Assessment of the isolated right atrium as a pump

Assessment of the isolated right atrium as a pump The ability of the hemodynamically isolated dog right atrium to pump against a resistance equivalent to normal pulmonary artery pressure was tested in an in vivo preparation. At a preload of 10 mm. Hg, the right atrium reached a peak systolic pressure of 18.2 mm. Hg (±1.3 S.E.M.) against an afterload of 12 mm. Hg, but it could eject a flow only equivalent to 53 percent (±6) of the dog's cardiac output (n = 6). At a preload of 15 mm. Hg, a peak systolic pressure of 21.5 mm. Hg (±0.8) was recorded against an afterload of 17 mm. Hg, but again, only 55 percent (±5) of the dog's cardiac output was ejected. It appears that the cardiac output from the nonhypertrophied right atrium is limited by the strength of the right at rial contraction and the physiological limits imposed by the venous driving pressure.

David A. Murphy, M . D . , Alan E. Marble, Ph.D., Rod Landymore, M.D., and Himmet Dajee, M . D . , Halifax, Nova Scotia, Canada With the technical assistance of Richard Leadon, Paul Pike, and Ralph Ricketts

X erformance of the right atrium as a pump is influenced by four variables: atrial contraction, venous return, venous pressure, and output resistance. The first three have been extensively studied. 1 The fourth variable, outflow resistance, particularly the capacity of the right atrium to pump against increasing outflow resistance, has not been examined. For convenience, we will refer to right atrial inflow pressures as preload and the outflow resistance as afterload. The purpose of the following study was to isolate the right atrium hemodynamically and, by applying the principles of the Frank-Starling evaluation of myocardial function, to study right atrial performance by varying inflow rate and pressure at varying outflow resistances. Materials and methods Mongrel dogs weighing 20 to 29 kilograms were premedicated with Innovar-Vet* (fentanyl and droperidine) and 0.4 mg. of atropine. Anesthesia was induced with 15 mg. of Pentothal sodium per kilogram, given intravenously, and maintained by continuous inFrom the Maritime Heart Center, Division of Cardiovascular Surgery, Dalhousie University, Halifax, Nova Scotia, Canada. Supported by the Nova Scotia Heart Foundation. Address for reprints: David A. Murphy, M.D., 5850 University Ave., Halifax, Nova Scotia, Canada B3J 3G9. Received for publication March 14, 1978. Accepted for publication July 13, 1978. *Pittman-Moore Co., Englewood, N. J.

fusion. Body temperature was maintained with a warming blanket. Mechanical volume ventilation was used until full cardiopulmonary bypass was established. Arterial pressure was monitored with a large-bore catheter inserted into the right femoral artery and was recorded with a Statham P23Db transducer and Statham SP1400 blood pressure monitor. This pressure signal was applied to a Model 8 801A DC amplifier and a Hewlett-Packard Model 7858B, eight-channel recorder. The right side of the chest was entered by excision of the fourth and fifth ribs. Cardiac output was measured three separate times by injecting 10 ml. of cool saline into the left atrium and recording output with a Kimray* thermodilution cardiac output computer. The thermistor previously had been inserted into the ascending aorta via the right carotid artery. The mean of these three values was then used to compare with the cardiac output of the isolated right atrium. Cardiopulmonary bypass at normothermia (38.5° C.) was initiated with a pediatric Bentley disposable oxygenator, primed with equal volumes of acid-citratedextrose reconstituted unmatched canine blood and Ringer's lactate solution. The arterial cannula was placed through the left femoral artery, and the venous cannulas were inserted directly into the superior vena cava (SVC) and through a purse-string suture on the inferior vena cava (IVC); the latter was approached through a thoracoabdominal incision in the tenth intercostal space. The SVC distal to the azygos vein was *Kimray Medical Associates, Oklahoma City, Okla.

0022-5223/78/100483+06$00.60/0 © 1978 The C. V. Mosby Co.

483

484

The Journal of Thoracic and Cardiovascular Surgery

Murphy et al.

