Medical Hypotheses Medical Hypluhesu (1991) Ml-6 0 LoCroup UK Ltd 1991
Hypothesis: Insulin is Responsible for the Vascular Complications of Diabetes G. GWINUP and A. N. ELIAS Department of Medicine, University of California at Irvine, Irvine, California 927 17, USA (Correspondence to: Dr. G. Gwinup, Professor of Medicine, Division of Endocrinology and Metabolism, University of Californie at Irvine, Medical Center, 10 7 City Drive South, Orange, CA 92668, USA)
Abstract - It is proposed that the systemic hyperinsulinemia and hepatic portal hypoinsulinemia that occurs with conventional injectable preparations of insulin currently used in the treatment of patients with diabetes mellitus is largely responsible for the morbidity associated with this disease. Epidemiological evidence and animal experimentation strongly support systemic hyperinsulinemia as a major factor in genesis of atherosclerosis in diabetic patients. In addition, in vitro studies demonstrate a direct effect of insulin on endothelial cell and arterial smooth muscle proliferation. On the other hand, inadequate hepatic delivery of insulin is associated with overproduction of renal vasoregulatory factors leading to glomerular hyperfiltration and ultimately to glomerulosclerosis and its clinical endpoint - end-stage renal disease. In the absence of widespread success of pancreatic and islet-cell transplantation as a means to deliver insulin physiologically into the hepatic portal circulation, methods must be devised and perfected to accomplish such delivery using approaches such as orally administering insulin in intestinal-enzyme protected capsules. Until such methods of delivery are available for safe and widespread use, one should abandon the illusory goal of rigid glucose control in favor of methods that reduce insulin requirement. Along these lines, dietary restriction and aerobic exercise should be the major life style changes advised for diabetic patients. Reduction of glomerular hyperfiltration in diabetic patients can be promoted with the use of low protein diets and/or angiotensin converting enzyme inhibitors.
Introduction While much attention has been directed toward glucose ‘toxicity’ the possibility of insulin ‘toxicity’ has received little concern. When a diabetic is treated with insulin, and particularly when Date received 29 September 1989 Date accepted 16 February 1990
an attempt is made to drive the blood glucose down to physiologic concentrations, circulating mean insulin concentrations are about twice normal (1 - 2). In other words a physiologic concentration of one substance (glucose) has been achieved at the expense of a very unphysiologic
2 concentration of another substance (insulin). This occurs because at the present time it is impossible to give insulin physiologically. The non-diabetic secretes insulin into the portal circulation where the major fraction acts on the liver, is extracted by it (3) and never gains entry into the systemic circulation. Since the monumental discovery of insulin in 1921, there has not been much good news for diabetics. Our increased understanding of the metabolic abnormalities in the different types of diabetes has not, unfortunately, been translated into meaningful advances in therapy. Even insulin failed to fulfill its early promise. The discovery of the ‘missing factor’ quite naturally gave rise to the expectation that its administration to the diabetic would restore optimum health. It was only after a number of years of insulin treatment that it was appreciated that although insulin prevented death and produced well-being, with time its use was associated with the development of slow but progressively severe disorders of the small and large blood vessels. In spite of every maneuver which has been attempted in an effort to alter the course of diabetic vascular disease, more diabetics develop blindness, manifest coronary heart disease and enter dialysis programs every year (4 - 9). Virtually all of the maneuvers have been directed at lowering the blood glucose is associated with increasing circulating insulin levels and that unphysiologic insulin rather than unphysiologic glucose might be responsible for diabetic vascular disease. Insulin and diabetic microangiopathy epidemiological evidence In the diabetic, small blood vessel disease is most often exhibited as diabetic retinopathy. In the 1930’s it was proposed that elevated blood glucose was responsible for diabetic retinopathy as well as other small blood vessel involvement. In the US the Joslin Clinic was at the forefront of groups emphasizing rigid blood glucose control as a means for preventing or retarding blood vessel problems. In spite of evangelical promotion, the benefits of rigid glucose control were unconvincing and a comprehensive review by Bondy and Felig (10) in the 70’s concluded the case for rigid glucose control was tenuous at best. In the 1980’s it became possible for the first time to conveniently monitor blood glucose concentrations using the newly developed glucose sensitive paper strips. This important innovation led to new-found en-
MEDICAL HYPOTHESES
thusiasm for rigid control of blood glucose and a number of controlled prospective studies were undertaken comparing more rigid with conventional blood glucose control. Although the Job study (11) and the initial Steno (12) results were originally interpreted as showing less retinopathy with rigid control, this interpretation was obviously the result of strong preconviction on the part of the investigators, and both misinterpretations were subsequently corrected (5,13). As additional results became available from the Steno group (5) as well as from other groups conducting similar studies (6,7), it became apparent that the only clear cut difference between the rigidly controlled and conventionally controlled diabetics was the occurrence of accelerated retinal disease which occurred in the rigidly controlled patients soon after rigid control was imposed. Although this finding came as a surprise, such rapid retinal deterioration with rigid control had previously been reported by Daneman (14). The phenomenon of rapid retinal deterioration and its relation to rigid control has now been documented by several other studies (15 - 19). The devastating effect of bringing down glucose to absolutely normal concentrations was emphasized in the case of a 16-year-old boy who received a pancreatic graft which drained into the systemic circulation and who evidenced spectacular retinal deterioration in the first four months after transplantation (20). Both the Steno group and the Kroc group (the other large study comparing rigid and conventional control) are abandoning their undertakings (21) and it seems that rigid control has nothing to offer the diabetic with established disease. Currently a multi-institutional study (The Diabetic Control and Complications study) is being undertaken in the US. In one arm of the study diabetics who are very early in the course of their disease will be treated with rigid control with the hope that extremely early blood glucose control might help to avert vascular disease. This will be a most important venture but the expectation for favorable outcome is fueled more by hope than scientific evidence. High circulating insulin levels are presumably produced by subcutaneous administration in Type II diabetic as well as in the Type I diabetic. In the Type II diabetic however, the blood vessels may be exposed to high insulin levels for many years before treatment is initiated. High fasting and glucose-stimulated insulin levels, despite nor-
INSULIN AND VASCULAR COMPLICATIONS
3
OF DIABETES
