Vox Sang. 33: 1 4 (1977)
Intensive Plasmapheresis as a Risk Factor for Arteriosclerotic Cardiovascular Disease? P . Lundsgaard-Hansen University Department of Experimental Surgery (Prof. P. Lundsguard-Hunsen), Inselspital, Berne
Abstract. Recent studies suggest that the analogy between intensive plasmapheresis and the nephrotic syndrome with respect to plasma protein loss includes events which are generally recognized as risk factors for the development of arteriosclerotic cardiovascular disease. Experimental nephrosis and plasmapheresis alike increase the synthesis of very low density and low density lipoproteins in response to the hypoproteinemic stimulation of plasma protein synthesis. During electrophoresis, these lipoproteins move with the a2-and /?-globulins,and the serum levels of these fractions are often elevated in humans subjected to intensive plasmapheresis. Hyperlipoproteinemias of the %- and /?-typesare recognized risk factors in the genesis of arteriosclerosis. Another such factor appears to be an increased synthesis of fibrinogen. This protein enhances plasma viscosity and red cell aggregation and may play a role in myocardial infarction. Intensive plasmapheresis thus not only creates an unphysiologicaland partly depletional state; it is also, like the nephrotic syndrome, capable of producing surplus abnormalities of plasma constituents with a potential for delayed, insidious injury.
In an earlier paper [7], we concluded that plasmapheresis donors should, as now recommended by European authorities, not deliver more than 10-15 liters of plasma per year, that not more than 5 O O m l of plasma should be withdrawn per session, and that the interval between two such sessions should not be less than 2 weeks. These conclusions were based on calculations pertaining to albumin metabolism, on the accepted safeguards for whole blood donations, and on the analogy between intensive plasmapheresis and the nephrotic syndrome
with respect to plasma protein loss. There is now evidence that the latter analogy includes events which are generally recognized as risk factors for the development of arteriosclerotic cardiovascular disease. This evidence will be presented here. It has long been known that the hypoproteinemia of the nephrotic syndrome is regularly accompanied by hyperlipoproteinemia and that an inverse quantitative relationship usually exists between those two phenomena. One possible reason for this association appeared to be a general or non-
2
specific increase of hepatic plasma protein synthesis as a compensatory response to the intravascular or interstitial protein - i.e. oncotic - deficit. This assumption was supported by the finding that the administration of plasma substitutes based on artificial colloids also reduces hyperlipidemia in nephrotic animals [l]. The significant new evidence stems from experiments performed by Shafrir and Brenner [lo] to test this oncotic hypothesis. They induced hypoproteinemia in rats by the administration of aminonucleoside or plasmapheresis and studied the hepatic synthesis of plasma lipoproteins by tracer methods. They found marked alterations of very low density lipoprotein (VLDL) and low density lipoprotein (LDL) metabolism, which are best summarized by quoting their own conclusions: ‘Our results demonstrate an increased incorporation of amino acids into apolipoproteins in rats with hypoproteinemia of nephrosis or after plasmapheresis in tight association with the increased channeling of preformed fatty acids and of citrate into the lipoprotein lipids. Thus, the increased lipoprotein synthesis is very likely a part of the general response to the loss of plasma oncotic pressure and a compensatory adjustment to hypoproteinemia. It is not evoked by protein seepage specifically through the kidney. The hyperlipogenesis in nephrosis or after plasmapheresis occurs to fulfil the immediate requirement of lipid complement for the apolipoproteins synthetized in excess.. . The overwhelming phenomenon which determines the demand for lipids is the increased rate of synthesis of the apoprotein carriers, whereas the nature of the lipid precursors depends on the availability of appropriate precursors in deference to
Lundsgaard-Hansen
