SEARCH

SEARCH BY CITATION

Keywords:

  • anatomic distribution;
  • computer tomography;
  • prognosis;
  • pulmonary artery obstruction;
  • pulmonary embolism;
  • troponin

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

Background.  The aim of the study was to investigate the association between the proximal level of the clot and the severity of pulmonary embolism (PE).

Methods.  The cohort consisted of 99 consecutive patients with PE diagnosed by multi-detector computed tomography. A new score was constructed by calculating the mean value of the largest affected vessel [sub-segmental = 1, segmental = 2, lobar = 3, main pulmonary artery (MPA) = 4] in each lung.

Results.  A significant association was found between the most proximal level of PE and pulmonary artery obstruction index (PAOI) (P < 0.0001), right ventricular (RV)/left ventricular (LV) ratio (P < 0.0001), and PaO2 (P = 0.004). No significant association was found between systolic blood pressure and the level of PE. Troponin-T was elevated in none of the sub-segmental, 5% of segmental, 20% of lobar, and in 56% of PEs in the MPA (P = 0.001). Significant association was found between the proposed score and PAOI (P < 0.0001), RV/LV ratio (P < 0.0001), PaO2 (P < 0.008). Troponin-T was elevated in 10% of level 1, 0% of level 2, 43% level of 3, 66% of level 4 PE (P < 0.0001). Cut-off level score 4 yielded a sensitivity of 84% and a specificity of 74% for the detection of elevated troponin-T.

Conclusions.  In conclusion, the study indicates that both the most proximal level of PE and the proposed score are related to the severity of PE as determined by blood oxygenation, biochemical and radiological parameters and could therefore be of value for rapid risk stratification of PE. However, the prognostic value of these classifications and their clinical significance needs to be evaluated in properly designed studies.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

Pulmonary embolism (PE) is a potentially fatal condition with a 3-month mortality rate reaching up to 15% [1]. Clinically, PE ranges from anatomically small and clinically asymptomatic emboli to massive, proximal emboli with cardiogenic shock.

Spiral computed tomography (CT) scanning has considerably modified the diagnostic approach for PE [2, 3] and is now widely applied as an initial imaging test for the diagnostic work-up of PE [4]. PE is diagnosed by demonstrating a filling defect in the pulmonary arteries. The extent of PE is commonly expressed by indicating the anatomic level of the most proximal vessel affected by a clot [5].

Rapid and effective diagnosis and treatment is essential to avoid high mortality. The key to appropriate therapy is risk stratification [6]. Low risk patients have an excellent prognosis with anticoagulation alone, whilst high risk patients, who comprise haemodynamically unstable patients and normotensive patients with signs of right ventricular (RV) dysfunction, might benefit form thrombolysis. Assessment of the haemodynamic status is the mainstay for risk stratification of patients with PE and for subsequent choice of therapy [7].

In normotensive patients, however, echocardiography is considered as the principle tool for risk stratification, but the limited round-the-clock availability in many hospitals, the relatively high cost, and the occasional poor imaging quality are major drawbacks of this modality [6]. Cardiac biomarkers have been validated and proposed as means for risk stratification, but are still not incorporated in management guidelines [8]. Moreover, the release of cardiac biomarkers could be delayed until 6–12 h following an episode of acute PE.

At least two radiological parameters have been shown to be associated with poor short-term outcome: high pulmonary artery obstruction index (PAOI) and increased right ventricle (RV)/left ventricle (LV) ratio [9, 10]. However, none of these markers have therapeutic implications.

Our aim of this study was to investigate the association between the most proximal level of the clot and the severity of PE as determined by clinical, biochemical and radiological parameters as well as to develop and to evaluate a new simple score that reflects the total clot burden.

Patients and methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

Study population

A total of 432 consecutive patients referred to the Emergency Department of the Østfold Hospital Trust, Fredrikstad, Norway, for suspected PE during February 2002 to December 2003, were found eligible to participate in a clinical trial evaluating a decision-based algorithm combining clinical probability assessment, D-dimer, and multi-detector computerized tomography. The study design and main outcomes of the study have previously been reported in detail [11]. After reviewing all CT images by a reviewing committee, PE was deemed present by consensus in 100 patients. One of these patients was excluded as thrombolysis was given before CT was performed.

