The major morbidity and mortality of diabetes mellitus is a consequence of the chronic complications of the disease. Therefore, of fundamental importance in diabetes is the relationship of the chronic complications of the disease to the deranged metabolism (hyperglycemia, insulinopenia, or associated metabolic alterations). If the development of chronic complications of diabetes is either a consequence of, or in large part influenced by, the metabolic derangements that characterize the diabetic state, then the degree of metabolic control achieved by therapy might be expected to influence the development of chronic complications. On the other hand, if the chronic complications of diabetes are determined to be independent of the disordered metabolism, other strategies would be required to alter outcome.
The chronic complications of diabetes include: accelerated vascular disease, neurological deficits (diabetic neuropathy), and other organ-specific degenerative processes. The vascular disease is of two types: 1) microangiopathy - characteristic capillary disease, associated more or less specifically with diabetes mellitus, and clinically manifested principally in the retina (diabetic retinopathy) and kidney (diabetic nephropathy); 2) microangiopathy - accelerated atherosclerotic disease of arteries, clinically manifested principally in coronary arteries, cerebral vasculature, and peripheral vessels in the lower extremities. Pathologically, macroangiopathy (atherosclerosis) in diabetes differs from that seen in nondiabetic individuals only quantitatively, not qualitatively. The most characteristic pathological feature of diabetic microangiopathy is thickening of capillary basement membranes. This may be generalized, segmental, or focal in character. Although such thickening is evident in capillaries throughout the body, including muscle and skin, it is clinically manifested primarily by thickening of renal glomerular basement membranes (diabetic nephropathy) and retinal capillary basement membranes (diabetic retinopathy). Although diabetes mellitus increases the risk of macroangiopathy, the relationship between deranged glucose metabolism and macroangiopathy is complicated by many other factors which influence the atherosclerotic process. Therefore, this discussion will focus on the microangiopathic and neurologic complications of diabetes. The available evidence favors the thesis that complications are secondary to the metabolic derangements.1"7 The evidence in support of this comes from epidemiological, clinical, and pathological studies involving human beings, as well as studies using animal models of diabetes. Additional support for the thesis is gained by the fact that a number of biochemical mechanisms are influenced directly by hyperglycemia. In this article, we will briefly consider a few representative studies involving human subjects.
STENO STUDY- FUNDUS PHOTOS AFTER ONE YEAR (VS BASELINE)
With regards to microangiopathy and neuropathy, a number of studies have reported correlations between degree of hyperglycemia (variously assessed) and frequency, severity, and rate of progression of complications. Some have noted a greater correlation between frequency of retinopathy and hyperglycemia during the first few years after diagnosis of diabetes than between retinopathy and control during later years.8-10
One recent epidemiological survey found higher levels of glycosylated hemoglobin in subjects with various angiopathic markers of diabetic retinopathy than in subjects without those markers.11 This is an example of newer studies using modern methods to assess integrated glycémie control (glycosylated hemoglobin) and quantitative evaluation of complications (angiopathy). In a large survey of diabetic children in Berlin, Weber et al reported a correlation between glycosylated hemoglobin levels and presence of retinopathy.12
Clinical studies of the relationship between diabetic control and diabetic complications have long been controversial.13'15 Besides the need for long-term, prospective, randomized studies, it is also necessary to carefully define eligibility criteria to obviate concern that some confounding feature may influence the results. One concern raised in the past about retrospective or nonrandomized studies is that some diabetic patients may be more prone to complications, and that these may also be patients with greater difficulty in achieving diabetic control.
