentation, the enlargement of the pulmonary artery (Fleischner’s sign) seen on this patient’s chest x-ray is a supportive finding that the V/Q mismatch represents a pulmonary embolus for which the patient received treatment. Enlargement of the pulmonary artery is due to pulmonary arterial hypertension and/or distention of vessels by bulky thrombosis. It is to be noted that prominence of the pulmonary artery may be seen in 20% of radiographs of patients with or without pulmonary emboli. As such, the sensitivity of Fleischner’s sign in predicting PE is low and many patients with PE do not show this sign.