- Email: [email protected]
Relative thresholds for acute intravascular and extravascular mechanical hemolysis
Relative thresholds for acute intravascular and extravascular mechanical hemolysis We tested the postulate that acute extravascular hemolysis and intravascular hemolysis occurring during and after in vivo pumping of blood in an extracorporeal circuit are manifestations of the same insult to the red blood cell in experiments designed to determine if the threshold pumping rate in the extracorporeal circuit is the same for intravascular and extravascular hemolysis. In 10 experiments performed on dogs anesthetized with pentobarbital, blood was pumped in an extracorporeal circuit for one hour and measurements of intravascular and extravascular hemolysis were made for a total of 4 hours. The presence of significant intravascular hemolysis was always associated with significant extravascular hemolysis. In experiments performed at relatively low pumping rates, neither intravascular nor extravascular hemolysis was detected. Thus the data support the viewpoint that extravascular hemolysis and intravascular hemolysis are related processes and that acute extravascular mechanical hemolysis does not occur in the absence of intravascular hemolysis.
Herbert W. Wallace, M.D., and Ronald F. Coburn, M.D., Philadelphia, Pa.
We recently described a method of quantitating intravascular and extravascular hemolysis when both mechanisms of erythrocyte destruction are occurring. 1 This method is based on measurements of rates of endogenous carbon monoxide production and influx and efflux of hemoglobin in plasma. We investigated mechanisms of hemolysis during and after pumping of blood in an extracorporeal circuit and found that extravascular hemolysis may be more significant, From the Departments of Physiology and Surgery, School of Medicine, University of Pennsylvania, Philadelphia, Pa. This work was supported in part by United States Public Health Service research grants HE 11276 and HE 10331, and by a contract with the United States Army. Dr. Coburn is the recipient of United States Public Health Service Research Career Development Award HE 11564. Received for publication May 3, 1974. Address for reprints: Dr. Herbert W. Wallace, Department of Surgery, Graduate Hospital of the University of Pennsylvania, 19th and Lombard Sts., Philadelphia, Pa. 19146.
792
in terms of red blood cell destruction, than intravascular hemolysis. The relationships of intravascular and extravascular hemolysis are unknown. One can pose the question whether the same insult causes loss of hemoglobin into plasma and alterations of the cell or membrane that result in sequestration and destruction. It seems possible that an initial loss of hemoglobin from the cells is followed by sequestration and destruction of the cell remnants. Another possible relationship of intravascular and extravascular hemolysis is that the number of damaged cells exceeds the ability of the reticuloendothelial system to catabolize hemoglobin and the hemoglobin leaks out of the reticuloendothelial system into the blood, as appears to occur following very large infusions of erythrocytes damaged by incubation with Nserhylmaleimide." To prove that intravascular and extravascular hemolysis are related is a difficult problem but one
Volume 68 Number 5 November, 1974
Intravascular and extravascular hemolysis
approach is to determine whether the threshold "force" resulting in the two types of hemolysis is the same. In the present study we evaluate the relationship between mechanically induced intravascular and extravascular hemolysis utilizing in vivo extracorporeal perfusion in the anesthetized dog. Since we have no way of directly estimating "force" exerted on the red blood cell, we have assumed that this is a function of the pumping rate of blood in the extracorporeal circuit. We have attempted to determine whether there are different threshold pumping rates for intravascular and extravascular hemolysis and to observe the relationship between the two types of hemolysis at pumping rates above threshold. Materials and methods
The animal preparation has been described previously. 1 Adult, male, mongrel dogs weighing between 17.7 and 22.3 kilograms were anesthetized with intravenous pentobarbital (30 mg. per kilogram). A tracheostomy was performed through a midline incision and a snugly fitting endotracheal tube, encircled by two heavy ligatures to insure an airtight connection, was inserted. Intravenous fluids (5 Gm. of dextrose per 100 ml. of water) were administered at a rate of approximately 100 ml. per hour. A long polyethylene catheter was inserted into an external jugular vein for withdrawal of blood samples. A closed rebreathing system with a mechanical ventilator was used to prevent loss of endogenously produced carbon monoxide. The respirator was set to give an arterial Pco, of approximately 40 mm. Hg and the oxygen tension in the rebreathing system to give an arterial POz of 70 to 100 mm. Hg. At the beginning of the experiment the animal was given heparin (30 U. per kilogram) intravenously and 14CO (3 to 5 !Joc) via the rebreathing system. In 10 experiments we measured carbon monoxide production during a 2 hour control period. Following this control time