Retrospective studies are usually difficult to interpret because even the investigator cannot fully assess confounding variables and other factors. The retrospective study of Johnsson is noteworthy because he tried to assess all patients in Malmo, Sweden with diabetes diagnosed before age 40, using two contrasting treatment groups.16 Series 1 consisted of patients diagnosed between 1922 and 1935, when treatment included strict dietary control and multiple daily injections of insulin. Series 2 consisted of patients diagnosed between 1936 and 1945, when control standards were relaxed and treatment included a less restricted diet and a single daily injection of longacting insulin. Surprisingly, patients in Series 1 had less nephropathy (18 of 56 patients [32%]) than patients in Series 2 (56 of 104 patients [54%]), despite the longer disease duration in Series 1 (24.5 years versus 15.9 years). The difference was even more obvious when only subjects with diabetes of greater than 15 years duration were compared: 5 of 56 (9%) of Series 1 patients had nephropathy, compared to 35 of 57 (61%) of Series 2 patients. Similar findings were noted for the occurrence of retinopathy.
An important prospective but nonrandomized study is the Brussels study of Jean Pirart.17 Pirart followed 4400 patients prospectively from 1947 to 1973, including a large inception cohort. He found a strong relationship between incidence and prevalence of complications and duration of disease. Poor control assessed cumulatively over the years was associated with a higher prevalence and incidence of microangiopathy and neuropathy, especially severe retinopathy. Pirart also found that the annual incidence of microangiopathy and neuropathy was clearly related to glycémie control during the preceding year, whatever the prior degree of control.
The Steno study (from the Steno Hospital in Gentofte, Denmark) involved 30 patients with type I diabetes and advanced background diabetic retinopathy, who were randomized to either unchanged conventional insulin therapy (UCIT) or to continuous subcutaneous insulin infusion (CSII).18"20 With treatment, the mean blood glucose and glycosylated hemoglobin were significantly lower in the CSII group than in the UCIT group. This statistical difference was maintained throughout two years of follow-up. Three functional retinal parameters were assessed - posterior vitreous fluorophotometry, macular recovery time, and oscillatory potential. After only six months, in the CSII group, there was decreased fluorescein leakage, improved macular recovery time, and increased oscillatory potential. In contrast, in the UCIT group, there was deterioration in all three parameters. These differences were sustained throughout the two years of the study, with further deterioration in the UCIT group. Retinal morphology at the start of the study did not differ significantly between the two groups. Yet, at one year, the Steno group reported the surprising finding that the CSII group showed greater deterioration of retinopathy than the UCIT group. 19 Moreover, of the ten subjects in the CSII group with the best control, there was a greater risk of deterioration. The deterioration was particularly characterized by the appearance of soft exudates (cotton wool spots), which represent areas of retinal infarction. The results are displayed in Table 1.
However, when the two year retinal photos were compared with baseline, there was a marginally significant trend toward more frequent improvement in retinopathy in the CSII group, with continued deterioration in the UCIT group.20 This indicates an improvement in morphology, and may represent a clearing of the soft exudates. The soft exudates may have been a consequence of rapid improvement in control and thus a temporary deterioration, more apparent than real (in terms of true retinopathy). The results at two years are shown in Table 2. However, the Steno study found that some subjects in both groups (4 in CSIl, 5 in UClT) did progress to proliferative retinopathy during the two years of the study. This occurred in CSII patients in spite of excellent control.
The Kroc collaborative multicenter study involved 65 patients with type I diabetes and mild to moderate background diabetic retinopathy, who were randomized to either conventional insulin treatment (CIT) or to continuous subcutaneous insulin infusion (CSII).21'25 With treatment, the mean blood glucose and glycosylated hemoglobin were significantly lower in the CSII group than in the CIT group. This statistical difference was maintained throughout 8 months of follow-up. All subjects showed a worsening of retinopathy over time. In this study, too, the worsening was greater in the CSII group (15 of 32) than in the CIT group (9 of 33). 21'23 The best single predictor of worsening retinopathy was lower plasma glucose. Again, worsening was associated with the appearance of retinal infarcts (soft exudates), as well as intraretinal microvascular abnormalities (IRMA). In this study, too, however, a two year follow-up of patients (who were no longer maintained on the original study protocol) revealed stabilization of retinopathy in the CSII group, and continued deterioration in the CIT group.24"25 This, again, is consistent with but a transient worsening of retinopathy accompanying rapid improvement in control.
