ORIGINAL ARTICLE
Computed Tomography and Echocardiography in Patients With Acute Pulmonary Embolism: Part 2 Prognostic Value Elizabeth George, MBBS,* Kanako K. Kumamaru, MD, PhD,* Nina Ghosh, MD,w Carlos Gonzalez Quesada, MD,z Nicole Wake, MS,w Arash Bedayat, MD,*y Ruth M. Dunne, MD,* Sachin S. Saboo, MD,* Ashish Khandelwal, MD,* Andetta R. Hunsaker, MD,* Frank J. Rybicki, MD, PhD, FAHA, FACR,* and Marie Gerhard-Herman, MDw
Purpose: The aim of the study was to compare the prognostic value of right ventricular (RV) dysfunction detected on computed tomography pulmonary angiography (CTPA) and transthoracic echocardiography (TTE) in patients with acute pulmonary embolism (PE). Materials and Methods: From all consecutive CTPAs performed between August 2003 and May 2010 that were positive for acute PE (n = 1744), those with TTE performed within 48 hours of CTPA (n = 785) were selected as the study cohort. Multivariate logistic regression analysis was performed to assess the association of CTPA RV/left ventricular (LV) diameter ratio and TTE RV strain with PE-related 30-day mortality, including other associated factors as covariates. The predictive ability (area under the curve) was compared between the model including the CT RV/LV diameter ratio and that including TTE RV strain. Test characteristics of the 2 modalities were calculated. Results: Both CT RV/LV diameter ratio and TTE RV strain were independently associated with PE-related 30-day mortality (adjusted odds ratio = 1.14, P = 0.023 for 0.1 increment of the CT RV/LV diameter ratio; and odds ratio = 2.13, P = 0.041 for TTE RV strain). History of congestive heart failure and malignancy were independent predictors of PE-related mortality, while there was significantly lower mortality associated with anticoagulation use. The model including TTE RV strain and that including CT RV/LV had similar predictive ability (area under the curve = 0.80 vs. 0.81, P = 0.50). The sensitivity, specificity, and positive and negative predictive values of TTE RV strain and CT RV/LV diameter ratio at a cutoff of Z1.0 were similar for PE-related 30-day mortality. Conclusions: Both RV strain on TTE and an increased CT RV/LV diameter ratio are predictors of PE-related 30-day mortality with similar prognostic significance. Key Words: computed tomography pulmonary angiography, transthoracic echocardiography, pulmonary embolism, prognosis, right ventricular strain
(J Thorac Imaging 2014;29:W7–W12)
From the *Applied Imaging Science Laboratory, Department of Radiology; wDepartment of Medicine, Cardiovascular Division; zDepartment of Medicine, Brigham and Women’s Hospital & Harvard Medical School, Boston; and yDepartment of Radiology, University of Massachusetts Medical School, Worcester, MA. The authors declare no conflicts of interest. Reprints: Frank J. Rybicki, MD, PhD, FAHA, FACR, Applied Imaging Science Laboratory, Department of Radiology, Brigham and Women’s Hospital & Harvard Medical School, 75 Francis Street, Boston, MA 02115 (e-mail:
[email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
J Thorac Imaging
Volume 29, Number 1, January 2014
A
cute pulmonary embolism (PE) is a common cardiopulmonary disease with an annual incidence of about 0.69/1000 in the United States.1 The clinical presentation can vary from massive thromboembolism with cardiogenic shock to incidentally detected asymptomatic subsegmental emboli. Although the overall short-term mortality in patients with acute PE is about 8% to 17%,2–7 there is a large variation (0% to 59%) in the fatality rates in different clinical scenarios.3,8–12 Computed tomography pulmonary angiography (CTPA), the current reference standard for the diagnosis of PE, has been shown to provide additional imaging-based prognostic information. Clot burden, thrombus distribution, dual-energy lung perfusion defects, and lung perfused blood volume quantification on CTPA are some of the pulmonary metrics explored for prognostic utility.13–16 In addition, various CT-derived signs of right ventricular (RV) dysfunction have been shown to be associated with mortality and adverse events, with RV to left ventricular (LV) diameter ratio being the most common.3–5,7,17–19 Although echocardiography is a poor diagnostic test for PE, it is excellent for risk stratification and guiding treatment strategies through identification of RV dysfunction, persistent pulmonary hypertension, and free-floating right heart thrombi.20 The combination of various echocardiographic indicators of RV size, function, and pressure as a sign of RV dysfunction has been shown to be associated with a worse prognosis,10,11,21–23 even in hemodynamically stable patients.11,24–27 Good correlation between CT-derived and TTE derived signs of RV dysfunction15,16,28–32 has been reported, although the imaging-based metrics used vary between studies. Although a few studies have compared these 2 modalities with respect to prognostic value,16,30,31 routinely available CT and echocardiography-based markers of RV dysfunction have not, to our knowledge, been compared for the prediction of mortality in a large population of patients presenting with acute PE. The purpose of this study is to compare the prognostic significance of CT-derived RV/LV diameter ratio and transthoracic echocardiographic (TTE) detection of RV strain in patients with acute PE.
