Cavitation is the rapid formation and collapse of vaporous bubbles caused by a momentarily sharp drop in local pressure below vapor pressure.1 Some clinically explanted mechanical heart valves (MHVs) show erosions and pittings, which are typical of the damage caused by the collapse of cavitation bubbles.2 The mechanism and origin of MHV cavitation is yet to be fully understood. Factors affecting MHV cavitation and associated flow characteristics have been extensively studied in the past decade. MHV occluder closure behavior is widely believed to be the key factor in MHV cavitation.3,4 As the occluder is approaching its fully closed position, the gap between the occluder tip and the solid orifice wall becomes a narrowing flow channel. Fluid contained between this gap has a tendency to be squeezed out of this channel, causing a high-speed jet flow. This squeeze-flow characteristic of local pressure reduction may initiate cavitation.5
The MHV closure flow relating to cavitation was studied both by experimental work and computational fluid dynamics (CFD). Lee et al.6 identified possible valve design factors for cavitation initiation when studying nine different MHVs. The occluder stop seat, where fluid squeezing may occur, was suggested to be most critical to cavitation because the valve with this feature showed a high intensity of cavitation and a low threshold loading rate. Further investigation of transient pressure measurements and cavitation bubble visualization was performed at seven locations of an MHV.7 The presence of negative pressure and bubbles around the seat stop suggested that the fluid squeezing effect was the main cause of cavitation. Chandran et al.8 gave a rough estimation of the squeeze flow velocity of 12 m/s based on the observation of cavitation bubbles in a flow loop simulating single closing of mitral valve. The squeeze flow velocity in MHV was also approximately calculated from the impacting speed and gap width in two impinging rods. An increase of cavitation risk and intensity was found with elevated squeeze flow velocity.9 Maymir et al.10 found fluid velocity spike in the minor orifice of a monostrut valve upon closure, with a magnitude of 1.67 m/s in the aortic position and 3.02 m/s in the mitral position, which was suggested as evidence of “squeeze flow.” Kini et al.11 studied the flow field both for forward closure and occluder rebound with particle image velocimetry (PIV) techniques in a single-shot experiment setup. Flow velocity in excess of 1.5 m/s was detected 3 mm away from the valve seat on the major orifice side, where cavitation has been observed. Manning et al.12 also observed high-speed regurgitant jet along with vortex in PIV images. It was suggested that near leaflet vortex may provide a low pressure environment for cavitation.
The calculation of the squeeze flow inside the gap can reach as high as 30 m/s near the leaflet surface in the mitral position.13 With considerable leaflet deceleration and rebound, relatively low squeeze flow velocity (about 9m/s) is observed.14 CFD simulation coupled with occluder motion and housing deformation demonstrated that the squeeze flow field is influenced by the valve mounting compliance and the geometry of the contact region.15
Detailed experimental investigation in squeeze flow is still limited. This study aimed to provide some quantitative observations for the flow field upon valve closure, because the cavitation potential in MHV can be predicted through the explanation of squeeze flow field.
Materials and Methods
Pulsatile Flow Test Loop
The pulsatile flow experimental setup is illustrated in Figure 1. The computer controlled servo-motor (SEM Controlled Motor Technology) drives an 8-cm-diameter piston to produce pulsatile flow in the test loop. The piston motion curve can be created, modified, and implemented by Motion Generator software. The piston head is connected to a liquid-filled Perspex box, inside which a collapsible left ventricular (LV) sac is installed. Upstream of the simulated LV is a left atrium chamber connected via a commercial tilting-disk valve. Downstream of the LV is a perspex test section simulating the aorta, inside which a transparent aortic valve model is installed. The afterload resistive and compliance elements are connected with the test section and then return back to the atrium. These elements can be tuned separately to obtain desirable physiological pressure and flow rate. Two pressure transducers (Model T4812AD-R) coupled with a calibrated multichannel amplifier (AM-PACK AP9991, Vivitro System Inc., Victorìa, B.C., Canada) were used to measure the ventricle and aortic pressure. The transducers are located respectively at the LV immediate exit and MHV five aorta diameters downstream, as illustrated in Figure 1. Figure 2 shows a typical pressure waveform recorded by an oscilloscope (Tektronix, TDS36). The ventricular and aortic pressures fall in the range of 0∼160 mm Hg and 60∼130 mm Hg, respectively. The heart rate was 63 beats per minute with 40% systolic duration. The flow rate was measured as 4.3 l/min by a calibrated rotameter. These parameters were selected to protect the transparent MHV model operating under prolonged physiologic condition. The inhouse-developed 1:1 bileaflet MHV model, as shown in Figure 3, has a geometry and opening and closing angles similar to that of the SJM 29 mm MHV. Its transparency enables the closure flow field measurement by the PIV technique.