PRELOAD • RESERVOIR

TRANSDUCER

Fig. 1. In vivo circuit used to test right atrial function. S.V.C., Superior vena cava. I.V.C., Inferior vena cay&.R.A., Right atrium. T.V., Partitioned tricuspid valve. ligated, and the IVC was ligated at the entry of this vessel through the diaphragm (Fig. 1). The mean aortic perfusion pressure was maintained at its prebypass level by the addition of whole blood or electrolyte solution, and the hematocrit value was not allowed to fall below 25 percent. The heart was fibrillated and the right ventricle was then opened in a curvilinear fashion adjacent to the interventricular septum. The chordae were transected, and the tricuspid orifice was then closed by suturing an oval plastic disc (2.5 by 1.5 cm.), with a fringe of thick felt, to the tricuspid anulus with a continuous 4-0 Prolene stitch. The integrity of this closure was tested at the conclusion of the experiment, and the animal was included in the study only if closure was complete. The tip of the auricle was opened prior to lowering the disc into place to prevent atrial distention. The right ventricle was closed and the heart defibrillated. The heart was then allowed to beat spontaneously or was atrially paced at 150 beats per minute. The range of the atrial rate did vary from 150 to 170 among animals but was uniform during a particular experiment. The single inflow into the right atrium was through a short segment (12 inches) of Vi inch Tygon tubing inserted into the IVC remnant. This tubing was connected to a blood reservoir, which was filled with warm blood from the main cardiopulmonary bypass reservoir by means of a calibrated roller pump. Before right atrial performance was tested, the amount of coronary sinus return was directly measured from the right atrial conduit for 1 minute. Flow was then increased into the stationary reservoir, which was at a selected level above the right

atrium, until blood backup indicated that the selected preload had been exceeded. This flow was then recorded, added to the measured coronary sinus flow, and compared to the previously measured base-line cardiac output of the dog. The value was expressed as a percentage. The blood reservoir was set at two levels above the right atrium, representing 10 and 15 mm. Hg. The former value was chosen as the maximum mean preload acceptable under normal hemodynamic conditions. The latter preload (15 mm. Hg) is an approximate value observed in patients with tricuspid atresia after the Fontan-Baudet2 procedure. 3,4 The pressure gradients across the whole circuit were assessed with a cadaveric heart and a preload of 10 mm. Hg, with the conduit at the level of the IVC. The gradients were assessed by use of whole blood at 38.5° C. and were recorded by pullback pressure tracings with a Millar catheter-tip transducer. The conduit was composed of a 14 mm. Lillehei-Kaster valve in a 16 mm. Dacron conduit.* A 2 mm. Hg gradient at flow rates equivalent to the animal's cardiac output allowed observations to be made of flow potential and peak right atrial pressures for a given afterload. For the measurement of right atrial pressures, the animal was rotated on its long axis, so that the right atrial conduit was at the same level as the medial skin incision and the IVC. Right atrial pressure and rate of pressure development were measured with a Millar Microtip (PC 350-F5) transducer catheter (24 KHz) and a derivative computer (Hewlett-Packard Model 8811 A). The transducer was inserted into the right atrium via the ligated azygos vein and positioned manually at the physiological reference point of Guyton and associates.5 Results Atrial function was assessed at a preload of 10 mm. Hg, with afterloads of 0 to 2, 5 to 7, and 10 to 12 mm. Hg (Table I, Fig. 2). The preload was then raised to 15 mm. Hg, and function was assessed at afterloads of 0 to 2, 5 to 7, 10 to 12, and 15 to 17 mm. Hg (Table I, Fig. 3). Predictably, with a 0 to 2 mm. Hg afterload and a preload of 10 mm. Hg, the right atrium added a pulse waveform to a flow equivalent to the base-line cardiac output, or even greater than base-line. At an afterload of 5 to 7 mm. Hg, the right atrial output was 89 ± 10 percent of base-line cardiac output and at 10 to 12 mm. Hg afterload, it was 53 ± 6 percent. The peak right atrial pressure was 18.2 ± 1.3 mm. Hg. This pressure *Earlier experiments using a 16 mm. porcine prosthesis developed pressure gradients of up to 7 mm. Hg across the valve and thus were not useful in these experiments.

