- Email: [email protected]
Hemodynamic effects of pacing-induced heart rate augmentation
Hemodynamic effects of pacing-induced heart rate augmentation A study in patients early after mitral and aortic valve operations Heart rate augmentation induced by atrial and ventricular pacing was carried out in 23 patients early after intracardiac valve operations. The series comprised two groups of patients: 11 with aortic and 12 with mitral valve surgery. A different pattern of hemodynamic responses emerged during heart rate augmentation in each group. Atrial pacing produced, in the aortic group, a significant increase in cardiac output and mean aortic pressure and a reduction in left atrial pressure. The changes in the mitral group were not significant. During ventricular pacing the left atrial pressure rose in both groups. Although there was a decrease in the mean cardiac output in the aortic group and an increase in the mitral group, these changes did not reach statistical significance. A similar distinction between the aortic and mitral groups was observed when comparing the results of atrial and ventricular pacing at the same heart rate. The hemodynamic benefit appeared to be confined to the aortic group. The different pattern of response in each group of patients was probably due to clinical, hemodynamic, and therapeutic factors related to the condition and management of patients with each type of valve surgery.
David A. S. Mary, M.B., Ch.B., M.R.C.P., Brojesh C. Pakrashi, M.B.B.S., M.R.C.P., and Marian I. Ionescu, M.D., F.A.C.S., Leeds, England
*L/ow cardiac output is one of the early complications after intracardiac valve operations and especially after mitral valve surgery.1"3 It appears to be related to the occurrence of subendocardial necrosis and it is responsible for the majority of deaths in the early postoperative period.4 Electrical pacing of the heart has become established as a technique in the treatment of bradyarrhythmias and ectopic rhythms.5' 6 It has been shown to increase the cardiac output in patients with atrioventricular conduction disturbances following open-heart operations.7' 8 Information concerning the influence of heart rate augmentation on cardiac output early after mitral and aortic valve surgery, in the absence of abnormal atrioventricular conduction or other correctable factors, is limited and variable. 6,7 ' 9 ' 10 The results of atrial and From the Department of Cardiothoracic Surgery, The General Infirmary, Leeds, England. Received for publication July 10, 1975. Address for reprints: Mr. M. I. Ionescu, Department of Cardiothoracic Surgery, The General Infirmary, Leeds, 1, Great Britain. 520
ventricular pacing following isolated aortic valve replacement10 and the over-all results in patients with aortic, mitral, and combined valve surgery with atrial9 pacing have been reported. A comparison of hemodynamic consequences of heart rate augmentation by atrial or ventricular pacing between patients with isolated aortic and those with isolated mitral valve surgery is lacking. The published reports on the effect of atrial and ventricular pacing at the same heart rate in patients after heart valve replacement8' 9 do not elucidate the difference in response between these two types of pacing. This study analyzes the influence of heart rate augmentation by atrial and ventricular pacing on the hemodynamic status in stable postoperative patients who had either isolated aortic or mitral valve surgery. The results are compared according to the valve operated upon (aortic or mitral) and the type of pacing used (atrial or ventricular). Materials and methods Twenty-three patients were investigated during the first 48 hours following open intracardiac operations.
Volume 71 Number 4 April, 1976
Postoperative cardiac pacing
52 1
Table I. Preoperative clinical data in 23 patients
Valve lesion
II
III
IV
S.R.
A.F.
L.V.H.
Range
Mean
Patients with digitalis therapy
Aortic Mitral
6 5
4 7
1 —
10 3
1 9
11 3
44-66 49-67
53 58
2 9
N.Y.H.A. grade
Cardiothoracic ratio
ECG findings*
*S.R., Sinus rhythm. A.F., Atrial fibrillation. L.V.H., Left ventricular hypertrophy.
There were 14 men and 9 women and their ages ranged from 24 to 69 (mean 49) years. Twelve patients had mitral valve surgery which comprised replacement with Starr prostheses in 5 patients, stented tissue valves in 3, Bjork prostheses in 1, and mitral annuloplasty in 3 patients. Eleven patients had aortic valve replacement. Stented tissue valves were used in 10 and Bjork prosthesis in 1 patient. The details of the operative procedure have been reported previously.11, 12 Following the completion of the operation, temporary pacemaker electrodes were inserted into the anterior wall of the right ventricle, the atrial wall at the junction with the superior vena cava, and an indifferent one in the abdominal muscle. The electrode wires were brought out through the skin at sites other than the surgical incision. Left atrial and femoral artery catheters were kept in place for postoperative pressure monitoring. In all the patients the pericardial sac was left open at the conclusion of the operation. The preoperative clinical data of the patients are shown in Table I. Hemodynamic studies. Investigations were carried out after a stable clinical status had been attained. This was evaluated by monitoring, among other parameters, the cardiac rhythm, central and peripheral skin temperature, urine output, intracardiac and arterial pressures, blood gases, electrolytes, hemoglobin, packed cell volume, and acid-base status. When applied, infusions and mechanical ventilators were kept at a constant rate throughout the procedure. Positive pressure ventilation was used in 22 patients. Four patients were lying flat and 19 were in a semirecumbant position during the study. Atrial and ventricular pacing was performed with a battery-operated fixed-rate generator. Square waves of 1 msec, duration were utilized. Cardiac output was measured by the indicator dilution techniques. One milligram of indocyanine green was injected rapidly into the left atrium and blood was aspirated from the femoral artery through the catheters inserted at operation. Catheter tubing of the same size was used in every patient. Constant blood withdrawal through a Water's SC-302 densitometer
cuvette was employed. The dye curves were recorded simultaneously with the elctrocardiogram (ECG), left atrial, and systemic artery pressures with an ultraviolet recording system. The dye curves were calibrated at the end of each procedure by three consecutive levels of dye concentration in fresh blood withdrawn from each patient. The Hamilton-Stewart semilogarithmic manual replot extrapolation method was used to calculate the cardiac output. Pressure measurements were performed with the zero reference point set at the level of the sternal angle and were taken as the average of several respiratory cycles. Cardiac index and left ventricular minute work were calculated according to conventional formulas. Cardiac output and pressure measurements were begun at the intrinsic heart rate of the patient and regarded as the data during the first control period. Pacing was then performed to increase the heart rate in two stages, each of about 20 beats per minue (b.p.m.), and further measurements were made after about 2 to 6 minutes from reaching each new heart rate level. The same measurements were repeated when the heart rate was allowed to return to the initial control level. The data during the initial and final control period were averaged and used for comparison with those during cardiac pacing. This sequence of stepwise increase in heart rate was completed in 13 patients. In the remaining 10 patients, a full sequence was not possible because of either a relatively faster intrinsic heart rate or the occurrence during pacing of atrioventricular block and palpitation. Statistical analysis of the differences was carried out with Student's t test for paired data and standard probability tables. Results There was no significant statistical difference between heart rate (P > 0.70), cardiac output (P > 0.40), aortic pressure (P > 0.90), and left atrial pressure (P > 0.50) recorded at the beginning and at the end of the investigation. Reproducibility of results. Cardiac output curves were obtained in duplicate at each stage in every patient and the average was used in the analysis. The
522
The Journal of Thoracic and Cardiovascular Surgery
Mary, Pakrashi, lonescu
Table II. The effect of atrial pacing on hemodynamic parameters in 16 patients Type of operation Aortic valve replacement
Mitral valve surgery
No. of measurements
Heart rate (b.p.m.)