the nutritional state and the efficiency of the fatty acid synthesis pathway.’ Previous studies of Bur-On undshufrir [ 3 ] had, in fact, revealed abnormalities of glucose metabolism in experimental nephrosis which were equally interpreted as a response to an oncotic stimulus of plasma protein synthesis. The conversion of glucose carbons into the proteins and lipids of the plasma-liver pool was enhanced at the expense of glucose incorporation into glycogen of all tissues and into peripheral fat and protein. In the chronically nephrotic animals studied by Shafrir and Brenner [lo], the serum concentrations of the VLDL and LDL increased approximately tenfold, but the serum levels were not measured in their short-term (24 h) plasmapheresis experiments. There is, however, suggestive evidence that the serum lipoproteins may frequently be elevated in humans subjected to intensive ‘harvesting’ of plasma. When classified by their electrophoretic mobility, the very low density lipoproteins (VLDL) behave as pre-8 or a, and the low density lipoproteins (LDL) as /?-lipoproteins.Friedman et al. [6] found the percentage of a,-globulins to increase significantly during the first 4 months of an intensive plasmapheresis program, and a recent survey of 175 donors participating in such programs disclosed abnormally high levels of a*- and 8-globulins in at least one serum sample from 45.6 and 60% of the donors, respectively [2]. In contrast to albumin and yglobulin, it is obviously by no means uncommon for the synthetic rates of the a*and ,%globulin fractions to exceed the rates of simultaneous loss during intensive plasmapheresis in man. The role of hyperlipoproteinemias involving the a, (VLDL) and j3 (LDL) moie-
Intensive Plasmapheresis as a Risk Factor for Arteriosclerotic Cardiovascular Disease?
ties as risk factors for the development of arteriosclerosis - also in patients suffering from a nephrotic syndrome - is now sufficiently established to be mentioned in standard textbooks [S, 111, and it continues to be accepted at the research level despite the numerous questions remaining to be answered [4]. The study of Shufrir and Brenner [lo] thus adds a new dimension to the medical reservations against intensive plasmapheresis, which have so far mainly centered on its depletional features, because it must now also be assumed that, like the nephrotic syndrome, it may generate a potentially harmful surplus of certain plasma constituents. It has, furthermore, been shown [3] that the synthetic rate and serum levels of fibrinogen parallel those of the lipoproteins in nephrotic animals. If this phenomenon is another part of the general, unspecific response of plasma protein synthesis to an oncotic stimulus, which appears to be the case, one would expect it to occur also in donors subjected to intensive plasmapheresis. Elevated plasma fibrinogen levels are now known to enhance plasma viscosity and red cell aggregation, and there is suggestive evidence that both of these mechanisms are causally related to myocardial infarction ~ 9 1 . To sum up, a plasmapheresis program of sufficient intensity to produce a continuous and prolonged acceleration of the hepatic synthesis of plasma proteins does not only create an unphysiological and partly depletional state which appears to be indefensible from an ethical viewpoint [7] ; it is also, like the nephrotic syndrome, capable of causing definable surplus abnormalities of several plasma constituents with potentially harmful sequels. Moreover, the
3
coupling between lipoprotein as well as fibrinogen synthesis and the general synthetic response to the stress of protein loss through plasmapheresis implies that the donor is caught between the pincers of depletion and surplus abnormality: the better he is able to dodge the former by a vigorous synthetic response, the more he will become exposed to the latter, and vice versa. It has thus once again become apparent that nature exacts its forfeits for any excessive interference with an equilibrium which is normally regulated by elaborate mechanisms. It will probably be difficult to obtain quantitative data on the importance of these events for the human plasmapheresis donor. In the recent past, however, the results of animal experiments revealing a noxious potential of chemicals have been considered a valid basis for conclusions affecting their use in man. Since the medical profession has, in such instances, adhered to the precept of primum nil nocere, why not the best in the protection of plasmapheresis donors against similar risks of delayed, insidious injury?