mal glucose levels, are seen even in individuals destined to develop Type II diabetes (22). Even higher insulin levels are observed when such individuals begin to develop abnormal blood glucose levels (23). Although retinopathy rarely begins to appear in Type I diabetics before 5 years after insulin therapy is instituted, it is not unusual to observe retinopathy at the time of presentation in Type II diabetics. There is yet another unusual type (or types) of diabetes which has generally been referred to as maturity onset diabetes of the young (24 - 26). Although this is without doubt a heterogenous group (27,28), the most characteristic pattern of abnormalities is the simultaneous occurrence of high blood glucose concentrations with normal insulin concentrations present for decades. This group of diabetics has been noted to be particularly resistant to the development of diabetic retinal disease in spite of very high glucose concentrations for many years (24 - 26). This small sub-group of diabetics would appear to offer an experimental model suggesting that high glucose concentrations in the absence of abnormal insulin concentration is not particularly damaging to small blood vessels. The mechanism by which insulin damages small blood vessels is not apparent, but it has recently been demonstrated that proliferation of the endothelium of microvessels can be accentuated by the addition of insulin to the culture medium (29). Insulin and atherosclerosis Large blood vessel disease in the diabetic is most often manifested by coronary heart disease (30). In a study of approximately 1000 Finnish policemen who were followed for a decade, it was found that myocardial infarction correlated best with a high insulin response to an oral glucose challenge (31). Another study of over 7000 male Parisians found a strong correlation between coronary heart disease and high fasting insulin levels (32). A striking relationship between high insulin concentrations and cardiovascular disease was found in a study comparing a group of normal individuals with Type II diabetics. The Type II diabetics were sub-divided into a group treated with insulin and another group treated with diet and oral hypoglycemics. In the non-insulin groups there was a significant correlation between the incidence of coronary heart disease and C-peptide levels (33). In the insulin treated group the C-peptide levels were obviously lower than in the other two groups
but the incidence of cardiovascular disease was the highest of all. Furthermore, in the insulin treated patients both the dose and plasma insulin concentrations were higher in subjects who developed blood vessel disease over the 5 year study period. Animal and in vitro effects of insulin on atherogenesis and microangiopathy There is firm evidence from animal experiments that insulin is atherogenic. A group of rabbits fed a high cholesterol diet after being made diabetic with alloxan experience a lower incidence of atherosclerosis than non-diabetic controls (34). In chickens insulin has been shown to prevent the regression of atheromatous changes when a high cholesterol diet is changed to a normal diet (35). Insulin infused into one femoral artery of dogs made diabetic with pancreatectomy produced ipsilateral atherosclerosis which was not present in the contralateral saline-infused artery (36). This study implicates insulin itself as a direct cause of atherosclerosis. Recently Sato has shown that the administration of insulin to normal rats for a period of one year produces extensive atherosclerotic changes in their aortas but not in the aortas of saline treated controls (37). Insulin has been shown to stimulate the proliferation and migration of arterial smooth muscle cells in tissue culture preparations. Receptors for the growth promoting effects of insulin on arterial smooth muscle cells differ from the receptors concerned with insulin’s metabolic effects so that even in states of insulin resistance may have a growth promoting propensity on the cells of the arterial wall (38). Hepatic hypoinsulinization and diabetic nephropathy In addition to small and large blood vessel problems, the diabetic is also likely to develop, with time, diabetic kidney disease. Although clinical renal disease occurs in less than half of all diabetics histologic abnormalities are present in the great majority who have had diabetes for many years (39). Although there are some prospective studies reporting benefits from rigid glucose control (40), in others (8) this has failed to influence the progression of nephropathy and the case for benefit for rigid glucose control is unconvincing (41). There is one remarkable report in which diseased kidneys from a diabetic patient were accepted by
4 a renal transplant group in Kuwait (42). Each of the kidneys was transplanted into two separate non-diabetic recipients. Before transplantation, biopsies of each kidney evidenced very significant abnormalities which had improved dramatically when they were re-biopsied approximately six months after transplantation. The authors of the report ascribe the healing process (and improving function) to the normoglycemic environment in the recipients. This interpretation is not, however, in keeping with the studies cited above in which aggressive management of blood glucose in diabetics has produced unconvincing benefits on the progress of nephropathy. A recent publication addressed the effects of pancreas transplantation on glomerular structure in patients with insulindependent diabetes mellitus (43). In this study the pancreatic endocrine secretion was delivered into the systemic circulation via the iliac veins. The authors concluded that pancreas transplantation arrested the progression of mesangial proliferation as assessed by post-transplant histological examination of needle biopsy specimens from the transplanted kidneys. However, previous publications from this same group have emphasized the marked variability in the rate of development of mesangial proliferation in the kidneys of pancreas transplant recipients (44). Although it is tempting to try to ascribe the pathogenesis of diabetic nephropathy to the abnormaly elevated concentrations of insulin acting on the small blood vessels of the kidneys, there is better evidence for another pathogenic mechanism. One of the most consistent findings in diabetics who are early in the course of their treatment is a supfunormaI glomerular filtration rate which presumably leads to nephron damage over time (45). The increased filtration may not be due to systemic hyperinsulinemia but due to another problem related to the unphysiologic way in which insulin secreted by the pancreas does not go directly to the liver in high concentration and therefore the insulin treated diabetic suffers from systemic hyperinsulinemia. His liver, however, is not properly insulinized and this produces production of excessive glomerulopressin and/or other related vasoactive peptides which act on the kidney to produce abnormal elevation of the glomerular filtration rate (46). In contrast to retinopathy the Type I diabetic is much quicker to develop retropathy than the Type II diabetic. Since the classic type I diabetic secretes virtually no insulin into his portal circulation, this would suggest that
MEDICAL HYPOTHESES
proper insulinization of the liver may prevent diabetic nephropathy. Conclusions and future directions The hypothesis set forward here is based on the circumstantial evidence detailed above but it can not be confirmed unless tested directly. How might that be accomplished? Pancreatic grafts are generally not drained into the portal vein. The Roux-en-Y technique we used 15 years ago (47) would allow the venous drainage of the pancreas to be diverted into the portal vein and we are considering offering such a procedure to diabetics with rapidly progressing vascular disease. Attempts to develop a practical oral insulin preparation continue. Saffrom has successfully delivered insulin coated with an azo-copolymer to the colon of dogs (48) and we have accomplished the same thing in humans using a copolymer which only dissolves in the pH range found in the colon (49). Attempts to transplant islets continue (50) and they might some day be grafted into tissue drained by the portal circulation. In the meantime, it would seem that there is enough evidence to abandon the illusory goal of rigid blood glucose control with its attendant dangers and disruptions of the day to day life of the diabetic and concentrate on measures which reduce insulin requirements. Therefore dietary restriction and aerobic exercise should be the keystones of life style changes advised for the Type I as well as for the Type II diabetic. Several studies suggest that diminishing glomerular filtration rate with a very low protein diet and/or ACE inhibitors will delay the progression of diabetic nephropathy (51- 53). References 1. 2.