Cardiac index
(L.lmin.lM.2)
Mean aortic pressure (mm. Hg) 86 ± 7 97 ± 9*
Mean left atrial pressure (mm. Hg)
8
79 ± 3 102 ± 2*
3.60 ± 0.47 4.21 ± 0.55*
11
100 ± 2 128 ± 1*
4.47 ± 0.42 4.48 ± 0.40
101 ± 7 98 ± 6
4 ± 1 5 ± 1
8
79 ± 3 128 ± 2*
3.60 ± 0.47 4.16 ± 0.51*
86 ± 7 92 ± 6
5 ± 1 4 ± 1*
4
69 ± 6 101 ± 1*
2.50 ± 0.54 2.83 ± 0.49
95 ± 9 99 ± 10
12 ± 1 12 ± 1
4
101 ± 1 128 ± 3*
2.98 ± 0.35 3.03 ± 0.37
103 ± 10 101 ± 9
14 ± 2 14 ± 2
3
68 ± 9 127 ± 4*
2.90 ± 0 . 5 1 3.39 ± 0.05
97 ± 13 102 ± 13
12 ± 1 12 ± 2
5 ± 1 3 ± 1*
indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
difference between the cardiac indices calculated from consecutive measurements was not significant (P > 0.6). The standard error of the mean difference between 31 pairs of consecutive measurements was 0.095 liter per minute per square meter (L./min./M. 2 ). It was slightly larger (0.127 L./min./M. 2 ) in 7 pairs of measurements with cardiac indices below 2.5 L./ min./M. 2 when compared with 0.118 L./min./M. 2 obtained in the remaining 24 estimations with higher cardiac indices, and this is in accord with previous reports.13 Atrial pacing. Heart rate increments by atrial pacing were possible in 16 patients, of whom 11 underwent aortic valve replacement and 5 mitral valve surgery. All these patients were in sinus rhythm during the postoperative period. In patients with aortic valve replacement there was a significant increase in cardiac output of 0.608 L./ min./M. 2 (17 per cent, P < 0.025) and 0.558 L.I min./M. 2 (16 per cent, P < 0.05) in two sets of investigations each of 8 pairs of measurements, on increasing the heart rate from 79 ± 3 to 102 ± 2 b.p.m. (29 per cent P < 0.001) and from 79 ± 3 to 128 ± 2 b.p.m. (63 per cent P < 0.001), respectively. There was a corresponding increase of mean aortic pressure by 11 mm. Hg (12 per cent P < 0.01) and 7 mm. Hg (8 per cent P > 0.05), respectively. The mean left atrial pressure decreased by 2.5 mm. Hg (49 per cent) and 1.5 mm. Hg (29 per cent), respectively. These changes reached statistical significance (P < 0.005 and P < 0.05). In patients with mitral valve surgery heart rate
increments from 69 ± 6 to 101 ± 1 b.p.m. (46 per cent P < 0.01) and from 68 ± 9 to 127 ± 4 b.p.m. (91 per cent P < 0.001) resulted in insignificant rises in cardiac indices (P > 0.20 and P > 0.40) and aortic pressures (P > 0.40 and P > 0.50), while the mean left atrial pressure remained nearly unchanged (Table II). In both the aortic and mitral groups, increments in heart rate from 100 to 128 b.p.m. revealed a smaller increase in cardiac index, which suggested an optimum heart rate of about 100 b.p.m. The left ventricular work, estimated only in patients with aortic valve replacement, showed an insignificant increase of 0.615 (14 per cent P > 0.20) and 1.81 Kg./min./M. 2 (26 per cent P > 0.05) with increments of heart rate from 79 ± 3 to 102 ± 2 b.p.m. and from 79 ± 3 to 128 ± 2 b.p.m., respectively. Ventricular pacing. Augmentation of heart rate by ventricular pacing was performed in 18 patients. Seven of these had aortic valve replacement and 11 had mitral valve surgery (valve replacement in 9 and repair in 2). All patients with aortic valve replacement were in sinus rhythm. In the mitral valve surgery group, there were 5 patients in atrial fibrillation, 5 in sinus rhythm, and 1 in nodal rhythm. In patients with aortic valve replacement studies of heart rate increments from 73 ± 2 to 103 ± 1 b.p.m. (41 per cent P < 0.001) and from 73 ± 2 to 131 ± 4 b.p.m. (81 per cent P < 0.001) produced a small reduction in the average of cardiac indices (0.11 L./min./M. 2 , 3 per cent, and 0.023 L./min./M. 2 , 1 per cent respectively) which did not reach statistical significance. The mean left atrial pressure increased in
Volume 71 Number 4 April, 1976
Postoperative cardiac pacing
523
Table III. The effect of ventricular pacing on hemodynamic parameters in 18 patients Type of operation
No. of measurements
Heart rate (b.p.m.)
