Comparative studies of pulsatile and nonpulsatile flow during cardiopulmonary bypass I. Pulsatile system employed and its hematologic effects A new, commercially available roller pump system able to deliver pulsatile and nonpulsatile flow has been studied in patients undergoing elective open-heart surgical procedures. The pulsatile pump (Stockert lnstrumente) may be used with standard extracorporeal circuit equipment and consistently produces a peripheral arterial pulse pressure of 25 to 30 mm. Hg at mean flow rates of 3.5 to 4.0 L. per minute. Twenty patients, arbitrarily allocated to pulsatile or nonpulsatile groups, have been studied. There were no significant differences between the groups in respect of age, weight, bypass time, pump flow, or mean arterial pressure during bypass. Comparative studies of the hematologic effects of pulsatile and nonpulsatile perfusion were carried out. There was no evidence of increased hemolysis with pulsatile flow, nor was there increased depletion of red blood cells (RBC s) or platelets in the pulsatile group. This pulsatile pump system may therefore be used to produce pulsatile perfusion during cardiopulmonary bypass without the fear of producing excessive blood cell trauma.
Kenneth M. Taylor, M.B., F.R.C.S., William H. Bain, M.D., F.R.C.S., Kenneth J. Maxted, B.Sc., Margaret M. Hutton, M.B., M.R.C.Path., William Y. McNab, F.I.M.L.S., and Philip K. Caves, M.B., F.R.C.S., Glasgow, Scotland
DesPite numerous advances in cardiac surgical techniques, there have been few changes in perfusion methods since the early days of open-heart surgery-in particular, the use of nonpulsatile flow during extracorporeal circulation (ECC) has continued. However, it has always been thought that such an unphysiological method of perfusion could be associated with disturbances of organ metabolism during the period of ECC. Much of the uncertainty regarding the possible dangers of nonpulsatile flow and the theoretical benefits of the more physiological pulsatile flow may be attributed to From the University Department of Cardiac Surgery, Royal Infirmary, Glasgow, and the Department of Haematology, Victoria Infirmary, Glasgow, Scotland. These studies were carried out with the aid of a grant from the British Heart Foundation. Received for publication Aug. 16, 1977. Accepted for publication Ocl. II, 1977. Address for reprints: Mr. Kenneth M. Taylor, University Department of Cardiac Surgery, Royal Infirmary, Glasgow G40SF, Scotland.
0022-5223/78/0475-0569$00.50/0 © 1978 The C. V. Mosby Co.
two principal factors: (I) the lack of clear objecti ve data on organ metabolism during nonpulsatile ECC and (2) the fear of increased hemolysis from pulsatile flow, as suggested in earlier studies. 1 In previously published studies, we2 - 4 have used the pituitary-adrenal system as an index of the metabolic changes which occur during open-heart surgery. We have shown that the normal stress response patterns become highly significantly disordered during nonpulsatile ECC, with restoration of normal response patterns in the early post-ECC phase. We 5 have also demonstrated a similar disorder in metabolism in the renin-angiotensin system, with the abnormalities developing during the period of nonpulsatile ECC. These abnormal pituitary-adrenal and renin-angiotensin responses during ECC have also been reported by Uozumi," Favre," and their colleagues. In view of these clear evidences of organ system dysfunction during nonpulsatile ECC, and with the recent commercial availability of a modified roller pump able to deliver pulsatile flow, we have undertaken an
569
The Journal of
570 Taylor et al.
Thoracic and Cardiovascular Surgery
..
.
..
•
-
.....
•
..
•
•
. S1CKkt ri
1M•
FIg. 2. The Stockert pump control module allows pulsatile flow to be triggered from the internal trigger of the control module or from the patient's electrocardiogram for synchronized pulsation.
Fig. 1. The Stockert pump, a roller pump able to deliver nonpulsatile or pulsatile flow.
evaluation of pulsatile flow perfusion during cardiopulmonary bypass.