3. 4. 5.
6.
Rasmussen SM, Heding LG, Parbst E: Serum IRI in insulin-treated diabetic during a 24-hour period. Diabetologia 11: ISI- 158, 1975. Werther GA, Jenkins PA, Turner RC, et al: Twenty-four hour metabolic profiles in diabetic children receiving insulin injections once or twice daily. Br Med J 2: 414 - 418, 1980. Rijjmark S, Bloom G, Chou MCY, Field JB: Hepatic extraction of exogenous insulin and glucagon in the dog. Endocrinology 102: 806-813, 1978. Paterson AD, Dornan TL: Diabetes and renal failure (letter) Lancet 1: 922, 1987. Lauritzen T, Frost-Larsen K, Larsen HW, Deckert T. Steno Study Group: Effect of 1 year of near-normal blood glucose levels on retinopathy in insulin-dependent diabetics. Lancet 1: 200-204, 1983. Brinchmann-Hansen 0, Dahl-Jorgensen K, Hanssen KF,
INSULIN AND VASCULAR COMPLICATIONS
I.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22. 23.
24.
OF DIABETES
Sandvilc L, Oslo Study Group: Effects of intensified insulin treatment on various lesions of diabetic retinopathy. Am J Ophthalmol loo: 644-653, 1985. The Kroc Collaborative Study Group: Blood glucose control and the evolution of diabetic retinopathy and albuminuria: a preliminary multicenter trial. N Eng J Med 311: 365 - 372, 1984. Viberti GG, Bilous RW, Mackintosh D, Bending JJ, Keen H: Long-term correction of hyperglycaemia and progression of renal failure in insulin dependent diabetes. Br Med J (Clin Res) 286: 598 - 602, 1983. Pyorala K: Relationship of glucose tolerance and plasma insulin to the incidence of coronary heart disease: Results from two population studies in Finland. Diabetes Care 2: 131- 141, 1979. Bondy PK. Felig P: Relation of diabetic control to the development of vascular complications. Med Clin North Am 55: 889-897, 1971. Job D, Eschwege E, Guyot-Argenten C et al: Effect of multiple daily insulin injections on the course of diabetic retinopathy. Diabetes 25: 463 - 469, 1976. Steno Study Group: Efect of six months of strict metabolic control on eye and kidney function in insulindependent diabetics with background retinopathy. Lancet 1: 121- 124, 1982. Ashikaga T, Borodic G, Sims EA: Multiple daily insulin injections in the treatment of diabetic retinopathy. The Job study revisited. Diabetes 27: 592 - 596, 1978. Daneman D, Drash AL, Lobes LA, Becker DJ, Baker LM, Travis LB: Progressive retinopathy with improved control in diabetic dwarfism (Mauriac’s syndrome). Diabetes Care 4: 360- 365, 1981. Van Ballegooie E, Hooymans JMM, Timmerman Z et al: Rapid deterioration of diabetic retinopathy during treatment with continuous subcontinuous insulin infusion. Diabetes Care 7: 236 - 242, 1984. Beak-Neilsen H, Richelsen B, Morgensen CE et al: Effect of insulin pump treatment for one year on renal function and retinal morphology in patients with IDDM. Diabetes Care 8: 585-589, 1985. Kelly TM, Sanborn GE, Haug PJ, Edwards CQ: Effect of insulin infusion pump use on diabetic retinopathy. Arch Ophthalmol 102: 1156- 1159, 1984. Friberg TR, Rosenstock J, Sanborn G, Vaghefi A, Raskin P: The effect of long-term normal glycemic control on mild diabetic retinopathy. Ophthalmology 92: 1051- 1058, 1985. Phelps RL, Sakol P, Metzger BE, Jampol LM, Freinkel N: Changes in diabetic retinopathy during pregnancy: correlations with regulation of hyperglycemia. Arch Ophthalmol 104: 1806- 1810, 1986. Ramsey RC, Goetz PC, Sutherland DER, Mauer SM, Robinson LL, Cantrill HL, Knobloch WH, Najarian JS: Progression of diabetic retinopathy after pancreas transplantation for insulin dependent diabetes mellitus. New Eng J Med 318: 208 -214, 1988. Lauritzen T and Sherwin R - personal communications. Porte DJ, Bagdade JD: Human insulin secretion: An integrated approach. Annu Rev Med 21: 219- 240, 1970. Saad MF. Knowles WG, Pettitt DJ, Nelson RG, Mott DM, Bennett PH: Sequential changes in serum insulin concentration during development of non-insulin-dependent diabetes. Lancet 1: 1356- 1358, 1989. Tattersall RB: Mild familial diabetes with dominant inheritance. Q. J. Med 43: 339-357, 1974.
