"Doctor, when will I be able to measure my child's blood sugar without pricking his finger? Can you tell if my other children will develop diabetes and can they be prevented from getting it? Is it possible to prevent the eye and kidney complications? When will they find a cure for diabetes?" These and similar questions are asked frequently by parents of children with diabetes mellitus. This article answers several common questions about current and future treatment of insulin-dependent diabetes mellitus (IDDM) and its complications.
PREDICTION AND PREVENTION OF IDDM
In his 1916 textbook Treatment of Diabetes Metiitus, Eltiott P. Joslin stated: "The prophylactic and etiologic treatment of diabetes will surely play an important role in the future, and it is already plain that progress will be made along two lines: toward the early detection of the disease, and toward the prevention of the development of the disease in those susceptible to it." Nearly 80 years after Joslin wrote these prophetic words, clinical investigators are capitalizing on advances in the understanding of the etiology and pathogenesis of IDDM made in the last two decades.
The risk of progression to IDDM has been studied extensively in first-degree relatives of IDDM patients; now, highly specific prediction is possible within this population. The consensus of a National Institutes of Health workshop in 1990 was that techniques are available now to identify, with near certainty, those individuals who will develop diabetes and that immune intervention before the onset of symptoms holds promise as a means to prevent the disease.1 The ability to predict reliably the development of IDDM in apparently nonna) individuala has ushered in an era of clinical trials to test interventions aimed at preventing the disease in susceptible individuals.
The American Diabetes Association (ADA) has published an official position statement concerning intervention studies and screening subjects.1 Intervention studies are best accomplished by randomized, controlled design; they should be attempted within defined clinical studies approved by human investigation committees. Testing of high-risk subjects is encouraged if those who test positive are referred to centers participating in defined intervention studies. All potential subjects should be tested initially by measurement of cytoplasmic islet cell antibody using a standardized assay. If positive, they should then undergo an oral glucose tolerance test to characterize the state of carbohydrate tolerance, an intravenous glucose tolerance test (IVGTT) to assess early insulin response, and measurement of insulin autoantibodies.
A number of interventions are being tested in prediabetic subjects; these include azathioprine, nicotinamide, and low-dose intravenous and subcutaneous insulin. The favorable result of a recent pilot study2 has provided the impetus to initiate a largescale multicenter clinical trial to test the efficacy of low-dose insulin therapy for preventing IDDM in high-risk individuals who have been identified by the presence of islet cell antibodies and low first-phase insulin release on an IVGTT.
Approximately 90% of new patients with IDDM will not have a close relative with the disease. Therefore, methods to identify individuals within the general population who are destined to develop IDDM will need to be developed before any intervention could have widespread applicability.3
Progress in elucidating the immunopathogenesis of type 1 diabetes during the past decade has been such that it is not unrealistic to anticipate that an intervention able to prevent the development of type 1 diabetes meilitus will be discovered during the next decade.
BLOOD GLUCOSE MONITORING
Noninvasive Monitoring Systems
Self- monitoring of blood glucose was hailed as a great step forward in the battle against diabetes. Now patients can measure their blood glucose concentrations accurately by pricking a finger and performing a simple test with a drop of blood. Soon, people with diabetes may be able to measure blood glucose concentrations without actually using a sample of blood.4
Most noninvasive monitoring systems in development are based on technology that uses a beam of near-infrared light directed through the skin. Nearinfrared light penetrates tissue to a depth of several centimeters. Part of the light at each of the wavelengths is absorbed by components of the tissue and blood. When the beam interacts with glucose, its wavelength is altered; glucose has a distinct spectral "signature" that can be analyzed to produce a measurement of the blood glucose concentration. The test is performed by shining a beam of light through a finger inserted into a cylinder that contains a sensor. Good correlations between blood glucose concentrations and near-infrared light transmission readings through the finger have been reported by several groups.5 All noninvasive monitoring systems under development have yet to be approved by the Food and Drug Administration. Continuous glucose monitoring is not yet possible with this technique, nor is the apparatus wearable.