79 3
period, venoarterial perfusion was carried out between the external jugular vein and a carotid artery for approximately 60 minutes at constant flow rates ranging from 500 to 1,040 ml. per minute. At the end of the perfusion period the tubing contents were infused into the animal and the cannulas were removed. Carbon monoxide production and plasma hemoglobin concentrations were determined over the next 3 hours. The extracorporeal circuit consisted of a small roller pump (Masterflex Tubing Pump, Cole-Partner) and Tygon tubing (% inch I.D.). New tubing-was used in each experiment in order to insure uniform cleanliness of the circuit. The priming volume of the system was 70 ml. A stainless steel arterial cannula (1;4 inch J.D.) and a large silicone rubber venous catheter were utilized. The pump was calibrated before each experiment. The mechanisms by which erythrocytes are damaged in extracorporeal circuits are poorly understood-" and we have not studied our circuit in this regard. Measurements and calculations
Rates of carbon monoxide production were computed from serial measurements of blood carboxyhemoglobin per cent saturation (12COHb) and "CO radioactivity (1·COHb) in venous blood as described previously.' We determined 12COHb by an infrared method" and 14COHb by the method of Luomanmaki and Coburn." Plasma hemoglobin concentrations were determined by the method of Crosby and Furth':' (±1 mg. per 100 ml.). Baseline values with this technique were higher in the current dog population than in those previously reported." Studies in which we added known quantities of hemoglobin to plasma showed that the measurement is linear over the range 1 to 100 mg. per 100 ml. The calculation of extravascular hemolysis (EH) and intravascular hemolysis (IH) has been described previously.' To summarize this calculation: IH was computed from the increase in plasma hemoglobin concentration during the one-hour period
The Journal af
794
Wallace and Coburn
Thoracic and Cardiovascular Surgery
Experiment 4 Pump Rote 1040 ml./ min.
Experiment 3 Pump Rote 680 ml./ min.
10
9
0.39 mll hr
Pump
Pump 2
3
4
6
7
8·
3
Time in hrs
9
8
~
t
4
Time in hrs
5
Experiment 5
Experiment 6
Pump rate 840 mI / hr
Pump rote 640 ml / min
erfl1/tl'
7
06
6
B4 ~.
8
6
•
0.25 ml /hr.
3 2 Pump
Pump
3
4 Time in hrs
6
7
8
2
3
4
5
6
8
9
Time in hrs.
Fig. 1. Carbon monoxide data in representative in vivo pumping experiments. This figure illustrates the method of plotting changes in body CO stores and thus deriving CO production data. Note that in Experiments 3 and 6 the rate of increase in body CO stores did not change following pumping, whereas in Experiments 4 and 5 there was a marked increase in the rate of increase of the CO stores. Excess CO was determined by measuring the differences between the actual CO stores at a time 3 hours following cessation of pumping and extrapolated baseline stores at that time.
when blood was pumped in the extracorporeal circuit, multiplied by a plasma dilution factor (77 ml. per kilogram). 1 No correction was made for "plasma hemoglobin" catabolism occurring during the time when blood was pumped in the extracorporeal circuit since evidence has been obtained that this results in only a very small error in the calculation of IH.I We assume that intravascular hemolysis does not continue after pumping has stopped and we have evidence that this assumption is justified.1 EH was computed from increases in the rate of endogenous carbon monoxide production (V'e o) compared to control values, correcting for carbon monoxide produced from catabolism of hemoglobin effluxing from plasma. EH and IH are given as grams of hemoglobin. Fig. 1 illustrates the method of plotting
carbon monoxide data. The baseline rate of endogenous carbon monoxide production is extrapolated, and "excess CO" is determined from the difference between the extrapolated baseline V'eo and the actual measurement of body CO stores. 1, 11 Assuming that 1 mole of heme catabolized results in I mole of carbon monoxide produced, we can convert "excess CO" into grams of hemoglobin catabolized. In the present experiments excess CO and efflux of hemoglobin were determined over a 3 hour period beginning immediately after cessation of pumping. Therefore EH represents extravascular hemolysis that occurred during that period. Simultaneous measurement of IH and EH is possible with this method since catabolism of "plasma hemoglobin" is so slow as to cause only a very small increase in the rate of carbon monoxide production in the
Volume 68 Number 5 November, 1974
Intravascular and extravascular hemolysis
anesthetized dog, compared to very large increases that occur during extravascular hemolysis, ~ The largest source of error in determining IH is the assumption of a "plasma hemoglobin" dilution space and this is not at all critical in the present study. The primary objective was to determine the presence or absence of intravascular and extravascular hemolysis; the carbon monoxide method is ideal for this type of study because it can detect an EH of 1 ml, of erythrocytes. An IH of 0.5 ml. of erythrocytes can be detected.
