- Email: [email protected]
Intraoperative hemodynamic evaluation of the Björk-Shiley tilting disc aortic valve
Intraoperative hemodynamic evaluation of the Bjork-Shiley tilting disc aortic valve Intraoperative cardiac output and aortic and left ventricular pressures were measured simultaneously in 15 consecutive patients before and after aortic valve replacement (A VR) with the Bjork-Shiley tilting disc valve. The predominant lesion in these patients was aortic stenosis. The following hemodynamic indices were calculated: aortic valve area (AVA), determined by Gorlin s formula, effective orifice area, and effective area index. Their applicability as hemodynamic criteria of the immediate hemodynamic performance of the replaced valve is discussed. By every criterion, AVR greatly improved the hemodynamic performance. The effective area index seemed more suitable than the other indices for the intraoperative hemodynamic evaluation of the replaced aortic valve. Teuvo K. I. Larmi, M.D., Matti I. Kairaluoma, M.D., Pentti Karkola, M.D., Seppo Tuononen, M.D., and Lauri Nuutinen, M.D., Oulu, Finland
JLn earlier years, hemodynamic assessment of the replaced valve required postoperative cardiac catheterization.1 Now, however, the simultaneous recording of cardiac output, aortic pressure, and left ventricular pressure by currently available flowmeters and pressure-recording systems has made possible the intraoperative hemodynamic evaluation of prosthetic valves during aortic valve replacement (AVR). However, it has been difficult to find a suitable criterion which would reliably describe the hemodynamic performance of the replaced valve prosthesis. The aortic valve area (AVA), calculated according to Gorlin's2 formula, has been commonly used as one such criterion for the stenotic aortic valve. Recently, Aaslid and colleagues3 have introduced a new criterion, the effective area index, which eliminates as effectively as possible the effect of the compromised postperfusion myocardium and the pulsatory flow on fluctuations of the pressure gradient. Using intraoperatively measured flow and pressure data, we calculated both the AVA and the effective area index in 15 consecutive patients, in whom the predominant lesion was aortic stenosis, who underwent AVR. Based on the established indices, we have examFrom the Department of Surgery and Anesthesiology, University of Oulu, Oulu, Finland. Received for publication Sept. 20, 1976. Accepted for publication Nov. 5, 1976. Address for reprints: Teuvo K. I. Larmi, M.D., Department of Surgery, University of Oulu, SF-90220 Oulu 22, Finland. 7 12
ined their applicability as criteria of the intraoperative hemodynamic performance of the Bjork-Shiley tilting disc valve. Patients and methods One hundred patients have undergone AVR with the Bjork-Shiley tilting disc valve at Oulu University Central Hospital. The study group consisted of 15 consecutive patients with AVR for stenosis of the aortic valve operated upon between September, 1974, and May, 1976. There were 12 men and 3 women. The mean age was 42 years, with a range 26 to 52 years. All patients belonged to the New York Heart Association Classes II and III. Table I shows the diagnoses, Table II the operations performed, and Table III the size and the area of the tissue annulus of the valves.4 We used bubble oxygenators, mild hypothermia, and hemodilation during cardiopulmonary bypass. The operations were carried out with the vented heart beating empty at 31° C. and with continuous perfusion of both left and right coronary arteries, as reported in detail elsewhere.5 All studies were carried out intraoperatively before and after AVR with the patient under general endotracheal anesthesia but not on cardiopulmonary bypass and at a point when his condition was judged to be hemodynamically stable. Cardiac output was measured by a Nycotron electromagnetic flowmeter (Nyecaard & Co., Oslo, Norway) with appropriately sized tranducers around the ascending aorta. The aortic and left ventricular pressures were recorded simultaneously by transmural puncture with a needle connected to
Volume 73 Number 5 May, 1977
Bjdrk-Shiley aortic valve
Bjork- Shi ley No. -SD
31 29 27 25 23
|_
713
Table I. Diagnosis
079
-I + S D
No.
Diagnosis AS AS, AI Paravalvular leakage* AS, AI + MI AS, AI + MS
1
2 AVA
Fig. 1. Relation of valve size to aortic valve area (AVA). Each bar represents an individual case. The height of the bar indicates the valve size.
Legend: AS, Aortic stenosis. AI, Aortic insufficiency. MS, Mitral stenosis. MI, Mitral insufficiency. *Smeloff-Cutter valve.
