Pulsatile Perfusion: Its Effects on Blood Flow Distribution in Hypertrophied Hearts Tomas A. Salerno, M.D., Harry M. Shizgal, M.D., and Anthony R. C. Dobell, M.D.

ABSTRACT Left ventricular hypertrophy was created in 15 pigs by banding the ascending aorta when they were young. The adult animals were placed on normothermic cardiopulmonary bypass and perfused with either nonpulsatile (two groups of pigs) or pulsatile (one group) flows. As long as the perfusion rate was maintained at 70 mllkglmin, myo- cardial blood flow distribution as determined by radioactive microspheres, was identical in the hearts with normal sinus rhythm and those with ventricular fibrillation irrespective of the type of perfusion. At low flow rates, however, subendocardial ischemia developed in all three groups, but was most severe in the fibrillating hearts, and was not reversed by pulsatile perfusion.

In normal hearts, less than 25% of myocardial perfusion occurs in systole [S ] , and this is di- rected to the subepicardial layer of the ven- tricular myocardium. The deeper layers are perfused in diastole. Overall perfusion is equal through all layers [7] . Thus, autoregulation leads to a greater diastolic flow to the suben- docardial region. This region is particularly vulnerable when there is coronary artery dis- ease, aortic stenosis, aortic insufficiency, or a combination of the last two factors.

Subendocardial ischemia occurs in fibrillat- ing hypertrophied hearts under cardiopulmo- nary bypass (CPB) when extracorporeal flow is marginal [l, 15, 18, 19, 251. It seemed that the combination of nonpulsatile CPB and ven- tricular fibrillation (VF) might completely upset the autoregulatory mechanism since only by such a combination would the coronary ar- terioles be subjected to nonpulsatile coronary perfusion. Nonpulsatile CPB alone does not produce nonpulsatile coronary perfusion with a

From the Departments of Surgery, Royal Victoria Hospital and McGill University, Montreal, Que, Canada. Accepted for publication Sept 14, 1978. Address reprint requests to Dr. Salerno, Department of Surgery, Queen’s University, Kingston, Ont, Canada.

beating heart since coronary resistance is then pulsatile.

We hypothesized that pulsatile perfusion during CPB with VF might restore a more even flow distribution to the various layers of the myocardium. The experiments reported here were designed to test that hypothesis.

Materick and Methods Hypertrophy of the left ventricular myocardium was induced in 15 pigs by banding the as- cending aorta when they were 4 k 0.3 weeks old. The animals were placed on normothermic CPB when they were 17 k 1.8 weeks old, and the distribution of blood flow to the myocar- dium was studied at different flows using four different radioactive carbonized microspheres* (15 f 5 p) labeled with strontium 85, cerium 141, iodine 125, and niobium 95. The animals were divided into three equal groups: Group 1, normal sinus rhythm (NSR) perfused with a Sarns roller pump; Group 2, ventricular fibrilla- tion (VF) perfused with the roller pump; and Group 3, VF perfused with a Bentley pulsatile pump. The pulsatile pump was powered by an inert compressed gas at a pressure of 75 pounds per square inch. The disposable ventricle was made of thin polyurethane with polycarbonate inlet fittings, with one-way valves at each end. Pump action was achieved when a diaphragm squeezed the ventricle inside the pump chamber. Stroke volume ranged from Zero to 100 ml per stroke. Stroke rate could be adjusted from 20 to 120 strokes per minute. The pulsatile ventricle was capable of functioning up to an average perfusion rate of 6 liters per minute. Pump flows were measured with a flowmeter in the arterial cannula. Pressure tracings from heart action, roller pump, and pulsatile pump are shown in Figure 1.

Normothermic cardiopulmonary bypass was

*3M Company, Nuclear Products, St. Paul, MN.

559 0003-49751791060559-05$01.25 @ 1978 by Tomas A. Salemo

560 The Annals of Thoracic Surgery Vol 27 No 6 June 1979

mm Hg

A

80

40 20

mmHg 60

C Fig 1 . (A) Pressure tracings f rom beating hearts before cardiopulmonary bypass, (B) during pulsatile perfusion, and (C) during nonpulsatile perfuszon.

instituted by cannulation of the venae cavae, which were later snared. Arterial inflow was through the right subclavian artery. The pul- monary artery was snared. The azygos and hemiazygos veins, which drain into the superior vena cava and the right atrium, re- spectively, were tied off. Vents were inserted into the left and right ventricles before CPB. Coronary blood flow was measured as the effluent from the right ventricle. Aortic root pressure, blood gases, hematocrit, and intra- myocardial temperature were monitored. Ven- tricular fibrillation, when needed, was induced by a single alternating-current shock pulse.