The TSI Insight PIV system (TSI Incorporated, Ithaca, NY) was used in this study. The double-pulsed Nd:YAG laser (maximum 50 mJ/pulse) provides illumination for the flow field plane. A CCD camera (1000*1016 pixel), coupled with an X-Y traveling mechanism (revolution step 0.01 mm) and Stemi 2000-C microscope (Zeiss), precisely position the observation of the minute channel flow upon MHV closure. The system magnification can be gradually adjusted and the maximum value is up to 3.2. The minimum velocity vector spacing is 0.09 mm under 50% overlapped Nyquist sampling.
Leaflet thickness (1 mm) is used to calibrate the velocity value when processing PIV images in Insight software. The corresponding vector files are then loaded into Tecplot to present PIV data with velocity, strain rate, and vorticity.
Working Fluid and Seeding Particle
The working fluid was composed of 79% saturated aqueous sodium iodide, 20% percent pure glycerol, and 1% of distilled water by volume. This composition yields a kinematics viscosity of 3.4 × 10–6 m2/s (3.4 cS), matching that of human blood at higher shear rates. The refractive index of this fluid was measured as 1.49, same as the test section (perspex), hence eliminating the refractive deformation when light passes through the curved liquid-solid interfaces. The seeding particle, 10 μm silver-coated glass sphere, was selected because of its good image contrast and suitable particle image diameter.
The frame rate of the CCD camera, 30 frames per second, is not adequate for analyzing one cycle of MHV operation, especially when its opening and closing process is <50 milliseconds. Hence a triggering method was developed. A photoelectric sensor (HPX-NT1, Yamatake), its position fixed as in Figure 4, captured the closing movement of MHV leaflet. The sensor signal was imputed into a Intel 82C54 Programmable Interval Counter/Timer (1600 Data Acquisition Board, Keithley Instruments) for time delay via a written program. The delayed signal was connected with synchronizer for PIV trigger with a time step of 10 μs. The PIV frame was taken only at specific instant in each cycle, and 100 cycles of images at this instant were saved for postprocessing
With the increase of system magnification, the sensor position was finely tuned closer to the leaflet full-closure position so that the time delay was triggered within a narrower orifice. The leaflet tip velocities within the decreasing gap dimensions were then obtained.
Local Flow Field Observation
A complete transparent MHV model enables the investigation of channel flow between the occluder and orifice wall. As the closing process proceeds, the channel becomes narrower until zero gap when the valve is considered fully closed. Figure 5 schematically depicts the local flow field when the MHV nearly impacts on the housing wall. Three profiles were selected within the gap: profile 1 along the leading tip, profile 3 along the trailing tip, and profile 2 between them.
Leaflet Closure Velocity
The MHV average tip velocity within Δt, i.e., from zero instant to the leaflet full-closure instant, can be calculated by analyzing leaflet position in the PIV image. Due to the variation from beat to beat, Δt was selected as the value at which the occluder reaches its full closure position over 85% of 100 continuous cardiac cycles. Table 1 summarizes the average leaflet tip velocities within various gap dimensions, in which α is the angle the leaflet oriented with the transverse direction, as in Figure 5. The results demonstrate that the leaflet tip velocity accelerates as the closing process proceeds.
Flow Field Results
Under magnification of 1.8, the channel dimension was narrowed from 1.1∼1.4 mm to 0.5∼0.8 mm. Typical velocity map, speed profile, and vorticity plot with different gap sizes are demonstrated in Figures 6–8. To further increase spatial resolution, magnification of 3.2 was applied. The dimension of observation field was thus decreased to about 2.70 × 2.70 mm. Related PIV measurements when the gap is 0.2∼0.3 mm and <0.1 mm are shown in Figures 9 and 10.
When the gap is about 1.1∼1.4 mm, as in Figure 6, the leaflet is pushed to close by the backwind flow with the fluid velocity perpendicular to the surface. In the upwind region, the flow adjacent to the leaflet is perpendicular to the solid surface, which is driven in motion by the valve rotation. While apart from the surface, the flow field is the combined result of the back flow through the gap and the occluder-driven flow. Thus, the fluid velocity in this area diverts parallel to the axial direction. A vortex flow is formed around the trailing tip, produced by the solid-fluid interaction (SFI) at the tip and backflow around the gap. The peak velocity is found around 0.27 m/s. The speed profile inside the gap shows the highest magnitude at the leading tip and lowest magnitude at the trailing tip.
