Hemodynamic performance of the Ionescu-Shiley valve prosthesis

J THORAC

CARDIOV ASC

S URG 80:613-620, 1980

Hemodynamic performance of the Ionescu-Shiley valve prosthesis The hemodynamic performance of the lonescu-Shiley bovine heterograft valve has been evaluated by intraoperative measurement of transvalvular gradients and cardiac outputs, Effective orifice areas (EOAs) were calculated and the data compared to those obtained by other investigators for other prostheses, In the aortic position. each valve, from 19 to 31 mm external diameter. produced a pressure gradient; mean EGA increased with increasing valve size. so that small (19 to 23 mm} valves were moderately stenotic and larger valves were only mildly stenotic. The presence of a mitral prosthesis produced larger transuortic valve gradients. probably as a result of aortic outflow obstruction by the mitral prosthesis, The lonescu valve appears to he hemodynamically superior to other biological valves in the aortic position and comparable to most mechanical prostheses. although the data for comparison are scant, Each mitral valve produced a pressure gradient and. on the average. larger (29 mm) valves performed no better than smaller (25 mm}, Mean EOAs for each valve size (25 to 29 mm} were adequate to provide satisfactory hemodynamics comparable to other available prosthetic valves, Mild obstruction of the left ventricular outflow by the prosthetic struts was seen to be related to the distance between ventricular septum and the struts, Most currently available prostheses seem to provide similar hemodynamics in the mitral position. and considerations such as thrombogenicity and durability may be relatively more important in the choice o] a mitral valve substitute than in the choice of an aortic valve substitute,

Ronald M. Becker, M.D., Joel Strom, M.D., William Frishman, M.D.,* Yasu Oka, M.D., Yen Tse Lin, M,D., Edward L. Yellin, Ph.D., and Robert W. M. Frater, M.B., Ch.B., Bronx, N. Y.

The popularity of porcine heterografts attests to the need for a reliable bioprosthesis relatively free from the complications of thromboembolism which have plagued all mechanical prostheses. Several studies have indicated that glutaraldehyde-preserved heterografts maintain functional and structural integrity for at least several years. 1-4 If the hemodynamic performance and durability of tissue valves can match those of mechanical valves, then their other advantages would make tissue valves the preferred choice for valve replacement. The Ionescu-Shiley bovine pericardial prosthetic

heart valve was designed to provide improved hemodynamics relative to other bioprostheses. In vitro studies in our laboratory': 6 have shown that the valve performs better in the mitral position than other bioprostheses. We report here hemodynamic studies of the IonescuShiley prosthesis in the mitral and aortic positions in patients.

Methods

Address for reprints: Dr. Ronald M. Becker, Department of Surgery • Hospital of the Albert Einstein College of Medicine. 1825 Eastchester Rd.• Bronx, N. Y. 10461.

Intraoperative hemodynamics were studied in 37 patients (19 aortic, 15 mitral, three double) when postcardiopulmonary bypass condition was stable. Left atrial and left ventricular pressures were obtained by direct catheterization of these chambers, and ascending aortic pressures measured through the aortic cannula. The catheters were connected via identical stiff tubing to Statham strain gauges balanced and calibrated to produce superimposed excursions throughout the full range of pressures. The pressures were recorded continuously on a multichannel recorder. * Simultaneous cardiac output was obtained by indocyanine green

"Dr. Frishman is a "teaching scholar" of the American Heart Association.

"Electronics for Medicine VR-6, White Plains. N. Y.

From the Departments of Surgery, Cardiology, and Anesthsiology, Albert Einstein College of Medicine, Bronx, N. Y. Supported in part by the Judith Harris Selig Memorial Fund. Received for publication Dec. 26. 1979. Accepted for publication April 2. 1980.

0022-5223/80/100613+08$00.80/0 © 1980 The C. V. Mosby Co.

613

The Journal of

6 14

Becker et al.