After the animal had been on CPB for 90 minutes at a flow of 75 mllkglmin, the flow was altered to range between 70 and 20 ml/kg/min, with stabilization for 20 minutes at each level. At that time, a known number of radioactive microspheres labeled with a single isotope were injected into the arterial line of the pump oxygenator. Since these microspheres do not pass through capillaries, their distribution in the myocardium is proportional to regional perfusion [ l , 221. At the end of 180 minutes on CPB, the animal was killed, and differential counting techniques were employed to deter-

mine the concentration of each isotope in the myocardium. The heart was removed from the chest, cleared of fat and pericardium, and weighed.

The heart was then fixed in buffered formal- dehyde for 12 hours and was divided into epicardial, midmyocardial, and endocardial thirds of equal thickness. The tissue was then allowed to dry at 373°C for 48 hours, was weighed, and was counted in a Packard deep- well scintillation counter. Countslmidgm of dry myocardium was determined for each indi- vidual layer and the ratio of activity in the epicardial and endocardial layers was calcu- lated. Blood flow distribution to the individual layers was determined by multiplying the per- cent of counts in each layer by the total mea- sured coronary blood flow.

Results The three groups of animals had a comparable degree of myocardial hypertrophy, with mean heart weight in grams of 457.0 f 8.7 (standard error of the mean) for NSR group (roller pump), 370.0 f 10.2 for VF group (roller), and 384.0 f 9.1 for VF group (pulsatile) compared with 210.0 f 3.9 for normal hearts. The dif- ference between normal and hypertrophied hearts was statistically significant (analysis of variance) ( p < 0.05).

Coronary artery flow is shown in Table 1. The fibrillating hearts had the highest coronary flow at all pump rates. The distribution of coronary flow to the free wall of the left ventri- cle at high pump flows of 70 mllkglmin (Fig 2) was not different in the three groups ( p > 0.05). However, as flows declined from 70 to 20 ml/ kglmin (Fig 3), the ratio of epicardium to en- docardium changed from 1.30 f 0.12 to 3.20 f 0.52 for NSR group (roller); from 1.80 f 0.07 to 7.0 f 1.67 for VF group (roller); and from 1.50 f 0.09 to 10.50 f 1.81 for VF group (pulsatile) ( p < 0.05), indicating underperfusion of the en- docardium compared with the epicardium. The epicardium to endocardium ratios for interme- diate flows are shown in Table 2.

Coronary artery resistance at all flows was higher for NSR group (0.90 k 0.08 resistance units at 70 mUkglmin and 1.40 k 0.09 at 20 ml/ kglmin) compared with fibrillating hearts (0.63

561 Salerno, Shizgal, and Dobell: Pulsatile Perfusion

Table 1 . Coronary Artery Blood Flow (mllmin)”

Pump Flow (mYkg/min)h

Group 70 50 40 20

NSR roller 80.0 f 4.7 62.4 f 3.0 46.4 f 2.1 27.4 f 1.84 84.0 f 7.8 67.0 f 7.3 52.0 f 7.1 VF roller 105.0 f 7.0

VF pulsatile 102.0 f 6.4 81.6 f 6.2 51.8 f 4.0 40.4 f 5.4

“At all levels of pump flows the fibrillating hearts had the highest coronary flows. b+ Standard error of the mean. NSR = normal sinus rhythm; VF = ventricular fibrillation.

DISTRIBUTION OF MYOCARDIAL FLOW (Pump flow 70 cc/Kg/min )

1 OEPI =MID UENOO ISE 60

NSR Roller VF Roller VF Pulsatile

Fig 2. Distribution of blood flow to the myocardium at high flow rates.

DISTRIBUTION OF MYOCARDIAL FLOW ( Pump flow 20cc/Kg/min )

60T ~ E P I UMIR NENDO I SE

NSR Roller VF Roller VF Pulsotile

Fig 3 . Distribution of blood flow to the myocardium at low flow rates.

f 0.08 and 0.54 f 0.06 for VF group [roller], and 0.43 f 0.03 and 0.84 f 0.08 for VF group [pul- satile]) ( p < 0.05).