Clinical data were recorded prospectively. Percentage oxygen (O2) saturation was measured in 90 patients and arterial blood gas analysis was performed in 86 patients. Alveolar–arterial oxygen gradient [P(A–a)O2gradient] was calculated using the following equation: P(A–a)O2 = [FIO2(PB − PAH2O) − 1.25(PaCO2) − PaO2], where FIO2 was estimated as 0.209, PB as 100 kPa, PAH2O as 6.3 kPa, PAO2 was the calculated partial pressure (kPa) of oxygen in the alveolar space, PaCO2 was the measured partial pressure of CO2 in arterial blood and PaO2 was the measured partial pressure of O2 in arterial blood.

Quantitative real-time troponin-T (ECLIA-Roche, Mannheim, Germany) was assayed during the first 48 h following the establishment of PE diagnosis. The assay has a detection limit of >0.01 ng mL−1, and levels >0.01 ng mL−1 were regarded as indicating myocardial injury. If more than one test result was available, the highest troponin-T value was chosen.

The indications for thrombolytic therapy were hypotension, severe hypoxaemia or transthoracic echocardiographic evidence of RV dysfunction. The entire cohort was followed-up for 3 months to assess the outcome. The follow-up was completed in 99% of the patients.

The study protocol was approved by the Regional Ethics Committee, and written informed consent was obtained from all participants in accordance with the revised Declaration of Helsinki.

CT imaging

A 4-detector row CT scanner (MX 8000 Marconi, Cleveland, OH, USA) was used. A total volume of 120 mL non-ionic contrast medium (Iomeron, Bracco, Milano, Italy) containing 300 mg iodine mL−1 was injected intravenously at a rate of 4 mL s−1. An automated bolus tracking system initiated scanning 4 s after the contrast medium reached the aortic arch. Scanning was carried out with 2.5 mm collimation, a pitch of 1.25, 120 kV, and 175 mAs with a gantry rotation time of 0.5 s.

CT images interpretation

Computed tomography images were retrieved from the picture archiving and communication system (PACS) for verification of diagnosis, determination of the most proximal level of the clot, and for the estimation of RV/LV ratio and PAOI. For patients with PE, 89 examinations were retrievable by PACS; the remaining examinations were reviewed on hard copies.

Pulmonary embolism was diagnosed if a filling defect or complete occlusion was seen in the pulmonary vasculature and was considered absent when the pulmonary arteries, including sub-segmental branches, was visualized and was free of filling defects. The proximal extension of the clot was determined in each lung and was classified into four levels: the main pulmonary arteries (MPA), lobar arteries, segmental, and sub-segmental vessels [5]. As the most proximal level of the clot reflects the clot burden on one side only, we proposed a new score that would accommodate for the bilateral burden of the clot. The score was constructed by assigning a number to each of the four main vessels: 1 for sub-segmental, 2 for segmental, 3 for lobar, and 4 for MPA. The mean score of the largest affected vessel in each lung, ranging from 1 to 4, was calculated and adjusted to the higher whole number.

Pulmonary artery obstruction index was estimated by one radiologist (L.O.H.) on the basis of the degree of obstruction and the location of the thrombus on CT according to the method described by Qanadli et al. [12]. RV/LV ratio was assessed by another radiologist (B.E.N.) by computing the ratio between the width of the right and the left ventricular (LV) cavities assessed on axial images obtained at the plane maximal distances between the ventricular endocardial free wall and the interventricular septum, perpendicular on the long axis [9, 13].