When the gap decreases to about 0.7∼1.1 mm, as in Figure 7, both the backflow momentum and SFI around the occluder become stronger as the peak flow velocity reaches 0.36 m/s. The intensity of the trailing vortex and the magnitude of the speed profiles also increase. When the gap continuously decreases to about 0.5∼0.8 mm, as in Figure 8, the flow momentum inside the gap further strengthens, possibly due to the increased aortic-ventricular pressure gradient and ever-narrowing gap dimension. The resultant flow in the upwind region, as described earlier, is stronger than that in the backwind region. The maximum velocity at the leading tip reaches 0.4 m/s. The vorticity also strengthens to about 420 per second.
Figure 9 shows the flow field when the gap is about 0.2∼0.3 mm. Due to the stronger backflow momentum from the gap and backwind region, together with the acceleration of occluder motion, the peak flow velocity increases to 0.7 m/s at the upwind region. When the gap is <0.1 mm, as in Figure 10, the jet with magnitude of 2.3 m/s is observed at the immediate exit of the gap. The zoom-in map shows that the high speed flow coming from the leading tip encounters a sudden geometry expansion. Then it mixes with relatively weak flow in the upwind region, which may result in local pressure reduction for possible cavitation occurrence. The vorticity around the leading tip reaches about 1400 per second.
For the first time, the development of squeeze flow is verified experimentally. The PIV system coupled with a microscope enables gap flow observation from the dimension of 1.1∼1.4 mm to <0.1 mm. As the gap between the tip of the valve occluder and housing wall becomes narrower, evidence of high-speed jet flow becomes more apparent. The flow velocity between 1 and 3 m/s is observed rushing out the gap when the gap dimension is <0.1 mm. This high-speed jet may cause a low-pressure center around the leading tip with the potential for cavitation occurrence.
The squeezing velocity shown in this study differs from other CFD works, possibly because of the varied valve position or working conditions. The limitation of our study is that there is a difference in material density between our clear MHV model and commercial ones, which may result in different MHV closure dynamics.
1. Young FR (ed): Cavitation
. London, McGraw-Hill Book Company, 1989.
2. Klepetko W, Moritz A: Leaflet fracture in Edwards Duromedics bileaflet valves. J Thorac Cardio Vasc Surg
97: 90–94, 1989.
3. Hwang NHC: Cavitation potential of pyrolytic carbon heart valve Prostheses: A review and current status. J Heart Valve Dis
7: 140–150, 1998.
4. He ZM, Xi BS, Zhu KQ, Hwang NHC: Mechanisms of mechanical heart valve cavitation: Investigation using a tilting disk valve model. J Heart Valve Dis
10: 666–674, 2001.
5. Rood EP: Review-mechanisms of cavitation inception. J Fluids Eng
113: 163–175, 1991.
6. Lee CS, Chandran KB, Chen LD: Cavitation dynamics of mechanical heart valve prostheses. Artif Organs
18: 138–142, 1994.
7. Lee CS, Chandran KB, Chen LD: Cavitation dynamics of Medtronic hall mechanical heart valve prosthesis: fluid squeezing effect. J Biomech Eng
118: 97–105, 1996.
8. Chandran KB, Chew YT, Low HT, et al:
In vitro demonstration of squeeze flow at the instant of mechanical heart valve closure. ASME Adv Bioeng
39: 75–76, 1998.
9. Lim WL, Chew YT, Low HT, Foo WL: Cavitation phenomenon in mechanical heart valves: The role of squeeze flow velocity and contact area on cavitation initiation between two impinging rods. J Biomech
36: 1269–1280, 2003.
10. Maymir JC, Deutsch S, Meyer RS, et al:
Mean velocity and Reynolds stress measurements in the regurgitant jets of tilting disk heart valves in an artificial heart environment. Ann Biomed Eng
26: 146–156, 1998.
11. Kini V, Bachmann C, Fontaine A, et al:
Flow visualization in mechanical heart valves: Occluder rebound and cavitation potential. Ann Biomed Eng
28: 431–441, 2000.