Thoracic and Cardiovascular Surgery

Table I. Hemodynamic data and calculated effective orifice areas in 18 patients following mitral valve replacement with the lonescu-Shiley valve Valve size (mm)

25

27

Patient

Mean pressure gradient (mm Hg)

Meanftow velocity (cc/sec)

Cardiac output (Llmm)

Effective orifice area (em')

15.3 3.9 2.5 13.7 10.0 10.0 3.2 2.7 1.1 3.1 2.6 13.0 7.2 5.1 4.1 2.4 3.3 1.8

156 174 128 83 103 85 80 156 121 178 97 188

3.08 5.43 2.89 3.07 2.59 1.98 2.18 5.20 3.85 3.74 1.90 3.93 2.39 4.36 2.62 3.56 3.38 5.11

0.8

L.c. L. R. J. V. S. M. M. V. A. R.

T. G. A. S. B. G. F. R.

S. L. 29

N. R. M. R. G. D. R. V. M. S. J. N. M. D.

(Cardio-Green) dye dilution, with a cardiac output computer* sampling from the radial artery. Pressure gradients were calculated by planimetric integration of the area between the simultaneously recorded atrial and ventricular, or ventricular and aortic, pressure tracings. Effective orifice areas (EOAs) were calculated according to our modification-:" of the Gorlin" equation: flow rate .

EOA =

51.6~

As discussed in previous papers from our laboratory."? the formula is used without a discharge coeffi-

cient since the discharge coefficient is unknown for any given prosthetic valve and cannot be theoretically predicted. Two indices are calculated from these data by comparing the calculated EOA first with the actual orifice area (AOA) and second with the external mounting area (EMA) of the particular size prosthetic valve. The first of these indices is the discharge coefficient (Cd). The formula is Cd

=

EOA AOA

and is a measure of the hydraulic efficiency of the orifice. *Electronics for Medicine, White Plains, N. Y.

77

210 83 100 139 185

1.7

1.6 0.5 0.7 0.5 0.8 1.8 2.3 2.0 1.1 1.0 0.5 1.8 0.8 1.3 1.5 2.8

The second index, the performance index (PI), compares the EOA with the EMA so that EOA

PI = EMA'

This is a measure of what effective orifice the surgeon can provide for the patient when he uses a valve of a particular size. In eight patients who had normal aortic valves preoperatively, the effect of the mitral prosthesis on left ventricular outflow properties was studied. Aortic pressure was measured via the aortic cannula and ventricular pressure as described earlier. Mean pressure gradients were calculated from the superimposed tracings. Ventricular diastolic and outflow diameters were determined from simultaneous echocardiograms, performed as previously described." Results Adequate data were obtained in 37 patients. The results are presented in Tables I and II along with the calculated EOAs. Mean EOAs, performance indices, and discharge coefficients are shown in Table III. In Table IV, expected pressure gradients for aortic valves at different cardiac outputs are shown. Our results are compared with those obtained by others with various other prosthetic valves in Tables V and VI. Mitral position. There is a wide scatter of calculated

Volume 80

lonescu-Shiley valve prosthesis

Number 4

6 15

October, 1980

Table II. Hemodynamic data and computed effective orifice areas in 22 patients following aortic valve replacement with the lonescu-Shiley valve Valve size (mm) 19 21

Mean pressure gradient (mm Hg)

Patient

18.7 37.3 24.6 8.7 10.6 6.7 10.6 3.6 10.9 7.6 10.0 6.3 2.8 20.3 9.7 3.2 3.1 3.3 13.3 3.1 3.4 4.2 1.7

G. J.

I. L.' J. V. S. L. M. S. M.W. M. R.t A. R.t S. L.t A. H. F. G. N. D. M.Q. J. M.

23

I. B. H. W. V. B. J. R. M. G. (I):j: M. G. (2):j: D. B. R. F. L. R.

25

27 31

I

Mean flow (mllsec)

I

Cardiac output (Llmin)