MATERIALS AND METHODS Study Population The institutional review board approved this HIPAAcompliant retrospective cohort study; informed consent was www.thoracicimaging.com |
W7
George et al
waived. All CTPA examinations performed at our institution from August 2003 to May 2010 were examined. From this group, 1744 examinations were identified as positive for acute PE. If there were multiple episodes of acute PE in a subject, only the latest episode was included in the study. All 804 subjects for whom TTE was performed within 48 hours of CTPA were included in the study cohort. For subjects who underwent multiple TTE examinations within 48 hours of the CTPA, the TTE closest in time to the CT scan was chosen for further evaluation.
CTPA Image Acquisition All CTPA examinations were performed in the craniocaudal direction with 16-, 64-, or 128-slice multidetector CT (MDCT) scanners at a single large, urban teaching hospital. The images were acquired with a slice thickness of 1.0 to 1.25 mm and reconstructed at 0.75 to 1.0 mm slice interval. A tube potential setting of 80 to 120 kVp with an effective milliampere (mA) of approximately 200 mA was used, after intravenous administration of 75 to 100 mL iodinated (370 mg iodine/mL) contrast media at 3 mL/s timed by bolus tracking at the main pulmonary artery. The RV/LV diameter ratios were obtained from 4-chamber reformatted images using a standardized double-oblique method33 on a dedicated 3D workstation (VitreafX; Vital Images, Minnetonka, MN). On the 4chamber reformatted image, RV and LV diameters were measured manually as the maximum distance from the interventricular septum to the endocardial border, perpendicular to the long axis of each ventricle. Official CT reports were reviewed to determine the proximal extent of PE (central, lobar, segmental, or subsegmental pulmonary artery). The 4-chamber RV/LV diameter ratios were obtained retrospectively, as these are clinically reported routinely at our institution.
Echocardiography Acquisition All clinical echocardiography examinations were performed in a laboratory accredited by the Intersocietal Accreditation Commission in accordance with the recommendations of the American Society for Echocardiography.34 Standard 2 to 5 MHz phased array transducers were used to perform transthoracic studies. Each patient was examined in the supine position, and the patient’s position was adjusted to the acoustic window being utilized. All clinical TTE reports were reviewed by a physician with 1-year experience in cardiovascular imaging for information regarding the RV. RV strain was considered present when any of the following signs were present: (1) reduced RV systolic function assessed qualitatively on the basis of the RV free wall motion or presence of RV hypokinesia/dyskinesia/akinesia; (2) pulmonary artery systolic pressure >36 mm Hg34 [determined as the sum of the estimated maximum pressure difference between the RV and right atrium (RA), measured by applying the simplified Bernoulli equation to the maximum velocity of the tricuspid regurgitation Doppler signal, and the estimated mean RA pressure, determined from the diameter and respirophasic variation of the inferior vena cava (IVC)34,35]; (3) moderate or severe dilatation of the RV determined qualitatively; and (4) abnormal interventricular septal movement. If RV wall hypertrophy was present, these signs were considered chronic, and RV strain from acute PE was excluded.11
W8 | www.thoracicimaging.com
J Thorac Imaging
Volume 29, Number 1, January 2014
Covariate Collection Information on patient demographics (age, sex), comorbidities [malignancy, congestive heart failure (CHF), coronary artery disease, atrial fibrillation, peripheral artery disease, chronic lung disease, sleep apnea, pulmonary hypertension, hypertension, diabetes mellitus, coagulopathies, chronic renal failure, and stroke or transient ischemic attack], history of recent surgery (r30 d before diagnosis of acute PE), and treatment (anticoagulation, thrombolysis, catheter/surgical thrombectomy, IVC filter placement) was obtained from the institution electronic medical records.