12. Manning KB, Kini V, Fontaine AA, et al:
Regurgitant flow field characteristics of the St. Jude bileaflet mechanical heart valve under physiologic pulsatile flow using particle image velocimetry. Artif Organs
27: 840–846, 2003.
13. Blustein D, Einav S, Hwang NHC: A squeeze flow phenomenon at closing of a bileaflet mechanical heart valve prosthesis. J Biomech
27: 1369–1378, 1994.
14. Makhijani VB, Yang HQ, Singhal AK, Hwang NHC: An experimental-computational analysis of MHV cavitation: Effect of leaflet squeezing and rebound. J Heart Valve Dis
3(Suppl I): 35–48, 1994.
15. Makhijani VB, Siegel JM, Hwang NHC: Numerical study of squeeze-flow in tilting disc mechanical heart valves. J Heart Valve Dis
5: 97–103, 1996.
16. Adrian RJ: Particle-imaging techniques for experimental fluid mechanics. Ann Rev Fluid Mech
23: 261–304, 1991.
17. Brucker C, Steinseifer U, Schroder W, Reul H: Unsteady flow through a new mechanical heart valve prosthesis analyzed by digital particle image velocimetry. Meas Sci Tech
13: 1043–1049, 2002.
18. Chandran KB, Lee CS, Chen LD: Pressure field in the vicinity of mechanical valve occluders at the instant of valve closure: Correlation with cavitation initiation. J Heart Valve Dis
3(Suppl I): 65–76,1994.
19. Chandran KB, Aluri S: Mechanical valve closing dynamics: relationship between velocity of closing, pressure transients, and cavitation initiation. Ann Biomed Eng
25: 926–938, 1997.
20. Chandran KB, Dexter EU, Aluri S, Richenbacher WE: Negative pressure transients with mechanical hear valve closure: correlation between in vitro and in vivo results. Ann Biomed Eng
26: 546–556, 1998.
21. Ellis JT, Healy TM, Fontaine AA, Weston MW: An in vitro investigation of the retrograde flow fields of two bileaflet mechanical heart valves. J Heart Valve Dis
5: 600–606, 1996.
22. Guo GX, Xu CC, Hwang NHC: Laser assessment of leaflet closing motion in prosthetic heart valve. J Biomed Eng
12: 447–481, 1990.
23. Lim WL, Chew YT, Chew TC, Low HT: Steady flow dynamics of prosthetic aortic hearts: A comparative evaluation with PIV techniques. J Biomech
31: 411–421, 1998.
24. Lim WL, Chew YT, Chew TC, Low HT: Pulsatile flow studies of a porcine bioprosthetic aortic valve in vitro: PIV measurement and shear induced blood damage. J Biomech
34: 1417–1427, 2001.
25. Maymir JC, Deutsch S, Meyer RS, et al:
Effects of tilting disk heart valve gap width on regurgitant flow through an artificial heart mitral valve. Artif Organs
21: 1014–1025, 1997.
26. Prasad AK, Adrian RJ, Landreth CC, Offutt PW: Effect of resolution on the speed and accuracy of particle image velocimetry interrogation. Exp Fluids
13: 105–116, 1992.
27. Raffel M, Willert C, Kompenhans J (eds): Particle Image Velocimetry: A Practical Guide
. Berlin, New York, Springer, 1998.
28. Shu MCS, Leuer LH, Armitage TL, et al:
In vitro observation of mechanical heart valve cavitation. J Heart Valve Dis
3(Suppl I): S85–S93, 1994.
29. Subramanian A, Mu H, Kadambi JR, et al:
Particle image velocity investigation of intravalvular flow fields of a bileaflet mechanical heart valve in a pulsatile flow. J Heart Valve Dis
9: 721–731, 2000.
30. Willert CE, Gharib M: Digital particle image velocimetry. Exp Fluids
10: 181–193, 1991.
31. Wu ZJ, Wang Y, Hwang NHC: Occluder closing behavior: A key factor in mechanical heart valve cavitation. J Heart Valve Dis
3(Suppl I): 25–34, 1994.
32. Wu ZJ, Gao BZ, Hwang NHC: Transient pressure at closing of a monoleaflet mechanical heart valve prosthesis: Mounting compliance effect. J Heart Valve Dis
4: 553–567, 1995.
33. Zhao JB, Shi YB, Yeo JH, Hwang NHC: Digital particle image velocimetry investigation of the pulsating flow around a simplified 2-D model of a bileaflet heart valve. J Heart Valve Dis
10: 239–253, 2001.