201 172 220 179 194 165 150 90 174 227 151 210 101 244 267 165 262 146 238 129 340 204 211

I

Effective orifice area (cm 2) 0.9 0.5 0.9 1.2 1.1 1.2 0.9 1.0 1.0 1.6 1.0 1.6 1.2 1.0 1.6 1.8 2.8 1.6 1.3 1.4 3.5 1.9 3.2

2.93 3.29 3.36 3.41 3.08 3.13 2.39 1.98 3.32 5.46 2.38 5.16 1.50 5.75 5.69 3.61 6.36 3.41 5.80 3.33 6.50 3.83 3.83

'Patient had idiopathic hypertrophic subaortic stenosis. t Patient had simultaneous mitral valve replacement.

t Patient was reoperated upon for paravalvular leak related to sinus aneurysm formation secondary to endocarditis.

Table III. Mean effective orifice areas (EOA), discharge coefficients (Cd), and performance indices (PI) for lonescu-Shiley valves Mounting diamter (mm) 19 25 25 29tt

Position Aortic Aortic Mitral Mitral

Mounting area (cm 2)

Actual orifice area' (cm 2)

2.83 4.90 4.90 6.61

1.27 2.09 2.09 2.83

PI 0.9 1.8

1.4 1.7

0.71 0.86 0.67 0.60

0.32 0.37 0.29 0.26

• Area available for flow at level of free margin of leaflets." tTwo patients with transvalvular flow rates < 100 cc/sec were not included in averages (see text for rationale).

Table IV. Expected mean pressure gradients for a range of cardiac outputs (CO) for the lonescu-Shiley valve. based on a systolic flow period of 22.5 sees/min, in the aortic position Mounting diameter (mm)

t1P at CO = 2 Llmin (mm Hg)

t1P at CO = 5 Llmin (mm Hg)

t1P at CO = 8 Llmin (mm Hg) 63 38 25 13 6.2 4.7

19

4.0

21

2.4

23 25 27 31

1.6 0.8

25 15 10 5.2

0.4

2.4

0.3

1.8

The Journal of Thoracic and Cardiovascular Surgery

6 I 6 Becker et al.

Table V. Comparison of calculated mean effective orifice area (EOA) for various commercially available valves in mitral position Valve

/--R-e-i-.N-o-.--

D

EOA In)

D

EOA In)

D

EOA In)

lonescu Starr 6400' Starr Lilleheit Lillehei Lillehei Carpentier Hancock

This study

25

2.3 (3)

25 25

1.1 (I) 1.4 (2)

25

1.7 (2)

27 27 27 27 27 27 27 27

2.0 (8) 2.0 (4) 1.4 (8) 1.6 (8) 1.6 (4) 1.8 (9) 2.3 (2) 1.7 (4)

29 30 30 29 29 29 29 29

1.9 (7) 2.2 (II) 1.4 (10) 1.7 (8) 2.1 (I) I. 9 (7) 2.3 (3) 2.4 (2)

19 20 20 21 16 22 23

Legend: In order to allow meaningful comparison our own results were recalculated, assuming a discharge coefficient of 0.6, as used by all the other authors. D, External (mounting) diameter. n, Number of patients. 'Starr 2M and 3M have anulus diameters of 27 and 30 mm, respectively. tLillehei 18M, 20M, and 22M valves have anulus diameters of 25,27, and 29 mm, respectively.

Table VI. Comparison of calculated mean orifice area for various commercially available aortic valves Valve lonescu lonescu Hancock Hancock Hancock Hancock Hancock MO' Carpentier Carpentier Lilleheit Lillehei Lillehei Lillehei Starr:j: Starr 2400§ Hall Bjork Bjork

EOA In) This study

13 24 25 23 26 27 28 29 16 20 21 30 20 19 31 32 31

19 19

1.0 (I) 1.0 (I)

19 19

1.0 (3) 0.8 (2)

19

0.9 (I)

21

1.3 (4)

21 21

1.1 (2) 1.0 (6)