Cardiac index (L./min./M.2)
Mean aortic pressure (mm. Hg)
Mean left atrial pressure (mm. Hg)
Aortic valve replacement
4
73 ± 2 103+1*
3.63 ± 0.56 3.52 ± 0 . 5 6
90 + 11 91 ± 11
6 ±2 10 ± 2*
7
100 ± 2 129 ± 3*
4.22 ± 0.45 3.84 ± 0.24
101 ± 9 97 ± 7
9 ± 1 11 ± 2
4
73 ± 2 131 ± 4 *
3.63 ± 0.57 3.61 + 0.37
90 ± 11 93 ± 12
6 ± 2 9 ± 3
8
69+4 101 + 1*
2.69 ± 0.30 2.99 ± 0.40
85 ± 5 86 ± 5
11+2 11 ± 1
7
99 ± 2 126 ± 3*
3.31 ± 0.37 3.23 ± 0.42
83 ± 6 84 ± 6
11 ± 2 12 ± 2
8
73 ± 4 127 ± 2*
2.95 + 0.29 3.15 ± 0.40
95 + 9 97 ± 9
12 ± 2 14 ± 2*
Mitral valve surgery
indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
Table IV. A comparison of the effect of atrial and ventricular pacing in the same patients and at the same heart rate Type of operation
No. of measurements
Aortic valve replacement Mitral valve surgery
11 9
Pacing A V A V
Heart rate (b.p.m.) 119 119 104 104
±4 ± 4 ± 9 ± 9
Cardiac index (L./min./M.2) 4.68 3.72 3.04 3.02
± ± ± ±
0.38 0.24* 0.22 0.20
Mean aortic pressure (mm. Hg) 103 95 103 96
± + ± ±
7 6* 6 5*
Mean left atrial pressure (mm. Hg) 4 11 12 12
± ± ± ±
1 1* 1 1
'Indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
each, and reached statistical significance (4.8 mm. Hg, 87 per cent P < 0.02) only in the first set of measurements (Table III). The changes in the mean aortic pressure were small. Two sets of heart rate increments, each with 8 measurements, were carried out in patients with mitral valve surgery. Heart rate increases from 69 ± 4 to 101 ± 1 (46 per cent P < 0.001) and from 73 ± 4 to 127 ± 2 (74 per cent P < 0.001) resulted in an insignificant rise in the average cardiac indices (0.301 L./min./M. 2 , 11 per cent P > 0 . 1 , and 0.194 L./ min./M.2, 7 per cent P > 0.40, respectively), mean aortic pressures (P > 0.50, P > 0.50), and mean left atrial pressures (P > 0.50, P > 0.05). These changes were not influenced by the type of the original cardiac rhythm. Increases in heart rate from the intermediate to the maximal stage were associated with a consistent reduction in cardiac index in both aortic and mitral groups (Table III). This result suggested an optimum rise in cardiac index at a ventricular rate of 101 b.p.m.
in patients with mitral valve surgery. In the aortic group a follow-up of cardiac index in the same patients at the three levels of heart rate suggested an optimum heart rate at any level between 69 and 100 b.p.m. when patients were in sinus rhythm. The left ventricular work was estimated only in patients with aortic valve replacement and revealed an insignificant reduction of 0.33 (8 per cent, P > 0.40) and 0.29 Kg./min./M. z (7 per cent, P > 0.40), with increments of heart rate from 73 ± 2 to 103 dt 1 b.p.m. and from 73 ± 2 to 131 ± 4 b.p.m., respectively. Atrial and ventricular pacing. In 10 patients, the hemodynamic data were measured during successive atrial and ventricular pacing at the same rate. A total of 19 studies were performed, 11 in patients with aortic valve replacement and 8 in patients with mitral valve surgery. In the aortic group, pacing at a rate of 119 ± 4 b.p.m. resulted in a higher cardiac index (0.956
524
Mary, Pakrashi, lonescu
L./min./M. 2 , 26 per cent, P < 0.001) and mean aortic pressure (8 mm. Hg, 8 per cent, P < 0.05) and a lower mean left atrial pressure (7 mm. Hg, 62 per cent, P < 0.001) during atrial, as compared to ventricular pacing (Table IV). The corresponding data in the mitral group paced at a rate of 104 ± 9 b.p.m. showed practically similar cardiac indices (P > 0.975) and mean left atrial pressures (P > 0.70) with both atrial and ventricular pacing, whereas the aortic pressure was higher (7 mm. Hg, 7 per cent, P < 0.01) during atrial pacing. In the aortic group, left ventricular work during atrial pacing was higher (2.01 Kg./min./M. 2 , 45 per cent, P < 0.001) than that measured during ventricular pacing. Discussion Published reports have shown that following openheart surgery the cardiac output is low in the immediate postoperative period, especially in patients undergoing mitral valve replacement.1-3 The low cardiac output state has been implicated in the development of myocardial necrosis and is the cause of a significant incidence of postoperative death.4 Atrioventricular conduction disturbances, dysrhythmias, hypoxia, hypokalemia, and altered acid-base status are identifiable factors which may be corrected by electrical pacing and pharmacological agents. Depressed cardiac output may often exist in the absence of—or after correction of—such adverse factors. In the absence of these factors, heart rate augmentation induced by atrial pacing has produced in this series a significant increase in cardiac output in the aortic group of patients and an insignificant one in the mitral group. Similar results may be observed in another report.9 In that study an increase in heart rate by atrial pacing has resulted in a larger average rise in cardiac index in 14 patients following aortic valve replacement, as compared with that in 4 patients with isolated mitral valve surgery. Woodson and Starr6 reported a rise of 43 per cent in the cardiac index with the use of atrial pacing in 15 patients with mitral valve surgery. The intrinsic rhythm in 8 of these patients was nodal and a comparison between these results and ours is, therefore, difficult. A further report on atrial pacing in 13 patients following isolated aortic valve replacement with tissue grafts has shown that at a comparable percentage increase in heart rate (20 to 40 per cent), there was a similar directional change in cardiac output, mean aortic, and left atrial pressures,10 as seen in the present series. The results of the present study suggest that heart rate augmentation by atrial pacing
The Journal of Thoracic and Cardiovascular Surgery
would produce a better hemodynamic improvement in patients following aortic valve replacement than in those with mitral valve surgery. By the use of ventricular pacing in this series, there was an insignificant decrease in the cardiac index in the aortic group and an increase in the mitral group, regardless of the intrinsic cardiac rhythm (sinus mechanism or atrial fibrillation). The mean left atrial pressure rose in each group of patients. Similar results have been shown in 13 patients with aortic valve replacement.10 The results of the present investigation which are supported by findings from other reports, therefore, suggest that early after intracardiac valve operations, heart rate augmentation by pacing induces a different pattern of hemodynamic changes in patients with aortic as compared to mitral valve surgery. The use of atrial pacing has resulted in a better hemodynamic improvement in the aortic group than in the mitral group. With the use of ventricular pacing the changes in cardiac output were not significant. Similarly, a comparison between the hemodynamic results of atrial and ventricular