Table II. Operation
0.47 Bjork- Shiley No.
-SD
Legend: AVR, Aortic valve replacement. MVR, Mitral valve replacement. ♦Includes one reoperation for paravalvular leakage.
31 -
29272523-
Table III. Size of the valves
0-2
0.4
0.6
L
Fig. 2. Relation of valve size to effective area index (la). Each bar represents an individual case. The height of the bar indicates the valve size. Statham P23 pressure transducers, after which the signals from the pressure-recording systems and flowmeter were amplified and recorded (Hewlett-Packard Co., Waltham, Mass.). The AVA was calculated according to Gorlin's2 formula for fixed orifices as modified for the aortic valve in man: AVA
13 2
AVR* AVR + M V R
+ SD
h
No.
Operation
Size
Area of tissue annulus (sq. cm.)
21 23 25 27 29 31
3.46 4.15 4.91 5.72 6.60 7.54
and the density of the blood, F ao is the instantaneous or peak ascending aortic flow (in liters per minute), and AP is the peak pressure gradient (in millimeters of mercury) at the instant of peak flow. The effective area index (Ia) is the ratio between the effective orifice area and the area of the tissue annulus of the sewing ring (A a ), reported by manufacturer.4
_ mean aortic valve flow (ml./sec.) 44.5 VpLVsm - pAo
T a
sm
where AVA is aortic valve area (in square centimeters), pLVsm is the mean left ventricular systolic pressure (in millimeters of mercury), and pAosm is the mean aortic systolic pressure (in millimeters of mercury). The effective orifice area was calculated according to the formula proposed by Aaslid and colleagues3:
No. of valve prostheses
_ Aeff Aa
'
All calculations were made in each patient both before and after AVR. The peak aortic flow used as cardiac output for calculation of the effective orifice area was corrected for any aortic insufficiency in combined valvular disease but not for coronary flow, since coronary flow was considered to be negligible during systole. Results
where Aeff is effective orifice area (in square centimeters), 0.32 is a constant determined by the units used
There were no hospital deaths. One patient (Patient 6) died of serum hepatitis 5 months postoperatively. AVR greatly improved the hemodynamic perfor-
The Journal of Thoracic and Cardiovascular Surgery
7 14 Larmi et al.
Table IV. Comparison of hemodynamic data before and after aortic valve replacement Aortic valve area (sq. cm.)
Effective orifice area (sq. cm.)
Effective area index
Pt. No.
Before
After
Before
After
Before
After
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.10 0.59 0.22 0.34 0.29 0.49 0.31 0.70 0.34 0.67 0.67 0.58 0.75 0.34 0.29
1.5 1.1 0.3 1.0 1.3 0.4 0.8 0.8 0.8 1.8 0.4 0.6 1.3 0.4 0.3
1.2 1.6 0.7 0.9 0.6 0.6 1.2 1.3 2.4 2.0 1.5 0.9 1.7 2.7 1.9
3.3 2.5 1.1 3.7 2.9 2.1 2.7 1.5 3.3 3.4 2.2 1.6 3.8 3.2 2.8
0.20 0.32 0.14 0.12 0.11 0.10 0.28 0.31 0.43 0.30 0.28 0.18 0.23 0.35 0.33
0.58 0.51 0.22 0.49 0.38 0.37 0.65 0.37 0.58 0.52 0.34 0.33 0.59 0.56 0.52
Mean ± S.D. p Values*
0.46 ± 0.20
0.79 ± 0.44
1.41 ± 0.64
2.67 ± 0.83
0.25 ± 0.10
<0.01
<0.001
0.47 ± 0.12 <0.001
'Student's t test.