21 21 21

1.6 (7) 1.3 (I) 1.4 (2)

21 21 21 21

0.8 0.77 0.8 1.0

21 21 21

(9) (5) (2) (5)

1.8 (3) 1.3 (5) 1.3 (3)

23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23

1.6 1.6 1.2 1.3 1.2 1.I 1.5 1.6 1.1 0.98 1.1 1.05 1.0 1.1 1.6 2.2 1.7 1.3

(6) (7) (2) (7)

(3) (6) (10) (6) (6) (2) (12) (3) (10) (7)

(8) (II) (II) (4)

25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25

2.2 1.6 1.8 1.4 1.2 1.4 1.7 1.3 1.5 1.1 1.3 1.50 1.4 1.3 1.9 2.5 2.2 1.8

(4) (5)

(2)

27

3.1

27 27 27

1.5 (I) 1.7 (2) 1.5 (3)

27 27 27 27 27 27 26.5

1.7 2.7 1.55 1.9 1.65 1.6 1.3

27 27

2.4 ( 10)

(3)

(6) (3) (12) (I)

(2) (6) (I) (15) (3) (3) (7)

(4) (3) (2) (4) (4) (3) (6)

(8) (7)

(II) (2)

Legend: In order to allow meaningful comparison, we recalculated our own data, assuming a discharge coefficient of 0.86, as used by all the other authors. D, External (mounting) diameter. n, Number of patients.

'Hancock modified orifice valves; EOAs are those reported for rest. tLillehei 14, 16, 18, and 20A valves correspond to anulus diameters of 21,23,25 and 27 mm, respectively. :j:Starr 8A, 9A, lOA, and IIA correspond to anulus diameters of 21, 23, 25, and 26.5, respectively. §Starr 9A and lOA correspond to anulus diameters of 22 and 24 mm.

EGAs for any given valve size, and as the number of patients is small this may account for the finding that the mean EGA for each size is approximately 1.4 ern". Further examination of the data reveals that patients with transvalvular flow rates less than 100 eel sec tended to have low calculated EGAs: Six of the seven patients with EGA < 1.0 ern" had flow rates < 103 eel sec, usually as a result of low cardiac output. These seven patients with low calculated EGA did not appear to share any other important characteristics. Two had predominant mitral stenosis, three predorni-

nant insufficiency, and two combined mitral and aortic valve disease, so that left ventricular cavity size was not uniformly small in this subgroup of patients. Their postoperative courses were not markedly different from those of the other II patients, and relief of symptoms in the six long-term survivors has been excellent, despite the low calculated EGAs. Since each size provided, on the average, an EGA of approximately 1.4 ern", larger sizes appeared to be far less efficient in their use of their internal orifices and mounting surfaces. This is reflected in the low dis-

Volume 80

lonescu-Shiley valve prosthesis

Number 4

6 17

October, 1980

\+

•

'"

:I:

E E

8

•

\

,:0.52 p:NS

+

....

z w C c(

a: C)

4

~

...

0

LL

....

:::l

0

'"w c(

r :0.92

p<0.05

\

+

\

\+ \+ +

e,

'!5 DIASTOLIC DIAMETER, em

OUTFLOW

I

2

DIAMETER, em

Fig. 1. Left ventricular-aortic pressure gradient versus echocardiographically determined diastolic diameter (no correlation) and distance between prosthetic stent and ventricular septum (good correlation) following mitral valve replacement with the Ionescu-Shiley valve.