References 1 Allen, J. C.;Baxter, J. H., and Goodman, H. C.: Effects of dextran, polyvinylpyrrolidone and gamma globulin on the hyperlipidemia of experimental nephrosis. J. din. Invest. 40: 499-508 (1961). 2 Barker, L. F.: Personal commun. 3 Bar-On, H. and Shafrir, E.: Conversion of glucose into lipids and proteins in experimental nephrotic syndrome. Israel I. med. Sci. I : 365-377 (1965). 4 Camerini-Davalos, R. A.; Bierman, E. L.; Redisch, W., and Zilversmit, D. B.: Atherogenesis. Ann. N.Y. Acad. Sci. 275 (1976).
Lundsgaard-Hansen
4
5 Dintenfass, L.: Blood microrheology. Viscosity
10 Shafrir, E. and Brenner, T.: Lipoprotein syn-
factors in blood flow, ischaemia and thrombosis (Butterworths, London 1971). Friedman, B. A.; Schork, M. A.; Mocniak, I. L., and Oberman, H. A.: Short-term and longterm effects of plasmapheresis on serum proteins and immunoglobulins. Transfusion 15: 467-472 (1975). Lundsgaard-Hansen, P.: Volume limitations of plasmapheresis. Vox. Sang. 32: 20-25 (1977). Schettler, G.:Innere Medizin; 3. Aufl., vol. 1, 2 (Thieme, Stuttgart 1972). Schmid-Schonbein, H. and Wells, R. E., jr.: Rheological properties of human erythrocytes and their influence upon the ‘anomalous’ viscosity of blood. Rev. Physiol. 63: 146-219 (1971).
thesis in hypoproteinemia of experimental nephrotic syndrome and plasmapheresis; in Bianchi, Mariani and McFarlane, Plasma protein turnover, p. 343-355 (MacMillan Press, London 1976). 11 Siegenthaler, W.;Klinische Pathophysiologie; 3. Aufl. (Thieme, Stuttgart 1976).
6
7 8 9
Received: January 27, 1977. Accepted: February 1, 1977 Prof. P. Lundsgaard-Hansen, Univ. Dept. of Exp. Surgery, Inselspital, CH-3010Berne (Switzerland)
Intensive Plasmapheresis as a Risk Factor for Arteriosclerotic Cardiovascular Disease? P . Lundsgaard-Hansen University Department of Experimental Surgery (Prof. P. Lundsguard-Hunsen), Inselspital, Berne
Abstract. Recent studies suggest that the analogy between intensive plasmapheresis and the nephrotic syndrome with respect to plasma protein loss includes events which are generally recognized as risk factors for the development of arteriosclerotic cardiovascular disease. Experimental nephrosis and plasmapheresis alike increase the synthesis of very low density and low density lipoproteins in response to the hypoproteinemic stimulation of plasma protein synthesis. During electrophoresis, these lipoproteins move with the a2-and /?-globulins,and the serum levels of these fractions are often elevated in humans subjected to intensive plasmapheresis. Hyperlipoproteinemias of the %- and /?-typesare recognized risk factors in the genesis of arteriosclerosis. Another such factor appears to be an increased synthesis of fibrinogen. This protein enhances plasma viscosity and red cell aggregation and may play a role in myocardial infarction. Intensive plasmapheresis thus not only creates an unphysiologicaland partly depletional state; it is also, like the nephrotic syndrome, capable of producing surplus abnormalities of plasma constituents with a potential for delayed, insidious injury.