Stockert pulsatile roller pump* (Fig . I) This pump is a modified roller pump based on a large "stepping" motor with low pump-head inertia which allows rapid acceleration and deceleration of the pump head. The pump motor is driven via a control module (Fig. 2) which can provide an internal trigger or can relay an external trigger from the patient 's electrocardiogram and thereby allow synchronized counterpulsation, e .g., in the cardiac recovery phase. The control module allows the operator to set the frequency, amplitude, and duration of pulsation delivered by the pump head. The pump can also run as a standard roller pump , delivering nonpulsatile blood flow. A digital record of instantaneous mean or phasic pump flow is relayed on the pump module, and the pump can be calibrated easily for different circuit tubing diameters. The system has been incorporated into our existing extracorporeal circuitry without necessitating any mod*St6ckert Instrumente, Munich, West Germany .
ifications. The radial artery pressure profile relayed to a pen recorder has been used to record the production of a true pulsatile arterial waveform (Fig. 3) . At a mean flow rate of 3.75 L per minute and a mean arterial blood pressure of 59 .6 mm. Hg (range '47 to 73), a pulse pressure of 25 to 30 mm. Hg has consistently been achieved . At a mean pump flow of 3.75 L., the peak pulsatile flow has been 5.0 to 8.4 L. and the trough flow 2.1 to 2.3 L. per minute. The system has proved easy to use and appears reliable to date in 50 clinical cases .
Hematologic effects: Pulsatile and nonpulsatile flow In conjunction with stud ies of the metabolic effects of pulsatile flow on pituitary-adrenal metabolism during ECC (Parts II and III), a study of hemoly sis and blood constituent depletion was undertaken .
Patients and methods Twenty consecutive adult patients admitted for elective cardiac surgical procedures were studied. Ten consecutive patients were subjected to nonpulsatile flow throughout ECC (nonpulsatile group). The other 10 patients (pulsatile group) were subjected to pulsatile flow during ECC according to the following regimen: (I) at the onset of ECC until left ventricular ejection ceased-nonpulsatile flow; (2) thereafter until left ventricular ejection recommenced in the cardiac recovery period-pulsatile flow; (3) thereafter until the end of ECC-nonpulsatile flow .
Volume 75 Number 4 April,1978
Studies offlow during cardiopulmonary bypass, I
Non Pulsatile Mode
57 I
Pulsatile Mode (Rate. 71/min)
mm Hg
150 100 50
0'-
--'-
_
Patient M.N. 45 yrs Pump Flow Rate: Mean = 3,75/1 min. (Pulsatile = Non Pulsatile) Peak Pulsatile .. 5·48 I/min Trough Pulsatile. 2·14 Ilmin
Fig. 3. Tracing from printout of right radial artery pressure waveform in Patient M. N. on total bypass with the Stockert pump in the nonpulsatile mode (left) and in the pulsatile mode (right). Thus no period of pulsatile counterpulsation was used in the pulsatile group-the only difference between the groups in relation to perfusion being the period of pulsatile versus nonpulsatile flow during the central period of ECC. As indicated in Table I, there were no significant differences between the pulsatile and nonpulsatile groups with respect to age, weight, bypass time, mean pump flow during ECC, or mean arterial blood pressure during ECC. The bypass regimen and anesthetic regimen were standard in all cases. The extracorporeal circuit was primed with 2 L. of Ringer's lactate solution, and oxygenation was achieved with a Temptrol bubble oxygenator. * Body temperature on bypass was maintained at a mean of 37.2° C. in both groups (range 36.5° to 38.2° C.). The arterial return blood was passed through a 40 [J- screen filter (Ultipor filtert) and was returned to the ascending aorta through a Sarns aortic arch cannula (adult size). Anesthesia was induced with sodium thiopentone and maintained with nitrous oxide, oxygen, and intravenous morphine. The morphine dose administered during ECC was calculated and was found to be less than 0.2 mg. per kilogram of body weight per patient, well below the level shown to interfere with pituitary-adrenal stress responses." The pulsatile group comprised seven cases of prosthetic valve replacement and three of coronary artery bypass grafting. The nonpulsatile group comprised six cases of valve replacement and four of coronary artery bypass grafting. These 20 patients formed the study populations for the hematologic studies described in Part I and for the metabolic studies in Parts II and III. *Bentley Laboratories, Inc., Santa Ana, Calif. tPall Corporation, Glen Cove, N. Y.
Table I. Patients studied: Pulsatile and control groups
No. of patients Mean age (yr.) Mean weight
Nonpulsatile
Pulsatile
10 46 63.8
10 47.6 65.7
84.7
87.2
(kg.)
Mean bypass time (min.) Mean pump flow (L./min.) Mean arterial BP during ECC (mm. Hg)
4.08
3.74
69.4 ± 6.8 S.E.M.
56.0 ± 5.9 S.E.M.
Legend: BP, Blood pressure. ECC, Extracorporeal circulation.