Similar findings were seen in the Aker study in Oslo, Norway, which involved randomization of 45 subjects with type I diabetes and background retinopathy into three groups - conventional insulin therapy (CIT), ie, twice daily injections; multiple injections (MI); and CSII.26"28 Again, there was initial deterioration of retinopathy with rapidly improved control (soft exudates), and subsequent stabilization in the multiple injection and CSII groups, and deterioration in the twice daily injection group.26'27 Deterioration was particularly related to a large and rapid decline in plasma glucose.27 Other features of patients who developed soft exudates included a longer duration of diabetes, more severe retinopathy at the outset, and more frequent episodes of hypoglycemia during treatment. However, two year follow-up data reported progression of retinopathy (from baseline) in the CIT group, but not in the CSII or MI groups.28
STENO STUDY- FUNDUS PHOTOS AFTER TWO YEARS (VS BASELINE)
The Aarhus (Denmark) group has initiated asimilar study, in which 24 subjects with type I diabetes were randomly allocated to conventional insulin therapy (CIT) or CSII.29 In contrast to the Steno, Kroc, and Aker studies, their subjects had minimal or no retinopathy at the outset. Their one year data showed minimal progression of retinopathy in four CIT subjects and three CSII subjects, with no statistical difference between the groups. However, they did not observe transient worsening of retinopathy, or the appearance of soft exudates, as seen in the three studies involving subjects with more severe retinopathy at baseline.
The Oxford study involved random assignment of 174 patients with insulin treated diabetes and background retinopathy into two groups - usual diabetic care (group U) and more intensive care (group A).30 Group A achieved lower glycosylated hemoglobin levels and had a trend toward decreased progression of retinopathy.
The diabetes control and complications trial (DCCT) is a multicenter study (27 centers in North America), which will enroll nearly 1500 patients with type I diabetes and minimal or no retinopathy, and follow these subjects for 8 to 10 years.31 They are randomly allocated either to experimental treatment involving either CSII or multiple injections, or to standard treatment involving one to two daily injections. Preliminary data from this trial show that statistically significant differences in mean blood glucose and glycosylated hemoglobin levels can be achieved and maintained over time. 32 It will be many years until the results of this study emerge, but it represents a model in terms of experimental design.
A number of small randomized studies have examined the influence of intensive control, usually with CSIl, versus conventional therapy on various parameters of renal function. In the Steno and Kroc studies of retinopathy, in CSII patients there was a decreased rate of albumin excretion in subjects with microalbuminuria at baseline.18,21,33
A study from Guy's Hospital randomized 12 patients with normal albumin excretion rates, but increased glomerular filtration rates, to CSII or UCIT.34 They found that improved control in the CSII group resulted in normalization of glomerular filtration rate, but persistence of increased renal size. In another study of 12 patients with intermittent clinical proteinuria, they found that improved control had no impact on renal function.35
A second study from the Steno group randomized to either UCIT or CSII 36 patients with type I diabetes who had "incipient nephropathy," defined as elevated urinary albumin excretion rates (ie, microalbuminuria) in two of three 24 hour urine specimens.36 At one year, they found no change in albumin excretion rate or glomerular filtration rate, although they did see a reduction in renal size in the CSII group. By two years, they reported the appearance of clinical diabetic nephropathy in five subjects in the UCIT group, but none in the CSII group. 37 In addition, they found that fractional albumin excretion had increased in the UCIT group but not in the CSII group. Diastolic blood pressure also rose in the UCIT group. Elevated glomerular filtration rate was normalized in the CSII group. Thus, with two years of strict control, progression of diabetic renal disease was arrested.