pacing, at the same heart rate, yielded a different pattern of response between the aortic and the mitral groups. In the aortic group, during atrial pacing, there was a higher cardiac index, mean aortic pressure, and left ventricular minute work and a lower mean left atrial pressure than during ventricular pacing. Similar results have been reported previously.10 In the mitral group, however, the only difference in response to atrial pacing was a higher mean aortic pressure. The improvement in cardiac output in the present series was associated with an optimum ventricular rate, beyond which there was either a reduction or no further change. These results are similar to other reports6'9 and have been attributed to the limitation of diastolic time available for ventricular filling and myocardial blood flow.14' 15 The present work was not designed to investigate the mechanism of changes in the hemodynamic parameters, and the explanation of the difference in results between groups of patients according to which valve has been operated upon remains conjectural. It is unlikely that the type of surgical approach in a given valve lesion could have significantly contributed to the difference in results and none of the patients in this series was suspected to have malfunction of the valve substitute. The clinical data attending each group of valve surgery, however, were different and might be significant in determining the type of response to heart rate augmentation. In the aortic group, the hemodynamic improvement, including reduction of left
Volume 71 Number 4 April, 1976
Postoperative
atrial pressure during atrial pacing, was probably effected by the enhancement of the contractile performance of the ventricular myocardium 16 ' 17 rather than the Starling mechanism. All these patients had evidence of left ventricular hypertrophy. This explains the difference in hemodynamic response to atrial and ventricular pacing which was confined to this group, as has been shown before in postoperative patients. 7 ' 18 The patients in the mitral group had a more advanced preoperative cardiac disability, which was also reflected by the low cardiac output 1, 3 and a more difficult period which required inotropic support, digitalis therapy, and the maintenance of a high left atrial pressure during the postoperative period. The extent of the influence of each of these regimens cannot be quantitated. It has been shown experimentally, however, that on the one hand, the usual forcefrequency relationship is abolished if the myocardium is operating near the upper limit of its contractile response 19 ' 20 and that, on the other hand, a high left atrial pressure combined with heart rate augmentation results in improvement of the cardiac output. 21 There are two practical implications which emanate from the results of the present study. Patients with aortic valve replacement may benefit hemodynamically, in the early postoperative period, when heart rate augmentation is induced by atrial pacing. In patients with mitral valve surgery, the changes in cardiac output would probably depend on factors other than the influence of heart rate augmentation on cardiac performance. We would like to thank Mrs. Shirley Sharp for help in the preparation of this paper. REFERENCES 1 Rouleau, C. A., Frye, R. L., and Ellis, F. H.: Hemodynamic State After Open Mitral Valve Replacment and Reconstruction, J. THORAC. CARDIOVASC. SURG. 58: 870,
1969.
2 Rastelli, G. C , and Kirklin, J. W.: Hemodynamic State Early After Prosthetic Replacement of Mitral Valve, Circulation 34: 448, 1966. 3 Rastelli, G. C , and Kirklin, J. W.: Hemodynamic State Early After Replacement of Aortic Valve With BallValve Prosthesis, Surgery 61: 873, 1967. 4 Roberts, W. C , Bulkley, B. H., and Morrow, A. G.: Pathologic Anatomy of Cardiac Valve Replacement: A Study of 224 Necropsy Patients, Prog. Cardiovasc. Dis. 15: 539, 1973. 5 Dreifus, L. S., Rabbino, M. D., Watanabe, Y., and Tabesh, E.: Arrhythmias in the Postoperative Period, Am. J. Cardiol. 12:431, 1963. 6 Woodson, R. D., and Starr, A.: Atrial Pacing After
cardiac pacing
525
Mitral Valve Surgery, Arch. Surg. 97: 984, 1968. 7 Friesen, W. G., Woodson, R. D., Ames, A. W., Herr, R. H., Starr, A., and Kassebaum, D. G.: A Hemodynamic Comparison of Atrial and Ventricular Pacing in Postoperative Cardiac Surgical Patients, J. THORAC. CARDIOVASC. SURG. 55: 271,
1968.
8 Litwak, R. S., Kuhn, L. A., Gadboys, H. L., Lukban, S. B., and Sakurai, H.: Support of Myocardial Performance After Open Cardiac Operations by Rate Augmentation, J. THORAC. CARDIOVASC. SURG. 56: 484,
1968.
9 Armstrong, P. W., Gold, H. K., Buckley, M. J., Willerson, J. T., and Sanders, C. A.: Hemodynamic Evaluation of Rate Augmentation Produced by Atrial Pacing and Isoproterenol in the Early Postoperative Phase of Cardiac Valve Surgery, Circulation 44: 649,1971. 10 Wisheart, J. D., Wright, J. E. C , Rosenfeldt, F. L., and Ross, J. K.: Atrial and Ventricular Pacing After Open Heart Surgery, Thorax 28: 9, 1973. 11 Wooler, G. H., Nixon, P. G. F., Grimshaw, V. A., and Watson, D. A.: Experiences With the Repair of the Mitral Valve in Mitral Incompetence, Thorax 27: 49,1962. 12 Ionescu, M. I., Pakrashi, B. C , Mary, D. A. S., Bartek, I. T., and Wooler, G. H.: Replacement of Heart Valves With Frame-Mounted Tissue Grafts, Thorax 29: 56, 1974. 13 Stenson, R., Crouse, L., and Harrison, D. C : Computer Measurement of Cardiac Output by Dye Dilution: Comparison of Computer, Fick, and Dowe Techniques, Cardiovasc. Res. 6: 449, 1972. 14 Benchimol, A., and Liggett, M. S.: Cardiac Hemodynamics During Stimulation of the Right Atrium, Right Ventricle, and Left Ventricle in Normal and Abnormal Hearts, Circulation 33: 933, 1966. 15 Knoebel, S. B., Elliott, W. C , Ross, E., and McHenry, P. L.: The Effect of Cardioacceleration by Right Atrial Pacing on Myocardial Blood Flow in Normal Human Subjects, Cardiovasc. Res. 4: 306, 1970. 16 Conell, J. W., Ross, J., Jr., Taylor, R., Sonnenblick, E. H., and Braunwald, E.: Effects of Increasing Frequency of Contraction on the Force Velocity Relation of Left Ventricle, Cardiovasc. Res. 1: 2, 1967. 17 Mitchell, J. H., Wallace, A. G., and Skinner, N. S. Jr.: Intrinsic Effects of Heart Rate on Left Ventricular Performance, Am. J. Physiol. 205: 41, 1963. 18 Braunwald, E.: The Hemodynamic Significance of Atrial Systole, Am. J. Med. 37: 665, 1964. 19 Hajdu, S., and Szent-Gyorgyi, A.: Action of Digitalis Glucosides on Isolated Frog Heart, Am. J. Physiol. 168: 171, 1952. 20 Tuttle, R. S., and Farah, A.: Effect of Ouabain on the Frequency-Force Relation and on Post-Stimulation Potentiation in Isolated Atrial and Ventricular Muscle, J. Pharmacol. Exp. Ther. 135: 142, 1962. 21 Sugimoto, T., Sagawa, K., and Guyton, A. C : Effect of Tachycardia on Cardiac Output During Normal and Increase Venous Return, Am. J. Physiol. 211: 288, 1966.