Statistical analysis

Our hypothesis was that the proximal level of PE is associated with the severity of PE determined by clinical, biochemical and radiological markers. Data were expressed by the median. The 95% confidence intervals (CI) for the proportions were calculated by the binomial distribution. Kruskall–Wallis test was used to compare the distribution of continuous variables and serial groups. Contingency tables were used to determine the association between the level of PE and between categorical variables. Spearman's rho correlation coefficient was determined to measure the strength of association between the level of PE and between PAOI [14]. P < 0.05 was considered statistically significant. The statistical analysis was performed with spss version 14.0 (SPSS Inc., Chicago, IL., USA). The sensitivity, specificity, positive and negative predictive values (NPV), were calculated for the conventional classification and new score; receiver operating characteristics curves (ROC) were constructed by plotting sensitivity versus 1 − specificity, and area under the curve (AUC) was calculated for both classification using openepi software (http://www.openepi.com, copyright ‘The Open Source Initiative’).

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

The final cohort of this study consisted of 99 patients with PE diagnosed by unequivocally positive CT finding. The anatomic distribution of PE is summarized in Table 1. The most proximal extension of the clot was the MPA in 53 patients (53%), lobar arteries in 14 (14%), segmental vessels in 26 (26%) and the sub-segmental vessels in six (6%).

Table 1.   The most proximal extension of the clot in both lungs and in each lung separately
 Proximal level of the PE
Most proximal extensionRight lungLeft lung
  1. MPA, main pulmonary artery.

Sub-segmental6716
Segmental262312
Lobar141312
MPA535036
Total999376

Most proximal level of PE (conventional classification)

A significant association was found between the most proximal level of PE and between the median values of PAOI, RV/LV ratio, PaO2, P(A–a)O2 gradient, as shown in Table 2. A strong correlation was found between PAOI and the most proximal level of PE (r = 0.86). The association between systolic blood pressure and the level of PE was not statistically significant although a clear trend of lower systolic blood pressure was found towards the more proximal emboli. Troponin-T was measured in 67 patients. It was elevated (>0.01 ng mL−1) in 25 patients (37%; 95% CI 26–50): in none with sub-segmental PE, in one of 19 (5%; 95% CI 0.1–26) patients with segmental PE, in one of five (20%; 95% CI 0.5–71) patients with lobar PE, and in 23 of 41 (56%; 95% CI 40–71) patients with PE in the MPA (P = 0.001). All the 12 patients who received thrombolytic treatment had PE extending to the MPA (P = 0.01) (Table 3). From the ROC curve (not shown) the MPA level was chosen as the optimal cut-off for this classification; the performance of this classification for the detection of RV/LV ratio >1.0 and >1.4 and elevated troponin-T is summarized in Table 4.

Table 2.   Median values of blood oxygenation parameters, systolic blood pressure and radiological markers according to the conventional classification (most proximal level of PE) and the proposed new score for estimation of total clot burden
MedianConventional classificationPNew scoreP
Subseg (N = 6)Seg (N = 26)Lob (N = 14)MPA (N = 53)1 (N = 18)2 (N = 31)3 (N = 8)4 (N = 42)
  1. Subseg, subsegmental; Seg, segmental; Lob, lobar; MPA, main pulmonary artery.

PAOI (%)2.57.516.250<0.00015153050<0.0001
RV/LV ratio0.950.91.01.2<0.00010.91.00.951.3<0.0001
PaO2 (kPa)10.41010.38.20.00410.49.88.08.30.008
P(A-a) (kPa)33.53.55.8<0.00013.13.94.75.9<0.0001
Systolic BP (mm Hg−1)150140135130NS140139138130NS
Table 3.   Number (%) of patients with hypotension, elevated PAOI, RV/LV ratio, and troponin-T, according to the conventional classification (most proximal level of PE) and the proposed new score for estimation of total clot burden
 Conventional classificationNew score
Subseg (N = 6)Seg (N = 26)Lob (N = 14)MPA (N = 53)1 (N = 18)2 (N = 31)3 (N = 8)4 (N = 42)
  1. Subseg, subsegmental; Seg, segmental; Lob, lobar; MPA, main pulmonary artery; SBP, systolic blood pressure. aModerately dilated right ventricle. bSeverely dilated right ventricle.