5
25. Tattersall RB, Fajans SS: A difference between the inheritance of classical juvenile onset and maturity onset type diabetes of young people. Diabetes 24: 44- 53, 1975. 26. Tattersall RB: The present status of maturity onset type of diabetes of the young people (MODY). In Genetics of Diabetes Mellitus. Kobberling J and Tattersall RB (Eds). London and New York, Academic Press, 1982: 1 - 19. 27. Fajans SS: Heterogeneity between various families with non-insulin dependent diabetes of the MODY type. In Genetics of Diabetes Mellitus. Kobberling J and Tattersall RB (Eds). London and New York, Academic Press. 1982: 251-260. 28. Fajans SS: MODY - A model for understanding the pathogenesis and natural history of type II Diabetes. Horm Metab Res 19: 591- 599, 1987. 29. Bar RS, Siddle K, Dolash S, Bees M, Drake M: Actions of insulin-like growth factors I and II in cultured microvessel endothelial cells from bovine adipose tissue. Metab 37: 714-720, 1988. 30. Jarrett RJ, McCartney P, Keen H: The Bedford Survey. Ten year mortality rates in newly diagnosed diabetics and normglycaemic controls and risk indices for coronary heart disease in borderline diabetics. Diabetologia 22: 79 - 86, 1982. 31. Pyorala K, Savolainen E, Kaukola S, Haapakoski J: High plasma insulin as a coronary disease risk factor. In: Eschwege E (ed): advances in diabetes epidemiology. INSERM Symposium No 22. Elsevier, Amsterdam 143, 1982. 32. Eschwege E, Richard JL, Thibult N, Ducimetiere P. Warnet JM, Claude JR, Rosselin GE: Coronary heart disease mortality in relation with diabetes, blood glucose and plasma insulin levels: the Paris prospective study, ten years later. Horm Metab Res (Suppl 15): 41-46, 1985. 33. Stand1 E, Janka HV: High insulin serum concentrations in relation to other cardiovascular risk factors in diabetes: Results of a prospective study. Horm Metab Res 17: 63 -68, 1985 (suppl) 34., Duff GL, Brechin DJH, Finkelstein WE: The effect of alloxan diabetes on experimental cholesterol atehrosclerosis in the rabbit: I. The inhibition of experimental cholesterol atherosclerosis in alloxan diabetes. II. The effect of alloxan diabetes on the retrogression of experimental cholesterol atherosclerosis. J Exp Med 89: 611 -630, 1949. 35. Stamler J, Pick R, Katz LN: Effect of insulin in the induction and regression of atherosclerosis in the chick. Circ Res 8: 572 - 576, 1960. 36. Cruz AB, Amatuzio DS, Grande F et al: Effect of intraarterial insulin on tissue cholesterol and fatty acids in alloxan-daibetic dogs. Circ Res 9: 39-43, l%l. 37. Sato Y, Shiralshi S, Oshida T, Sakomoto N: Experimental athrosclerosis-like lesions induced by hyperinsulinism in Wistor rats. Diabetes 38: 91- %, 1989. 38. Stout RW, Bierman FL, Ross R: Effect of insulin on the proliferation of primate arteriol smooth muscle cells. Cir Res 36: 319-327, 1975. 39. McCray RF, Pitts TO, Puschett JB: Diabetic nephropathy: Natural course, survivorship and therapy. Am J Nephrol 1: 206-218, 1981. 40. Holman RR, Dornan TL, Maynon-White V et al: Prevention of deterioration of renal and sensory-nerve function by more intensive management of insulin-dependent daibetic patients: a two year randomized prospective study. lancet 1: 204-208, 1983.
MEDICAL HYPOTHESES
6 41.
42.
43.
44.
45.
46.
47.
The DCCT Research Group: Are continuing studies of metabolic control and microvascular complications in insulindependent diabetes justified? The Diabetes Control and Complication Trial. N. Eng J Med 318: 246 - 250, 1988. Abouna GM, Dremer GD, Daddah SK, Al-Adanani MS, Kumar SA, Kusma G: Reversal of diabetic nephropathy in human cadeveric kidneys after transplantation into nondiabetic recipients. Lancet 2: 1274- 1276, 1983. Bilous RW. Mauer SM, Sutherland DER, Najarian JS, Goetx FC, Steffes Mw: The effects of pancreas transplantation on the glomerular structure of renal ahografts in patients with insulin-dependent diabetes. N Eng J Med 321: 80-85, 1989. Mauer SM. Goetz FC, McHugh LE et al: Long-term study of normal kidneys transplanted into patients with type I diabetes. Diabetes 38: 516-523, 1981. Brenner BM: Hemodynamically mediated glomerular injury and the progressive nature of kidney disease. Kidney Int. 23: 647-655, 1983. Del Castillo ER, Fuenxalida R. Uranga J: Increased glomerular filtration rate and glomemlopressin activity in diabetic dogs. Horm Metab Res 9: 46 - 53, 1977. Connolly JE, Martin DC, Steinberg T, Gwinup G.
48.
49.
50.
51.
52. 53.
Garzaniga AB, Bartlett RH: Clinical experience with pancreatico-duodenal transplantation. Arch Surg 106: 489 - 494, 1973. Saffran M, Field JB, Pena J, Jones RH, Okuda Y: Oral Delivery of insulin in pancreatectomized diabetic dogs (abstract). Program of the Am Diabetes Assoc Detroit Mich. Diabetes 38 (suppl2): pp 81A, 324, 1989. Gwinup G, Elias AN, Domurat ES: Insulin and C-peptide levels following oral administration of insulin in intestinal-enzyme protected capsules. Clin Res 38: 8A. 1990. Mullen Y, Clare-Salzler M, Stein E, Clark W: Islet transplantation for the cure of diabetes. Pancreas 4(l): 123 - 135, 1989. Taguma Y, Kitamovo Y, Futaki G: Effect of captopril on heavy proteinuria in axotemic diabetes. N Eng J Med 313: 1717- 1620, 1985. Wiseman MJ, Dodds R, Bending JJ et al: Dietary protein and the diabetic kidney. Diabetic Med 4: 144- 146, 1987. Evanoff GV, Thompson CS, Brown J et al: The effect of dietary protein restriction on the progression of diabetic nephropathy: A 1Zmonth follow-up. Arch Intern Med 147: 492-495, 1987.
