Cor Vasa 2018, 60(2):e105-e113 | DOI: 10.1016/j.crvasa.2017.04.003

Role of echocardiography and cardiac biomarkers in prediction of in-hospital mortality and long-term risk of brain infarction in pulmonary embolism patients

David Vindiš^a, Martin Hutyra^a,*, Daniel Šaňák^b, Michal Král^b, Eva Čecháková^c, Jana Zapletalová^d, Simona Littnerová^e, Tomáš Adam^f, Jan Přeček^a, Miloš Táborský^a

^a I. interní klinika - kardiologická, Univerzita Palackého a Fakultní nemocnice Olomouc, Olomouc, Česká republika

^b Komplexní cerebrovaskulární centrum, Neurologická klinika, Univerzita Palackého a Fakultní nemocnice Olomouc, Olomouc, Česká republika

^c Radiologická klinika, Univerzita Palackého a Fakultní nemocnice Olomouc, Olomouc, Česká republika

^d Ústav lékařské biofyziky, Univerzita Palackého, Olomouc, Česká republika

^e Institut biostatistiky a analýz Lékařské fakulty Masarykovy univerzity, Brno, Česká republika

^f Oddělení klinické biochemie, Fakultní nemocnice Olomouc, Olomouc, Česká republika

Introduction: The aim of prospective study was to evaluate the ability of echocardiography and cardiac biomarkers to predict in-hospital mortality and the risk of brain infarction during a 12-month follow-up period (FUP) with anticoagulation in pulmonary embolism (PE) patients.

Methods: Eighty-eight consecutive acute PE patients (39 men, mean age 63 years) were enrolled; 78 underwent baseline echocardiography and brain magnetic resonance imaging (MRI). After a 12-month FUP, 58 underwent brain MRI. In-hospital mortality and the rates of new ischemic brain lesions (IBL) on MRI with clinical ischemic stroke (IS) events were predicted based on echocardiography (patent foramen ovale presence with right-to-left shunt - PFO/RLS; right/left ventricle diameter ratio - RV/LD; tricuspid annulus plane systolic excursion - TAPSE; tricuspid annulus systolic velocity - ST; pulmonary artery systolic pressure - PASP) and biomarkers results (amino-terminal fragment of brain natriuretic peptide - NT-proBNP and cardiac troponin T - cTnT).

Results: Our series involved 88 patients, of whom 11 (12.5%) presented high-risk PE, 24 (27.3%) intermediate-high risk PE, 19 (21.6%) intermediate-low risk PE and 34 (38.6%) patients had low risk PE.
Nine patients (10.2%) died during hospitalization including high-risk PE [6/9 (66.6%)] and intermediate--high-risk PE [3/24 (12.5%)]. cTnT [odds ratio (OR) 4.3; 95% confidence interval 0.59-31.3, p = 0.014], NT-pro-BNP (OR 14.2 [1.5-133.4], p = 0.02), RV/LD ≥0.79 (OR 36.6 [4.2-316.4], p = 0.001), TAPSE (OR 0.55 [0.34-0.92], p = 0.022) and PASP ≥51.5 mmHg (OR 33.3 [3.8-292.6], p = 0.022) were predictors of in-hospital mortality.
Seventeen patients (19.3%) experienced IS (n = 8) or new IBL (n = 9). On multivariate analysis, PFO/RLS (OR 27.1 [3.0-245.3], p = 0.003) and ST ≤14.5 cm/s (OR 34.1 [CI 3.4-344.0], p = 0.003) were independent predictors of IS and IBL risk.

Conclusions: High blood troponin T, NT-proBNP, RV dilatation/systolic dysfunction and pulmonary hypertension predicted in-hospital mortality. PFO/RLS presence and ST were predictors of clinically apparent/silent brain infarction.

Keywords: Brain infarction; Cardiac troponin; Echocardiography; Magnetic resonance imaging; Mortality; Natriuretic peptides; Pulmonary embolism

Received: April 17, 2017; Revised: April 24, 2017; Accepted: April 25, 2017; Published: April 1, 2018  Show citation