SBP < 100 mmHg (N = 5)01 (20)04 (80)01 (20)04 (80)
PAOI > 40% (N = 39)00039 (100)002 (5)37 (95)
RV/LV ratio > 1.0 (N = 38)a02 (5)5 (13)31 (81)2 (5)5 (13)3 (8)28 (74)
RV/LV ratio > 1.4 (N = 18)b001 (5)17 (95)001 (5)17 (95)
Troponin-T > 0.01 ng mL−1 (N = 25)01 (4)1 (4)23 (92)1 (4)03 (12)21 (84)
Thrombolysis (N = 12)00012 (100)001 (8)11 (92)
Table 4.   Performance (95% CI) of the conventional classification (cut-off level = main pulmonary artery) and the proposed new score (cut-off level = score 4)
 Conventional classificationNew score
SensitivitySpecificityPPVNPVAUCSensitivitySpecificityPPVNPVAUC
  1. PPV, positive predictive value; NPV, negative predictive value; AUC, area under the ROC curve. aModerately dilated right ventricle. bSeverely dilated right ventricle.

RV/LV ratio > 1.0a82 (67–91)64 (51–75)58 (45–71)85 (72–92)0.7574 (58–85)77 (65–86)67 (52–79)82 (71–90)0.77
RV/LV ratio > 1.4b94 (74–99)56 (45–66)32 (21–45)97 (89–100)0.7694 (74–99)69 (58–78)40 (27–55)98 (91–100)0.83
Troponin-T  > 0.01 ng mL−192 (75–98)58 (42–71)56 (41–70)92 (76–98)0.7584 (65–94)74 (59–85)65 (48–80)89 (74–95)0.82

Total clot burden (new score)

Pulmonary embolism was classified as score 1 in 18 (18%), score 2 in 31 (31%), score 3 in eight (8%), and score 4 in 42 (42%) patients. A significant association was found between the new score and the median values of PAOI, RV/LV ratio, PaO2, P(A–a)O2 gradient, as shown in Table 2. A highly strong correlation was found between PAOI and the proposed score (r = 0.90). The histograms (Fig. 1) present a comparison between the median values of PAOI (Fig. 1a), RV/LV ratio (Fig. 1b) and PaO2 (Fig. 1c) for the different levels of the two scales. Troponin-T was elevated in one of 10 (10%; 95% CI 0.2–44) patients with score 1, in 0 of 18 (0%; 95% CI 0–15) with score 2, in three of seven patients (43%; 95% CI 10–81) with score 3, and in 21 of 32 (66%; 95% CI 47–81) patients with score 4 (P < 0.0001). Thrombolytic therapy was given to one patient with score 3 and 11 patients with score 4 (P = 0.003). Table 4 summarizes the performance of this score at a cut-off level score 4 that was chosen from the ROC curve for the detection RV/LV ratio >0.1 and >1.4 and elevated troponin-T.

image

Figure 1.  Median value of pulmonary artery obstruction index (PAOI) (a), right ventricular/left ventricular (RV/LV) ratio (b), PaO2 (c) according to most proximal level of pulmonary embolism (white boxes) and the proposed new score (black boxes).

Download figure to PowerPoint

Four patients died during the first 3 months: three patients had PE extending to the MPA (new score = 4), whilst the fourth patient had PE extending to the lobar arteries (new score = 2), (P > 0.05).

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

The current study indicates that the most proximal level of the clot is related to the severity of PE as determined by clinical, biochemical and radiological parameters. The classification of PE according to its most proximal level is frequently reported in the clinical as well as in the academic settings [11, 15, 16]. However, the clinical significance of this classification is not well known. Although it was not unexpected to find associations between the most proximal level of the clot and severity of PE, there is limited data in the literature to support this [17].

Our results show a highly significant association between the most proximal level of the clot and PAOI and RV/LV ratio as prognostic markers for PE [18]. These markers have been validated in several studies, and most of these studies have demonstrated a significantly worse outcome in patients with PAOI exceeding 40–60% and/or RV/LV ratio exceeding 1.0–1.4 [9, 10, 13, 19]. Moreover, the proximal level of the clot correlates with the blood oxygenation parameters. Despite the presence of a trend for lower systolic blood pressure, this association did not reach statistical significance owing to the small number of patients in various categories.