Collectively, these studies suggest that there may be an early phase of nephropathy in which renal function may be influenced by control, but that later in the course of the disease, this may not be the case. This conclusion must be tempered with the caveat that all of the studies to date have involved small numbers of subjects followed for short durations. More studies are needed.
Neurological endpoints have been included in several of the prospective trials of retinopathy and nephropathy. The endpoint most commonly used has been nerve conduction velocity, shown to be a sensitive and reliable indicator of neuropathy.38 In the Steno study, cardiovascular autonomic neuropathy deteriorated in the UCIT group, while remaining stable in the CSIl group.18 In the Aker study, after two years, motor nerve conduction velocity had improved in the CSII group, while deteriorating in the CIT group.28 In addition, in a very small series from the Mayo Clinic, there was small but significant improvement in a number of neurological parameters on CSII but not on conventional therapy.39
Pathological studies have focused primarily on muscle capillaries and renal pathology. Although microangiopathic changes are clinically reflected principally in the retina and kidney, they are evident throughout the body, including muscle and skin capillaries. Muscle capillaries have been used because muscle (usually quadriceps) is a tissue readily accessible for repeated biopsies; most studies have suggested that muscle may be used to gain insight into capillary changes throughout the body.
Several groups have shown that capillary basement membrane (CBM) thickening is related to duration of carbohydrate intolerance and severity of diabetes.40'42 Studies by Jackson's group clearly show a relationship between severity of diabetes in children, as reflected by degree of metabolic control, and CBM width, consistent with a metabolic basis for CBM changes.43 Sosenko et al have found a relationship between CBM width and metabolic control in postpuberal subjects, but do not find such a relationship in prepubertal subjects, in whom CBM changes are trivial.44
In several (but not all) studies of identical twins (or triplets) discordant for diabetes, the diabetic twins had increased CBM width in comparison with their nondiabetic co-twins.4549 In such studies, differences in age, gender, and confounding genetic factors are all eliminated. Thus, these findings are consistent with CBM thickening as a consequence of, or influenced by, the metabolic derangement. The data argue strongly against an independent genetic basis for CMB changes.
Several investigators have noted that with improved glycemic control, there can even be thinning of thickened CBMs.50"53 The largest such report is a prospective nonrandomized study of patients treated with CSlI versus conventional treatment.52'53 In the CSII patients, glycosylated hemoglobin markedly improved and a progressive reduction in CBM width was seen. In contrast, in conventionally treated patients, there was no change in either glycosylated hemoglobin levels or CBM width.
Osterby and coworkers have examined renal biopsy specimens in children with diabetes. 54'59 They have found: 1) normal biopsies at onset of the disease; 2) very slight glomerular changes within the first few months of onset; 3) increased prevalence of changes after 3 to 5 years duration, with the degree of abnormality a clear function of duration of disease; 4) progressive changes in serial biopsies of individual patients, suggesting that this progression is more rapid in patients less well controlled.
Steffes and coworkers evaluated biopsy specimens from seven pairs of identical twins who were discordant for type I diabetes.49 AU of the twins of the diabetic patients had normal glomerular basement membrane widths and normal fractional volumes of the glomerular mesangium. Values for glomerular basement membrane width, tubular basement membrane width, and mesangium volume in each diabetic twin exceeded the values in the respective sibling, even when the value in the diabetic twin was within established normal ranges. These observations suggest that the metabolic abnormalities of diabetes are necessary (although not sufficient) for the development of glomerular abnormalities.