David A. S. Mary, M.B., Ch.B., M.R.C.P., Brojesh C. Pakrashi, M.B.B.S., M.R.C.P., and Marian I. Ionescu, M.D., F.A.C.S., Leeds, England
*L/ow cardiac output is one of the early complications after intracardiac valve operations and especially after mitral valve surgery.1"3 It appears to be related to the occurrence of subendocardial necrosis and it is responsible for the majority of deaths in the early postoperative period.4 Electrical pacing of the heart has become established as a technique in the treatment of bradyarrhythmias and ectopic rhythms.5' 6 It has been shown to increase the cardiac output in patients with atrioventricular conduction disturbances following open-heart operations.7' 8 Information concerning the influence of heart rate augmentation on cardiac output early after mitral and aortic valve surgery, in the absence of abnormal atrioventricular conduction or other correctable factors, is limited and variable. 6,7 ' 9 ' 10 The results of atrial and From the Department of Cardiothoracic Surgery, The General Infirmary, Leeds, England. Received for publication July 10, 1975. Address for reprints: Mr. M. I. Ionescu, Department of Cardiothoracic Surgery, The General Infirmary, Leeds, 1, Great Britain. 520
ventricular pacing following isolated aortic valve replacement10 and the over-all results in patients with aortic, mitral, and combined valve surgery with atrial9 pacing have been reported. A comparison of hemodynamic consequences of heart rate augmentation by atrial or ventricular pacing between patients with isolated aortic and those with isolated mitral valve surgery is lacking. The published reports on the effect of atrial and ventricular pacing at the same heart rate in patients after heart valve replacement8' 9 do not elucidate the difference in response between these two types of pacing. This study analyzes the influence of heart rate augmentation by atrial and ventricular pacing on the hemodynamic status in stable postoperative patients who had either isolated aortic or mitral valve surgery. The results are compared according to the valve operated upon (aortic or mitral) and the type of pacing used (atrial or ventricular). Materials and methods Twenty-three patients were investigated during the first 48 hours following open intracardiac operations.
Volume 71 Number 4 April, 1976
Postoperative cardiac pacing
52 1
Table I. Preoperative clinical data in 23 patients
Valve lesion
II
III
IV
S.R.
A.F.
L.V.H.
Range
Mean
Patients with digitalis therapy
Aortic Mitral
6 5
4 7
1 —
10 3
1 9
11 3
44-66 49-67
53 58
2 9
N.Y.H.A. grade
Cardiothoracic ratio
ECG findings*
*S.R., Sinus rhythm. A.F., Atrial fibrillation. L.V.H., Left ventricular hypertrophy.
There were 14 men and 9 women and their ages ranged from 24 to 69 (mean 49) years. Twelve patients had mitral valve surgery which comprised replacement with Starr prostheses in 5 patients, stented tissue valves in 3, Bjork prostheses in 1, and mitral annuloplasty in 3 patients. Eleven patients had aortic valve replacement. Stented tissue valves were used in 10 and Bjork prosthesis in 1 patient. The details of the operative procedure have been reported previously.11, 12 Following the completion of the operation, temporary pacemaker electrodes were inserted into the anterior wall of the right ventricle, the atrial wall at the junction with the superior vena cava, and an indifferent one in the abdominal muscle. The electrode wires were brought out through the skin at sites other than the surgical incision. Left atrial and femoral artery catheters were kept in place for postoperative pressure monitoring. In all the patients the pericardial sac was left open at the conclusion of the operation. The preoperative clinical data of the patients are shown in Table I. Hemodynamic studies. Investigations were carried out after a stable clinical status had been attained. This was evaluated by monitoring, among other parameters, the cardiac rhythm, central and peripheral skin temperature, urine output, intracardiac and arterial pressures, blood gases, electrolytes, hemoglobin, packed cell volume, and acid-base status. When applied, infusions and mechanical ventilators were kept at a constant rate throughout the procedure. Positive pressure ventilation was used in 22 patients. Four patients were lying flat and 19 were in a semirecumbant position during the study. Atrial and ventricular pacing was performed with a battery-operated fixed-rate generator. Square waves of 1 msec, duration were utilized. Cardiac output was measured by the indicator dilution techniques. One milligram of indocyanine green was injected rapidly into the left atrium and blood was aspirated from the femoral artery through the catheters inserted at operation. Catheter tubing of the same size was used in every patient. Constant blood withdrawal through a Water's SC-302 densitometer
cuvette was employed. The dye curves were recorded simultaneously with the elctrocardiogram (ECG), left atrial, and systemic artery pressures with an ultraviolet recording system. The dye curves were calibrated at the end of each procedure by three consecutive levels of dye concentration in fresh blood withdrawn from each patient. The Hamilton-Stewart semilogarithmic manual replot extrapolation method was used to calculate the cardiac output. Pressure measurements were performed with the zero reference point set at the level of the sternal angle and were taken as the average of several respiratory cycles. Cardiac index and left ventricular minute work were calculated according to conventional formulas. Cardiac output and pressure measurements were begun at the intrinsic heart rate of the patient and regarded as the data during the first control period. Pacing was then performed to increase the heart rate in two stages, each of about 20 beats per minue (b.p.m.), and further measurements were made after about 2 to 6 minutes from reaching each new heart rate level. The same measurements were repeated when the heart rate was allowed to return to the initial control level. The data during the initial and final control period were averaged and used for comparison with those during cardiac pacing. This sequence of stepwise increase in heart rate was completed in 13 patients. In the remaining 10 patients, a full sequence was not possible because of either a relatively faster intrinsic heart rate or the occurrence during pacing of atrioventricular block and palpitation. Statistical analysis of the differences was carried out with Student's t test for paired data and standard probability tables. Results There was no significant statistical difference between heart rate (P > 0.70), cardiac output (P > 0.40), aortic pressure (P > 0.90), and left atrial pressure (P > 0.50) recorded at the beginning and at the end of the investigation. Reproducibility of results. Cardiac output curves were obtained in duplicate at each stage in every patient and the average was used in the analysis. The
522
The Journal of Thoracic and Cardiovascular Surgery
Mary, Pakrashi, lonescu
Table II. The effect of atrial pacing on hemodynamic parameters in 16 patients Type of operation Aortic valve replacement
Mitral valve surgery
No. of measurements
Heart rate (b.p.m.)