charge coefficients and performance indices shown in Table 1II. Comparison of the results of this study with others (Table V) shows that most mitral prostheses provide approximately equal EOA, and in other series no apparent advantage of the larger prostheses was found. Fig. I shows that there is a direct correlation between the gradient produced across the left ventricular outflow and the distance between prosthetic stent and ventricular septum, as measured by echocardiography. This confirms that high-profile prostheses may indeed cause mild outflow obstruction. Aortic position. As noted in Table II, Patient 1. L. had subaortic muscular stenosis. Although a myectomy was performed at the time of valve replacement, subaortic obstruction was clearly present at autopsy 1 year later, the patient having died of a ruptured thoracic aortic aneurysm. This accounts for the exceptionally low EOA in this patient. All three patients in the 23 mm group with simultaneous mitral valve replacements had lower EOAs than the 23 mm group as a whole. The presence of a mitral prosthesis may contribute to the aortic outflow gradient, as we have shown in Fig. I, and as discussed earlier. Gradients were no more than moderate in any case, but the higher gradients were largely found with the 19 and 21 mm mounting diameter sizes. This is expected, as the thickness of the stents is the same for the small size valves as it is for the larger ones. There is thus more encroachment on the area available for flow by the mounting mechanism in the smaller sizes.

In Table IV the results are portrayed differently. For each mounting size the average gradient for a range of cardiac outputs is given so that the surgeon can predict what general level of gradient will be produced by a particular size valve at a given cardiac output and therefore how much hemodynamic benefit can be expected from valve replacement. Tables V and VI compare the results of this study with those obtained for other valves in other studies. For these tables our own results were recalculated with the use of the standard Gorlin equations, as had been used in all other studies.

Discussion Evaluation of valve function. Proper data for the quantitative assessment of natural and prosthetic valve function must include, at the least, simultaneous measurements of upstream and downstream pressures and cardiac output. The pressure measurements should be obtained with identically balanced and calibrated strain gauges equidistant from their sites of pressure measurement. The methods available as a result of our monitoring in the operating room make it possible for us to come far closer to the ideal that we use in the experimental laboratory than we can in the clinical catheterization laboratory. We are quite satisfied that the techniques described for obtaining the pressures and cardiac outputs are accurate and reproducible. We 5- 7 have elsewhere discussed in detail the proper form for the Gorlin equation in calculating the EOAs of prosthetic valves. The essential point is that no dis-

6 I 8 Becker et

at.

The Journal of Thoracic and Cardiovascular Surgery

charge coefficient is used in the formula since, in contrast to natural valves studied in the catheterization laboratory, the actual orifice area is known but the discharge coefficient is unknown. The discharge coefficient must be derived by comparison of the known orifice area with the EOA obtained from hydraulic data. The discharge coefficient may not be the same from one design to the next, and indeed the discharge coefficients for different designs provide comparisons of the hemodynamic efficiency of different orifices and occluding mechanisms. For these reasons the Gorlin equation used in the evaluation of prosthetic valves should be used without a discharge coefficient. (In practical terms, this is the equivalent of using a discharge coefficient of I. In fact, our in vitro data" confirm that the Ionescu valve and other bioprostheses have discharge coefficients of approximately I when these are calculated on the basis of the area measured at the free margin of the leaflets.) Comparison with other studies. There are a number of difficulties in attempting to compare the data of various studies: (I) Gradients measured from ventricle to radial artery I() may not reflect the ventriculoaortic gradient; (2) studies of gradients measured without simultaneous cardiac outputs provide useless information; (3) data must be considered suspect when calculated EOAs are larger than actual orifice areas," when there is a wide scatter of results for a given size, when results differ greatly from those of other studies, II or when large numbers of valves are reported to have' 'no gradients" at normal outputs'"; (4) when data have been lumped together for all sizes of valves, 12 detailed comparisons are impossible. Taking these factors into consideration, we have culled the available appropriate information from the literature to develop Tables V and VI. Not all published studies are included in the averages for the reasons given above. Performance in the aortic position. The information contained in Table VI shows that (I) the results of our study are in good agreement with those previously reported by Tandon and associates I:!; (2) the hemodynamic performance of the Ionescu valve appears to be slightly superior to that of the other bioprostheses; this is to be expected since the porcine valves must include the aortic valve anulus in their mounting mechanisms, and this reduces the area available for flow; (3) the Ionescu valve performs as well as most current mechanical prostheses for which data are available; (4) all prosthetic valves produce some restriction to outflow. Performance in the mitral position. In common with several other series (Table V) involving both bio-