mance of the diseased aortic valve, as shown in Table IV. In patients with a long perfusion time (Patients 3, 6, and 11), the AVA index showed little or no improvement in valve hemodynamics after AVR, but both the effective orifice area and the effective area index demonstrated a definite increase in valve performance. In Patient 8, who underwent AVR for paravalvular leakage, all indices showed equally only a small improvement. Figs. 1 and 2 show the relation between the different valve sizes and the AVA and effective area index. The standard deviation in the AVA is somewhat larger than in effective area index. However, in both there are also large individual differences in hemodynamic performance in valves of equal size. Discussion The intraoperative assessment of valve hemodynamic function ensures that the immediate hemodynamic performance of the valve is satisfactory and provides a data base for comparison with postoperative catheterization results. This method can also be used for comparison of different types of valve prostheses.3 An ideal index is one in which the influence of inertial components and compromised postperfusion myocardial contractibility is minimal; it also takes into consideration how effectively the valve prosthesis utilizes the available space in the aortic root. At present, two criteria are available: the well-accepted AVA calculated by Gorlin's formula and the new effective area
index suggested by Aaslid and colleagues.3 Before valve replacement, when the diseased valve is still in place, the AVA and effective orifice area can be calculated only for stenotic valves. This fact forced us to conduct our investigation in patients in whom the predominant lesion was stenosis of the aortic valve. However, after valve replacement, it is possible to compare the immediate hemodynamic results of the replaced valves in patients with both aortic stenosis and aortic regurgitation, because every prosthetic valve is somewhat stenotic. In the present study, we determined both criteria before and after AVR with the Bjork-Shiley tilting disc prosthesis. Each criterion showed a marked improvement in valve hemodynamics after aortic valve replacement, as also reported by Fletcher and colleagues.6 There were no postoperative deaths in the present series, but in Fletcher's series 2 patients died postoperatively of valve malfunction. In both, the intraoperative hemodynamic studies after replacement demonstrated abnormal findings. Our results suggest that the effective orifice area and the effective area index are probably less easily affected by the depressed postperfusion myocardial function than is the AVA as determined by Gorlin's formula, and thus they are more suitable for assessing the hemodynamic performance of the inserted prosthesis. Their greater suitability is also quite evident of a theoretical basis. For determination of AVA by Gorlin's formula, the mean aortic flow was used as cardiac output. After perfusion, cardiac output
Volume 73 Number 5 May, 1977
measured in this way is generally quite low owing to myocardial depression. Furthermore, this formula does not take into consideration the size of the prosthesis used. The effective orifice area is calculated by use of the instant of the peak aortic flow and the instantaneous pressure gradient at the instant of peak flow, which, according to Aaslid and associates,3 are less sensitive to inertial components and to the depressed myocardial function. The effective area index describes the postoperative hemodynamics even more effectively, because it also takes into consideration the area occupied by the valve. Fletcher and colleagues6 have described a very similar criterion called efficiency index, but they used Gorlin's formula for determination of AVA. In the present study, the mean effective area index for the Bjork-Shiley tilting disc prosthesis was 0.44 ± 0.12, and in the series of Aaslid and associates3 0.32 ± 0.06. The individual differences in the hemodynamic performance of the valves of equal size were marked, as also reported by Aaslid,3 Starek,7 and their colleagues. However, the standard deviation in the effective area index was smaller than that in the aortic valve area, indicating that the former index is more applicable as a hemodynamic criterion. In conclusion, on the basis of our results, we suggest the use of the effective area index as a hemodynamic criterion to assess the immediate hemodynamic function of the inserted valve prosthesis. However, to ascer-
Bjork-Shiley aortic valve
11 5
tain the applicability and predictability of the proposed index, long-term clinical follow-up studies are required. This program is now in progress in our clinic.
REFERENCES 1 Henze, A.: The Bjork-Shiley Tilting Disc Valve in Aortic Valvular Disease, Scand. J. Thorac. Cardiovasc. Surg. Suppl. 14, 1974. 2 Gorlin, R., and Gorlin, S. G.: Hydraulic Formula for Calculation of the Area of Stenotic Mitral Valve, Other Cardiac Valves, and Central Circulatory Shunts, Am. Heart J. 41: 1, 1951. 3 Aaslid, R., Levang, O., Froysaker, T., Skagseth, E., and Hall, K. V.: "In Situ" Evaluation of the Aortic Pivoting Disc Valve Prosthesis, Scand. J. Thorac. Cardiovasc. Surg. 9: 81, 1975. 4 Bjork, V. O.: A New Tilting Disc Valve Prosthesis, Scand. J. Thorac. Cardiovasc. Surg. 3: 1, 1969. 5 Larmi, T. K. I., Karkola, P., Kairaluoma, M. I., and Takkunen, J.: Four Years' Experience of Valve Replacement With the Bjork-Shiley Tilting-Disc Prosthesis, Ann. Chir. Gynaecol. 65: 110, 1976. 6 Fletcher, J. R., Ash worth, H. E., Shepard, B. M., and Mills, M.: Intraoperative Hemodynamics After Insertion of Starr-Edwards Ball-Valve Prostheses for Acquired Valvular Heart Disease, J. THORAC. CARDIOVASC. SURG. 67: 285, 1974. 7 Starek, P. J. K., Wilcox, B. R., and Murray, G. F.: Hemodynamic Evaluation of the Lillehei-Kaster Pivoting Disc Valve in Patients, J. THORAC. CARDIOVASC. SURG. 71: 123, 1976.