Volume 76 Number 4 October, 1978

485

Assessment of isolated right atrium as a pump

after toad 15 mm Hg + valve

w 12-

4H—*—*-*-*" 0.4

Q8

j ~*

T

12 1.6 2D VENOUS INFLOW ( L min )

'

| afterload " valve alone

2.4

n

4 —}

Fig. 2. Pressure-flow curves for a preload of 10 mm. Hg at three afterloads values (n = 6). R.A., Right atrial. represents a systolic peak of 8.2 mm. Hg greater than the input pressure. In the same six experiments, preload was increased to 15 mm. Hg and afterload to 17 mm. Hg (15 ± 2 mm. Hg valve gradient). Again, the output of the right atrium with no afterload was 100 percent of the dog's cardiac output. With an afterload of 5 to 7, it decreased to 94 percent. At 12 and 17 mm. Hg, the output was 86 ± 15 percent and 55 ± 5 percent, respectively. The peak right atrial pressure achieved was 21.5 ± 0.8 mm. Hg at an afterload of 17 mm. Hg. This peak pressure is only 6.5 mm. Hg greater than input pressure. The peak rate of rise of pressure (dp/dt), in millimeters of mercury per second, for each of the previous two afterloads is listed in Table I, and typical curves are illustrated in Fig. 4. The fact that the standard errors of the mean for preloads of 10 and 15 mm. Hg overlap at high afterloads indicates considerable variation in this measurement from animal to animal. This variation could be due to differences in heart rate or differences in the experimental preparation. Two interesting observations were made during these experiments. Small defects in the integrity of the right atrial chamber greatly influenced peak right atrial pressure. In three experiments, a defect of right atrial partitioning was found (two atrial septal defects, one leak in tricuspid valve closure). The effect on right atrial pressure and dp/dt are illustrated in Figs. 5 and 6. In Fig. 5, the usual increment in right atrial pressure and dp/dt with an increase in right atrial inflow is demonstrated. In Fig. 6, both the right atrial pressure and dp/dt are decreased because of a septal defect. The implications of this phenomenon will be discussed. The influence of ventricular movement on the right atrial curves was recorded twice in these experiments by allowing the atrium and ventricles to beat synchro-

08

12 VENOUS

1B INFLOW

} afterload valve alone 2.8

20 2.4 (L/min)

Fig. 3. Pressure-flow curves for a preload of 15 mm. Hg at four afterloads values (n = 6). R.A., Right atrial.

i | 150

0.4

0.8

1.2 1.6 20 2.4 VENOUS INFLOW (L/min)

2.8

a2

Fig. 4. Rate of pressure development (dp/dt) plotted against venous inflow at afterloads of 0 to 2, 5 to 7, and 15 to 17 mm. Hg with a 15 mm. Hg preload (n = 6). nously; then, without changing the preload or afterload, the ventricles were fibrillated while the atrium continued to beat (Fig. 6). The reason for the increase in right atrial pressure and dp/dt is unknown. It appears unrelated to the movement of the right tricuspid disc, since the disc could be held firmly to a support or allowed free movement without similar changes occurring. Discussion The purpose of this investigation was to attempt to define the capacity of the nonhypertrophied right atrium to pump blood against resistance equivalent to that normally present in the pulmonary circulation. The surgical successes of Fontan and Baudet demonstrated that the hypertrophied human right atrium in tricuspid atresia was capable of pumping a large portion

The Journal of Thoracic and Cardiovascular Surgery

4 8 6 Murphy et al.

20 RAP (mm Hg)

„, r

10

•Lirwi""! ft

IULJUU

One-Tenth Cardiac Output

Total Cardiac Output

CARDIAC OUTPUT = 3.02 Liters/min PRELOAD = 15 mm Hg AFTERLOAD = 5 mm Hg

DOG WT.= 23Kg HEART RATE=T76

Fig. 5. Typical tracings of right atrial pressure (RAP) and dp/dt in a normal animal at one-tenth (left) and full cardiac output (right). Note typical increments and compare with animal with atrial septal defect (Fig. 6). Table I. Right atrial performance at preloads of 10 and 15 mm. Hg with afterloads of 12 and 17 mm. Hg and with low-profile disc-valve conduits Preload 15, afterload 17

Preload 10, afterload 12 RACO

%CO

Peak RAP

Peak dp/dt

RACO

%CO

Peak RAP

Peak dp/dt

2.90 3.02 2.37 1.78 3.52 2.35

0.94 1.38 1.34 1.14 1.63 1.74

32 46 57 64 46 74

17 18 22 13 21 17

190 285 390 150 280 280

1.12 1.88 1.73 0.98 1.52 1.51

39 62 73 55 43 64

20 19 25 21 22 22

210 175 375 190 275 333

2.66 ±0.61 0.25

1.36 ±0.30 0.12

18.2 ±3.2 1.3

262 ±84 34.3

1.46 ±0.35 0.14

55 ±13 5.3

21.5 ±2.1 0.85

260 ±81 33

Exp.