Cardiac index
(L.lmin.lM.2)
Mean aortic pressure (mm. Hg) 86 ± 7 97 ± 9*
Mean left atrial pressure (mm. Hg)
8
79 ± 3 102 ± 2*
3.60 ± 0.47 4.21 ± 0.55*
11
100 ± 2 128 ± 1*
4.47 ± 0.42 4.48 ± 0.40
101 ± 7 98 ± 6
4 ± 1 5 ± 1
8
79 ± 3 128 ± 2*
3.60 ± 0.47 4.16 ± 0.51*
86 ± 7 92 ± 6
5 ± 1 4 ± 1*
4
69 ± 6 101 ± 1*
2.50 ± 0.54 2.83 ± 0.49
95 ± 9 99 ± 10
12 ± 1 12 ± 1
4
101 ± 1 128 ± 3*
2.98 ± 0.35 3.03 ± 0.37
103 ± 10 101 ± 9
14 ± 2 14 ± 2
3
68 ± 9 127 ± 4*
2.90 ± 0 . 5 1 3.39 ± 0.05
97 ± 13 102 ± 13
12 ± 1 12 ± 2
5 ± 1 3 ± 1*
indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
difference between the cardiac indices calculated from consecutive measurements was not significant (P > 0.6). The standard error of the mean difference between 31 pairs of consecutive measurements was 0.095 liter per minute per square meter (L./min./M. 2 ). It was slightly larger (0.127 L./min./M. 2 ) in 7 pairs of measurements with cardiac indices below 2.5 L./ min./M. 2 when compared with 0.118 L./min./M. 2 obtained in the remaining 24 estimations with higher cardiac indices, and this is in accord with previous reports.13 Atrial pacing. Heart rate increments by atrial pacing were possible in 16 patients, of whom 11 underwent aortic valve replacement and 5 mitral valve surgery. All these patients were in sinus rhythm during the postoperative period. In patients with aortic valve replacement there was a significant increase in cardiac output of 0.608 L./ min./M. 2 (17 per cent, P < 0.025) and 0.558 L.I min./M. 2 (16 per cent, P < 0.05) in two sets of investigations each of 8 pairs of measurements, on increasing the heart rate from 79 ± 3 to 102 ± 2 b.p.m. (29 per cent P < 0.001) and from 79 ± 3 to 128 ± 2 b.p.m. (63 per cent P < 0.001), respectively. There was a corresponding increase of mean aortic pressure by 11 mm. Hg (12 per cent P < 0.01) and 7 mm. Hg (8 per cent P > 0.05), respectively. The mean left atrial pressure decreased by 2.5 mm. Hg (49 per cent) and 1.5 mm. Hg (29 per cent), respectively. These changes reached statistical significance (P < 0.005 and P < 0.05). In patients with mitral valve surgery heart rate
increments from 69 ± 6 to 101 ± 1 b.p.m. (46 per cent P < 0.01) and from 68 ± 9 to 127 ± 4 b.p.m. (91 per cent P < 0.001) resulted in insignificant rises in cardiac indices (P > 0.20 and P > 0.40) and aortic pressures (P > 0.40 and P > 0.50), while the mean left atrial pressure remained nearly unchanged (Table II). In both the aortic and mitral groups, increments in heart rate from 100 to 128 b.p.m. revealed a smaller increase in cardiac index, which suggested an optimum heart rate of about 100 b.p.m. The left ventricular work, estimated only in patients with aortic valve replacement, showed an insignificant increase of 0.615 (14 per cent P > 0.20) and 1.81 Kg./min./M. 2 (26 per cent P > 0.05) with increments of heart rate from 79 ± 3 to 102 ± 2 b.p.m. and from 79 ± 3 to 128 ± 2 b.p.m., respectively. Ventricular pacing. Augmentation of heart rate by ventricular pacing was performed in 18 patients. Seven of these had aortic valve replacement and 11 had mitral valve surgery (valve replacement in 9 and repair in 2). All patients with aortic valve replacement were in sinus rhythm. In the mitral valve surgery group, there were 5 patients in atrial fibrillation, 5 in sinus rhythm, and 1 in nodal rhythm. In patients with aortic valve replacement studies of heart rate increments from 73 ± 2 to 103 ± 1 b.p.m. (41 per cent P < 0.001) and from 73 ± 2 to 131 ± 4 b.p.m. (81 per cent P < 0.001) produced a small reduction in the average of cardiac indices (0.11 L./min./M. 2 , 3 per cent, and 0.023 L./min./M. 2 , 1 per cent respectively) which did not reach statistical significance. The mean left atrial pressure increased in
Volume 71 Number 4 April, 1976
Postoperative cardiac pacing
523
Table III. The effect of ventricular pacing on hemodynamic parameters in 18 patients Type of operation
No. of measurements
Heart rate (b.p.m.)
Cardiac index (L./min./M.2)
Mean aortic pressure (mm. Hg)
Mean left atrial pressure (mm. Hg)
Aortic valve replacement
4
73 ± 2 103+1*
3.63 ± 0.56 3.52 ± 0 . 5 6
90 + 11 91 ± 11
6 ±2 10 ± 2*
7
100 ± 2 129 ± 3*
4.22 ± 0.45 3.84 ± 0.24
101 ± 9 97 ± 7
9 ± 1 11 ± 2
4
73 ± 2 131 ± 4 *
3.63 ± 0.57 3.61 + 0.37
90 ± 11 93 ± 12
6 ± 2 9 ± 3
8
69+4 101 + 1*
2.69 ± 0.30 2.99 ± 0.40
85 ± 5 86 ± 5
11+2 11 ± 1
7
99 ± 2 126 ± 3*
3.31 ± 0.37 3.23 ± 0.42
83 ± 6 84 ± 6
11 ± 2 12 ± 2
8
73 ± 4 127 ± 2*
2.95 + 0.29 3.15 ± 0.40
95 + 9 97 ± 9
12 ± 2 14 ± 2*
Mitral valve surgery
indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
Table IV. A comparison of the effect of atrial and ventricular pacing in the same patients and at the same heart rate Type of operation
No. of measurements
Aortic valve replacement Mitral valve surgery
11 9
Pacing A V A V
Heart rate (b.p.m.) 119 119 104 104
±4 ± 4 ± 9 ± 9
Cardiac index (L./min./M.2) 4.68 3.72 3.04 3.02
± ± ± ±
0.38 0.24* 0.22 0.20
Mean aortic pressure (mm. Hg) 103 95 103 96
± + ± ±
7 6* 6 5*
Mean left atrial pressure (mm. Hg) 4 11 12 12
± ± ± ±
1 1* 1 1
'Indicates a statistically significant change. Hemodynamic data are expressed as the mean ± standard error of the mean.
