Cardiac computed tomography tissue tracking of the right ventricle longitudinal strain in patients with pulmonary thromboembolism

Abstract

Abstract Background Pulmonary embolism (PE) represents a common entity with a broad clinical spectrum. Right ventricle (RV) dysfunction remains the leading cause of mortality. Current guidelines recommend a multimodal stratification, including echocardiographic or computed tomography and high-sensitive cardiac troponin (hs-cTn) for stratification of the risk. It is of utmost importance to identify those patients at highest risk for clinical deterioration. Longitudinal Strain (LS) is a new generation technique that quantifies the myocardial deformation without geometric assumptions, allowing the assessment of the RV myocardial function. Purpose Firstly, we aim to detect early RV abnormalities in patients with acute PE, with the LS value. Secondly, evaluate if there is a significant difference between Right Ventricle Longitudinal Strain (RVLS) in patients with acute PE versus controls without any cardiac pathology. Methods Observational retrospective single-centre study from patients 18 years or older admitted to the ER with proved PE by multiphase CT angiogram from 2016 to 2020. We included a control group from non-cardiac chest pain causes with multiphase CT angiogram negative for PE, without evidence of valvular disease and a coronary calcium score of 0. We evaluated the RV in the four-chamber views to obtain the LS value using tissue tracking CT software. A ROC curve assessed the LS performance for RV dysfunction detection. We calculated the Youden index for every value to identify the best cut-point. We considered a P value of <0.05 for statistical significance with a confidence interval of 95%. Results From a total of 53 patients with PE diagnostic only 10 fulfilled all the inclusion criteria. For the control group there were 500 patients with CT angiogram of which only 33 had a complete multiphase sequence. 63% of the patients included in the PE group were considered of high-risk PE with expression of hs-cTn, the rest were classified as intermediate risk. Average hs-cTn in the high-risk group was 149.3 ng/dL, which correlated with a lower RVLS in comparison to the intermediate risk PE without being statistically significative. Brain Natriuretic Peptide (BNP) had a higher correlation between lower RVLS in patients with intermediate and high-risk PE, respectively (80.56 pg/mL vs 466.02 pg/ml). There is an overt tendency for alteration of the RVLS in patients with PE compared to the control group (−8.41% vs. −21.09%; P<0.001). The AUC for the cut-point of −18.25% for the LS was 0.89 (CI 95%, 0.70–1.00), with a sensibility of 90.9% and specificity of 90% for RV dysfunction. Conclusion There is no reference value of RVLS in the general population. We propose the cut-point value of −18.25% for LS by cardiac CT to detect early RV dysfunction in patients with PE. Based on our findings we proposed a LS value of −21.09% cut-off point for patients with no cardiac disease. The correlation and clinical impact as prognostic factors, is yet to be studied. Funding Acknowledgement Type of funding sources: None.