Study Outcomes PE-related death within 30 days of CTPA was the study outcome. The Social Security Death Index was used to confirm patient death. The cause of death was determined by consensus of 3 observers (with 7, 1, and 1 y experience, respectively) blinded to the CTPA and TTE results, after independent review of autopsy reports, death certificates, and electronic medical records. A subject was considered to have died of PE-related causes if (a) the autopsy report, death certificate, or electronic medical record stated PE as a cause of death, or (b) respiratory failure, cardiopulmonary arrest, or shock was the immediate cause of death in the absence of other cardiopulmonary diseases or precipitating factors.8,36
Statistical Analysis Univariate comparison of clinical characteristics was performed using the w2 statistics for categorical variables and the Student t test for continuous variables. Separate multivariate logistic regression models were run using TTE RV strain (Model 1) and the CT RV/LV diameter ratio (Model 2) as indicators of RV dysfunction along with the clinical characteristics associated with PE-related mortality identified in univariate analysis (P < 0.15). The predictive ability of the 2 models was compared using c statistics. The optimal CT RV/LV diameter ratio cutoff was identified as the point with balanced sensitivity and specificity, rounded to the closest clinically reasonable single decimal value. The point estimates of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were determined for both modalities. All statistical analyses were performed on STATA version 11.2 (Stata Corp., College Station, TX).
RESULTS Population Characteristics Among the 804 subjects, 19 were excluded. Nine subjects were excluded because the CT images did not adequately depict the RV/LV diameter ratio (inappropriate field of view, poor LV enhancement, or artifact from metallic leads). The other 10 subjects were excluded because TTE was indeterminate for RV strain. Among the final study population of 785 subjects, 104 (13.2%) subjects died within 30 days of CTPA, of which 36 (4.6%) were because of PE-related causes. The cause of death was unknown in 7 subjects.
Univariate Analysis Increasing age, higher RV/LV diameter ratio on CTPA, presence of RV strain on TTE, malignancy, and lack of anticoagulation were significantly associated with r
2013 Lippincott Williams & Wilkins
J Thorac Imaging
Volume 29, Number 1, January 2014
Prognosis by CT and TTE in Pulmonary Embolism
higher PE-related 30-day mortality, whereas the presence of CHF approached statistical significance (Table 1).
Multivariate Analysis On multivariate logistic regression, both TTE RV strain [adjusted odds ratio (OR) = 2.13, 95% confidence interval = 1.03-4.42, P = 0.041] and CT RV/LV diameter ratio (adjusted OR for 0.1 increment = 1.14, 95% confidence interval = 1.02-1.27, P = 0.023) were independently associated with PE-related 30-day mortality (Table 2). Other variables found to be independent positive predictors include the presence of malignancy and CHF. Anticoagulation was identified as an independent negative predictor. The model including CT RV/LV diameter ratio and the one including TTE assessment of RV strain have similar overall predictive ability (area under the curve = 0.81 vs. 0.80, P = 0.50) (Fig. 1).
TABLE 1. Clinical Characteristics of the Population
Age (mean ± SD) (y) Female Proximal extent of clot Central Lobar Segmental Subsegmental CT RV/LV ratio (mean ± SD) TTE RV strain Enlargement Reduced systolic function Treatment for PE Anticoagulation IVC filter Thrombolysis/ thrombectomy Comorbidities Atrial fibrillation CHF Coronary artery disease Chronic lung disease Diabetes mellitus Hypertension Peripheral artery disease Coagulopathies Stroke/transient ischemic attack Chronic renal failure Sleep apnea Pulmonary hypertension Malignancy Recent surgery
PE-related 30 d Death (n = 36)* (%)
No PE-related 30 d Death (n = 742) (%)
64.6 ± 14.5 61.1
58.4 ± 16.8 53.2
27.8 19.4 41.7 11.1 1.15 ± 0.34
36.7 21.6 30.6 11.2 1.05 ± 0.26
63.9 44.1 60.0
43.3 19.7 29.2
80.6 27.8 5.56
94.5 16.4 8.22
0.001 0.077 0.567
11.1 13.9 22.2
9.16 5.93 13.1
0.694 0.055 0.117
11.1
13.3
0.700
16.7 38.9 2.78
17.7 42.5 3.10
0.879 0.672 0.913
2.78 5.56
4.31 5.53
0.655 0.994
2.78
2.70
0.976
2.78 2.78
2.29 1.75
0.850 0.651
80.6 27.8
43.3 35.0
*Excluding 7 cases with unknown cause of death.