Dog Wt. (Kg.)

CO

1 2 3 4 5 6

23 23 23 20 29 23

Mean S.D. S.E.M.

23.5 ±2.9 1.2

53 ±14.9 6.1

Legend: CO, Cardiac output (L./min.). RACO, Right atrial cardiac output (L./min.). Peak RAP, Peak right atrial pressure (mm. Hg). dp/dt, Rate of pressure development (mm. Hg/sec).

of the right-sided cardiac output directly into the pulmonary artery. Subsequent reports 6-8 demonstrated that the right atrium could pump all of the right-sided cardiac output. The patients treated by the FontanBaudet procedure have all been older children. The capacity of an infant right atrium to perform this work is less well known. Several attempts to perform a repair of the Fontan-Baudet type in very young subjects have been made (right atrium to pulmonary artery and right atrial partitioning operation, described by Lamberti7) but, with the exception of Lamberti's case, all have

been unsuccessful owing to acute right atrial dilation.9-12 Puga and McGoon,13 in dog experiments, partitioned the tricuspid valve and attached the right auricle to the pulmonary artery by means of a large nonvalved conduit. They found that the cardiac output decreased to 60 percent of the base-line values, and all animals died within 24 hours. These experiments were designed to evaluate the possibility of bypassing the right ventricle briefly, with a view of using this technique to repair certain congenital heart conditions.

Volume 76 Number 4 October, 1978

Assessment of isolated right atrium as a pump

4 87

■l/UUl (mm Hg/sec)

Inflow = 710 ml/min

Inflow =1275 ml/min

AFTERLOAD=5mmHg PRELOAD = 10 mm Hg

Fig. 6. Tracings of right atrial pressure (RAP) and dp/dt in an animal later found to have a small atrial septal defect. Note that there is no increment in these pressures following the increase in venous inflow to therightatrial chamber. Dog weight 24 kilograms. Heart rate 167 beats per minute. 20 RAP 10 (mmHg) 0L 200

ILAAA. LOAD J

I sec n

PRELOAD=15 mmHg AFTERLOAD = 15mmHg

Fig. 7. An unexpected finding was an increase inrightatrial pressure (RAP) and dp/dt when the ventricles were fibrillated while the atria were allowed to beat. EKG, Electrocardiogram. Dog weight 23 kilograms. Heart rate 176 beats per minute. A considerable amount of information is available concerning the influence of right atrial pressure on venous return. Guyton and associates1 developed a partial right atrial bypass circuit wherein the pressure in the right atrium could be precisely controlled. Small rises in right atrial pressure had a profound effect on venous return to the right side of the heart. Guyton's group demonstrated in this circuit that all venous return to the right heart ceases when the right atrial pressure is raised to 8 mm. Hg in areflexic animals. This information has importance in understanding the late sequelae of venous hypertension seen following the Fontan-Baudet procedure but not in understanding the pumping capacity of the right atrium.

A preload of 10 mm. Hg was chosen in these experiments despite Guyton's observations, since this figure is one which would be considered a high normal right atrial pressure in the human subject. With this preload, the right atrium could reach a peak systolic pressure of 18.2 mm. Hg, but it was capable of pumping only 53 percent of the cardiac output at this afterload. There is a problem of expressing the output as a percentage, since the prebypass cardiac output is used for comparison. A more reliable method might have been to use the index figures of Wiggers.14 However, both his and other reports15' 16 of cardiac output (expressed as liters per minute) for animals of equivalent size compare well with ours. Nonetheless, a high or low value for the

The Journal of Thoracic and Cardiovascular Surgery

4 8 8 Murphy et al.