JLn earlier years, hemodynamic assessment of the replaced valve required postoperative cardiac catheterization.1 Now, however, the simultaneous recording of cardiac output, aortic pressure, and left ventricular pressure by currently available flowmeters and pressure-recording systems has made possible the intraoperative hemodynamic evaluation of prosthetic valves during aortic valve replacement (AVR). However, it has been difficult to find a suitable criterion which would reliably describe the hemodynamic performance of the replaced valve prosthesis. The aortic valve area (AVA), calculated according to Gorlin's2 formula, has been commonly used as one such criterion for the stenotic aortic valve. Recently, Aaslid and colleagues3 have introduced a new criterion, the effective area index, which eliminates as effectively as possible the effect of the compromised postperfusion myocardium and the pulsatory flow on fluctuations of the pressure gradient. Using intraoperatively measured flow and pressure data, we calculated both the AVA and the effective area index in 15 consecutive patients, in whom the predominant lesion was aortic stenosis, who underwent AVR. Based on the established indices, we have examFrom the Department of Surgery and Anesthesiology, University of Oulu, Oulu, Finland. Received for publication Sept. 20, 1976. Accepted for publication Nov. 5, 1976. Address for reprints: Teuvo K. I. Larmi, M.D., Department of Surgery, University of Oulu, SF-90220 Oulu 22, Finland. 7 12
ined their applicability as criteria of the intraoperative hemodynamic performance of the Bjork-Shiley tilting disc valve. Patients and methods One hundred patients have undergone AVR with the Bjork-Shiley tilting disc valve at Oulu University Central Hospital. The study group consisted of 15 consecutive patients with AVR for stenosis of the aortic valve operated upon between September, 1974, and May, 1976. There were 12 men and 3 women. The mean age was 42 years, with a range 26 to 52 years. All patients belonged to the New York Heart Association Classes II and III. Table I shows the diagnoses, Table II the operations performed, and Table III the size and the area of the tissue annulus of the valves.4 We used bubble oxygenators, mild hypothermia, and hemodilation during cardiopulmonary bypass. The operations were carried out with the vented heart beating empty at 31° C. and with continuous perfusion of both left and right coronary arteries, as reported in detail elsewhere.5 All studies were carried out intraoperatively before and after AVR with the patient under general endotracheal anesthesia but not on cardiopulmonary bypass and at a point when his condition was judged to be hemodynamically stable. Cardiac output was measured by a Nycotron electromagnetic flowmeter (Nyecaard & Co., Oslo, Norway) with appropriately sized tranducers around the ascending aorta. The aortic and left ventricular pressures were recorded simultaneously by transmural puncture with a needle connected to
Volume 73 Number 5 May, 1977
Bjdrk-Shiley aortic valve
Bjork- Shi ley No. -SD
31 29 27 25 23
|_
713
Table I. Diagnosis
079
-I + S D
No.
Diagnosis AS AS, AI Paravalvular leakage* AS, AI + MI AS, AI + MS
1
2 AVA
Fig. 1. Relation of valve size to aortic valve area (AVA). Each bar represents an individual case. The height of the bar indicates the valve size.
Legend: AS, Aortic stenosis. AI, Aortic insufficiency. MS, Mitral stenosis. MI, Mitral insufficiency. *Smeloff-Cutter valve.
Table II. Operation
0.47 Bjork- Shiley No.
-SD
Legend: AVR, Aortic valve replacement. MVR, Mitral valve replacement. ♦Includes one reoperation for paravalvular leakage.
31 -
29272523-
Table III. Size of the valves
0-2
0.4
0.6
L
Fig. 2. Relation of valve size to effective area index (la). Each bar represents an individual case. The height of the bar indicates the valve size. Statham P23 pressure transducers, after which the signals from the pressure-recording systems and flowmeter were amplified and recorded (Hewlett-Packard Co., Waltham, Mass.). The AVA was calculated according to Gorlin's2 formula for fixed orifices as modified for the aortic valve in man: AVA
13 2
AVR* AVR + M V R
+ SD
h
No.
