p.19 抽出テキスト

............................................................................................................................................................................. mortality in patients with acute PE.67,158 A patent foramen ovale also increases the risk of ischaemic stroke due to paradoxical embolism in patients with acute PE and RV dysfunction.182,183

5.2.2 Computed tomographic pulmonary angiography CTPA parameters used to stratify the early risk of patients with PE are summarized in Supplementary Data Table 3. Fourchamber views of the heart by CT angiography can detect RV enlargement (RV end-diastolic diameter and RV/LV ratio measured in the transverse or four-chamber view) as an indicator of RV dysfunction. The prognostic value of an enlarged RV is supported by the results of a prospective multicentre cohort study in 457 patients.184 In that study, RV enlargement (defined as an RV/LV ratio >_0.9) was an independent predictor of an adverse inhospital outcome, both in the overall population with PE [hazard ratio (HR) 3.5, 95% CI 1.67.7] and in haemodynamically stable patients (HR 3.8, 95% CI 1.310.9).184 A meta-analysis of 49 studies investigating >13 000 patients with PE confirmed that an increased RV/LV ratio of >_1.0 on CT was associated with a 2.5fold increased risk for all-cause mortality [odds ratio (OR) 2.5, 95% CI 1.83.5], and with a five-fold risk for PE-related mortality (OR 5.0, 95% CI 2.79.2).185 Mild RV dilation (RV/LV slightly above 0.9) on CT is a frequent finding (>50% of haemodynamically stable PE patients186), but it probably has minor prognostic significance. However, increasing RV/ LV diameter ratios are associated with rising prognostic specificity,187,188 even in patients considered to be at ‘low’ risk on the basis of clinical criteria.186 Thus, RV/LV ratios >_ 1.0 (instead of 0.9) on CT angiography may be more appropriate to indicate poor prognosis. Apart from RV size and the RV/LV ratio, CT may provide further prognostic information based on volumetric analysis of the heart chambers189191 and assessment of contrast reflux to the inferior vena cava (IVC).185,192,193

5.3 Laboratory biomarkers

5.3.1 Markers of myocardial injury Elevated plasma troponin concentrations on admission may be associated with a worse prognosis in the acute phase of PE. Cardiac troponin I or T elevation are defined as concentrations above the normal limits, and thresholds depend on the assay used; an overview of the cut-off values has been provided by a meta-analysis.194 Of patients with acute PE, between 30 (using conventional assays)194,195 and 60% (using highsensitivity assays)196,197 have elevated cardiac troponin I or T concentrations. A meta-analysis showed that elevated troponin concentrations were associated with an increased risk of mortality, both in unselected patients (OR 5.2, 95% CI 3.38.4) and in those who were haemodynamically stable at presentation (OR 5.9, 95% CI 2.713.0).195 On their own, increased circulating levels of cardiac troponins have relatively low specificity and positive predictive value for early mortality in normotensive patients with acute PE. However, when interpreted in combination with clinical and imaging findings, they may improve the identification of an elevated PE-related risk and the further prognostic stratification of such patients (Supplementary Data Table 4). At the other end of the severity spectrum, high-sensitivity troponin assays possess a high negative predictive value in the setting of acute PE.197 For example, in a prospective multicentre cohort of 526 normotensive patients, high-sensitivity troponin T concentrations <14 pg/mL had a negative predictive value of 98% for excluding an adverse in-hospital clinical outcome.63 Age-adjusted high-sensitivity troponin T cutoff values (>_14 pg/mL for patients aged <75 years and >_45 pg/mL for those >_75 years) may further improve the negative predictive value of this biomarker.196

Heart-type fatty acid-binding protein (H-FABP), an early and sensitive marker of myocardial injury, provides prognostic information in acute PE, both in unselected198,199 and normotensive patients.200,201 In a meta-analysis investigating 1680 patients with PE, H-FABP concentrations >_6 ng/mL were associated with an adverse short-term outcome (OR 17.7, 95% CI 6.051.9) and all-cause mortality (OR 32.9, 95% CI 8.8123.2).202

5.3.2 Markers of right ventricular dysfunction RV pressure overload due to acute PE is associated with increased myocardial stretch, which leads to the release of B-type natriuretic peptide (BNP) and N-terminal (NT)-proBNP. Thus, the plasma levels of natriuretic peptides reflect the severity of RV dysfunction and haemodynamic compromise in acute PE.203 A meta-analysis found that 51% of 1132 unselected patients with acute PE had elevated BNP or NT-proBNP concentrations on admission; these patients had a 10% risk of early death (95% CI 8.013%) and a 23% (95% CI 2026%) risk of an adverse clinical outcome.204 Similar to cardiac troponins (see above), elevated BNP or NTproBNP concentrations possess low specificity and positive predictive value (for early mortality) in normotensive patients with PE,205 but low levels of BNP or NT-proBNP are capable of excluding an unfavourable early clinical outcome, with high sensitivity and a negative predictive value.180 In this regard, an NT-proBNP cut-off value <500 pg/mL was used to select patients for home treatment in a multicentre management study.206 If emphasis is placed on increasing the prognostic specificity for an adverse early outcome, higher cut-off values >_600 pg/mL might be more appropriate.207

5.3.3 Other laboratory biomarkers Lactate is a marker of imbalance between tissue oxygen supply and demand, and consequently of severe PE with overt or imminent haemodynamic compromise. Elevated arterial plasma levels >_2 mmol/L predict PE-related complications, both in unselected208 and in initially normotensive209,210 PE patients. Elevated serum creatinine levels and a decreased (calculated) glomerular filtration rate are related to 30 day all-cause mortality in acute PE.211 Elevated neutrophil gelatinase-associated lipocalin and cystatin C, both indicating acute kidney injury, are also of prognostic value.212 A recent meta-analysis investigating 18 616 patients with acute PE found that hyponatraemia predicted in-hospital mortality (OR 5.6, 95% CI 3.49.1).213 Vasopressin is released upon endogenous stress, hypotension, and low CO. Its surrogate marker, copeptin, has been reported to be useful for risk stratification of patients with acute PE.214,215 In a singlecentre derivation study investigating 268 normotensive PE patients, copeptin levels >_24 pmol/L were associated with a 5.4-fold (95% CI 1.717.6) increased risk of an adverse outcome.216 These results Downloaded from academic.oup.com/eurheartj/article/41/4/543/5556136 by guest on 07 October 2026