Evaluation of Pulmonary Embolism
Clinical Workup
The initial workup of the patient with suspected pulmonary embolism, like most diseases, begins with a proper history and physical exam and may include additional testing such as D-Dimer or EKG. It is important to remember this step when considering the imaging of these patients, because in most cases we are already dealing with a subset of patients for which there is clinical uncertainty.Imaging of Suspected PE
The classic workup algorithm for PE was chest film, V/Q scan and pulmonary arteriography (as needed). The emergence of good CT imaging of the pulmonary arteries has modified this algorithm. There is no universal agreement about what the best algorithm is, but I will present the method I use and teach. Thus far it has served me well, though as technology changes so will the appropriateness of different tests in different scenarios.The Plain Radiograph
Most patients presenting with signs and/or symptoms suggesting PE should probably have a plain chest radiograph (PA and lateral if possible) as their first imaging study. While there are several plain radiograph signs of PE (e.g. Hampton's hump, Westermark's sign, Fleischner's sign, etc.) these are exceptionally rare and will seldom, if ever, provide a difinitive diagnosis. The true reason for the chest film is to look for alternate diagnoses (such as pneumonia, heart failure, etc) that may be causing the patient's problems. Most patients who do have a PE will have either a normal chest radiograph, or a very non-specific abnormal radiograph (atelectasis, small effusion) that does not give the diagnosis. The additional advantages of the plain fim are (1) it can aid in the interpretation of a V/Q scan, and (2) it can help determine if the patient should get further evaluation with V/Q or CT scanning.The Next Step -- V/Q or CT?
The next, and most controversial step, in the imaging workup of this patient is to determine if it is better to get a V/Q (scintigraphy) scan or a CT scan. In my experience, the major problem with V/Q scans is the subset that are interpreted as "intermediate probability" for pulmonary embolism. Often, the problem is underlying lung disease (emphysema, etc.). On the other hand, there is extensive literature and understanding about the clinical significance of "normal," "low- probability" and "high-probability" V/Q studies. For my algorithm, patients who have known underlying cadiopulmonary disease or those with extensive changes on their plain film of the chest, should have CT as their next imaging study. Otherwise healthy patients (i.e. those for whom you do not expect an "intermediate probability" V/Q) should have a V/Q scan as their next study.Ventillation/Perfusion (V/Q) Scintigraphy
The original Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study investigated the role of V/Q scanning in patients for suspected PE and it has also outlined the criteria used in the interpretation of these studies. The major lesson to be learned from that study is that patients for whom the clinical and scintographic suspicion of the presence of PE matched (i.e. both have a high suspicion, or both have a low suspicion) then the two were almost always correct. When the clinical suspicion and scintographic interpretation did not match, it is very unclear if the patient does or does not have the disease. In patients for whom the V/Q does not give a definitive result (i.e. those with intermediate probability studies or those whose clinical and scintographic probabilities do not agree) further imaging with CT is usually the best next step.Computed Tomography
Though almost all CT scans are now performed with "spiral" (or helical) techique, and all CT angiographic studies are performed that way, many people still strongly associate the term "spiral CT" with the evaluation for PE. CT scanning for PE is proving to be extremely useful clinically, and it has replaced both V/Q and pulmonary angiography in many cases. The study is performed while contrast is strongly-concentrated in the pulmonary arteries. The diagnosis of PE is made by identfying areas within the vessel lumen that do not opacify with contrast ("filling defects"). The sensitivity and specificity of CT for PE depends greatly upon where in the pulmonary arterial system the clot may or may not be. For the central PAs, CT scanning is on the order of 95% or more sensitive and specific for PE. On further-order branches, the sensitivity and specificity decline. There is no real agreement about how far down the pulmonary arterial system CT can reliably see PEs.Clearly there are small PEs that will not be seen on CT scanning. Debate continues about whether these small PEs are significant clinically. Likely the answer depends upon how small of a PE we are talking about. It should be remembered, however, that since the major goal of treatment of PE is prevention of the next (and potentially fatal) PE, those small PEs may prove to be important as a way to identify patients at risk for subsequent PE.
Another advantage of CT that should not be forgotten is that this study will often provide an alternate diagnosis (much like a plain film might).
Pulmonary Arteriography
While pulmonary arteriography is still considered the "gold standard" for the clinical evaluation of PE, it is use is decreasing over time. In so many cases, CT provides a definitive diagnosis and no further imaging is pursued. Also, it is clear from animal studies that there are small PEs that get missed by angiography, so all those caveats that were said about CT also apply to angiography. Of course pulmonary arteriography is an invasive study, with risks to the patient that must at least be considered when one is contemplating its use. The major role for angiography in PE evaluation is slowing becoming one of problem- solving. Often the arteriogram is obtained in patients with a questionable area on a CT scan. In these cases, directed angiography, focused at the vessel or vessels in question, can be performed.The other advanage of the angiography suite is that it is possible to provide intervention, when clinically indicated. IVC filter placement can be performed at the time of angiography. Also, in patients with evidence of right heart strain on echo, catheter-directed thrombolysis may be indicated. Since sedation will usually be used for the diagnostic angiogram, anyone contemplating such a study should decide beforehand if an intervention, such as filter placement, should be performed so that informed consent can be obtained prior to the induction of sedation.