Hypothesis: Insulin is Responsible for the Vascular Complications of Diabetes G. GWINUP and A. N. ELIAS Department of Medicine, University of California at Irvine, Irvine, California 927 17, USA (Correspondence to: Dr. G. Gwinup, Professor of Medicine, Division of Endocrinology and Metabolism, University of Californie at Irvine, Medical Center, 10 7 City Drive South, Orange, CA 92668, USA)
Abstract - It is proposed that the systemic hyperinsulinemia and hepatic portal hypoinsulinemia that occurs with conventional injectable preparations of insulin currently used in the treatment of patients with diabetes mellitus is largely responsible for the morbidity associated with this disease. Epidemiological evidence and animal experimentation strongly support systemic hyperinsulinemia as a major factor in genesis of atherosclerosis in diabetic patients. In addition, in vitro studies demonstrate a direct effect of insulin on endothelial cell and arterial smooth muscle proliferation. On the other hand, inadequate hepatic delivery of insulin is associated with overproduction of renal vasoregulatory factors leading to glomerular hyperfiltration and ultimately to glomerulosclerosis and its clinical endpoint - end-stage renal disease. In the absence of widespread success of pancreatic and islet-cell transplantation as a means to deliver insulin physiologically into the hepatic portal circulation, methods must be devised and perfected to accomplish such delivery using approaches such as orally administering insulin in intestinal-enzyme protected capsules. Until such methods of delivery are available for safe and widespread use, one should abandon the illusory goal of rigid glucose control in favor of methods that reduce insulin requirement. Along these lines, dietary restriction and aerobic exercise should be the major life style changes advised for diabetic patients. Reduction of glomerular hyperfiltration in diabetic patients can be promoted with the use of low protein diets and/or angiotensin converting enzyme inhibitors.
Introduction While much attention has been directed toward glucose ‘toxicity’ the possibility of insulin ‘toxicity’ has received little concern. When a diabetic is treated with insulin, and particularly when Date received 29 September 1989 Date accepted 16 February 1990
an attempt is made to drive the blood glucose down to physiologic concentrations, circulating mean insulin concentrations are about twice normal (1 - 2). In other words a physiologic concentration of one substance (glucose) has been achieved at the expense of a very unphysiologic
2 concentration of another substance (insulin). This occurs because at the present time it is impossible to give insulin physiologically. The non-diabetic secretes insulin into the portal circulation where the major fraction acts on the liver, is extracted by it (3) and never gains entry into the systemic circulation. Since the monumental discovery of insulin in 1921, there has not been much good news for diabetics. Our increased understanding of the metabolic abnormalities in the different types of diabetes has not, unfortunately, been translated into meaningful advances in therapy. Even insulin failed to fulfill its early promise. The discovery of the ‘missing factor’ quite naturally gave rise to the expectation that its administration to the diabetic would restore optimum health. It was only after a number of years of insulin treatment that it was appreciated that although insulin prevented death and produced well-being, with time its use was associated with the development of slow but progressively severe disorders of the small and large blood vessels. In spite of every maneuver which has been attempted in an effort to alter the course of diabetic vascular disease, more diabetics develop blindness, manifest coronary heart disease and enter dialysis programs every year (4 - 9). Virtually all of the maneuvers have been directed at lowering the blood glucose is associated with increasing circulating insulin levels and that unphysiologic insulin rather than unphysiologic glucose might be responsible for diabetic vascular disease. Insulin and diabetic microangiopathy epidemiological evidence In the diabetic, small blood vessel disease is most often exhibited as diabetic retinopathy. In the 1930’s it was proposed that elevated blood glucose was responsible for diabetic retinopathy as well as other small blood vessel involvement. In the US the Joslin Clinic was at the forefront of groups emphasizing rigid blood glucose control as a means for preventing or retarding blood vessel problems. In spite of evangelical promotion, the benefits of rigid glucose control were unconvincing and a comprehensive review by Bondy and Felig (10) in the 70’s concluded the case for rigid glucose control was tenuous at best. In the 1980’s it became possible for the first time to conveniently monitor blood glucose concentrations using the newly developed glucose sensitive paper strips. This important innovation led to new-found en-
MEDICAL HYPOTHESES
thusiasm for rigid control of blood glucose and a number of controlled prospective studies were undertaken comparing more rigid with conventional blood glucose control. Although the Job study (11) and the initial Steno (12) results were originally interpreted as showing less retinopathy with rigid control, this interpretation was obviously the result of strong preconviction on the part of the investigators, and both misinterpretations were subsequently corrected (5,13). As additional results became available from the Steno group (5) as well as from other groups conducting similar studies (6,7), it became apparent that the only clear cut difference between the rigidly controlled and conventionally controlled diabetics was the occurrence of accelerated retinal disease which occurred in the rigidly controlled patients soon after rigid control was imposed. Although this finding came as a surprise, such rapid retinal deterioration with rigid control had previously been reported by Daneman (14). The phenomenon of rapid retinal deterioration and its relation to rigid control has now been documented by several other studies (15 - 19). The devastating effect of bringing down glucose to absolutely normal concentrations was emphasized in the case of a 16-year-old boy who received a pancreatic graft which drained into the systemic circulation and who evidenced spectacular retinal deterioration in the first four months after transplantation (20). Both the Steno group and the Kroc group (the other large study comparing rigid and conventional control) are abandoning their undertakings (21) and it seems that rigid control has nothing to offer the diabetic with established disease. Currently a multi-institutional study (The Diabetic Control and Complications study) is being undertaken in the US. In one arm of the study diabetics who are very early in the course of their disease will be treated with rigid control with the hope that extremely early blood glucose control might help to avert vascular disease. This will be a most important venture but the expectation for favorable outcome is fueled more by hope than scientific evidence. High circulating insulin levels are presumably produced by subcutaneous administration in Type II diabetic as well as in the Type I diabetic. In the Type II diabetic however, the blood vessels may be exposed to high insulin levels for many years before treatment is initiated. High fasting and glucose-stimulated insulin levels, despite nor-
INSULIN AND VASCULAR COMPLICATIONS
3
OF DIABETES