each, and reached statistical significance (4.8 mm. Hg, 87 per cent P < 0.02) only in the first set of measurements (Table III). The changes in the mean aortic pressure were small. Two sets of heart rate increments, each with 8 measurements, were carried out in patients with mitral valve surgery. Heart rate increases from 69 ± 4 to 101 ± 1 (46 per cent P < 0.001) and from 73 ± 4 to 127 ± 2 (74 per cent P < 0.001) resulted in an insignificant rise in the average cardiac indices (0.301 L./min./M. 2 , 11 per cent P > 0 . 1 , and 0.194 L./ min./M.2, 7 per cent P > 0.40, respectively), mean aortic pressures (P > 0.50, P > 0.50), and mean left atrial pressures (P > 0.50, P > 0.05). These changes were not influenced by the type of the original cardiac rhythm. Increases in heart rate from the intermediate to the maximal stage were associated with a consistent reduction in cardiac index in both aortic and mitral groups (Table III). This result suggested an optimum rise in cardiac index at a ventricular rate of 101 b.p.m.
in patients with mitral valve surgery. In the aortic group a follow-up of cardiac index in the same patients at the three levels of heart rate suggested an optimum heart rate at any level between 69 and 100 b.p.m. when patients were in sinus rhythm. The left ventricular work was estimated only in patients with aortic valve replacement and revealed an insignificant reduction of 0.33 (8 per cent, P > 0.40) and 0.29 Kg./min./M. z (7 per cent, P > 0.40), with increments of heart rate from 73 ± 2 to 103 dt 1 b.p.m. and from 73 ± 2 to 131 ± 4 b.p.m., respectively. Atrial and ventricular pacing. In 10 patients, the hemodynamic data were measured during successive atrial and ventricular pacing at the same rate. A total of 19 studies were performed, 11 in patients with aortic valve replacement and 8 in patients with mitral valve surgery. In the aortic group, pacing at a rate of 119 ± 4 b.p.m. resulted in a higher cardiac index (0.956
524
Mary, Pakrashi, lonescu
L./min./M. 2 , 26 per cent, P < 0.001) and mean aortic pressure (8 mm. Hg, 8 per cent, P < 0.05) and a lower mean left atrial pressure (7 mm. Hg, 62 per cent, P < 0.001) during atrial, as compared to ventricular pacing (Table IV). The corresponding data in the mitral group paced at a rate of 104 ± 9 b.p.m. showed practically similar cardiac indices (P > 0.975) and mean left atrial pressures (P > 0.70) with both atrial and ventricular pacing, whereas the aortic pressure was higher (7 mm. Hg, 7 per cent, P < 0.01) during atrial pacing. In the aortic group, left ventricular work during atrial pacing was higher (2.01 Kg./min./M. 2 , 45 per cent, P < 0.001) than that measured during ventricular pacing. Discussion Published reports have shown that following openheart surgery the cardiac output is low in the immediate postoperative period, especially in patients undergoing mitral valve replacement.1-3 The low cardiac output state has been implicated in the development of myocardial necrosis and is the cause of a significant incidence of postoperative death.4 Atrioventricular conduction disturbances, dysrhythmias, hypoxia, hypokalemia, and altered acid-base status are identifiable factors which may be corrected by electrical pacing and pharmacological agents. Depressed cardiac output may often exist in the absence of—or after correction of—such adverse factors. In the absence of these factors, heart rate augmentation induced by atrial pacing has produced in this series a significant increase in cardiac output in the aortic group of patients and an insignificant one in the mitral group. Similar results may be observed in another report.9 In that study an increase in heart rate by atrial pacing has resulted in a larger average rise in cardiac index in 14 patients following aortic valve replacement, as compared with that in 4 patients with isolated mitral valve surgery. Woodson and Starr6 reported a rise of 43 per cent in the cardiac index with the use of atrial pacing in 15 patients with mitral valve surgery. The intrinsic rhythm in 8 of these patients was nodal and a comparison between these results and ours is, therefore, difficult. A further report on atrial pacing in 13 patients following isolated aortic valve replacement with tissue grafts has shown that at a comparable percentage increase in heart rate (20 to 40 per cent), there was a similar directional change in cardiac output, mean aortic, and left atrial pressures,10 as seen in the present series. The results of the present study suggest that heart rate augmentation by atrial pacing
The Journal of Thoracic and Cardiovascular Surgery
would produce a better hemodynamic improvement in patients following aortic valve replacement than in those with mitral valve surgery. By the use of ventricular pacing in this series, there was an insignificant decrease in the cardiac index in the aortic group and an increase in the mitral group, regardless of the intrinsic cardiac rhythm (sinus mechanism or atrial fibrillation). The mean left atrial pressure rose in each group of patients. Similar results have been shown in 13 patients with aortic valve replacement.10 The results of the present investigation which are supported by findings from other reports, therefore, suggest that early after intracardiac valve operations, heart rate augmentation by pacing induces a different pattern of hemodynamic changes in patients with aortic as compared to mitral valve surgery. The use of atrial pacing has resulted in a better hemodynamic improvement in the aortic group than in the mitral group. With the use of ventricular pacing the changes in cardiac output were not significant. Similarly, a comparison between the hemodynamic results of atrial and ventricular pacing, at the same heart rate, yielded a different pattern of response between the aortic and the mitral groups. In the aortic group, during atrial pacing, there was a higher cardiac index, mean aortic pressure, and left ventricular minute work and a lower mean left atrial pressure than during ventricular pacing. Similar results have been reported previously.10 In the mitral group, however, the only difference in response to atrial pacing was a higher mean aortic pressure. The improvement in cardiac output in the present series was associated with an optimum ventricular rate, beyond which there was either a reduction or no further change. These results are similar to other reports6'9 and have been attributed to the limitation of diastolic time available for ventricular filling and myocardial blood flow.14' 15 The present work was not designed to investigate the mechanism of changes in the hemodynamic parameters, and the explanation of the difference in results between groups of patients according to which valve has been operated upon remains conjectural. It is unlikely that the type of surgical approach in a given valve lesion could have significantly contributed to the difference in results and none of the patients in this series was suspected to have malfunction of the valve substitute. The clinical data attending each group of valve surgery, however, were different and might be significant in determining the type of response to heart rate augmentation. In the aortic group, the hemodynamic improvement, including reduction of left
Volume 71 Number 4 April, 1976
Postoperative
atrial pressure during atrial pacing, was probably effected by the enhancement of the contractile performance of the ventricular myocardium 16 ' 17 rather than the Starling mechanism. All these patients had evidence of left ventricular hypertrophy. This explains the difference in hemodynamic response to atrial and ventricular pacing which was confined to this group, as has been shown before in postoperative patients. 7 ' 18 The patients in the mitral group had a more advanced preoperative cardiac disability, which was also reflected by the low cardiac output 1, 3 and a more difficult period which required inotropic support, digitalis therapy, and the maintenance of a high left atrial pressure during the postoperative period. The extent of the influence of each of these regimens cannot be quantitated. It has been shown experimentally, however, that on the one hand, the usual forcefrequency relationship is abolished if the myocardium is operating near the upper limit of its contractile response 19 ' 20 and that, on the other hand, a high left atrial pressure combined with heart rate augmentation results in improvement of the cardiac output. 21 There are two practical implications which emanate from the results of the present study. Patients with aortic valve replacement may benefit hemodynamically, in the early postoperative period, when heart rate augmentation is induced by atrial pacing. In patients with mitral valve surgery, the changes in cardiac output would probably depend on factors other than the influence of heart rate augmentation on cardiac performance. We would like to thank Mrs. Shirley Sharp for help in the preparation of this paper. REFERENCES 1 Rouleau, C. A., Frye, R. L., and Ellis, F. H.: Hemodynamic State After Open Mitral Valve Replacment and Reconstruction, J. THORAC. CARDIOVASC. SURG. 58: 870,
1969.