r
2013 Lippincott Williams & Wilkins
P 0.031 0.355 0.537
0.026 0.015 0.001 < 0.001
< 0.001 0.371
Test Characteristics Table 3 summarizes the test characteristics of the CT RV/LV diameter ratio and TTE RV strain. At the optimal CT RV/LV diameter ratio cutoff of Z1.0, the sensitivity, specificity, PPV, and NPV of the 2 modalities were comparable (sensitivity: 61% to 64%, specificity: 52% to 57%, PPV: 6% to 7%, NPV: 97%).
DISCUSSION This second paper in the 2-part series extends the work from the initial correlative studies in Part 1 to prognosis.37 RV dysfunction detected by CTPA and echocardiography are both predictors of PE-related 30-day mortality with similar prognostic significance. Although CT and TTE have been assessed independently for its prognostic value, to our knowledge, this is the largest study to compare CT RV/LV diameter ratio and a TTE-based composite clinical measure of RV strain for prediction of PE-related mortality in the same population of patients with acute PE. The majority of prior studies have assessed a composite outcome of mortality and intensive treatment.16,30,38,39 Quiroz et al30 reported similar predictive value of 4-chamber RV/LV diameter ratio on TTE and CTPA for adverse clinical events (area under the curve: 0.69 and 0.75, respectively) with a reported sensitivity of 71% to 83% and specificity of 49% to 56%. In a study by Park et al,16 qualitative abnormality in systolic RV free wall motion was found to have a lower sensitivity but higher specificity as compared with the axial CT RV/LV diameter ratio for adverse outcome prediction, whereas the point estimates of the 2 modalities were similar on using a combination of CT parameters (axial RV/LV diameter ratio, septal bowing, and proximal PE). In a recent large prospective study of hemodynamically stable patients, neither CT nor TTE was significantly predictive of a complicated clinical course after adjusting for other factors.38 In a prospective study of normotensive PE patients by Ozsu et al,31 it was observed that, although the prevalence of echocardiographic RV dysfunction was significantly higher among those who died due to PE, neither CTPA-derived metrics of RV size nor TTE-based RV measurements was predictive of the incidence of PE-related death from univariable Cox regression analysis, perhaps because of the limited number of PErelated deaths in this cohort (n = 6). In addition to being the reference standard for diagnosing PE, there is extensive literature on the prognostic utility of CTPA.3–7,17,19,40 We demonstrated that the CT RV/LV ratio is an independent predictor of PE-related mortality, with every 0.1 increment in ventricular ratio being associated with an adjusted OR of 1.14. This is in keeping with multiple previous reports of increasing mortality associated with increase in the CT RV/LV ratio.3,17,19 Various CT RV/LV diameter ratio cutoff values ranging from 0.9 to 1.5 have been reported,18 and a cutoff of Z1.0 was identified in our study on the basis of optimal sensitivity and specificity. Previous studies on a 4-chamber CT RV/LV diameter ratio report a sensitivity of 78% to 97% and a specificity of 38% to 59% for PE-related 30-day mortality.5,6,40 The point estimates from our study, however, demonstrate a lower sensitivity but similar specificity. This could be explained by the selective population of subjects who underwent both CTPA and TTE; it is reasonable to expect a higher prevalence of underlying cardiac disease and other comorbidities. Increase in LV size may result in a relatively normal RV/LV www.thoracicimaging.com |
W9
J Thorac Imaging
George et al
Volume 29, Number 1, January 2014
TABLE 2. Multivariate Analysis for PE-related 30 d Mortality
Model 1 Variables
OR (95% CI)
Age TTE RV strain CT RV/LV (0.1 increment) IVC filter Anticoagulation Coronary artery disease CHF Malignancy
1.01 (0.99-1.04) 2.13 (1.03-4.42) — 1.07 (0.45-2.57) 0.22 (0.08-0.61) 1.33 (0.53-3.34) 3.65 (1.19-11.2) 6.10 (2.52-14.8)
Model 2 P
P
OR (95% CI)
0.310 0.041 — 0.871 0.004 0.542 0.024 < 0.001
1.01 (0.98-1.04) — 1.14 (1.02-1.27) 1.06 (0.44-2.55) 0.19 (0.07-0.54) 1.69 (0.68-4.22) 4.09 (1.33-12.6) 5.79 (2.40-14.0)
0.453 — 0.023 0.888 0.002 0.259 0.014 < 0.001
CI indicates confidence interval.