Studies of transplanted kidneys have permitted observations of the influence on renal structure of the metabolic milieu in which the kidney exists. Mauer and coworkers have demonstrated that typical diabetic glomerular changes and immunohistological changes specific for diabetes invariably develop in normal kidneys transplanted into patients with diabetes, usually within two years of transplantation.60'61 Comparable changes are not seen in kidneys transplanted into nondiabetic individuals. In addition, in a single patient reported by Gliedman et al, renal histology remained normal four years after combined renal and pancreatic transplantation, suggesting that the maintenance of near-normal metabolic function by the pancreatic transplant may have contributed to the prevention of typical diabetic changes in the kidneys.62 Bohman et al have reported a similar experience in two cases.63 Also, in two patients reported by Sutherland et al, pancreas transplantation resulted in regression of recurrent early diabetic nephropathy in kidney grafts transplanted several years earlier.64
Abouna and coworkers have reported dramatic changes in two kidneys taken from a diabetic donor and transplanted into nondiabetic recipients.65'66 At the time of transplantation, the kidneys showed typical histological features of diabetic nephropathy. The renal changes in the donated organs were reversed on biopsies taken seven months post-transplantation, during which the recipients remained euglycemic.65 Fifteen months after transplantation, one of the recipients developed hyperglycemia, and required initiation of insulin therapy. One year later, he developed clinical evidence of renal disease, and on rebiopsy 30 months after his initial transplant,66 the kidney showed recurrence of diabetic nephropathy.
The evidence for a relationship between glucose control and diabetic complications is overwhelming. There appears to be no doubt that hyperglycemia (or other associated metabolic derangements) is necessary for the emergence of diabetic microangiopathy and diabetic neuropathy. Yet, it is possible that other factors may influence the onset, severity, or progression of these complications. One such factor is coexisting hypertension, which adversely influences both retinopathy and nephropathy. Other factors likely exist as well.
An implication that follows from the above conclusion is that the degree of metabolic control achieved by therapy should be expected to influence the development of these complications, and that normalization of metabolic aberrations should prevent, delay, or substantially reduce the severity of these complications. This appears to be the case, from the available data, but has not been established with certainty through long-term prospective randomized studies.
The clinical corollary of such implications is that we should expend our best efforts in devising and implementing therapeutic strategies that most closely approximate metabolic normality. This will involve appropriate patient selection, focusing on those with longest life expectancy and avoiding those with risk factors for devastating problems consequent to hypoglycemia.
Unfortunately, we do not yet know the degree of control necessary to minimize the risk of complications while avoiding an unacceptable risk of hypoglycemic consequences. Therefore, it is not yet possible to establish treatment goals with certainty. Nevertheless, it seems a prudent approach to strive for the best control reasonably attainable in any given patient, while awaiting the results of the studies necessary to confirm the above implications.
Figure 1. Fundus photograph showing early background diabetic retinopathy (a few scattered microaneurysms).
Figure 2. Fundus photograph showing proliferative diabetic retinopathy with peripheral (away from optic disc) neovascularization. This eye does not nave high risk characteristics for photocoagulation.
Figure 3. Fundus photograph showing advanced diabetic eye disease. This eye is a candidate for vitrectomy.
Figure 4. Fundus photograph showing involuted diabetic retinopathy with residual fibrous proliferation, but no active neovascularization or hemorrhages.
1 . Skyler JS: Complications of diabetes mellitus: Relationship to metabolic dysfunction. Diabetes Core 1979; 2:499-509.
2. Tchobroutsky G: Relation of diabetic control to development of microvascular complications. Diabetologa 1978; 15:143-152.
3. Brownlee M, Canili GF: Diabetic control and vascular complications. Arteriosclerosis Review 1979; 4:29-70.
4. Sherwin RS, Tamborlane WV: Metabolic control and diabetic complications, in Olefsky JM, Sherwin RS (eds): Diabetes Mellitus - Management and Complications. New York. Churchill Livingstone. 1985. pp 1-29.
5. Raskin P, Rosenstock J: Blood glucose control and diabetic complications. Ann Intern Med 1986: 105:254-263.
6. Leslie ND, Sperling MA: Relation of metabolic control to complications in diabetes mellitus. Pediatrics 1986; 108:491-497.
7. Hanssen KF, Dahl-Jorgensen K, Laurirzen T, et al: Diabetes control and microvascular complications: The near-nomoglycemic experience. Diabetalogia 1986; 29:677-684.
