CT pulmonary angiography for pulmonary embolism: role beyond diagnosis?

Abstract

With the introduction of multi-detector CT scanners, CT pulmonary angiography (CTPA) is now the method of choice for diagnosis of pulmonary embolism (PE). CTPA detects all degrees PE-incidental to fatal and sub-segmental to massive central embolism. Availability of excellent medical and interventional treatments has reduced the mortality associated with acute PE. The main cause of early death in acute pulmonary embolism (PE) is right ventricular failure. In acute pulmonary embolism, there is abrupt and steep rise in pulmonary vascular resistance that increases afterload. As right ventricular afterload increases, tension in the right ventricular wall rises and may lead to dilatation, dysfunction and ischaemia of the right ventricle (1). This can result in sudden death often before the patient getting medical attention; however, in most patients who survive the initial insult, circulatory failure results within hours of PE secondary to failing right ventricle (2). Treatment aimed at rapidly reversing the right ventricular failure will reduce the risk of recurrence and death. According to the European Society of Cardiology guidelines (3), acute PE can be stratified according to prognosis based on the presence of risk markers. High risk is a life-threatening situation where shock and hypotension are present, and carries short-term (30-day) mortality of > 15%. Non-high-risk PE can be further stratified according presence of markers of right ventricular dysfunction (RVD) and/or myocardial injury into intermediate-risk and low-risk groups. An intermediate risk is diagnosed if at least one right RVD or myocardial injury marker is positive and has 3–15% risk of 30-day mortality. Low-risk PE is diagnosed when all RVD and myocardial injury markers are negative and has excellent prognosis with < 1% risk of 30-day mortality. High-risk PE patients should be treated aggressively with thrombolysis or embolectomy. Low-risk PE patients may benefit from early discharge and/or home treatment. The intermediate-risk group that makes up a majority of the patients probably needs hospital admission and treatment depending on severity. A quick assessment of severity of PE is therefore crucial for selecting appropriate treatment. There is increasing evidence that presence of RVD identifies a subgroup of normotensive patients with much more guarded prognosis who may benefit from thrombolysis (4) or embolectomy (5). The RVD markers useful for risk stratification of acute PE include RV dilatation, hypokinesis or pressure overload on echocardiography, RV dilatation on CT, elevated right heart pressure at right heart catheterisation and elevated brain natriuretic peptide (3). Echocardiography is a good diagnostic test for evaluating RVD; however, echocardiography will only be useful if the RV free wall can be clearly defined, which may not be possible in some patients with poor acoustic window or due to obesity. Further limitations include non-availability on a 24-h basis, experienced readers can detect RVD more reliably on echocardiography compared with inexperienced readers and echocardiography criteria for RVD differ among the many published studies. In addition, echocardiography has a poor predictive value for diagnosis of acute PE (6). Positive tests for serum biomarkers such as troponin and brain-natriuretic peptide are related to intermediate risk of short-term mortality in acute PE, but are more useful for excluding RVD in the setting acute PE (7). Computed tomography is usually available round the clock and therefore gives a unique opportunity to assess cardiac parameters in the same sitting, thereby avoiding other investigations for assessment of severity of PE. CTPA provides opportunity for not only assessing the clot burden but also evaluating the signs of right ventricular pressure overload. The signs include RV dilatation, increased RV/LV ratio, deviation of interventricular septum, dilated superior vena cava, dilated azygos vein, reflux of contrast into inferior vena cava. CTPA assessment of severity of PE is based on morphological criteria – quantification of obstruction of arterial bed (8–12) and signs indicative of RVD (13–18). Most of the studies on clot burden quantification (6,10–12,16–18) have shown that clot index is an indicator of severity of acute pulmonary embolism. Although some studies have shown that clot index as a predictor of short-term mortality due to PE (10–12,17), other studies (18,19) did not find any significant association between clot burden and PE death. Differences in populations studied and the different scoring system may have affected the results in these studies. Quantification of obstruction of arterial bed can be a complex task, and a simple way of assessing clot burden is always useful. Recently, a study has shown that assessing only the central clot burden by evaluating the central pulmonary arteries is an independent predictor of mortality from acute PE (12). Right ventricular dilatation on CT predicts early death in patients with acute PE (13–16). The criterion for right ventricle dilatation in most studies has been a RV/LV ratio > 0.9 or ≥ 1.0 (Figure 1). Ventricular CT measurements obtained from reconstructed four-chamber views are superior to those from axial views for identifying high-risk patients and worse prognosis (15). The benefit of separate ECG-gated CT scan for the evaluation of RVD and volume measurements is also minimal and does not justify its routine clinical use as there is substantial increase in patient radiation exposure with only limited incremental diagnostic improvement (20,21). Computed tomography pulmonary angiography in a 57-year-old male with acute PE. Representative sections at the time of diagnosis of acute PE (A, B) and 7 days following thrombolytic therapy (C, D). At presentation, emboli seen in main pulmonary arteries (arrow heads) with an obstruction index of at least 50% and RV/LV ratio is > 2, indicative of right ventricular dysfunction. Following thrombolytic therapy, resolution of thrombi and reversal of RV/LV ratio to 0.8 are observed. On CTPA, RV dilatation correlates with the thrombotic load. A greater proportion of patients with thrombus in main pulmonary arteries had evidence of RV dilatation compared with those with segmental and subsegmental thrombus alone (22). Some studies have reported the combination of signs of RVD and clot score for assessment of outcome. A RV/LV > 1.0 and Obstruction index > 40% increased the positive predictive value for 90-day PE-related mortality to 18.8%. The predictive value of an RV/LV ratio < +1.0 for an uneventful outcome was 100% (17). In another study, pulmonary artery obstruction index of more than 50% and RV/LV ratio > 1.5 were found to be useful diagnostic criteria for severe PE and poor patient outcome (23). The role of CTPA in assessing severity and predicting outcome of acute PE is promising. Although it remains inconclusive whether CTPA-derived obstructive indices are useful for risk stratification, a consensus approach on the index to use may yield similar results across studies and therefore a meaningful conclusion about its utility can be derived. On the other hand, RV dilatation detected on CTPA alone can be a marker for poor outcome independent of severity. RV/LV ratio > 1 is a simple feature that can be readily identified by both clinicians and general radiologists and is suitable for use in clinical practice. Prospective studies with a large group of patients are awaited for establishing the role of other CT-derived parameters for risk stratification. No potential conflicts to declare.