The following hematologic parameters were measured: Plasma free hemoglobin. Samples were taken before the onset of ECC (pre-ECC) and at the end of the period of ECC (end-ECC). In addition to the absolute values, the rise in plasma free hemoglobin (ll-free hemoglobin) was calculated, and the ratio of plasma free hemoglobin to total bypass time in minutes was determined. Free hemoglobin was measured by the method of Cripps" by means of the SP600 spectrofluorimeter with a I ern. cell. The optical density (O.D.) of the sample was measured in triplicate and the average was used in the equation: Calculated 0.0. = 2 x 0.0. 576 (0.0. 560
+
0.0. 592)
The plasma hemoglobin (in milligrams per 100 ml.) was read off the reference graph of calculated O.D. against plasma hemoglobin.
The Journal of
572
Taylor et al.
Thoracic and Cardiovascular Surgery
Table II. Plasma free hemoglobin: Pulsatile and nonpulsatile groups
I
Nonpulsatile (mean ± S.E.M.)
Free hemoglobin (mg./100 ml.) Pre-ECC End-ECC 8 Free hemoglobin (mg./100 ml.) 8 Free hemoglobin (mg./ 100 ml.) Bypass time (min.) Platelet count 24 hours postop. RBC count 24 hours postop. (X \()6) WBC count 24 hours postop. (X \()3)
Pulsatile (mean ± S.E.M.)
t Value
I
p Value
4.7 ± 0.7 74.7 ± 16.3 69.7 ± \6.2
2.5 ± 0.5 45.7 ± 5.5 43.2 ± 5.5
1.83 1.68 1.55
N.S. N.S. N.S.
1.49 ± 0.71
0.64 ± 0.17
1.16
N.S.
85,000 ± 91 4.3±0.12 13.6 ± 0.92
114,000 ± 262 4.6 ± 0.\4 11.14 ± 1.\
1.03 1.81 1.66
N.S. N.S. N.S.
Legend: ECC, Extracorporeal circulation. RBC, Red blood cell.
wac, White blood cell.
Red cell, white cell and platelet counts. Samples were obtained on the day prior to operation and 24 hours after operation. The preoperative levels of red blood cells (RBC), white blood cells (WBC), and platelets were normal in all patients in both the pulsatile and nonpulsatile groups. No platelet transfusions were administered during or after operation in any of the 20 patients studied. The blood counts were performed on a Coulter counter model'S'. Platelet counts were performed by use of the Coulter thrombocounter and the sedimentation method. Statistical analyses were performed by means of Student's t test. Results (Table II) Plasma free hemoglobin. Plasma free hemoglobin levels rose in both groups during the period of ECC. The mean end-ECC level in the nonpulsatile group was 74.7 mg. per 100 ml. ± 16.3 S.E.M., and in the pulsatile group it was only 45.7 mg. per 100 ml. ± 5.5 S .E.M. The lower level of free hemoglobin in the pulsatile group was an interesting finding, although it was not statistically significant (t = 1.55, N.S.). In terms of absolute plasma levels of free hemoglobin, all patients in the pulsatile group had an end-ECC level of less than 70 mg. per 100 mi., which must be considered an acceptably low hemolysis index for a mean bypass time of more than 80 minutes with a bubble oxygenator. Although it is not likely that the rise in plasma free hemoglobin is linear in respect to time, we have also expressed our results as the ratio of .i Free hemoglobin (mg.lIOO m\.) Bypass time (min.)
This ratio was 1.49 mg. per 100 ml. per minute ± 0.71 S.E.M. in the nonpulsatile group and 0.64 mg. per 100 ml. per minute ± 0.17 S.E.M. in the pulsatile group.
Again, this difference is not statistically significant (t = 1.16, N.S.). The free hemoglobin values in the pulsatile group indicate that the degree of hemolysis associated with pulsatile flow is at least as acceptable as that associated with conventional nonpulsatile perfusion. RBC, WBC, and platelet counts. The RBC count 24 hours after the operation will obviously reflect the volume of blood lost during and after the operation and the adequacy of blood replacement. There were no significant differences between the groups in terms of the total amount of blood transfused during and after operation (nonpulsatile mean = 4:85 units; pulsatile mean = 4.62 units). The RBC and WBC 24 hours postoperatively also showed no significant differences between the groups. The mean platelet count 24 hours postoperatively in the nonpulsatile group was 85,000 ± 91 S.E.M.; it was higher in the pulsatile group, 114,000 ± 262 S.E.M. This difference was not, however, statistically significant (t = 1.03, N.S.). Discussion In this paper we have described a new, commercially available pump system able to deliver pulsatile perfusion during cardiopulmonary bypass. We have compared the degree of RBC destruction and blood cell and platelet depletion found after pulsatile flow perfusion with that found after conventional nonpulsatile ECC. We have found the system to be reliable, simple to operate, and able to produce consistently satisfactory arterial pulsatile flow at conventional flow rates. The studies clearly indicate that pulsatile perfusion with this modified roller pump system (Stockert pulsatile pump) does not produce any increased index of hemolysis or platelet and blood cell depletion when compared with nonpulsatile perfusion at the same mean flow rate and mean arterial pressure and over the same mean bypass time.