In the present study we propose a new CT angiographic score for PE based on the largest affected vessel in each lung regardless of the degree of obstruction. In contrast to the conventional classification which represents the most proximal level of the clot, the proposed score accommodates for the clot burden also in the contralateral lung. Figure 1a illustrates the change in the median value of PAOI in relation to the different levels of the proposed score when compared with the conventional classification. A slightly stronger correlation was found between PAOI and the proposed score (r = 0.90) than between PAOI and the conventional classification (r = 0.86) confirming the visual impression conveyed by the histogram (Fig. 1a). Compared with PAOI, the proposed new score may be easier to remember, and in addition, it can be calculated by the physician as the proximal extension is frequently mentioned in radiology reports.

Having proved the association between the proximal level of the clot and the various markers for the severity of PE, we proceeded to determine the potential clinical value of this finding. Several prognostic radiological markers have been developed to assess the severity of PE [18]. The current recommendations for risk stratification of PE in normotensive patients involve an initial screening with cardiac biomarkers followed by transthoracic echocardiography (TTE) in those with elevated cardiac biomarkers [8]. RV dysfunction determined by TTE is known to be associated with increased mortality and is commonly considered to be an indication for thrombolytic treatment [6, 20]. Moreover, RV/LV ratio has been shown to be correlated with RV dysfunction [17]. In our study, six of the 25 patients (24%) with elevated troponin-T had normal RV/LV ratio and five patients (20%) had moderately increased RV/LV ratio (1.1–1.4). The remaining patients (n = 14, 56%) had RV/LV ratio > 1.4. However, our study shows that 84–92% of patients with elevated troponin I were score 4 or MPA embolism respectively (NPV 90%). This indicates that the level of PE could be a better predictor for troponin-T elevation than the RV/LV ratio. It is logical to assume that the majority of patients with echocardiographic manifestations of RV dysfunction have PE extending to the MPA or score 4. Furthermore, all the patients who received thrombolytic treatment had PE extending to the MPA or score 4 apart from one patient with score 3 PE.

Our study demonstrates that the proximal extension of PE can be of unrecognized value as a prognostic marker for rapid risk stratification of patients with PE. It could be utilized to restrict the use of cardiac biomarkers and to select patients for echocardiographic examination, hence, to streamline the management of PE and to direct rational use of resources. The prognostic value of these classifications and their clinical significance needs however to be evaluated in properly designed prospective studies that should include cardiac biomarkers and evaluation with echocardiography before it can be applied in clinical practice. Moreover, whether the proposed score has any clinically significant additional advantage over the conventional classification will remain to be shown in future studies.

The advantage of this score over the conventional classification lies in its ability to reflect the bilateral extension of the clot. For example, a unilateral PE extending to right pulmonary artery will score 2, but it will still be interpreted as MPA embolus. In this condition, the new score might be more accurate to reflect the severity of the condition, as it is unlikely that such emboli would result in RV dilatation. The new score displays a consistently higher AUC compared with conventional classification as shown in Table 4, indicating its greater discriminating power. Moreover, patients with score 3 and 4 together constitute 50% of the patients in the new score, compared with 65% of the patients having PE extending in the two most proximal levels of the conventional classification. This would result in a lower number of patients needed to be tested, should one consider further evaluation of patients with troponin-T or TTE. However, our cohort is too small to demonstrate a statistically significant difference between the two scores.

Our study has some limitations. First, troponin-T was not measured in all patients with confirmed PE. Secondly, the study did not include TTE for comparison, which is still regarded as the principle tool for risk stratification in normotensive patients [6, 20]. Thirdly, the number of patients in each category was relatively small which resulted in wide confidence intervals. Fourthly, none of these two classifications takes into account the pre-existing cardiac reserve which has been previously shown to be an important determinant of patients' outcome [6, 21]. Thus, it is unknown how these classifications fare in patients with comorbid cardiorespiratory disease. Finally, the low mortality in our material and the nonsignificant association between systolic blood pressure and the level of PE preclude us from making definite conclusion regarding the predictive value of this classification.