In an earlier paper [7], we concluded that plasmapheresis donors should, as now recommended by European authorities, not deliver more than 10-15 liters of plasma per year, that not more than 5 O O m l of plasma should be withdrawn per session, and that the interval between two such sessions should not be less than 2 weeks. These conclusions were based on calculations pertaining to albumin metabolism, on the accepted safeguards for whole blood donations, and on the analogy between intensive plasmapheresis and the nephrotic syndrome
with respect to plasma protein loss. There is now evidence that the latter analogy includes events which are generally recognized as risk factors for the development of arteriosclerotic cardiovascular disease. This evidence will be presented here. It has long been known that the hypoproteinemia of the nephrotic syndrome is regularly accompanied by hyperlipoproteinemia and that an inverse quantitative relationship usually exists between those two phenomena. One possible reason for this association appeared to be a general or non-
2
specific increase of hepatic plasma protein synthesis as a compensatory response to the intravascular or interstitial protein - i.e. oncotic - deficit. This assumption was supported by the finding that the administration of plasma substitutes based on artificial colloids also reduces hyperlipidemia in nephrotic animals [l]. The significant new evidence stems from experiments performed by Shafrir and Brenner [lo] to test this oncotic hypothesis. They induced hypoproteinemia in rats by the administration of aminonucleoside or plasmapheresis and studied the hepatic synthesis of plasma lipoproteins by tracer methods. They found marked alterations of very low density lipoprotein (VLDL) and low density lipoprotein (LDL) metabolism, which are best summarized by quoting their own conclusions: ‘Our results demonstrate an increased incorporation of amino acids into apolipoproteins in rats with hypoproteinemia of nephrosis or after plasmapheresis in tight association with the increased channeling of preformed fatty acids and of citrate into the lipoprotein lipids. Thus, the increased lipoprotein synthesis is very likely a part of the general response to the loss of plasma oncotic pressure and a compensatory adjustment to hypoproteinemia. It is not evoked by protein seepage specifically through the kidney. The hyperlipogenesis in nephrosis or after plasmapheresis occurs to fulfil the immediate requirement of lipid complement for the apolipoproteins synthetized in excess.. . The overwhelming phenomenon which determines the demand for lipids is the increased rate of synthesis of the apoprotein carriers, whereas the nature of the lipid precursors depends on the availability of appropriate precursors in deference to
Lundsgaard-Hansen
the nutritional state and the efficiency of the fatty acid synthesis pathway.’ Previous studies of Bur-On undshufrir [ 3 ] had, in fact, revealed abnormalities of glucose metabolism in experimental nephrosis which were equally interpreted as a response to an oncotic stimulus of plasma protein synthesis. The conversion of glucose carbons into the proteins and lipids of the plasma-liver pool was enhanced at the expense of glucose incorporation into glycogen of all tissues and into peripheral fat and protein. In the chronically nephrotic animals studied by Shafrir and Brenner [lo], the serum concentrations of the VLDL and LDL increased approximately tenfold, but the serum levels were not measured in their short-term (24 h) plasmapheresis experiments. There is, however, suggestive evidence that the serum lipoproteins may frequently be elevated in humans subjected to intensive ‘harvesting’ of plasma. When classified by their electrophoretic mobility, the very low density lipoproteins (VLDL) behave as pre-8 or a, and the low density lipoproteins (LDL) as /?-lipoproteins.Friedman et al. [6] found the percentage of a,-globulins to increase significantly during the first 4 months of an intensive plasmapheresis program, and a recent survey of 175 donors participating in such programs disclosed abnormally high levels of a*- and 8-globulins in at least one serum sample from 45.6 and 60% of the donors, respectively [2]. In contrast to albumin and yglobulin, it is obviously by no means uncommon for the synthetic rates of the a*and ,%globulin fractions to exceed the rates of simultaneous loss during intensive plasmapheresis in man. The role of hyperlipoproteinemias involving the a, (VLDL) and j3 (LDL) moie-
Intensive Plasmapheresis as a Risk Factor for Arteriosclerotic Cardiovascular Disease?