Operation
Size
Area of tissue annulus (sq. cm.)
21 23 25 27 29 31
3.46 4.15 4.91 5.72 6.60 7.54
and the density of the blood, F ao is the instantaneous or peak ascending aortic flow (in liters per minute), and AP is the peak pressure gradient (in millimeters of mercury) at the instant of peak flow. The effective area index (Ia) is the ratio between the effective orifice area and the area of the tissue annulus of the sewing ring (A a ), reported by manufacturer.4
_ mean aortic valve flow (ml./sec.) 44.5 VpLVsm - pAo
T a
sm
where AVA is aortic valve area (in square centimeters), pLVsm is the mean left ventricular systolic pressure (in millimeters of mercury), and pAosm is the mean aortic systolic pressure (in millimeters of mercury). The effective orifice area was calculated according to the formula proposed by Aaslid and colleagues3:
No. of valve prostheses
_ Aeff Aa
'
All calculations were made in each patient both before and after AVR. The peak aortic flow used as cardiac output for calculation of the effective orifice area was corrected for any aortic insufficiency in combined valvular disease but not for coronary flow, since coronary flow was considered to be negligible during systole. Results
where Aeff is effective orifice area (in square centimeters), 0.32 is a constant determined by the units used
There were no hospital deaths. One patient (Patient 6) died of serum hepatitis 5 months postoperatively. AVR greatly improved the hemodynamic perfor-
The Journal of Thoracic and Cardiovascular Surgery
7 14 Larmi et al.
Table IV. Comparison of hemodynamic data before and after aortic valve replacement Aortic valve area (sq. cm.)
Effective orifice area (sq. cm.)
Effective area index
Pt. No.
Before
After
Before
After
Before
After
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.10 0.59 0.22 0.34 0.29 0.49 0.31 0.70 0.34 0.67 0.67 0.58 0.75 0.34 0.29
1.5 1.1 0.3 1.0 1.3 0.4 0.8 0.8 0.8 1.8 0.4 0.6 1.3 0.4 0.3
1.2 1.6 0.7 0.9 0.6 0.6 1.2 1.3 2.4 2.0 1.5 0.9 1.7 2.7 1.9
3.3 2.5 1.1 3.7 2.9 2.1 2.7 1.5 3.3 3.4 2.2 1.6 3.8 3.2 2.8
0.20 0.32 0.14 0.12 0.11 0.10 0.28 0.31 0.43 0.30 0.28 0.18 0.23 0.35 0.33
0.58 0.51 0.22 0.49 0.38 0.37 0.65 0.37 0.58 0.52 0.34 0.33 0.59 0.56 0.52
Mean ± S.D. p Values*
0.46 ± 0.20
0.79 ± 0.44
1.41 ± 0.64
2.67 ± 0.83
0.25 ± 0.10
<0.01
<0.001
0.47 ± 0.12 <0.001
'Student's t test.