Peripheral vascular resistance had values of 1.02 f 0.06 at 70 ml/kg/min and 1.91 f 0.1 at 20 ml/kg/min for NSR group, compared with 0.95 k 0.09 and 1.4 f 0.06 for VF group (roller) and 0.62 f 0.07 and 1.7 f 0.09 for VF group (pul- satile) ( p < 0.05).

The relationship between pressure and flow rates is shown in Fig 4. The pulsatile groups had the lowest pressures although pump flows were relatively high, indicating that the ani- mals were peripherally dilated.

Table 2. EpicardiallEndocardial Ratio of the Distribution of Blood Flow to the Free Wall of the Left Ventricle

Comment The deleterious effects of ventricular fibrillation on the distribution of blood flow away from the subendocardium have been borne out in several animal and human studies 14, 11-14, 211. Myo- cardial hypertrophy is a potentiating factor [4-6, 10, 15,181. Allowing the heart to beat while empty appears to be the most desirable way to protect the perfused subendocardium [12, 231.

Myocardial blood flow occurs primarily in diastole but systolic flow also has been noted

Pump Flow (mYkg/min)a

Group 70 50 40 20

1.30 f 0.12 2.14 f 0.15 2.55 f 0.88 3.22 f 0.52 NSR roller 3.71 f 0.72 4.03 f 0.51 7.0 f 1.67 VF roller 1.80 f 0.07

VF pulsatile 1.50 f 0.09 3.77 f 0.53 8.55 f 1.60 10.50 f 1.81

a+ Standard error of the mean. NSR = normal sinus rhythm; VF = ventricular fibrillation.

562 The Annals of Thoracic Surgery Vol 27 No 6 June 1979

I" - 80 I A NSR VF ROLLER T m I

w 40

0 20

VF SE

u + k i 0 20 40 50 70 a

PUMP FLOW (cc/Kg/min)

Fig 4 . Relationship between perfusion pressure and p u m p f l o w s . (NSR = normal sinus rhythm; VF = ven- tricular fibrillation; SE = standard error.)

and appears to be directed to the epicardial zones [2, 8, 9, 16, 271. Loss of this phasic flow during VF or during nonpulsatile perfusion can be important in causing ischemia to the suben- docardium, especially if there is compensatory vasodilatation or constriction, or both, of the different areas of the myocardium during sys- tole and diastole. If the systolic autoregulation is related to and depends on diastolic au- toregulation, absence of pulsatile flow could cause abnormal regulation of the vascular tone and thus lead to abnormal distribution of blood flow. Because pulsatile perfusion during CPB appears to be more physiological than non- pulsatile perfusion [17, 20, 24, 261, we hypothesized that perfusion of fibrillating hearts with pulsatile flow would restore normal blood flow distribution across the myocardium.

Our results failed to support our hypothesis. At high flow rates (70 mllkglmin), no significant maldistribution was demonstrated despite se- vere left ventricular hypertrophy and VF, irre- spective of the type of CPB pump used. At low flow rates (less than 50 mllkglmin), suben- docardial ischemia was apparent in all three groups and was more profound in the fibrillat- ing hearts. The deleterious effects of ventricular fibrillation were not reversed by pulsatile ex- tracorporeal perfusion and subendocardial ischemia was not prevented by pulsatile flow. It can be postulated that although pulsatile flow was achieved in the coronary circulation, the coronary vessels were constantly being com-

pressed by the fibrillating myocardium, im- peding flow to the deeper areas, as has been shown [3]. This force may be overcome when the perfusion pressure is high. At low flow rates, however, the perfusion pressure whether pulsatile or nonpulsatile is not sufficient to overcome myocardial compression of the coro- nary circulation.

From this study, we conclude that at low flow rates pulsatile perfusion does not restore the abnormal flow distribution observed in fibril- lating hearts. At adequate flow rates of 70 ml/ kglmin, there is no advantage in the use of pul- satile perfusion in terms of blood flow dis- tribution across the myocardium.