logical and mechanical prosthetic valves, this series demonstrates a wide variability of apparent EOA, little apparent advantage of larger size valves, and an average EOA of approximately 2.0 ern- (when calculated with a discharge coefficient of 0.6). Since the same valves seem to function more predictably in the aortic position, the apparently poor performance of mitral valves may be related to inadequate opening of the prosthesis, especially at low flows. In general, low EOAs were produced in patients with low transvalvular flow rates. In pulse duplicator work, we" 6 have observed that mitral bioprostheses open incompletely at flow rates <90 cc/sec. However, there is no measurable gradient across the prosthesis at these low flow rates, so that this phenomenon is not related to cusp stiffness but is rather a reflection of the response of flexible cusps to low flow: They open only as much as is necessary to transmit the available flow. Thus our in vitro work suggests that low EOAs at low flow rates are not related to incomplete opening of the prosthesis. A more satisfactory explanation of low EOAs at low flow rates is the inadequacy of the Gorlin equation in this situation. This is related to the fact that the use of mean rather than phasic flow measurements produces falsely low EOAs,5 and this problem is magnified under conditions of low flow. 7 The apparent lack of advantage of larger size valves in this and other series can be explained partially by the factors discussed earlier, since EOAs with larger valves were greater if one excluded those patients with low trans valvular flow rates (Table III). Additional factors related to ventricular geometry or rigid fixation of the anulus also might limit the apparent orifice regardless of the size of the prosthesis. These same arguments may explain the apparently greater efficiency of these valves in the aortic position than in the mitral. Since even the smaller valves provide generally satisfactory performance which compares well to other prostheses available, and since other factors may limit the apparent orifice, one need not feel obligated to "stuff in" the largest possible prosthesis, especially as larger valves may increase the potential for such complications as posterior wall rupture.':': I.'; or late prosthetic valve dysfunction. 16 Left ventricular outflow obstruction. We have demonstrated mild degrees of outflow tract obstruction with the Ionescu-Shiley prosthesis in the mitral position. The potential for this complication was recognized in the very earliest days of prosthetic mitral valve development. It may be related either to proximity of the prosthetic ring to the septum or to the projection of the occluding mechanism across the outflow tract, the

Volume 80 Number 4

Ionescu-Shiley valve prosthesis

6I9

October, 1980

latter depending on the variably oblique angle that the plane of the atrioventricular ring makes into the plane of the aortic ori fice . 17 Other factors. The Ionescu valve is more difficult to use than most other prostheses. In the aortic position, the struts interfere with knot-tying and aortotomy closure. In the mitral position, one must be cautious in inserting the valve to avoid laceration of the posterior wall of the ventricle by a strut, and also to avoid "catching" a strut with a loose stitch. Although these factors do make for a more tedious valve replacement, they are certainly not overwhelming. In our experience '" the Ionescu valve has been virtually free of catastrophic thromboembolic complications even in the absence of anticoagulation. Although its long-term durability is unknown, this is also true of other bioprostheses. From a hemodynamic point of view, it should be the bioprosthesis of choice, especially for use in small aortic roots.

Conclusions The Ionescu aortic valve is moderately obstructive at all sizes. Nevertheless, the EOAs even at the smaller sizes are not critically stenotic and the gradients at normal resting cardiac indices are not clinically significant. In this respect the Ionescu valve is better than most other prosthetic valves for which data are available, and it can be used interchangeably with the best of the well-established mechanical valves down to at least the 21 mm mounting size. The presence of a mitral valve prosthesis may produce outflow obstruction giving higher ventriculoaortic gradients for a given size of aortic valve prosthesis. In the mitral position, the Ionescu valve provides generally satisfactory performance comparable to other prostheses available. Mild left ventricular outflow obstruction is related to the proximity of the prosthetic stent to the septum. Apparent variability of performance, relatively poor performance compared to the same prosthesis in the aortic position, and apparent lack of hemodynamic advantage of larger sizes may be explained by the difficulties involved in accurately assessing prosthetic valve function in the mitral position. The expert help of Stuart Thompson, R.N., and Richard Paul, R.N., in data collection, and of Ms. Marsha Savarese in manuscript preparation is gratefully acknowledged. REFERENCES Hannah H, Reis RL: Current status of porcine heterograft prostheses: A 5 year appraisal. Circulation 54:Suppl 3:27-31, 1975 2 Pipkin RD, Buch WS, Fogarty TS: Evaluation of aortic