mance of the diseased aortic valve, as shown in Table IV. In patients with a long perfusion time (Patients 3, 6, and 11), the AVA index showed little or no improvement in valve hemodynamics after AVR, but both the effective orifice area and the effective area index demonstrated a definite increase in valve performance. In Patient 8, who underwent AVR for paravalvular leakage, all indices showed equally only a small improvement. Figs. 1 and 2 show the relation between the different valve sizes and the AVA and effective area index. The standard deviation in the AVA is somewhat larger than in effective area index. However, in both there are also large individual differences in hemodynamic performance in valves of equal size. Discussion The intraoperative assessment of valve hemodynamic function ensures that the immediate hemodynamic performance of the valve is satisfactory and provides a data base for comparison with postoperative catheterization results. This method can also be used for comparison of different types of valve prostheses.3 An ideal index is one in which the influence of inertial components and compromised postperfusion myocardial contractibility is minimal; it also takes into consideration how effectively the valve prosthesis utilizes the available space in the aortic root. At present, two criteria are available: the well-accepted AVA calculated by Gorlin's formula and the new effective area
index suggested by Aaslid and colleagues.3 Before valve replacement, when the diseased valve is still in place, the AVA and effective orifice area can be calculated only for stenotic valves. This fact forced us to conduct our investigation in patients in whom the predominant lesion was stenosis of the aortic valve. However, after valve replacement, it is possible to compare the immediate hemodynamic results of the replaced valves in patients with both aortic stenosis and aortic regurgitation, because every prosthetic valve is somewhat stenotic. In the present study, we determined both criteria before and after AVR with the Bjork-Shiley tilting disc prosthesis. Each criterion showed a marked improvement in valve hemodynamics after aortic valve replacement, as also reported by Fletcher and colleagues.6 There were no postoperative deaths in the present series, but in Fletcher's series 2 patients died postoperatively of valve malfunction. In both, the intraoperative hemodynamic studies after replacement demonstrated abnormal findings. Our results suggest that the effective orifice area and the effective area index are probably less easily affected by the depressed postperfusion myocardial function than is the AVA as determined by Gorlin's formula, and thus they are more suitable for assessing the hemodynamic performance of the inserted prosthesis. Their greater suitability is also quite evident of a theoretical basis. For determination of AVA by Gorlin's formula, the mean aortic flow was used as cardiac output. After perfusion, cardiac output
Volume 73 Number 5 May, 1977
measured in this way is generally quite low owing to myocardial depression. Furthermore, this formula does not take into consideration the size of the prosthesis used. The effective orifice area is calculated by use of the instant of the peak aortic flow and the instantaneous pressure gradient at the instant of peak flow, which, according to Aaslid and associates,3 are less sensitive to inertial components and to the depressed myocardial function. The effective area index describes the postoperative hemodynamics even more effectively, because it also takes into consideration the area occupied by the valve. Fletcher and colleagues6 have described a very similar criterion called efficiency index, but they used Gorlin's formula for determination of AVA. In the present study, the mean effective area index for the Bjork-Shiley tilting disc prosthesis was 0.44 ± 0.12, and in the series of Aaslid and associates3 0.32 ± 0.06. The individual differences in the hemodynamic performance of the valves of equal size were marked, as also reported by Aaslid,3 Starek,7 and their colleagues. However, the standard deviation in the effective area index was smaller than that in the aortic valve area, indicating that the former index is more applicable as a hemodynamic criterion. In conclusion, on the basis of our results, we suggest the use of the effective area index as a hemodynamic criterion to assess the immediate hemodynamic function of the inserted valve prosthesis. However, to ascer-
Bjork-Shiley aortic valve
11 5
tain the applicability and predictability of the proposed index, long-term clinical follow-up studies are required. This program is now in progress in our clinic.
REFERENCES 1 Henze, A.: The Bjork-Shiley Tilting Disc Valve in Aortic Valvular Disease, Scand. J. Thorac. Cardiovasc. Surg. Suppl. 14, 1974. 2 Gorlin, R., and Gorlin, S. G.: Hydraulic Formula for Calculation of the Area of Stenotic Mitral Valve, Other Cardiac Valves, and Central Circulatory Shunts, Am. Heart J. 41: 1, 1951. 3 Aaslid, R., Levang, O., Froysaker, T., Skagseth, E., and Hall, K. V.: "In Situ" Evaluation of the Aortic Pivoting Disc Valve Prosthesis, Scand. J. Thorac. Cardiovasc. Surg. 9: 81, 1975. 4 Bjork, V. O.: A New Tilting Disc Valve Prosthesis, Scand. J. Thorac. Cardiovasc. Surg. 3: 1, 1969. 5 Larmi, T. K. I., Karkola, P., Kairaluoma, M. I., and Takkunen, J.: Four Years' Experience of Valve Replacement With the Bjork-Shiley Tilting-Disc Prosthesis, Ann. Chir. Gynaecol. 65: 110, 1976. 6 Fletcher, J. R., Ash worth, H. E., Shepard, B. M., and Mills, M.: Intraoperative Hemodynamics After Insertion of Starr-Edwards Ball-Valve Prostheses for Acquired Valvular Heart Disease, J. THORAC. CARDIOVASC. SURG. 67: 285, 1974. 7 Starek, P. J. K., Wilcox, B. R., and Murray, G. F.: Hemodynamic Evaluation of the Lillehei-Kaster Pivoting Disc Valve in Patients, J. THORAC. CARDIOVASC. SURG. 71: 123, 1976.