ties as risk factors for the development of arteriosclerosis - also in patients suffering from a nephrotic syndrome - is now sufficiently established to be mentioned in standard textbooks [S, 111, and it continues to be accepted at the research level despite the numerous questions remaining to be answered [4]. The study of Shufrir and Brenner [lo] thus adds a new dimension to the medical reservations against intensive plasmapheresis, which have so far mainly centered on its depletional features, because it must now also be assumed that, like the nephrotic syndrome, it may generate a potentially harmful surplus of certain plasma constituents. It has, furthermore, been shown [3] that the synthetic rate and serum levels of fibrinogen parallel those of the lipoproteins in nephrotic animals. If this phenomenon is another part of the general, unspecific response of plasma protein synthesis to an oncotic stimulus, which appears to be the case, one would expect it to occur also in donors subjected to intensive plasmapheresis. Elevated plasma fibrinogen levels are now known to enhance plasma viscosity and red cell aggregation, and there is suggestive evidence that both of these mechanisms are causally related to myocardial infarction ~ 9 1 . To sum up, a plasmapheresis program of sufficient intensity to produce a continuous and prolonged acceleration of the hepatic synthesis of plasma proteins does not only create an unphysiological and partly depletional state which appears to be indefensible from an ethical viewpoint [7] ; it is also, like the nephrotic syndrome, capable of causing definable surplus abnormalities of several plasma constituents with potentially harmful sequels. Moreover, the
3
coupling between lipoprotein as well as fibrinogen synthesis and the general synthetic response to the stress of protein loss through plasmapheresis implies that the donor is caught between the pincers of depletion and surplus abnormality: the better he is able to dodge the former by a vigorous synthetic response, the more he will become exposed to the latter, and vice versa. It has thus once again become apparent that nature exacts its forfeits for any excessive interference with an equilibrium which is normally regulated by elaborate mechanisms. It will probably be difficult to obtain quantitative data on the importance of these events for the human plasmapheresis donor. In the recent past, however, the results of animal experiments revealing a noxious potential of chemicals have been considered a valid basis for conclusions affecting their use in man. Since the medical profession has, in such instances, adhered to the precept of primum nil nocere, why not the best in the protection of plasmapheresis donors against similar risks of delayed, insidious injury?
References 1 Allen, J. C.;Baxter, J. H., and Goodman, H. C.: Effects of dextran, polyvinylpyrrolidone and gamma globulin on the hyperlipidemia of experimental nephrosis. J. din. Invest. 40: 499-508 (1961). 2 Barker, L. F.: Personal commun. 3 Bar-On, H. and Shafrir, E.: Conversion of glucose into lipids and proteins in experimental nephrotic syndrome. Israel I. med. Sci. I : 365-377 (1965). 4 Camerini-Davalos, R. A.; Bierman, E. L.; Redisch, W., and Zilversmit, D. B.: Atherogenesis. Ann. N.Y. Acad. Sci. 275 (1976).
Lundsgaard-Hansen
4
5 Dintenfass, L.: Blood microrheology. Viscosity
10 Shafrir, E. and Brenner, T.: Lipoprotein syn-
factors in blood flow, ischaemia and thrombosis (Butterworths, London 1971). Friedman, B. A.; Schork, M. A.; Mocniak, I. L., and Oberman, H. A.: Short-term and longterm effects of plasmapheresis on serum proteins and immunoglobulins. Transfusion 15: 467-472 (1975). Lundsgaard-Hansen, P.: Volume limitations of plasmapheresis. Vox. Sang. 32: 20-25 (1977). Schettler, G.:Innere Medizin; 3. Aufl., vol. 1, 2 (Thieme, Stuttgart 1972). Schmid-Schonbein, H. and Wells, R. E., jr.: Rheological properties of human erythrocytes and their influence upon the ‘anomalous’ viscosity of blood. Rev. Physiol. 63: 146-219 (1971).
thesis in hypoproteinemia of experimental nephrotic syndrome and plasmapheresis; in Bianchi, Mariani and McFarlane, Plasma protein turnover, p. 343-355 (MacMillan Press, London 1976). 11 Siegenthaler, W.;Klinische Pathophysiologie; 3. Aufl. (Thieme, Stuttgart 1976).
6
7 8 9
Received: January 27, 1977. Accepted: February 1, 1977 Prof. P. Lundsgaard-Hansen, Univ. Dept. of Exp. Surgery, Inselspital, CH-3010Berne (Switzerland)
