Aims: In vitro and in vivo studies were performed to evaluate the diagnostic accuracy of the different Doppler-echocardiographic parameters proposed in the American Society of Echocardiography guidelines to identify dysfunction of bileaflet mechanical valves (BMV) in the aortic position.
Methods and results: Two models of BMV (St Jude HP, MCRI On-X) of different sizes (21;23;25;27 mm) were tested in vitro under a wide range of cardiac outputs (3-7 L/min). The motion of one or both leaflets was restricted to induce a mild (25% restriction in total valve orifice area) and moderate-to-severe (50% restriction in total valve area). Doppler-echocardiographic parameters of valve function were also measured in 17 patients with BMV of whom 4 had valve dysfunction confirmed by cinefluoroscopy. The specificity of all the parameters was high (in vitro: 83-100%; in vivo: 69-100%), but the sensitivity was low (range: 0-83% and 25-100%, respectively). A higher cut-off value for the ratio of peak left ventricular outflow tract velocity to peak aortic velocity or Doppler velocity index (DVI) (<0.35 instead of 0.3 or 0.25) improved the sensitivity (>90%) for the detection of moderate-to-severe dysfunction but remained low for mild dysfunction (50%). Furthermore, a difference of normal reference effective orifice area (EOA) minus measured EOA (EOA-D) >1 standard deviation identified mild and moderate-to-severe dysfunction with sensitivity of 61 and 100%, respectively.
Conclusion: The Doppler-echocardiographic parameters and criteria proposed in the guidelines lack sensitivity for the detection of BMV dysfunction. The utilization of a DVI < 0.35 or an EOA-D > 1 SD improved the sensitivity (>90%) for the detection of moderate-to-severe dysfunction, but the sensitivity remained suboptimal (<65%) for detection of mild dysfunction.
Keywords: Aortic stenosis; Doppler echocardiography; Prosthetic heart valve.