8. Burditt AF, Caird Fl, DraperGJ: The natural history of diabetic retinopathy. QJ Med 1968; 303-317.
9. Caird FI: Control of diabetes and diabetic retinopathy, in Goldberg M P, Fine SL (eds): Symposium on the Treatment nf Diabete Reimnpany. Washington, USPHS, 1969, pp 107-I14.
10. Constarti GR: Zur Spatprognose des Diabetes Mellitus. HeIv Med Acta 1965; 32:287-306.
11. Doft BH, Kingsley LA, Orchard TF, et al: The association between long-term diabetic control and early retinopathy. Ophthalmology 1984; 91:763-768.
12. Weber B, Burger W, Hartmann R , et ah Risk factors for the development of retinopathy in children and adolescents with type I (insulin-dependent) diabetes mellitus. Diabetologie 1986; 29:23-29.
13. Canili GF. Etzwilet DD. Freinkel N: Blood glucose control in diabetes. N Engl ] Med 1976; 294:1004-1005.
14. Siperstein MD, Foster DW, Knowles HC, et al: Control of blood glucose and diabetic vascular disease. N Engl J Med 1977; 296:1060-1063.
15. lngelfinger FJ: Deban» on diabetes. N Engl J Med 1977; 296:1228-1230.
16. Johnsson S: Retinopathy and nephropathy in diabetes mellitus: Comparison of the effects of two forms of treatment. Diabetes 1960; 9:1-8.
17- Pirart J: Diabetes mellitus and its degenerative complications: A prospective study of 4400 patients observed between 1947 and 1973. Diabetes Care 1978; 1:168-188. 252-263.
18. Steno Study Group; Effect of 6 months of srrict metabolic control on eye and kidney function in insulin-dependent diabetics with background retinopathy. Lincei 1982; 1:121-124.
19. Lauritzen T, Frost-Larsen K, Larsen HW, et al: Effect of 1 yeat of near-normal blood glucose levels on retinopathy in insulin-dependent diabetes. Lancet 1983; 1:200-204.
20. Lauritzen T, Frost-Larsen K, Larsen HW, et al: Two-year experience with continuous subcutaneous insulin infusion in relation to retinopathy and neuropathy. Diabeies 1985; 34(suppl 3):74-79.
21. Kroc Collaborative Study Group: Blood glycose concrol and the evolution of diabetic retinopathy and albuminuria. N Engl J Med 1984; 311:365-372.
22. Rodger NW (ed): Proceedings of a conference on insulin pump therapy in diabetes: Multicenter study of effect on microvascular disease. Diabetes 1985; 35(suppl 3):1-91.
23 . Canny CLB, Kohner EM, Trautman J, et al: Comparison of stereofundus photographs in patients with insulin-dependent diabetes during conventional insulin treatment or continuous subcutaneous insulin infusion. Diäteres 1985; 35(suppl 3):50-55.
24. Kroc Collaborative Study Group: The Kroc study patients at two years: A report on further retinal changes. Diatetes 1985; 34(suppl 1):39A.
25. Kroc Study Group: Effect of diabetic control On retinopathy: Follow-up report of the Kroc randomized clinical trial. Invest Ophthalmol Vis Sci 1985; 26(suppl 1):85.
26. Dahl-Jorgensen K, Brinchmann-Hansen O1 Hanssen KF, et al: Near-normoglycemia retards the progression of early diabetic retinopathy and neuropathy. Br Med J 1985; 290:811-815.
27. Brinchmann-Hansen Q, Dahl-Jorgensen K, Hanssen KF, et al: Effects of intensified insulin treatment on various lesions of diabetic retinopathy. AmJ Ophthalmol 1985; 100:644-653.
28. Dahl-Jorgensen K, Brinchmann-Hansen O, Hanssen KF, et al: Effect of near normoglycemia for two years on progression of early diabetic retinopathy, nephropathy, and neuropathy: The Oslo study. Br Med) 1986; 293:1195-1199.