References 1. Allard JR, Shizgal HM, Dobell ARC: Distribution

of myocardial blood flow during cardiopulmo- nary bypass in normal and hypertrophied left ventricles. Surg Forum 24:178, 1973

2. Anrep GV, Davis JC, Volhard E: The effect of pulse pressure upon the coronary blood flow. J Physiol (Paris) 73:405, 1931

3. Baird RJ, Goldbach MM, de la Rocha AG: Intra- myocardial pressure: the persistence of its transmural gradient in the empty heart and its relationship to myocardial oxygen consumption. J Thorac Cardiovasc Surg 64635, 1972

4. Becker RM, Shizgal HM, Dobell ARC: Distribu- tion of coronary blood flow during cardiopulmo- nary bypass in pigs: possible implications for left ventricular hemorrhagic necrosis. Ann Thorac Surg 16:228, 1973

5. Buckberg GD, Towers B, Paglia DE, et al: Sub- endocardia1 ischemia after cardiopulmonary by- pass. J Thorac Cardiovasc Surg 64:669, 1972

6. Cooley DA, Reul GJ, Wukasch DC: Ischemic contracture of the heart: "stone heart." Am J Cardiol 29:575, 1972

7. Domenech RJ, Hoffman JIE, Noble MIM, et al: Total and regional coronary blood flow measured by radioactive microspheres in conscious and anesthetized dogs. Circ Res 25:581, 1969

8. Gregg DE, Green HD: Registration and inter- pretation of normal phasic inflow into a left coronary artery by an improved differential manometric method. Am J Physiol 130:114, 1940

9. Gregg DE, Sabiston DC: Current research and problems of the coronary circulation. Circulation 13:916, 1956

10. Horn H, Field LE, Dack S, et al: Acute coronary insufficiency: pathological and physiological as- pects: an analysis of twenty-five cases of suben- docardial necrosis. Am Heart J 40:63, 1950

563 Salemo, Shizgal, and Dobell: Pulsatile Perfusion

11. Hottenrott C, Buckberg G: Studies of the effects of ventricular fibrillation on the adequacy of re- gional myocardial flow: 11. Effects of ventricular distention. J Thorac Cardiovasc Surg 68:626, 1974

12. Hottenrott C, Maloney JV, Buckberg GD: Stud- ies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow: I. Electri- cal vs spontaneous fibrillation. J Thorac Car- diovasc Surg 68:615, 1974

13. Hottenrott C, Maloney JV, Buckberg G: Stud- ies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow: 111. Mech- anisms of ischemia. J Thorac Cardiovasc Surg 68:634, 1974

14. Hottenrott CE, Towers B, Kurkji HJ, et al: The hazard of ventricular fibrillation in hyper- trophied ventricles during cardiopulmonary bypass. J Thorac Cardiovasc Surg 66:742, 1973

15. Huang SN, Masse S: Pathogenesis of hemor- rhagic myocardial necrosis following cardiac surgery. Circulation 41, 42:Suppl 3:III-17, 1970

16. Johnson JR, Di Palma JR: Intramyocardial pres- sure and its relation to aortic blood pressure. Am J Physiol 125:234, 1939

17. Mandelbaum I, Bums WH: Pulsatile and non- pulsatile blood flow. JAMA 191:657, 1965

18. Najafi H, Henson D, Dye WS, et al: Left ven- tricular hemorrhagic necrosis. Ann Thorac Surg 7:550, 1969

19. Najafi H, La1 R, Khalili M, et al: Left ventricular

hemorrhagic necrosis: experimental production and pathogenesis. Ann Thorac Surg 12:400, 1971

20. Nakayama K, Tamiya T, Yamamoto K, et al: High-amplitude pulsatile pump in extracor- poreal circulation with particular reference to hemodynamics. Surgery 54:798, 1963

21. Reis RL, Cohn LH, Morrow AG: Effects of in- duced ventricular fibrillation on ventricular per- formance and cardiac metabolism. Circulation 36:Suppl 1:234, 1967

22. Rudolph AM, Heymann MA: Circulation of the fetus in utero: methods for studying distribution of blood flow, cardiac output and organ blood flow. Circ Res 21:163, 1967

23. Sabiston DC, Gregg DE: Effect of cardiac con- traction on coronary blood flow. Circulation 15:14, 1957

24. Shepard RB, Kirklin JW: Relation of pulsatile flow to oxygen consumption and other variables during cardiopulmonary bypass. J Thorac Car- diovasc Surg 58:694, 1969

25. Taber RE, Morales AR, Fine G: Myocardial ne- crosis and the postoperative low-cardiac-output syndrome. Ann Thorac Surg 4:12, 1967

26. Trinkle JK, Helton NE, Bryant LR, et al: Pulsatile cardiopulmonary bypass: clinical evaluation. Surgery 68:1074, 1970

27. Wiggers CJ, Cotton FS: Studies on the coronary circulation: 11. The systolic and diastolic flow through the coronary vessels. Am J Physiol 106:597, 1933