795
30 0
Total 0 EH • IH 0
D 0
0
•
0
0
•
•• 0
0
Results
In six of the 10 experiments there was no change in the plasma hemoglobin concentration following one hour of extracorporeal perfusion; in these experiments there also was no increase in 'Vco. In the four other experiments plasma hemoglobin increased markedly during perfusion and there was an increase in vee, Thus when intravascular hemolysis occurred extravascular hemolysis also occurred. Whether or not intravascular and extravascular hemolysis occurred in these experiments depended upon the flow rate of blood in the extracorporeal circuit: Five of the six experiments in which neither type of hemolysis occurred were run at flow rates below 720 ml, per minute, whereas flow rates exceeding 700 ml. per minute were employed in all four experiments in which both types were observed. Data obtained in these experiments are plotted in Fig. 1. Calculated rates of IH and EH are shown in Fig. 2. EH accounted for an average of 61.2 per cent of the total hemoglobin catabolized in the 4 hour management period (range, 37 to 81 per cent) . Discussion
We were successful in finding a pumping rate in our system, approximately 700 ml. per minute, above which large quantities of hemoglobin appeared in plasma and the rate of endogenous production of carbon monoxide increased and below which these findings were absent. At the higher pumping rates elevations in plasma hemoglobin were
Flow Rate ml/ min
Fig. 2. Total extravascular and intravascular hemolysis. Data are given in grams of hemoglobin and are plotted vs. pump rate used in each experiment. Total hemolysis only is plotted in experiments where IH and EH were zero.
always associated with increases in vco and one interpretation might be that the elevated V'co resulted from catabolism of plasma hemoglobin. In computing EH a correction is made for efllux of hemoglobin from the "plasma" compartment and catabolism of this quantity of hemoglobin to CO is barely significant compared to the large increases in CO which occur. The possibility that this correction underestimated catabolism of "plasma" hemoglobin has been excluded by dog experiments in which hemoglobin solution was injected intravenously and the effect on vco was determined- and by experiments in which radioactive hemoglobin was injected prior to pumping blood in an extracorporeal circuit.' Thus we conclude that the threshold pumping rate for IH appears to be the same as that for EH during the 4 hour measurement period. The data are consistent with the concept that intravascular hemolysis and extravascular hemolysis are related processes under these conditions. Whether extravascular hemolysis was a result of sequestration and destruction of cells which had undergone partial intravascular hemolysis during passage through the extracorporeal circuit was not defined in these experiments but this is the most likely possi-
796
The Journol of Thorocic ond Cordiovosculor Surgery
Wallace and Coburn
bility. The influx of hemoglobin into plasma which occurs following overloading of the reticuloendothelial system with sequestered erythrocytes- is associated with a higher vco than was observed in our experiments, arguing against this mechanism being operative. The data obtained in the present study do not definitively prove that acute extravascular and intravascular hemolysis result from the same insult to the red blood cell; it could be that alterations of the red blood cell leading to extravascular hemolysis and to intravascular hemolysis are entirely different and it is coincidental that both mechanisms of hemolysis have the same pumping rate threshold. It is also possible that some red cells incur membrane damage, leading to subsequent sequestration and catabolism without changes in membrane permeability to hemoglobin resulting in loss of hemoglobin into plasma. However, in our experiments this could have occurred in significant amounts only at pump rates above the threshold for intravascular hemolysis. Our data relate only to "acute" mechanical hemolysis. There is evidence that the life spans of red blood cells decrease after pumping blood in an extracorporeal circuit." 12 These measurements were made over days and weeks and it is unknown if the "threshold" for damage associated with this phenomenon is the same as for IH. Our data, however, do provide initial evidence that intravascular hemolysis and extravascular hemolysis may be related processes. If the data can be extrapolated to man and to other pumping systems and other causes of mechanical hemolysis, we suggest that in the absence of an elevated plasma hemoglobin concentration, there is no syndrome of acute extravascular mechanical hemolysis.