mal glucose levels, are seen even in individuals destined to develop Type II diabetes (22). Even higher insulin levels are observed when such individuals begin to develop abnormal blood glucose levels (23). Although retinopathy rarely begins to appear in Type I diabetics before 5 years after insulin therapy is instituted, it is not unusual to observe retinopathy at the time of presentation in Type II diabetics. There is yet another unusual type (or types) of diabetes which has generally been referred to as maturity onset diabetes of the young (24 - 26). Although this is without doubt a heterogenous group (27,28), the most characteristic pattern of abnormalities is the simultaneous occurrence of high blood glucose concentrations with normal insulin concentrations present for decades. This group of diabetics has been noted to be particularly resistant to the development of diabetic retinal disease in spite of very high glucose concentrations for many years (24 - 26). This small sub-group of diabetics would appear to offer an experimental model suggesting that high glucose concentrations in the absence of abnormal insulin concentration is not particularly damaging to small blood vessels. The mechanism by which insulin damages small blood vessels is not apparent, but it has recently been demonstrated that proliferation of the endothelium of microvessels can be accentuated by the addition of insulin to the culture medium (29). Insulin and atherosclerosis Large blood vessel disease in the diabetic is most often manifested by coronary heart disease (30). In a study of approximately 1000 Finnish policemen who were followed for a decade, it was found that myocardial infarction correlated best with a high insulin response to an oral glucose challenge (31). Another study of over 7000 male Parisians found a strong correlation between coronary heart disease and high fasting insulin levels (32). A striking relationship between high insulin concentrations and cardiovascular disease was found in a study comparing a group of normal individuals with Type II diabetics. The Type II diabetics were sub-divided into a group treated with insulin and another group treated with diet and oral hypoglycemics. In the non-insulin groups there was a significant correlation between the incidence of coronary heart disease and C-peptide levels (33). In the insulin treated group the C-peptide levels were obviously lower than in the other two groups
but the incidence of cardiovascular disease was the highest of all. Furthermore, in the insulin treated patients both the dose and plasma insulin concentrations were higher in subjects who developed blood vessel disease over the 5 year study period. Animal and in vitro effects of insulin on atherogenesis and microangiopathy There is firm evidence from animal experiments that insulin is atherogenic. A group of rabbits fed a high cholesterol diet after being made diabetic with alloxan experience a lower incidence of atherosclerosis than non-diabetic controls (34). In chickens insulin has been shown to prevent the regression of atheromatous changes when a high cholesterol diet is changed to a normal diet (35). Insulin infused into one femoral artery of dogs made diabetic with pancreatectomy produced ipsilateral atherosclerosis which was not present in the contralateral saline-infused artery (36). This study implicates insulin itself as a direct cause of atherosclerosis. Recently Sato has shown that the administration of insulin to normal rats for a period of one year produces extensive atherosclerotic changes in their aortas but not in the aortas of saline treated controls (37). Insulin has been shown to stimulate the proliferation and migration of arterial smooth muscle cells in tissue culture preparations. Receptors for the growth promoting effects of insulin on arterial smooth muscle cells differ from the receptors concerned with insulin’s metabolic effects so that even in states of insulin resistance may have a growth promoting propensity on the cells of the arterial wall (38). Hepatic hypoinsulinization and diabetic nephropathy In addition to small and large blood vessel problems, the diabetic is also likely to develop, with time, diabetic kidney disease. Although clinical renal disease occurs in less than half of all diabetics histologic abnormalities are present in the great majority who have had diabetes for many years (39). Although there are some prospective studies reporting benefits from rigid glucose control (40), in others (8) this has failed to influence the progression of nephropathy and the case for benefit for rigid glucose control is unconvincing (41). There is one remarkable report in which diseased kidneys from a diabetic patient were accepted by
4 a renal transplant group in Kuwait (42). Each of the kidneys was transplanted into two separate non-diabetic recipients. Before transplantation, biopsies of each kidney evidenced very significant abnormalities which had improved dramatically when they were re-biopsied approximately six months after transplantation. The authors of the report ascribe the healing process (and improving function) to the normoglycemic environment in the recipients. This interpretation is not, however, in keeping with the studies cited above in which aggressive management of blood glucose in diabetics has produced unconvincing benefits on the progress of nephropathy. A recent publication addressed the effects of pancreas transplantation on glomerular structure in patients with insulindependent diabetes mellitus (43). In this study the pancreatic endocrine secretion was delivered into the systemic circulation via the iliac veins. The authors concluded that pancreas transplantation arrested the progression of mesangial proliferation as assessed by post-transplant histological examination of needle biopsy specimens from the transplanted kidneys. However, previous publications from this same group have emphasized the marked variability in the rate of development of mesangial proliferation in the kidneys of pancreas transplant recipients (44). Although it is tempting to try to ascribe the pathogenesis of diabetic nephropathy to the abnormaly elevated concentrations of insulin acting on the small blood vessels of the kidneys, there is better evidence for another pathogenic mechanism. One of the most consistent findings in diabetics who are early in the course of their treatment is a supfunormaI glomerular filtration rate which presumably leads to nephron damage over time (45). The increased filtration may not be due to systemic hyperinsulinemia but due to another problem related to the unphysiologic way in which insulin secreted by the pancreas does not go directly to the liver in high concentration and therefore the insulin treated diabetic suffers from systemic hyperinsulinemia. His liver, however, is not properly insulinized and this produces production of excessive glomerulopressin and/or other related vasoactive peptides which act on the kidney to produce abnormal elevation of the glomerular filtration rate (46). In contrast to retinopathy the Type I diabetic is much quicker to develop retropathy than the Type II diabetic. Since the classic type I diabetic secretes virtually no insulin into his portal circulation, this would suggest that
MEDICAL HYPOTHESES
proper insulinization of the liver may prevent diabetic nephropathy. Conclusions and future directions The hypothesis set forward here is based on the circumstantial evidence detailed above but it can not be confirmed unless tested directly. How might that be accomplished? Pancreatic grafts are generally not drained into the portal vein. The Roux-en-Y technique we used 15 years ago (47) would allow the venous drainage of the pancreas to be diverted into the portal vein and we are considering offering such a procedure to diabetics with rapidly progressing vascular disease. Attempts to develop a practical oral insulin preparation continue. Saffrom has successfully delivered insulin coated with an azo-copolymer to the colon of dogs (48) and we have accomplished the same thing in humans using a copolymer which only dissolves in the pH range found in the colon (49). Attempts to transplant islets continue (50) and they might some day be grafted into tissue drained by the portal circulation. In the meantime, it would seem that there is enough evidence to abandon the illusory goal of rigid blood glucose control with its attendant dangers and disruptions of the day to day life of the diabetic and concentrate on measures which reduce insulin requirements. Therefore dietary restriction and aerobic exercise should be the keystones of life style changes advised for the Type I as well as for the Type II diabetic. Several studies suggest that diminishing glomerular filtration rate with a very low protein diet and/or ACE inhibitors will delay the progression of diabetic nephropathy (51- 53). References 1. 2.