diameter ratio, even in the presence of RV enlargement. Due to acute pressure overload associated with PE, it is possible that RV dysfunction could precede significant RV dilation in the early stages of the disease process and thus not be reflected in the RV/LV ratio at the time of CTPA acquisition, possibly resulting in lower sensitivity. The prognostic significance of echocardiographic detection of RV afterload stress in the setting of PE has been described. Various combinations of metrics for RV size (end-diastolic diameter >30 mm, end-diastolic RV/LV diameter ratio, subjective enlargement), function (hypokinesis), and pressure (pulmonary artery systolic pressure, paradoxical systolic septal motion, RV-RA pressure gradient, RA pressure) are associated with short-term overall and PE-related mortality, long-term PE-related mortality, clinical adverse events, and the need for aggressive treatment,10,11,21–23 even in hemodynamically stable subjects.24–27 Although previous studies report a higher sensitivity (72% to 100%) for in-hospital mortality,11,23 direct comparison is difficult because of variations in the definition of the TTE-derived RV dysfunction variable and outcomes. Reported specificity (52% to 58%) is similar to that reported with the current data, and our results are in keeping with prior studies showing that RV dysfunction by TTE is an independent predictor of death.10,22–24
Our study has the following limitations. First, as a retrospective study, there could be potential selection bias, with TTE being performed in those with more severe PE (ie, a high fraction of central PE) and higher likelihood of RV dysfunction. Second, a quantitative continuous CT-based metric evaluating RV size relative to that of the LV was compared with a composite TTE-derived binary assessment of RV strain. The rationale for this comparison is that these are metrics routinely available to the physicians and hence relevant to clinical practice. Although CT RV/LV is routinely reported at our institution, we acknowledge that this may not be the case at other centers. Third, data on other CT variables such as septal bowing and contrast reflux into IVC were not available, and quantitative metrics such as obstruction index and clot volume were not included as several studies have demonstrated the lack of correlation of these variables with clinical outcome.2,17,41,42 In addition, quantitative echocardiographic right heart assessments such as RV dimensions, fractional area change, tricuspid annular plane systolic excursion, RV S0 (systolic excursion velocity), pulse Doppler myocardial performance index, and speckle tracking–based RV strain were not included in this study, as the purpose was to compare the prognostic significance of routinely available clinical information on RV dysfunction by the 2 imaging modalities. Future studies should incorporate these additional CT and TTE variables. Data on clinical presentation, stage of malignancy, and PE-related biomarkers, all of which have prognostic significance, were not available. About 21.2% of TTEs were performed before confirmation of diagnosis of PE by CTPA. Although the exact time of onset of symptoms and clinical diagnosis in these patients are unknown, it is unlikely that these TTEs were performed for unrelated reasons, as about half of these studies were performed within 6 hours before CTPA. The logistic regression model was used with about 5 events per variable.43 When the weak predictors (age, coronary artery disease, IVC filter use) were excluded, the events per variable increased to about 10, and the results for CT RV/LV
TABLE 3. Test Characteristics for Prediction of PE-related 30 d Mortality
TTE RV Strain FIGURE 1. Receiver operating characteristic curves of the 2 multivariate logistic regression models for PE-related 30-day mortality. Model 1 includes TTE RV strain. Model 2 includes the CT RV/LV diameter ratio. AUC indicates area under the curve.