Volume 75 Number 4
Studies offlow during cardiopulmonary bypass, I
573
April,1978
The fears that pulsatile perfusion would be associated with increased hemolysis are widespread, although not well supported by objective data. It is important to realize that much of the conventional acceptance of nonpulsatile perfusion and the rejection of the more physiological pulsatile flow dates back to the earliest days of open-heart surgery. These attitudes stemmed from (1) the technical simplicity and reliability of constantly revolving pump heads and (2) fears that the already high hemolysis index associated with early cardiopulmonary bypass circuits would be further increased by pulsatile flow. In addition, attempts to achieve true pulsatile arterial flow were often complicated by the use of femoral artery return and the use of narrow-diameter arterial return cannulas. In view of the continuous expansion in industrial technology in recent years, one might have anticipated the development of pumps able to deliver adjustable and reliable pulsatile flow. Not all systems currently available produce pulsatility in the same way, e.g., the occlusive pulsatile systems (PAD* and Avco systems) are fundamentally different in design from the pulsatile flow-generating system used in our studies. The hematologic effects of all such systems must be studied before any possible metabolic benefit associated with their use can be considered. It is a readily acceptable hypothesis that pulsatile flow, being essentially physiological, should allow better organ metabolism than conventional nonpulsatile perfusion; the latter has been shown increasingly to be associated with multi organ dysfunction during ECC. Such comparative metabolic studies are reported in Parts II and III. On the basis of this present study, we conclude that the particular pulsatile perfusion system employed in
these comparative studies (Stockert pulsatile pump) allows the operator to achieve satisfactory pulsatile perfusion at conventional mean flow rates with no increase in hemolysis and no increased depletion of RBC, WBC, or platelets when compared with non pulsatile perfusion.
2 3 4
5
6
7
8
9 *Datascope Corp., Paramus, N. J.
REFERENCES Dunn J, Kirsh MM, Harness J, Carroll M, Straker J, Sloan H: Hemodynamic, metabolic, and hematologic effects of pulsatile cardiopulmonary bypass. J THORAC CARDIOVASC SURG 68:138-147,1974 Taylor KM, Jones JV, Walker MS, Rao LGS, Bain WH: The cortisol response during heart-lung bypass. Circulation 54:20-25, 1976 Taylor KM, Bain WH, Jones JV, Walker MS: The effect of hemodilution on plasma levels of cortisol and free cortisol. J THoRAc CARDIOVASC SURG 72:57-61, 1976 Taylor KM, Walker MS, Rao LGS, Jones JV, Gray CE: Plasma levels of cortisol, free cortisol and corticotrophin during cardiopulmonarybypass. J Endocrinol 67:29P-30P, 1975 Taylor KM, Morton 11, Brown 11, Bain WH, Caves PK: Hypertension and the renin-angiotensin system following open-heart surgery. J THORAC CARDIOVASC SURG 74:840845, 1977 Uozumi T, Manabe H, Kawashima Y, Hamanaka Y, Monden Y, Matsumoto K: Plasma cortisol, corticosterone and non-protein-bound cortisol in extracorporeal circulation. Acta Endocrinol :517-525, 1972 Favre L, Vallotton MB, Muller AF: Relationship between plasma concentrations of angiotensin I, Angiotensin II and plasma renin activity during cardiopulmonary bypass in man. Eur J Clin Invest 4:135-138, 1976 George JM, Reier CE, Lanesse RR, Rower JM: Morphine Anesthesia blocks cortisol and growth hormone response to surgical stress in humans. J Clin Endocrinol Metab 38: 736-741, 1974 Cripps CM: Rapid method for the estimation of plasma haemoglobin levels. J Clin Pathol 21:110-112, 1968