3. 4. 5.
6.
Rasmussen SM, Heding LG, Parbst E: Serum IRI in insulin-treated diabetic during a 24-hour period. Diabetologia 11: ISI- 158, 1975. Werther GA, Jenkins PA, Turner RC, et al: Twenty-four hour metabolic profiles in diabetic children receiving insulin injections once or twice daily. Br Med J 2: 414 - 418, 1980. Rijjmark S, Bloom G, Chou MCY, Field JB: Hepatic extraction of exogenous insulin and glucagon in the dog. Endocrinology 102: 806-813, 1978. Paterson AD, Dornan TL: Diabetes and renal failure (letter) Lancet 1: 922, 1987. Lauritzen T, Frost-Larsen K, Larsen HW, Deckert T. Steno Study Group: Effect of 1 year of near-normal blood glucose levels on retinopathy in insulin-dependent diabetics. Lancet 1: 200-204, 1983. Brinchmann-Hansen 0, Dahl-Jorgensen K, Hanssen KF,
INSULIN AND VASCULAR COMPLICATIONS
I.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22. 23.
24.
OF DIABETES
Sandvilc L, Oslo Study Group: Effects of intensified insulin treatment on various lesions of diabetic retinopathy. Am J Ophthalmol loo: 644-653, 1985. The Kroc Collaborative Study Group: Blood glucose control and the evolution of diabetic retinopathy and albuminuria: a preliminary multicenter trial. N Eng J Med 311: 365 - 372, 1984. Viberti GG, Bilous RW, Mackintosh D, Bending JJ, Keen H: Long-term correction of hyperglycaemia and progression of renal failure in insulin dependent diabetes. Br Med J (Clin Res) 286: 598 - 602, 1983. Pyorala K: Relationship of glucose tolerance and plasma insulin to the incidence of coronary heart disease: Results from two population studies in Finland. Diabetes Care 2: 131- 141, 1979. Bondy PK. Felig P: Relation of diabetic control to the development of vascular complications. Med Clin North Am 55: 889-897, 1971. Job D, Eschwege E, Guyot-Argenten C et al: Effect of multiple daily insulin injections on the course of diabetic retinopathy. Diabetes 25: 463 - 469, 1976. Steno Study Group: Efect of six months of strict metabolic control on eye and kidney function in insulindependent diabetics with background retinopathy. Lancet 1: 121- 124, 1982. Ashikaga T, Borodic G, Sims EA: Multiple daily insulin injections in the treatment of diabetic retinopathy. The Job study revisited. Diabetes 27: 592 - 596, 1978. Daneman D, Drash AL, Lobes LA, Becker DJ, Baker LM, Travis LB: Progressive retinopathy with improved control in diabetic dwarfism (Mauriac’s syndrome). Diabetes Care 4: 360- 365, 1981. Van Ballegooie E, Hooymans JMM, Timmerman Z et al: Rapid deterioration of diabetic retinopathy during treatment with continuous subcontinuous insulin infusion. Diabetes Care 7: 236 - 242, 1984. Beak-Neilsen H, Richelsen B, Morgensen CE et al: Effect of insulin pump treatment for one year on renal function and retinal morphology in patients with IDDM. Diabetes Care 8: 585-589, 1985. Kelly TM, Sanborn GE, Haug PJ, Edwards CQ: Effect of insulin infusion pump use on diabetic retinopathy. Arch Ophthalmol 102: 1156- 1159, 1984. Friberg TR, Rosenstock J, Sanborn G, Vaghefi A, Raskin P: The effect of long-term normal glycemic control on mild diabetic retinopathy. Ophthalmology 92: 1051- 1058, 1985. Phelps RL, Sakol P, Metzger BE, Jampol LM, Freinkel N: Changes in diabetic retinopathy during pregnancy: correlations with regulation of hyperglycemia. Arch Ophthalmol 104: 1806- 1810, 1986. Ramsey RC, Goetz PC, Sutherland DER, Mauer SM, Robinson LL, Cantrill HL, Knobloch WH, Najarian JS: Progression of diabetic retinopathy after pancreas transplantation for insulin dependent diabetes mellitus. New Eng J Med 318: 208 -214, 1988. Lauritzen T and Sherwin R - personal communications. Porte DJ, Bagdade JD: Human insulin secretion: An integrated approach. Annu Rev Med 21: 219- 240, 1970. Saad MF. Knowles WG, Pettitt DJ, Nelson RG, Mott DM, Bennett PH: Sequential changes in serum insulin concentration during development of non-insulin-dependent diabetes. Lancet 1: 1356- 1358, 1989. Tattersall RB: Mild familial diabetes with dominant inheritance. Q. J. Med 43: 339-357, 1974.
