Clinical study on feasibility and repeatability of left ventricular systolic function assessment in patients with chronic kidney disease using artificial intelligence-based automatic strain technique
Main Article Content
Abstract
Background: Studies have shown that cardiovascular damage is a common complication in patients with chronic kidney disease (CKD), and left ventricular longitudinal strain measurement (GLS) is superior to LVEF in evaluating left ventricular systolic function. However, it has not yet been widely accepted as a routine clinical examination because it requires proficiency and is time-consuming. Therefore, the aim of this study is to investigate the feasibility, reproducibility and its predictive value of automated GLS measurement compared with standard manual measurement of GLS, so as to provide an important reference for clinical treatment and reduction of cardiovascular events in CKD patients.
Methods: A total of 285 CKD patients (aged 52 ± 12.85) were selected from Hainan Provincial People’s Hospital, all of whom had not received dialysis treatment. All CKD patients were measured by three different GLS evaluation methods using the same apical three-cavity, two-cavity and four-cavity heart images. (1) Entirely automatic GLS, directly analyzed by on-machine functions, (2) Semi-automatic GLS, corrected by experienced researchers on the basis of fully automatic measurements, and (3) manual GLS, standard manual measurements made by experienced researchers. Five patients were excluded due to poor image quality and could not be automatically measured and analyzed. Clinical outcomes were followed up with patients by telephone and outpatient review.
Results: After automatic GLS measurement, about 35% of the measurements were considered to need manual correction, and there was a statistically significant difference between automated, semi-automated and manual GLS (P <0.01). The correlation and consistency between semi-automated GLS and manual GLS were higher than automated GLS (P <0.01). At 2-year follow-up, 55 CKD patients (19.6%) experienced adverse cardiovascular events. Automated GLS was able to predict adverse cardiovascular events, but its predictive value was lower than semi-automated GLS. The automatic measurement and analysis time (15.23±0.75s/patient) and the semi-automatic measurement and analysis time (75.06±19.01s/patient) were significantly shorter than that of manual group (236.81±45.41s/patient). (P <0.01).
Conclusions: Automatic GLS assessment of left ventricular systolic function in CKD patients is feasible and reproducible. However, there are still some images that require manual correction at this stage, so a semi-automated approach using this new automated software to evaluate left ventricular systolic function in CKD patients and provide predictive value seems to be a superior option...
Article Details
References
[1] Joachim, J., et al., Cardiovascular Disease in Chronic Kidney Disease: Pathophysiological Insights and Therapeutic Options. 2021. 143(11): p. 1157-1172.
[2] Hensen, L.C.R., et al., Left Ventricular Mechanical Dispersion and Global Longitudinal Strain and Ventricular Arrhythmias in Predialysis and Dialysis Patients. Journal of the American Society of Echocardiography., 2018. 31(7): p. 777-783.
[3] Hensen, L.C.R., et al., Prevalence of left ventricular systolic dysfunction in pre-dialysis and dialysis patients with preserved left ventricular ejection fraction. . European journal of heart failure, 2018. 20(3): p. 560-568.
[4] Hensen, L.C.R., et al., Prognostic Implications of Left Ventricular Global Longitudinal Strain in Predialysis and Dialysis Patients. . The American journal of cardiology, 2017. 120(3): p. 500-504.
[5] Valocikova, I., et al., Incremental value of global longitudinal strain in prediction of all-cause mortality in predialysis and dialysis chronic kidney disease patients. . Wiener klinische Wochenschrift, 2016. 128(13-14): p. 495-503.
[6] Mitchell, C., et al., Guidelines for Performing a Comprehensive Transthoracic Echocardiographic Examination in Adults: Recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr., 2019. 32(1): p. 1-64.
[7] Badano, L.P., et al., Standardization of left atrial, right ventricular, and right atrial deformation imaging using two-dimensional speckle tracking echocardiography: a consensus document of the EACVI/ASE/Industry Task Force to Standardize Deformation Imaging. . Eur Heart J Cardiovasc Imaging., 2018. 19(6): p. 591-600.
[8] Hemmelgarn, B.R., et al., Overview of the Alberta Kidney Disease Network. BMC Nephrol., 2009. 10(1): p. 30.
[9] Kitano, T., et al., Prognostic value of automated longitudinal strain measurements in asymptomatic aortic stenosis. . Heart. , 2020.
[10] Gonzalez-Manzanares, R., et al., Automated Global Longitudinal Strain Assessment in Long-Term Survivors of Childhood Acute Lymphoblastic Leukemia. . Cancers (Basel). 2022. 14(6): p. 1513.
[11] Kawakami, H., et al., Feasibility, Reproducibility, and Clinical Implications of the Novel Fully Automated Assessment for Global Longitudinal Strain. J Am Soc Echocardiogr., 2021. 34(2): p. 136-145.
[12] Salte, I.M., et al., Artificial Intelligence for Automatic Measurement of Left Ventricular Strain in Echocardiography. JACC Cardiovasc Imaging., 2021. 14(10): p. 1918-1928.
[13] Knackstedt, C., et al., Fully automated versus standard tracking of left ventricular ejection fraction and longitudinal strain: the FAST-EFs multicenter study. J Am Coll Cardiol., 2015. 66(13): p. 1456-1466.
[14] Guijuan, P., et al., Feasibility and reproducibility of semi-automated longitudinal strain analysis: a comparative study with conventional manual strain analysis. Cardiovasc Ultrasound. , 2023. 21(1): p. 12.