REFERENCES
2 3
4
5
6 7
8
9
10
11 12
Wallace, H. W., Coburn, R. F., Habboushe, F., Blakemore, W. S., and Shepard, c.: Mechanically Induced Intravascular and Extravascular Hemolysis in Dogs, Circ. Res. 26: 347, 1970. Coburn, R. F., and Kane, P.: Maximal Erythrocyte and Hemoglobin Catabolism, J. Clin. Invest. 47: 1435, 1968. Blackshear, P. L., Dorman, F. D., and Steinbach, J. H.: Some Mechanical Effects That Influence Hemolysis, Trans. Am. Soc. Artif. Int. Organs 11: 112, 1965. Blackshear, P. L., Dorman, F. D., Steinbach, J. H., Maybach, E. I., Singh, A., and Collingham, R. E.: Shear Wall Interaction and Hemolysis, Trans. Am. Soc. Artif, Int. Organs 12: 113, 1966. Garfin, S. R., Indeglia, R. A., Shea, M. A., and Bernstein, E. F.: Effect of Albumin Coated Surfaces on Erythrocyte Mechanical Destruction, Surg. Forum 19: 135, 1968. Nevaril, C. G., Hellums, I. D., Alfrey, C. P., Jr., and Lynch, E. C.: Physical Effects in Red Blood Cell Trauma, AIChE J. 15: 707, 1969. Indeglia, B. A., and Bernstein, E. F.: Selective Lipid Loss Following Mechanical Erythrocyte Damage, Trans. Am. Soc. Artif. Int. Organs 16: 37, 1970. Coburn, R. F., Danielson, G. K., Blakemore, W. S., and Forster, R. E.: Carbon Monoxide in Blood; Analytical Method and Sources of Error, I. Appl, Physiol. 19: 510, 1964. Luomanmaki, K., and Coburn, R. F.: Effects of Metabolism and Distribution of Carbon Monoxide on Blood and Body Stores, Am. J. Physiol. 217: 354, 1969. Crosby, W. H., and Furth, F. W.: Modification of the Benzidine Method for Measurement of Hemoglobin in Plasma and Urine, Blood 11: 380, 1956. Coburn, R. F.: Endogenous Carbon Monoxide Production and Body CO Stores, Acta Med. Scand. (SuppI.) 472: 269, 1967. Wallace, H. W., and Blakemore, W. S.: Intravascular and Extravascular Hemolysis Accompanying Extracorporeal Circulation: A Clinical Study, Circulation 42: 521, 1970.
Herbert W. Wallace, M.D., and Ronald F. Coburn, M.D., Philadelphia, Pa.
We recently described a method of quantitating intravascular and extravascular hemolysis when both mechanisms of erythrocyte destruction are occurring. 1 This method is based on measurements of rates of endogenous carbon monoxide production and influx and efflux of hemoglobin in plasma. We investigated mechanisms of hemolysis during and after pumping of blood in an extracorporeal circuit and found that extravascular hemolysis may be more significant, From the Departments of Physiology and Surgery, School of Medicine, University of Pennsylvania, Philadelphia, Pa. This work was supported in part by United States Public Health Service research grants HE 11276 and HE 10331, and by a contract with the United States Army. Dr. Coburn is the recipient of United States Public Health Service Research Career Development Award HE 11564. Received for publication May 3, 1974. Address for reprints: Dr. Herbert W. Wallace, Department of Surgery, Graduate Hospital of the University of Pennsylvania, 19th and Lombard Sts., Philadelphia, Pa. 19146.