5
25. Tattersall RB, Fajans SS: A difference between the inheritance of classical juvenile onset and maturity onset type diabetes of young people. Diabetes 24: 44- 53, 1975. 26. Tattersall RB: The present status of maturity onset type of diabetes of the young people (MODY). In Genetics of Diabetes Mellitus. Kobberling J and Tattersall RB (Eds). London and New York, Academic Press, 1982: 1 - 19. 27. Fajans SS: Heterogeneity between various families with non-insulin dependent diabetes of the MODY type. In Genetics of Diabetes Mellitus. Kobberling J and Tattersall RB (Eds). London and New York, Academic Press. 1982: 251-260. 28. Fajans SS: MODY - A model for understanding the pathogenesis and natural history of type II Diabetes. Horm Metab Res 19: 591- 599, 1987. 29. Bar RS, Siddle K, Dolash S, Bees M, Drake M: Actions of insulin-like growth factors I and II in cultured microvessel endothelial cells from bovine adipose tissue. Metab 37: 714-720, 1988. 30. Jarrett RJ, McCartney P, Keen H: The Bedford Survey. Ten year mortality rates in newly diagnosed diabetics and normglycaemic controls and risk indices for coronary heart disease in borderline diabetics. Diabetologia 22: 79 - 86, 1982. 31. Pyorala K, Savolainen E, Kaukola S, Haapakoski J: High plasma insulin as a coronary disease risk factor. In: Eschwege E (ed): advances in diabetes epidemiology. INSERM Symposium No 22. Elsevier, Amsterdam 143, 1982. 32. Eschwege E, Richard JL, Thibult N, Ducimetiere P. Warnet JM, Claude JR, Rosselin GE: Coronary heart disease mortality in relation with diabetes, blood glucose and plasma insulin levels: the Paris prospective study, ten years later. Horm Metab Res (Suppl 15): 41-46, 1985. 33. Stand1 E, Janka HV: High insulin serum concentrations in relation to other cardiovascular risk factors in diabetes: Results of a prospective study. Horm Metab Res 17: 63 -68, 1985 (suppl) 34., Duff GL, Brechin DJH, Finkelstein WE: The effect of alloxan diabetes on experimental cholesterol atehrosclerosis in the rabbit: I. The inhibition of experimental cholesterol atherosclerosis in alloxan diabetes. II. The effect of alloxan diabetes on the retrogression of experimental cholesterol atherosclerosis. J Exp Med 89: 611 -630, 1949. 35. Stamler J, Pick R, Katz LN: Effect of insulin in the induction and regression of atherosclerosis in the chick. Circ Res 8: 572 - 576, 1960. 36. Cruz AB, Amatuzio DS, Grande F et al: Effect of intraarterial insulin on tissue cholesterol and fatty acids in alloxan-daibetic dogs. Circ Res 9: 39-43, l%l. 37. Sato Y, Shiralshi S, Oshida T, Sakomoto N: Experimental athrosclerosis-like lesions induced by hyperinsulinism in Wistor rats. Diabetes 38: 91- %, 1989. 38. Stout RW, Bierman FL, Ross R: Effect of insulin on the proliferation of primate arteriol smooth muscle cells. Cir Res 36: 319-327, 1975. 39. McCray RF, Pitts TO, Puschett JB: Diabetic nephropathy: Natural course, survivorship and therapy. Am J Nephrol 1: 206-218, 1981. 40. Holman RR, Dornan TL, Maynon-White V et al: Prevention of deterioration of renal and sensory-nerve function by more intensive management of insulin-dependent daibetic patients: a two year randomized prospective study. lancet 1: 204-208, 1983.
MEDICAL HYPOTHESES
6 41.
42.
43.
44.
45.
46.
47.
The DCCT Research Group: Are continuing studies of metabolic control and microvascular complications in insulindependent diabetes justified? The Diabetes Control and Complication Trial. N. Eng J Med 318: 246 - 250, 1988. Abouna GM, Dremer GD, Daddah SK, Al-Adanani MS, Kumar SA, Kusma G: Reversal of diabetic nephropathy in human cadeveric kidneys after transplantation into nondiabetic recipients. Lancet 2: 1274- 1276, 1983. Bilous RW. Mauer SM, Sutherland DER, Najarian JS, Goetx FC, Steffes Mw: The effects of pancreas transplantation on the glomerular structure of renal ahografts in patients with insulin-dependent diabetes. N Eng J Med 321: 80-85, 1989. Mauer SM. Goetz FC, McHugh LE et al: Long-term study of normal kidneys transplanted into patients with type I diabetes. Diabetes 38: 516-523, 1981. Brenner BM: Hemodynamically mediated glomerular injury and the progressive nature of kidney disease. Kidney Int. 23: 647-655, 1983. Del Castillo ER, Fuenxalida R. Uranga J: Increased glomerular filtration rate and glomemlopressin activity in diabetic dogs. Horm Metab Res 9: 46 - 53, 1977. Connolly JE, Martin DC, Steinberg T, Gwinup G.
48.
49.
50.
51.
52. 53.
Garzaniga AB, Bartlett RH: Clinical experience with pancreatico-duodenal transplantation. Arch Surg 106: 489 - 494, 1973. Saffran M, Field JB, Pena J, Jones RH, Okuda Y: Oral Delivery of insulin in pancreatectomized diabetic dogs (abstract). Program of the Am Diabetes Assoc Detroit Mich. Diabetes 38 (suppl2): pp 81A, 324, 1989. Gwinup G, Elias AN, Domurat ES: Insulin and C-peptide levels following oral administration of insulin in intestinal-enzyme protected capsules. Clin Res 38: 8A. 1990. Mullen Y, Clare-Salzler M, Stein E, Clark W: Islet transplantation for the cure of diabetes. Pancreas 4(l): 123 - 135, 1989. Taguma Y, Kitamovo Y, Futaki G: Effect of captopril on heavy proteinuria in axotemic diabetes. N Eng J Med 313: 1717- 1620, 1985. Wiseman MJ, Dodds R, Bending JJ et al: Dietary protein and the diabetic kidney. Diabetic Med 4: 144- 146, 1987. Evanoff GV, Thompson CS, Brown J et al: The effect of dietary protein restriction on the progression of diabetic nephropathy: A 1Zmonth follow-up. Arch Intern Med 147: 492-495, 1987.
