
Chapter 27 VENOUS THROMBOEMBOLISM 191
18. What noninvasive imaging methods are available for the diagnosis of DVT?
n
Duplex ultrasound: The sensitivity and specificity are operator dependent and related to
patient symptomatology, but this test can detect more than 95% of acute symptomatic
proximal DVTs. It should be noted that its specificity for acute thrombosis decreases in the
settings of chronic or recurrent VTE.
n
Radio fibrinogen leg scanning: Good for detecting distal clots, including clots in the calf,
popliteal ligament, and distal thigh vein, but relatively poor for more proximal clots.
n
Impedance plethysmography: The diagnostic sensitivity and specificity depend on the
technical expertise of the person doing the study, but in many centers this test detects
more than 95% of acute proximal lower extremity DVT.
n
Spiral computed tomography (CT) venography. Although rarely used and not extensively
studied, reports show promise for this modality, with a sensitivity and specificity
comparable to ultrasound.
n
Magnetic resonance imaging (MRI) venography. Can be useful, particularly for patients
with inconclusive ultrasound studies or a contraindication to radiation or contrast dye (i.e.,
pregnant patients). It has proved accurate for both lower extremity and pelvic DVT.
19. Can a single duplex ultrasound exclude DVT in isolation?
No. In patients with a moderate to high pretest probability for DVT, a negative D-dimer or a
repeat duplex ultrasound in 5 to 7 days is indicated to definitively exclude the diagnosis.
20. Are there classic chest X-ray (CXR) findings in patients with PE?
No. The chest radiograph may be normal in up to 30%. Subtle abnormalities such as focal
atelectasis, slight elevation of a hemidiaphragm, or focal hyperlucency of the lung
parenchyma, may be present. Specifically, local oligemia of vascular markings (Westermark’s
sign) or a plural-based wedge-shaped infiltrate suggestive of pulmonary infarct (Hampton’s
hump) is relatively uncommon.
21. Are there classic electrocardiogram (ECG) findings in patients with PE?
No. Normal or near-normal ECGs with sinus tachycardia or nonspecific ST-T wave changes
may be seen up to 30%. The findings classically associated with PE (i.e., S1, Q3, T3 pattern or
a new right bundle-branch block) occur in less than 15% of patients and occur with the same
frequency in patient work-up whether or not they are diagnosed with PE.
22. What imaging studies can be used to evaluate PE?
n
Computed tomography angiography (CTA) scan. CTA is rapid and generally the test of
choice. It is highly sensitive in diagnosing central or segmental emboli and other
intrathoracic pathology, however not as sensitive in ruling out subsegmental clots.
Outcomes data using newer generation multirow detector CTAs are showing higher
sensitivities. Diagnostic accuracy can be improved by performing CTA with a CT
venography (CTV), if indicated.
n
Ventilation/perfusion (V/Q) scan. Traditionally, a normal V/Q scan has been used to
essentially rule out a diagnosis of PE with a posttest probability of disease of ,4%.
Likewise, a high-probability scan is considered to rule in the diagnosis. Unfortunately,
upward of 60% of V/Q scans are read as nondiagnostic (low or intermediate probability),
particularly if the chest radiograph is abnormal or the patient has underlying
cardiopulmonary disease. A nondiagnostic scan should be followed up with further
diagnostic work-up. Limitations of the V/Q scan include tech support, availability, and
interpretation variability.
n
Pulmonary angiogram. This test has been the traditional gold standard for the diagnosis,
even though its inter-rater agreement on interpretation has been reported to be as low as
65%. Limitations include contraindications to contrast dye injection, interventional
radiology support, interpretation variability, and the need for expertise.
n
Magnetic resonance angiogram (MRA). Limited studies have shown MRA has sensitivities
and specificities comparable to standard pulmonary angiogram. Although often not