We conclude that the most proximal level of PE and the total clot burden as determined with the new score is related to the severity of PE as determined by blood oxygenation, biochemical and radiological parameters. They could hence be of unrecognized value as prognostic markers that can be used for rapid risk stratification of patients PE and to streamline the management of PE.

Conflict of interest statement

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References

This study was financially supported by a research grant from the Eastern Norway Health Authority.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conflict of interest statement
  8. References
  • 1
    Goldhaber SZ, Visani L, De RM. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet 1999; 353: 13869.
  • 2
    Fedullo PF, Tapson VF. Clinical practice. The evaluation of suspected pulmonary embolism. N Engl J Med 2003; 349: 124756.
  • 3
    Goodman LR, Curtin JJ, Mewissen MW et al. Detection of pulmonary embolism in patients with unresolved clinical and scintigraphic diagnosis: helical CT versus angiography. AJR Am J Roentgenol 1995; 164: 136974.
  • 4
    Schoepf UJ, Goldhaber SZ, Costello P. Spiral computed tomography for acute pulmonary embolism. Circulation 2004; 109: 21607.
  • 5
    Uflacker R. Atlas of Vascular Anatomy: an Angiographic Approach, 1st edn. Baltimore, MD: Lippencott Williams & Wilkins, 1997.
  • 6
    Goldhaber SZ. Pulmonary embolism. Lancet 2004; 363: 1295305.
  • 7
    Buller HR, Agnelli G, Hull RD, Hyers TM, Prins MH, Raskob GE. Antithrombotic therapy for venous thromboembolic disease: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126 (Suppl.): 401S28S.
  • 8
    Kucher N, Goldhaber SZ. Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation 2003; 108: 21914.
  • 9
    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: 798803.
  • 10
    Wu AS, Pezzullo JA, Cronan JJ, Hou DD, Mayo-Smith WW. CT pulmonary angiography: quantification of pulmonary embolus as a predictor of patient outcome–initial experience. Radiology 2004; 230: 8315.
  • 11
    Ghanima W, Almaas V, Aballi S et al. Management of suspected pulmonary embolism (PE) by D-dimer and multi-slice computed tomography in outpatients: an outcome study. J Thromb Haemost 2005; 3: 192632.
  • 12
    Qanadli SD, El HM, Vieillard-Baron A et al. New CT index to quantify arterial obstruction in pulmonary embolism: comparison with angiographic index and echocardiography. AJR Am J Roentgenol 2001; 176: 141520.
  • 13
    Ghuysen A, Ghaye B, Willems V et al. Computed tomographic pulmonary angiography and prognostic significance in patients with acute pulmonary embolism. Thorax 2005; 60: 95661.
  • 14
    Altman DG. Practical Statistics for Medical Research, 1st edn. London, FL: Chapman & Hall/CRC, 1999.
  • 15
    Perrier A, Roy PM, Aujesky D et al. Diagnosing pulmonary embolism in outpatients with clinical assessment, D-dimer measurement, venous ultrasound, and helical computed tomography: a multicenter management study. Am J Med 2004; 116: 2919.
  • 16
    Perrier A, Roy PM, Sanchez O et al. Multidetector-row computed tomography in suspected pulmonary embolism. N Engl J Med 2005; 352: 17608.
  • 17
    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: 12603.
  • 18
    Ghaye B, Ghuysen A, Bruyere PJ, D'Orio V, Dondelinger RF. Can CT pulmonary angiography allow assessment of severity and prognosis in patients presenting with pulmonary embolism? What the radiologist needs to know. Radiographics 2006; 26: 2339.
  • 19
    Schoepf UJ, Kucher N, Kipfmueller F, Quiroz R, Costello P, Goldhaber SZ. Right ventricular enlargement on chest computed tomography: a predictor of early death in acute pulmonary embolism. Circulation 2004; 110: 327680.
  • 20
    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: 281722.
  • 21
    Wicki J, Perrier A, Perneger TV, Bounameaux H, Junod AF. Predicting adverse outcome in patients with acute pulmonary embolism: a risk score. Thromb Haemost 2000; 84: 54852.