29. Beck-Nielsen H, Richelsen B, Mogensen CE, et al: Effect of insulin pump treatment for one year on renal function and retinal morphology in patients with IDDM. Diabetes Core 1985; 8:585-589.
30. Holman RR. Doman TL, Mayon- White V1 et al: Prevention of deterioration of renal and sensory-nerve function by more intensive management of insulin-dependent diabetic patients. Lancet 1983; 1:204-208.
31. The DCCT Research Group: The diabetes control and complications trial (DCCT). Design and methodologie considerations for the feasibility phase. Diabetes 1986; 35:530-545.
32. The DCCT Research Group: Diabetes control and complications trial (DGCT): Results of feasibility study. Diabetes Care 1987; 10:1-19.
33. Bending Jj, Viberti GC, Bilous RW, et al: Eight-month correction of hyperglycemia in insulin-dependent diabetes mellitus is associated with a significant and sustained reduction of urinary albumin excretion rates in patients with microalbuminuria. Diabetes 1985; 35(suppl 3):69-73.
34. Wiseman MJ, Saunders AJ, Keen H, et al: Effect of blood glucose control on increased glomerular filtration rate and kidney size in insulin-dependent diabetes. N Engl J Med 1985; 312:617-621.
35. Bending JJ, Viberti GC, Watkins PJ, et al: Intermittent clinical proteinuria and renal function in diabetes: Evolution and the effect of glycémie control. Br Med J 1986; 292:83-86.
36. Fekk-Rasmussen B, Mathiesen ER, Hegedus L, et ah Kidney function during 12 months of strict metabolic control in insulin-dependent diabetic patients with incipient nephropathy. N Engl J Med 1986; 314:665-670.
37. Feldt-Rasmussen B, Mathiesen ER, Decken T: Effect of two years of strict metabolic control on the progression of incipient nephropathy in insulin-dependent diabetes. Lancet 1986; 2:1300-1304.
38. Dyck PJ, Karnes JL, Daube J, et al: Clinical and neuropathological criteria for the diagnosis and staging of diabetic polyneuropathy. Brom 1985; 708:861-880.
39. Service FJ, Rizz RA, Daube JR, et al: Near normoglycemia improved nerve conduction and vibration sensation in diabetic neuropathy- Diobetoiogia 1985; 28:722-727.
40. Danowski TS, Fisher ER, Khurana RC, et al: Muscle capillary basement membrane in juvenile diabetes mellitus. Metabolism 1972; 21:1125-1132.
41. Pardo V, Perez-Stable E, Alzamora DB, et al: Incidence and significance of muscle capillary basal lamina thickness in juvenile diabetes. AmJ Pathol 1972; 68:67-77.
42. Seiss EA, Nathke HE1 Dexel T, et al: Dependency of muscle capillary basement membrane thickness on duration of diabetes. Diabetes Care 1979; 2:472-478.
43. Jackson RL, Ide CH, Guthrie RA, et al: Retinopathy in adolescents and young adults with onset of insulin-dependent diabetes in childhood. Ophthalmology 1982; 89:7-1 3.
44- Sosenko JM. Miettinen OS, Williamson JR, et al: Muscle capillary basementmembrane thickness and long-term glycemia in type I diabetes mellitus. N Enfi J Med 1984; 311.694-698.
45 . Karam J, Rosenthal M , O'Donnell J, et al: Discordance of diabetic microangiopathy in identical twins. Diabetes 1976; 25:24-28.
46. Ganda OP, Soeldner JS, Gleason RE1 et al: Monozygotic triplets with discordance for diabetes mellitus and diabetic microangiopathy. Diabetes 1977; 26:469-479.
47. Ganda OP, Williamson JR, Soeldner JS, et al: Muscle capillary basement membrane width and its relationship to diabetes mellitus in monozygotic twins. Diabetes 1983; 32:549-556.