792
in terms of red blood cell destruction, than intravascular hemolysis. The relationships of intravascular and extravascular hemolysis are unknown. One can pose the question whether the same insult causes loss of hemoglobin into plasma and alterations of the cell or membrane that result in sequestration and destruction. It seems possible that an initial loss of hemoglobin from the cells is followed by sequestration and destruction of the cell remnants. Another possible relationship of intravascular and extravascular hemolysis is that the number of damaged cells exceeds the ability of the reticuloendothelial system to catabolize hemoglobin and the hemoglobin leaks out of the reticuloendothelial system into the blood, as appears to occur following very large infusions of erythrocytes damaged by incubation with Nserhylmaleimide." To prove that intravascular and extravascular hemolysis are related is a difficult problem but one
Volume 68 Number 5 November, 1974
Intravascular and extravascular hemolysis
approach is to determine whether the threshold "force" resulting in the two types of hemolysis is the same. In the present study we evaluate the relationship between mechanically induced intravascular and extravascular hemolysis utilizing in vivo extracorporeal perfusion in the anesthetized dog. Since we have no way of directly estimating "force" exerted on the red blood cell, we have assumed that this is a function of the pumping rate of blood in the extracorporeal circuit. We have attempted to determine whether there are different threshold pumping rates for intravascular and extravascular hemolysis and to observe the relationship between the two types of hemolysis at pumping rates above threshold. Materials and methods
The animal preparation has been described previously. 1 Adult, male, mongrel dogs weighing between 17.7 and 22.3 kilograms were anesthetized with intravenous pentobarbital (30 mg. per kilogram). A tracheostomy was performed through a midline incision and a snugly fitting endotracheal tube, encircled by two heavy ligatures to insure an airtight connection, was inserted. Intravenous fluids (5 Gm. of dextrose per 100 ml. of water) were administered at a rate of approximately 100 ml. per hour. A long polyethylene catheter was inserted into an external jugular vein for withdrawal of blood samples. A closed rebreathing system with a mechanical ventilator was used to prevent loss of endogenously produced carbon monoxide. The respirator was set to give an arterial Pco, of approximately 40 mm. Hg and the oxygen tension in the rebreathing system to give an arterial POz of 70 to 100 mm. Hg. At the beginning of the experiment the animal was given heparin (30 U. per kilogram) intravenously and 14CO (3 to 5 !Joc) via the rebreathing system. In 10 experiments we measured carbon monoxide production during a 2 hour control period. Following this control time
79 3
period, venoarterial perfusion was carried out between the external jugular vein and a carotid artery for approximately 60 minutes at constant flow rates ranging from 500 to 1,040 ml. per minute. At the end of the perfusion period the tubing contents were infused into the animal and the cannulas were removed. Carbon monoxide production and plasma hemoglobin concentrations were determined over the next 3 hours. The extracorporeal circuit consisted of a small roller pump (Masterflex Tubing Pump, Cole-Partner) and Tygon tubing (% inch I.D.). New tubing-was used in each experiment in order to insure uniform cleanliness of the circuit. The priming volume of the system was 70 ml. A stainless steel arterial cannula (1;4 inch J.D.) and a large silicone rubber venous catheter were utilized. The pump was calibrated before each experiment. The mechanisms by which erythrocytes are damaged in extracorporeal circuits are poorly understood-" and we have not studied our circuit in this regard. Measurements and calculations
Rates of carbon monoxide production were computed from serial measurements of blood carboxyhemoglobin per cent saturation (12COHb) and "CO radioactivity (1·COHb) in venous blood as described previously.' We determined 12COHb by an infrared method" and 14COHb by the method of Luomanmaki and Coburn." Plasma hemoglobin concentrations were determined by the method of Crosby and Furth':' (±1 mg. per 100 ml.). Baseline values with this technique were higher in the current dog population than in those previously reported." Studies in which we added known quantities of hemoglobin to plasma showed that the measurement is linear over the range 1 to 100 mg. per 100 ml. The calculation of extravascular hemolysis (EH) and intravascular hemolysis (IH) has been described previously.' To summarize this calculation: IH was computed from the increase in plasma hemoglobin concentration during the one-hour period
The Journal af
794
Wallace and Coburn
Thoracic and Cardiovascular Surgery
Experiment 4 Pump Rote 1040 ml./ min.
Experiment 3 Pump Rote 680 ml./ min.
10
9
0.39 mll hr
Pump
Pump 2
3
4
6
7
8·
3
Time in hrs
9
8
~
t
4
Time in hrs
5
Experiment 5
Experiment 6
Pump rate 840 mI / hr
Pump rote 640 ml / min
erfl1/tl'
7
06
6
B4 ~.
8
6
•
0.25 ml /hr.
3 2 Pump
Pump
3
4 Time in hrs
6
7
8
2
3
4
5
6
8
9
Time in hrs.