W10 | www.thoracicimaging.com
Sensitivity Specificity PPV NPV
63.9 56.7 6.70 97.0
r
(46.2-79.2) (53.1-60.3) (4.30-9.90) (94.9-98.4)
CT RV/LVZ1.0 61.1 52.0 5.80 96.5
(43.5-76.9) (48.4-55.7) (3.70-8.70) (94.2-98.1)
2013 Lippincott Williams & Wilkins
J Thorac Imaging
Volume 29, Number 1, January 2014
diameter ratio and TTE RV strain were qualitatively similar. Balanced sensitivity and specificity was used to identify the optimal cutoff for CT RV/LV because there is no widely recognized standard cutoff. Although the 1.0 cutoff that we chose is the most frequently used value in the previous literature, we acknowledge that this may not be ideal and would depend on the clinical implication of false positives and false negatives. In summary, both the CT RV/LV diameter ratio and RV strain on TTE are significant predictors of PE-related short-term mortality with similar prognostic significance.
Prognosis by CT and TTE in Pulmonary Embolism
16.
17.
18.
REFERENCES 1. Silverstein MD, Heit JA, Mohr DN, et al. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based study. Arch Intern Med. 1998;158: 585–593. 2. Araoz PA, Gotway MB, Harrington JR, et al. Pulmonary embolism: prognostic CT findings. Radiology. 2007;242: 889–897. 3. Ghuysen A, Ghaye B, Willems V, et al. Computed tomographic pulmonary angiography and prognostic significance in patients with acute pulmonary embolism. Thorax. 2005;60: 956–961. 4. Kang DK, Thilo C, Schoepf UJ, et al. CT signs of right ventricular dysfunction: prognostic role in acute pulmonary embolism. JACC Cardiovasc Imaging. 2011;4:841–849. 5. Kumamaru KK, Hunsaker AR, Wake N, et al. The variability in prognostic values of right ventricular-to-left ventricular diameter ratios derived from different measurement methods on computed tomography pulmonary angiography: a patient outcome study. J Thorac Imaging. 2012;27:331–336. 6. Lu MT, Demehri S, Cai T, et al. Axial and reformatted fourchamber right ventricle-to-left ventricle diameter ratios on pulmonary CT angiography as predictors of death after acute pulmonary embolism. Am J Roentgenol. 2012;198:1353–1360. 7. Schoepf UJ, Kucher N, Kipfmueller F, et al. Right ventricular enlargement on chest computed tomography: a predictor of early death in acute pulmonary embolism. Circulation. 2004;110:3276–3280. 8. Casazza F, Becattini C, Bongarzoni A, et al. Clinical features and short term outcomes of patients with acute pulmonary embolism. The Italian Pulmonary Embolism Registry (IPER). Thromb Res. 2012;130:847–852. 9. Collomb D, Paramelle PJ, Calaque O, et al. Severity assessment of acute pulmonary embolism: evaluation using helical CT. Eur Radiol. 2003;13:1508–1514. 10. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet. 1999;353: 1386–1389. 11. Grifoni S, Olivotto I, Cecchini P, et al. Short-term clinical outcome of patients with acute pulmonary embolism, normal blood pressure, and echocardiographic right ventricular dysfunction. Circulation. 2000;101:2817–2822. 12. Vieillard-Baron A, Page B, Augarde R, et al. Acute cor pulmonale in massive pulmonary embolism: incidence, echocardiographic pattern, clinical implications and recovery rate. Intensive Care Med. 2001;27:1481–1486. 13. Apfaltrer P, Bachmann V, Meyer M, et al. Prognostic value of perfusion defect volume at dual energy CTA in patients with pulmonary embolism: correlation with CTA obstruction scores, CT parameters of right ventricular dysfunction and adverse clinical outcome. Eur J Radiol. 2012;81:3592–3597. 14. Meinel FG, Graef A, Bamberg F, et al. Effectiveness of automated quantification of pulmonary perfused blood volume using dual-energy CTPA for the severity assessment of acute pulmonary embolism. Invest Radiol. 2013;48:563–569. 15. Apfaltrer P, Henzler T, Meyer M, et al. Correlation of CT angiographic pulmonary artery obstruction scores with right r
2013 Lippincott Williams & Wilkins
19.