48. Bamett AH, Spiliopoulos AJ1 Pyke DA, et al: Muscle capillary basement membrane in identical twins discordant for insulin-dependent diabetes. Diabetes 1983; 32:857-864.
49. Steffes MW, Sutherland DER, Goetz FC, et al: Studies of kidney and muscle biopsy specimens from identical twins discordant for type I diabetes mellitus. N Engl J Med 1985; 312:1282-1287.
50. Peterson CM, Jones RL1 EsterlyJA, et al: Changes in basement membrane thickening and pulse volume concomitant with improved glucose control and exercise in patients with insulin-dependent diabetes mellitus. Diabetes Care 1980; 3:586-589.
51. Danowski TS, Ohlsen P, Fisher ER: Diabetic complications and their prevention or reversal. Diabetes Care 1980; 3:94-99.
52. Raskin P, Pietri AO, Unger R, et ah The effect of diabetic control on the width of skeletal-muscle capillary basement membrane in patients with type I diabetes mellitus. N Engt) Med 1983; 309:1546-1550.
53. Rosenstock J, Friberg T, Raskin P: Effect of glycémie control on microvascular complications in patients with type I diabetes mellitus. AmJ Med 1986; 81:1012-1018.
54. Mogensen CE, Osterby R, Gundersen HJG: Early functional and morphologic vascular renal consequences of the diabetic state. Diobetoiogia 1979; 17:71-76.
55. Osterby R: Morphometric studies of the peripheral glomerular basement membrane in early juvenile diabetes. Diobetoiogia 1972; 8:84-92.
56. Osterby R: Kidney structural abnormalities in early diabetes, in Camerini-Davaios RA, Cole HS (eds): Vascular and Neurological Changes in Early Diabetes. New York, Academic Press, 1973, pp 323-332.
57. Osterby R: A* quantitative light and electron microscopic study of mesangial regions of glomeruli from patients with short-term juvenile diabetes mellitus. Lob Invest 1973; 29:99-103.
58. Osterby R: Early phases in the development of diabetic glomerulopathy. A quantitative electron microscopic study. Arta Med ScW iSuppt] 1975; 574:1-80.
59. Gundersen HJG, Osterby R: Glomerular size and structure in diabetes mellitus. U. Late abnormalities. Diobetoiogia 1977; 13:43-48.
60. Mauer SM, Barbosa J, Vernier RL, et al: Development of diabetic vascular lesions in normal kidneys transplanted into patients with diabetes mellitus. N Engl J Med 1976; 295:916-920.
61. Mauer SM, Steffes MW, Gannett ], et al: The development of lesions in the glomerular basement membrane and mesangium after transplantation of normal kidneys to diabetic patients. Diabetes 1983; 32:948-952.
62. Gliedman ML, Tellis VA, Soberman R, et al: Long-term effects of pancreatic transplant function in patients with advanced juvenile-onset diabetes. Diabetes Care 1978; 1:1-9.
63. BohmanSO, TydenG, Wilczek H, et al: Prevention of kidney graft diabetic nephropathy by pancreas transplantation in man. Diabetes 1985; 34:306-308.
64. Sutherland DER, Kendall DM, Najarían JS: A single institution's experience with pancreas transplantation. West J Med 1985; 136:838-844.
65. Abouna GM, Al-Adnani MS, Kremer GD, et al: Reversal of diabetic nephropathy m human cadaveric kidneys after transplantation into nondiabetic recipients. Lancet 1983; 2:1274-1276.
66. Abouna GM, Al-Adnani MS, Kumar MSA, et al: Fate of transplanted kidneys with diabetic nephropathy. Lancet 1986; 1:622-623.
STENO STUDY- FUNDUS PHOTOS AFTER ONE YEAR (VS BASELINE)
STENO STUDY- FUNDUS PHOTOS AFTER TWO YEARS (VS BASELINE)