valve replacement with a porcine xenograft without longterm anticoagulation. J THORAC CARDIOVASC SURG 71:179-186, 1976 3 Zuhdi N, Hawley W, Voehl V, Hancock W, Carey J, Greer A: Porcine aortic valves as replacement for human heart valves. Ann Thorac Surg 17:479-491, 1974 4 Davila JC, Magilligan OJ Jr, Lewis JW Jr: Is the Hancock porcine valve the best cardiac valve substitute today? Ann Thorac Surg 26:303-316, 1978 5 Gabbay S, McQueen OM, Yellin EL, Becker RM, Frater RWM: In vitro hydrodynamic comparison of mitral valve prostheses at high flow rates. J THORAC CARDIOVASC SURG 76:771-787, 1978 6 Gabbay S, McQueen OM, Yellin EL, Frater RWM: In vitro hydrodynamic comparison of mitral valve bioprostheses. Circulation 60:Suppl 1:62-70, 1979 7 Yellin EL, McQueen ON, Gabbay S, Strom J, Becker RM, Frater RWM: Pressure-flow relations and energy losses across prosthetic mitral valves. In vivo and in vitro studies, Cardiac Dynamics, J Baan, A Artzenius, E Yellin, eds., The Hague, 1980, Martinus Nijhoff, pp 509-519 8 Godin R, Godin NG: Hydraulic formula for the calculation of the area of the stenotic mitral valve, other cardiac valves, and central circulatory shunts. Am Heart J 41:129, 1951 9 Strom J, Becker RM, Frishman W, Elkayam U, Sonnenblick E, Oka Y, Lin YT, Patel J, Frater RWM: Effects of hypothermic hyperkalemic cardioplegic arrest on ventricular performance during cardiac surgery. Assessment by intraoperative echocardiography. NY State J Med 78:2210-2213, 1978 10 Levine FH, Buckley MJ, Austen WG: Hemodynamic evaluation of the Hancock modified orifice bioprosthesis in the aortic position. Circulation 58:Suppl 1:33-35, 1978 II Starek PJK, Wilcox BR, Murray GF: Hemodynamic evaluation of the Lillehei-Kaster pivoting disc valve in patients. J THORAC CARDIOVASC SURG 71: 123-128, 1976 12 Larmi TKI, Kairaluoma MI, Karkola P, Tuononen S, Nuutinen L: Intraoperative hemodynamic evaluation of the Bjork-Shiley tilting disc aortic valve. J THORAC CARDIOVASC SURG 73:712-715, 1977 13 Tandon AP, Smith DR, Mary DAS, Ionescu MI: Sequential hemodynamic studies in patients having aortic valve replacement with the Ionescu-Shiley pericardial xenograft. Ann Thorac Surg 24: 149-155, 1977 14 Nunez L, Gil-Aguado M, Cerron M, Celcmin P: Delayed rupture of the left ventricle after mitral valve replacement with bioprosthesis. Ann Thorac Surg 27:465-467, 1978 15 Wolpowitz A, Bernard MS, Sanchez HE, Barnard CN: Intraoperative posterior left ventricular wall rupture associated with mitral valve replacement. Ann Thorac Surg 25:551-554, 1978 16 Mitha AS, Matisonn MB, leRoux BT, Chesler E: Clinical experience with the Lillehei-Kaster cardiac valve prosthesis. J THORAC CARDIOVASC SURG 72:401-407, 1976 17 Frater RWM: Mitral valve anatomy and prosthetic valve design. Mayo Clin Proc 36:582-592, 1961