measured base-line cardiac output would tend to influence the calculated percentage right atrial output recorded. At a preload of 15 mm. Hg, the right atrium was capable of pumping 55 ± 5 percent of the cardiac output and of reaching a peak systolic pressure equivalent to the pulmonary artery pressure of 21.5 ± 0.8 mm. Hg. At this pressure, however, little fiber shortening was observed. Fiber shortening has been measured with sonomicrometers 17 placed on the left atrium of dogs during volume loading. As pressure and diameter are raised initially, atrial systolic shortening increases; however, with a left atrial pressure of 10 to 27 mm. Hg, fiber shortening decreases. These findings indicate that the nonhypertrophied atria has a rather narrow range of pressure and flow capability before it loses its contractile properties. Our experiments did not specifically examine fiber shortening of the right atrium, although others have. 1 8 ' 19 Objections could be raised to comparing flow capacity from animal to animal in cases in which the atrial rates were not identical. We observed no significant change in either peak right atrial pressure or dp/dt in a particular experiment, whether the atrial pacing rate was 150 or 180 beats per minute. Williams and associates 19 while studying atrial contractile force, also did not observe a change in dp/dt at rates between 150 and 240 beats per minute. In those experiments in which an atrial septal defect was subsequently identified, the magnitude of the right-to-left shunt was probably accentuated by the low left atrial pressures, the latter owing to the right heart bypass circuit. To what extent shunting might occur if small defects were inadvertently left during the Fontan-Baudet procedure is not known. In these experiments, however, it greatly influenced the performance of the right atrium. With the expected rise in right atrial pressure following the Fontan-Baudet procedure, shunting could be surgically significant. We do not understand why ventricular fibrillation should enhance the systolic pressure and dp/dt in the beating right atrium. The plastic diaphragm separating the right atrium from the right ventricle was attached to a metal rod brought through the right ventricular closure, and no change in right atrial pressures was noted whether the tricuspid disc was kept immobile or allowed movement with beating. Pressure does not appear to be related to disc movement. Coronary flow (not measured) may vary with ventricular fibrillation and regular atrial beating, and such changes might explain this phenomenon.

We gratefully acknowledge the assistance of Dr. James Love, Dr. G. Klassen, Ann Fenton, Melanie Briand, and the Audiovisual Department, Dalhousie Medical School. REFERENCES 1 Guyton AC, Lindsay AW, Abernathy JB, Richardson T: Venous return at various right atrial pressures and the normal venous return curve. Am J Physiol 189:609, 1957 2 Fontan F, Baudet E: Surgical repair of tricuspid atresia. Thorax 26:240, 1971 3 Serratto M, Miller RA, Tatooles C, Ardekani R: Hemodynamic evaluation of Fontan operation in tricuspid atresia. Circulation 54:Suppl 3:99, 1976 4 Walker DR, Sbokos CG, Lennox SC: Correction of tricuspid atresia. Br Heart J 37:282, 1975 5 Guyton AC, Jones CE, Coleman TG: Circulatory Physiology, ed. 2, Philadelphia and Toronto, 1973, W. B. Saunders Company, p 190 6 Tatooles CJ, Ardekani RG, Miller RA, Seratto M: Results following physiological repair for tricuspid atresia. Ann Thorac Surg 22:578, 1976 7 Lamberti JJ, Thilenius O, de la Fuente D, et al: Right atrial partition and right ventricular exclusion. J THORAC CARDIOVASC SURG 71:386, 1976

8 Henry JN, Devloo RAE, Ritten DG, et al: Tricuspid atresia. Mayo Clin Proc 49:803, 1974 9 Murphy DA: Unreported case 10 Dobell ARC: Unreported case 11 Sade RM, Castaneda AR: The dispensable right ventricle. Surgery 77:624, 1975 12 Hurwitt ES, Young D, Escher DJW: The rationale of anastomosis of the right auricular appendage to the pulmonary artery in the treatment of tricuspid atresia. J THORAC SURG 30:503, 1955

13 Puga FJ, McGoon DC: Exclusion of the right ventricle from the circulation. Surgery 73:607, 1973 14 Wiggers HC: Cardiac output and total peripheral resistance measurements in experimental dogs. Am J Physiol 140:519, 1944 15 Altman PL, Dittmer DS: Respiration and Circulation, Bethesda, Md., 1971, Federation of American Societies for Experimental Biology, pp 328, 329 16 Kontos HA, Mauck HP, Richardson DW, Petterson JL: Mechanism of circulatory responses to systemic hypoxia in anesthesized dog. Am J Physiol 209:397, 1965 17 Payne RM, Stove HL, Engelken EJ: Atrial function during volume loading. J Appl Physiol 31:326, 1971 18 Eliahou HE, Clarke SD, Bull GM: Atrial pulsation during acute distention and its possible significance in the regulation of blood volume. Clin Sci 19:377, 1960 19 Williams JF, Sonnenblick EH, Braunwald E: Determinants of atrial contractile force in the intact heart. Am J Physiol 209:1061, 1965