Fig. 1. Carbon monoxide data in representative in vivo pumping experiments. This figure illustrates the method of plotting changes in body CO stores and thus deriving CO production data. Note that in Experiments 3 and 6 the rate of increase in body CO stores did not change following pumping, whereas in Experiments 4 and 5 there was a marked increase in the rate of increase of the CO stores. Excess CO was determined by measuring the differences between the actual CO stores at a time 3 hours following cessation of pumping and extrapolated baseline stores at that time.
when blood was pumped in the extracorporeal circuit, multiplied by a plasma dilution factor (77 ml. per kilogram). 1 No correction was made for "plasma hemoglobin" catabolism occurring during the time when blood was pumped in the extracorporeal circuit since evidence has been obtained that this results in only a very small error in the calculation of IH.I We assume that intravascular hemolysis does not continue after pumping has stopped and we have evidence that this assumption is justified.1 EH was computed from increases in the rate of endogenous carbon monoxide production (V'e o) compared to control values, correcting for carbon monoxide produced from catabolism of hemoglobin effluxing from plasma. EH and IH are given as grams of hemoglobin. Fig. 1 illustrates the method of plotting
carbon monoxide data. The baseline rate of endogenous carbon monoxide production is extrapolated, and "excess CO" is determined from the difference between the extrapolated baseline V'eo and the actual measurement of body CO stores. 1, 11 Assuming that 1 mole of heme catabolized results in I mole of carbon monoxide produced, we can convert "excess CO" into grams of hemoglobin catabolized. In the present experiments excess CO and efflux of hemoglobin were determined over a 3 hour period beginning immediately after cessation of pumping. Therefore EH represents extravascular hemolysis that occurred during that period. Simultaneous measurement of IH and EH is possible with this method since catabolism of "plasma hemoglobin" is so slow as to cause only a very small increase in the rate of carbon monoxide production in the
Volume 68 Number 5 November, 1974
Intravascular and extravascular hemolysis
anesthetized dog, compared to very large increases that occur during extravascular hemolysis, ~ The largest source of error in determining IH is the assumption of a "plasma hemoglobin" dilution space and this is not at all critical in the present study. The primary objective was to determine the presence or absence of intravascular and extravascular hemolysis; the carbon monoxide method is ideal for this type of study because it can detect an EH of 1 ml, of erythrocytes. An IH of 0.5 ml. of erythrocytes can be detected.
795
30 0
Total 0 EH • IH 0
D 0
0
•
0
0
•
•• 0
0
Results
In six of the 10 experiments there was no change in the plasma hemoglobin concentration following one hour of extracorporeal perfusion; in these experiments there also was no increase in 'Vco. In the four other experiments plasma hemoglobin increased markedly during perfusion and there was an increase in vee, Thus when intravascular hemolysis occurred extravascular hemolysis also occurred. Whether or not intravascular and extravascular hemolysis occurred in these experiments depended upon the flow rate of blood in the extracorporeal circuit: Five of the six experiments in which neither type of hemolysis occurred were run at flow rates below 720 ml, per minute, whereas flow rates exceeding 700 ml. per minute were employed in all four experiments in which both types were observed. Data obtained in these experiments are plotted in Fig. 1. Calculated rates of IH and EH are shown in Fig. 2. EH accounted for an average of 61.2 per cent of the total hemoglobin catabolized in the 4 hour management period (range, 37 to 81 per cent) . Discussion
We were successful in finding a pumping rate in our system, approximately 700 ml. per minute, above which large quantities of hemoglobin appeared in plasma and the rate of endogenous production of carbon monoxide increased and below which these findings were absent. At the higher pumping rates elevations in plasma hemoglobin were
Flow Rate ml/ min
Fig. 2. Total extravascular and intravascular hemolysis. Data are given in grams of hemoglobin and are plotted vs. pump rate used in each experiment. Total hemolysis only is plotted in experiments where IH and EH were zero.