620 Becker et at.

The Journal of Thoracic and Cardiovascular Surgery

18 Becker RM, Borg M, Frater RWM: Short tenn clinical followup of patients with biological valve replacements with special reference to transient ischemic attacks with mitral prostheses (abstr). NY State J Med 80:609-610, 1980 19 McAnulty JH, Morton M, Rahimtoola SH, Kloster FE, Ahuja N, Starr AE: Hemodynamic characteristics of the composite strut ball valve prostheses (Starr-Edwards track valves) in patients on anticoagulants. Circulation 58: Suppl I: 159-161, 1978 20 Pyle RP, Mayer JE Jr, Lindsay WG, Jorgenson CR, Wang Y, Nicoloff DM: Hemodynamic evaluation of Lillehei-Kaster and Starr-Edwards prostheses. Ann Thorae Surg 26:336-343, 1978 21 Forman R, Gersh BJ, Fraser R, Beck W: Hemodynamic assessment of Lillehei-Kaster tilting disc aortic and mitral prostheses. J THoRAc CARDIOVASC SURG 75:595-598, 1978 22 Edwards Laboratories Clinical Report, February, 1978 23 Lurie AJ, Miller RR, Maxwell KS, et al: Hemodynamic assessment of the glutaraldehyde preserved porcine heterograft in the aortic and mitral positions. Circulation 56:Suppl 2: 104-110, 1977 24 Cohn LH, Sanders JH Jr, Collins 11: Aortic valve replacement with the Hancock porcine xenograft. Ann Thorac Surg 22:221-227, 1976 25 Morris DC, King SB III, Douglas JS Jr, Wickliffe EW, Jones EL: Hemodynamic results of aortic valvular replacement with the porcine xenograft valve. Circulation 56:841-844, 1977

26 Johnson A, Thompson S, Vieweg WVR, Oury J, Peterson K: Evaluation of the in vivo function of the Hancock porcine xenograft in the aortic position. J THORAC CARD10VASC SURG 75:599-605, 1978 27 Craver JM, King SB, Douglas JS, Franch RH, Jones EL, Morris DC, Kopchak J, Hatcher CR: Late hemodynamic evaluation of Hancock modified orifice aortic bioprosthesis. Circulation 60:Suppl 1:93-97, 1979 28 Lee G, Grehl TM, Joye JA, Kaku RF, Harter W, DeMaria AN, Mason DT: Hemodynamic assessment of the new aortic Carpentier-Edwards bioprosthesis. Cath Cardiovasc Diag 4:373-381, 1978 29 Rothkopf M, Davidson T, Lipscoma K, Narahara K, Hillis LD, Willerson rr, Estrera A, Platt M, Mills L: Hemodynamic evaluation of the Carpentier-Edwards bioprosthesis in the aortic position. Am J Cardiol 44:209-214, 1979 30 Sigwart V, Schmidt H, Gleichman V, Borst HG: In vivo evaluation of the Lillehei-Kaster heart valve prosthesis. Ann Thorac Surg 22:213-220, 1976 31 Nitter-Hauge S, Erge I, Sembe BKH, Hall KV: Primary clinical experience with the Hall-Kaster valve in the aortic position. Circulation 60:Suppl 1:55-62, 1979 32 Bjork YO, Holmgren A, Olin C, Ovenfors C-O: Clinical and hemodynamic results of aortic valve replacement with the Bjork-Shiley tilting disc valve prosthesis. Scand J Thorac Cardiovasc Surg 5: 177-191, 1971 33 Tandon AP, Smith DR, lonescu MI: Hemodynamic evaluation of the lonescu-Shiley pericardial xenograft in the mitral position. Am Heart J 95:595-601, 1978