2 Rastelli, G. C , and Kirklin, J. W.: Hemodynamic State Early After Prosthetic Replacement of Mitral Valve, Circulation 34: 448, 1966. 3 Rastelli, G. C , and Kirklin, J. W.: Hemodynamic State Early After Replacement of Aortic Valve With BallValve Prosthesis, Surgery 61: 873, 1967. 4 Roberts, W. C , Bulkley, B. H., and Morrow, A. G.: Pathologic Anatomy of Cardiac Valve Replacement: A Study of 224 Necropsy Patients, Prog. Cardiovasc. Dis. 15: 539, 1973. 5 Dreifus, L. S., Rabbino, M. D., Watanabe, Y., and Tabesh, E.: Arrhythmias in the Postoperative Period, Am. J. Cardiol. 12:431, 1963. 6 Woodson, R. D., and Starr, A.: Atrial Pacing After
cardiac pacing
525
Mitral Valve Surgery, Arch. Surg. 97: 984, 1968. 7 Friesen, W. G., Woodson, R. D., Ames, A. W., Herr, R. H., Starr, A., and Kassebaum, D. G.: A Hemodynamic Comparison of Atrial and Ventricular Pacing in Postoperative Cardiac Surgical Patients, J. THORAC. CARDIOVASC. SURG. 55: 271,
1968.
8 Litwak, R. S., Kuhn, L. A., Gadboys, H. L., Lukban, S. B., and Sakurai, H.: Support of Myocardial Performance After Open Cardiac Operations by Rate Augmentation, J. THORAC. CARDIOVASC. SURG. 56: 484,
1968.
9 Armstrong, P. W., Gold, H. K., Buckley, M. J., Willerson, J. T., and Sanders, C. A.: Hemodynamic Evaluation of Rate Augmentation Produced by Atrial Pacing and Isoproterenol in the Early Postoperative Phase of Cardiac Valve Surgery, Circulation 44: 649,1971. 10 Wisheart, J. D., Wright, J. E. C , Rosenfeldt, F. L., and Ross, J. K.: Atrial and Ventricular Pacing After Open Heart Surgery, Thorax 28: 9, 1973. 11 Wooler, G. H., Nixon, P. G. F., Grimshaw, V. A., and Watson, D. A.: Experiences With the Repair of the Mitral Valve in Mitral Incompetence, Thorax 27: 49,1962. 12 Ionescu, M. I., Pakrashi, B. C , Mary, D. A. S., Bartek, I. T., and Wooler, G. H.: Replacement of Heart Valves With Frame-Mounted Tissue Grafts, Thorax 29: 56, 1974. 13 Stenson, R., Crouse, L., and Harrison, D. C : Computer Measurement of Cardiac Output by Dye Dilution: Comparison of Computer, Fick, and Dowe Techniques, Cardiovasc. Res. 6: 449, 1972. 14 Benchimol, A., and Liggett, M. S.: Cardiac Hemodynamics During Stimulation of the Right Atrium, Right Ventricle, and Left Ventricle in Normal and Abnormal Hearts, Circulation 33: 933, 1966. 15 Knoebel, S. B., Elliott, W. C , Ross, E., and McHenry, P. L.: The Effect of Cardioacceleration by Right Atrial Pacing on Myocardial Blood Flow in Normal Human Subjects, Cardiovasc. Res. 4: 306, 1970. 16 Conell, J. W., Ross, J., Jr., Taylor, R., Sonnenblick, E. H., and Braunwald, E.: Effects of Increasing Frequency of Contraction on the Force Velocity Relation of Left Ventricle, Cardiovasc. Res. 1: 2, 1967. 17 Mitchell, J. H., Wallace, A. G., and Skinner, N. S. Jr.: Intrinsic Effects of Heart Rate on Left Ventricular Performance, Am. J. Physiol. 205: 41, 1963. 18 Braunwald, E.: The Hemodynamic Significance of Atrial Systole, Am. J. Med. 37: 665, 1964. 19 Hajdu, S., and Szent-Gyorgyi, A.: Action of Digitalis Glucosides on Isolated Frog Heart, Am. J. Physiol. 168: 171, 1952. 20 Tuttle, R. S., and Farah, A.: Effect of Ouabain on the Frequency-Force Relation and on Post-Stimulation Potentiation in Isolated Atrial and Ventricular Muscle, J. Pharmacol. Exp. Ther. 135: 142, 1962. 21 Sugimoto, T., Sagawa, K., and Guyton, A. C : Effect of Tachycardia on Cardiac Output During Normal and Increase Venous Return, Am. J. Physiol. 211: 288, 1966.