Validation of a hand-held ultrasound device in the evaluation of aortic stenosis

Abstract Introduction Hand-held Ultrasound devices (HHUD) are increasingly used in routine clinical practice, although until recently they did not have continuous Doppler. There is no evidence on the usefulness of an HHUD for the assessment of aortic stenosis (AS) in real life. Objective To validate the use of a new HHUD device in a real-life scenario. Material and methods Observational, single-center and prospective study, including patients diagnosed with AS with a previous standard echocardiography. Patients were randomly recruited from a tertiary cardiac imaging laboratory. Immediately after the reference echocardiography performed by an imaging expert, a Cardiology trainee with ASE level II repeated the study blindly with an HHUD. Agreement between the two tests was evaluated by measuring the mean aortic valve gradient (mAG). An intraclass correlation coefficient (ICC) test, linear regression, and Bland-Altman test were performed to compare the two methods. Results 101 patients were recruited. The reference test calculated a mAG of 29 mmHg (19.8 - 42.2) versus a mAG of 27.2 mmHg (16.2 - 43.9) in HHUD. The statistical analysis showed a good correlation between both tests (ICC: 0.87 (95% CI: 0.79 -0.94) and r=0.89 (95% CI: 0.85-0.94) (Figure). The Bland-Altman test showed a non-significant underestimation (1.61 ± 0.9 mmHg; p= 0.09). The HHUD classified 29 cases as severe or major AS compared to 31 in the reference test (kappa= 0.81; 95% CI: 0.68 - 0.94). The HHUD showed an excellent agreement for classification of AS in at least severe AS grade. The kappa analysis had a value of 0.81 (95% CI 0.68 – 0.94) with a global agreement of 92.1%. The positive predictive value was 89.6%. The negative predictive value was 93.1%. Sensitivity and specificity were 83.8% and 95.7%. The rest of the variables evaluated showed poorer agreement. In those patients with obesity, worse acoustic window, atrial fibrillation and critical AS, the agreement was worse. Conclusions The new HHUD evaluated by an operator with limited experience showed good agreement with the reference echocardiographic device in determining the mean aortic gradient.Central picture