always associated with increases in vco and one interpretation might be that the elevated V'co resulted from catabolism of plasma hemoglobin. In computing EH a correction is made for efllux of hemoglobin from the "plasma" compartment and catabolism of this quantity of hemoglobin to CO is barely significant compared to the large increases in CO which occur. The possibility that this correction underestimated catabolism of "plasma" hemoglobin has been excluded by dog experiments in which hemoglobin solution was injected intravenously and the effect on vco was determined- and by experiments in which radioactive hemoglobin was injected prior to pumping blood in an extracorporeal circuit.' Thus we conclude that the threshold pumping rate for IH appears to be the same as that for EH during the 4 hour measurement period. The data are consistent with the concept that intravascular hemolysis and extravascular hemolysis are related processes under these conditions. Whether extravascular hemolysis was a result of sequestration and destruction of cells which had undergone partial intravascular hemolysis during passage through the extracorporeal circuit was not defined in these experiments but this is the most likely possi-
796
The Journol of Thorocic ond Cordiovosculor Surgery
Wallace and Coburn
bility. The influx of hemoglobin into plasma which occurs following overloading of the reticuloendothelial system with sequestered erythrocytes- is associated with a higher vco than was observed in our experiments, arguing against this mechanism being operative. The data obtained in the present study do not definitively prove that acute extravascular and intravascular hemolysis result from the same insult to the red blood cell; it could be that alterations of the red blood cell leading to extravascular hemolysis and to intravascular hemolysis are entirely different and it is coincidental that both mechanisms of hemolysis have the same pumping rate threshold. It is also possible that some red cells incur membrane damage, leading to subsequent sequestration and catabolism without changes in membrane permeability to hemoglobin resulting in loss of hemoglobin into plasma. However, in our experiments this could have occurred in significant amounts only at pump rates above the threshold for intravascular hemolysis. Our data relate only to "acute" mechanical hemolysis. There is evidence that the life spans of red blood cells decrease after pumping blood in an extracorporeal circuit." 12 These measurements were made over days and weeks and it is unknown if the "threshold" for damage associated with this phenomenon is the same as for IH. Our data, however, do provide initial evidence that intravascular hemolysis and extravascular hemolysis may be related processes. If the data can be extrapolated to man and to other pumping systems and other causes of mechanical hemolysis, we suggest that in the absence of an elevated plasma hemoglobin concentration, there is no syndrome of acute extravascular mechanical hemolysis.
REFERENCES
2 3
4
5
6 7
8
9
10
11 12
Wallace, H. W., Coburn, R. F., Habboushe, F., Blakemore, W. S., and Shepard, c.: Mechanically Induced Intravascular and Extravascular Hemolysis in Dogs, Circ. Res. 26: 347, 1970. Coburn, R. F., and Kane, P.: Maximal Erythrocyte and Hemoglobin Catabolism, J. Clin. Invest. 47: 1435, 1968. Blackshear, P. L., Dorman, F. D., and Steinbach, J. H.: Some Mechanical Effects That Influence Hemolysis, Trans. Am. Soc. Artif. Int. Organs 11: 112, 1965. Blackshear, P. L., Dorman, F. D., Steinbach, J. H., Maybach, E. I., Singh, A., and Collingham, R. E.: Shear Wall Interaction and Hemolysis, Trans. Am. Soc. Artif, Int. Organs 12: 113, 1966. Garfin, S. R., Indeglia, R. A., Shea, M. A., and Bernstein, E. F.: Effect of Albumin Coated Surfaces on Erythrocyte Mechanical Destruction, Surg. Forum 19: 135, 1968. Nevaril, C. G., Hellums, I. D., Alfrey, C. P., Jr., and Lynch, E. C.: Physical Effects in Red Blood Cell Trauma, AIChE J. 15: 707, 1969. Indeglia, B. A., and Bernstein, E. F.: Selective Lipid Loss Following Mechanical Erythrocyte Damage, Trans. Am. Soc. Artif. Int. Organs 16: 37, 1970. Coburn, R. F., Danielson, G. K., Blakemore, W. S., and Forster, R. E.: Carbon Monoxide in Blood; Analytical Method and Sources of Error, I. Appl, Physiol. 19: 510, 1964. Luomanmaki, K., and Coburn, R. F.: Effects of Metabolism and Distribution of Carbon Monoxide on Blood and Body Stores, Am. J. Physiol. 217: 354, 1969. Crosby, W. H., and Furth, F. W.: Modification of the Benzidine Method for Measurement of Hemoglobin in Plasma and Urine, Blood 11: 380, 1956. Coburn, R. F.: Endogenous Carbon Monoxide Production and Body CO Stores, Acta Med. Scand. (SuppI.) 472: 269, 1967. Wallace, H. W., and Blakemore, W. S.: Intravascular and Extravascular Hemolysis Accompanying Extracorporeal Circulation: A Clinical Study, Circulation 42: 521, 1970.