20. 21.
22.
23.
24.
25. 26. 27.
28.
29.
30.
31.
32. 33.
ventricular dysfunction and clinical outcome in patients with acute pulmonary embolism. Eur J Radiol. 2012;81:2867–2871. Park JR, Chang SA, Jang SY, et al. Evaluation of right ventricular dysfunction and prediction of clinical outcomes in acute pulmonary embolism by chest computed tomography: comparisons with echocardiography. Int J Cardiovasc Imaging. 2012;28:979–987. Ghaye B, Ghuysen A, Willems V, et al. Severe pulmonary embolism:pulmonary artery clot load scores and cardiovascular parameters as predictors of mortality. Radiology. 2006; 239:884–891. Kumamaru KK, Lu MT, Ghaderi Niri S, et al. Right ventricular enlargement in acute pulmonary embolism derived from CT pulmonary angiography. Int J Cardiovasc Imaging. 2013;29:705–708. van der Meer RW, Pattynama PM, van Strijen MJ, et al. Right ventricular dysfunction and pulmonary obstruction index at helical CT: prediction of clinical outcome during 3-month follow-up in patients with acute pulmonary embolism. Radiology. 2005;235:798–803. Goldhaber SZ. Echocardiography in the management of pulmonary embolism. Ann Intern Med. 2002;136:691–700. Kasper W, Konstantinides S, Geibel A, et al. Prognostic significance of right ventricular afterload stress detected by echocardiography in patients with clinically suspected pulmonary embolism. Heart. 1997;77:346–349. Ribeiro A, Lindmarker P, Juhlin-Dannfelt A, et al. Echocardiography Doppler in pulmonary embolism: right ventricular dysfunction as a predictor of mortality rate. Am Heart J. 1997;134:479–487. Fremont B, Pacouret G, Jacobi D, et al. Prognostic value of echocardiographic right/left ventricular end-diastolic diameter ratio in patients with acute pulmonary embolism: results from a monocenter registry of 1,416 patients. Chest. 2008;133: 358–362. Kucher N, Rossi E, De Rosa M, et al. Prognostic role of echocardiography among patients with acute pulmonary embolism and a systolic arterial pressure of 90 mm Hg or higher. Arch Intern Med. 2005;165:1777–1781. Zhu L, Yang Y, Wu Y, et al. Value of right ventricular dysfunction for prognosis in pulmonary embolism. Int J Cardiol. 2008;127:40–45. Kostrubiec M, Pruszczyk P, Bochowicz A, et al. Biomarkerbased risk assessment model in acute pulmonary embolism. Eur Heart J. 2005;26:2166–2172. Pieralli F, Olivotto I, Vanni S, et al. Usefulness of bedside testing for brain natriuretic peptide to identify right ventricular dysfunction and outcome in normotensive patients with acute pulmonary embolism. Am J Cardiol. 2006;97:1386–1390. Golpe R, Perez-de-Llano LA, Castro-Anon O, et al. Right ventricle dysfunction and pulmonary hypertension in hemodynamically stable pulmonary embolism. Respir Med. 2010;104: 1370–1376. Mansencal N, Joseph T, Vieillard-Baron A, et al. Diagnosis of right ventricular dysfunction in acute pulmonary embolism using helical computed tomography. Am J Cardiol. 2005;95: 1260–1263. Quiroz R, Kucher N, Schoepf UJ, et al. Right ventricular enlargement on chest computed tomography: prognostic role in acute pulmonary embolism. Circulation. 2004;109: 2401–2404. Ozsu S, Karaman K, Mentese A, et al. Combined risk stratification with computerized tomography/echocardiography and biomarkers in patients with normotensive pulmonary embolism. Thromb Res. 2010;126:486–492. Henzler T, Roeger S, Meyer M, et al. Pulmonary embolism: CT signs and cardiac biomarkers for predicting right ventricular dysfunction. Eur Respir J. 2012;39:919–926. Lu MT, Ersoy H, Whitmore AG, et al. Reformatted fourchamber and short-axis views of the heart using thin section (