A reliable method for long-term continuous in vivo monitoring of blood glucose concentrations has long been sought. The need for such a system has become even more urgent since publication of the results of the Diabetes Control and Complications Trial (DCCT), which showed that maintenance of strict glycemie control reduces the risk of microangiopathy at the cost of a threefold increase in the frequency of severe hypoglycemia.6 Continuous blood glucose monitoring would eliminate frequent fingerpricks, detect impending hypoglycemia, and could activate an alarm when hypoglycemia needs to be treated. An automatic feedback system to control the rate of insulin delivery could be achieved ultimately by use of blood glucose measurements to adjust the infusion rate of a portable or implanted pump.
Most methods used to measure blood glucose concentration share basic principles with systems for home blood glucose monitoring. The test strip contains glucose oxidase, the enzyme that reacts with glucose to produce gluconic acid and hydrogen peroxide. Home monitors measure a color change caused by the production of hydrogen peroxide. Implantable sensors measure changes in the concentrations of peroxide or oxygen, or in the level of acidity. Percutaneous devices measure glucose concentration Jn the subcutaneous interstitial fluid where its concentration is almost identical to that of plasma.7 A device currently in development consists of a flexible sensor that is the size and shape of a fine needle with an exterior monitor. The sensor is injected under the skin, and a wire tail protruding from the skin is connected to a monitor. If investigators succeed in creating telemetry systems that do not require the sensor to be physically connected to an external monitor, the sensor could be completely implanted under the skin. When the enzyme becomes inactive, the sensor would be refilled with a fresh supply of enzyme injected through the skin, much as people now fill implanted insulin pumps.
Enzyme electrodes implanted subcutaneously work well in vitro. Long-term clinical monitoring using this technology has been thwarted, largely because of unpredictably decreased sensitivity (necessitating calibration procedures) and drift in vivo.8 This most likely is caused by coating of the electrode with cells and protein, and the effects of unknown inhibitors on electrochemistry. Sensors are available that are stable in vivo for several days and are capable of recording glucose concentrations similar to those in plasma. Long-term use of enzyme-based sensors in the body will be possible when solutions are found to these technical problems. It seems reasonable to expect that an implantable glucose sensor linked to an alarm system might be available for general use within 10 years. A continuous glucose monitor could emit a signal when blood glucose concentrations drop below a critical level. Such a system would have considerable benefit for intensively treated patients who are at increased risk of hypoglycemia unawareness and severe hypoglycemia. Ultimately, such glucose sensors would be an integral component of an artificial pancreas system.
The year 1993 marked the 100th anniversary of the first attempt to transplant pancreatic tissue into a person with diabetes. In 1893, Dr FW. Williams reported his unsuccessful effort to transplant fragments of a sheep's pancreas into a 15-year-old child with diabetes. Today, pancreas transplantation can restore glucose metabolism and glycosylated hemoglobin values to normal. And yet, even if organs were abundantly available - and they are not - its use would be restricted by the need for permanent immunosuppression because type 1 diabetes is an autoimmune disease. Even an identical graft, such as that of a hemipancreas from a homozygotic twin, is rejected by autoimmune aggression in the absence of immunosuppression.9 The ADA consensus guidelines recommend pancreas transplantation only in conjunction with kidney transplants. Thus, its clinical indication is limited to a small number of IDDM patients who require another graft, usually a kidney for end-stage renal failure caused by diabetic nephropathy. This procedure obviously is not attractive for otherwise healthy young patients with IDDM.
Islet Transplantation for Treatment of Diabetes
In 1972, nearly 80 years after Dr Williams' felled attempt, Dr Paul E. Lacy successfully transplanted insulin-producing islet cells into laboratory rats. Since that landmark event, research on islet cell transplan' tation has progressed considerably.10 With improved isolation techniques, it is now possible to recover several hundred thousand islets from one human pancreas.1' This has led to a new phase of clinical trials using isolated human islet allografts. Until recently, trials of islet transplants in humans were all unsuccessful, but in the past few years, several successful transplants have been reported. Although few recipients of islet cells are independent of exogenous insulin 2 years after the transplant, a small number have enjoyed independence from exogenous insulin administration for varying periods of time. Recipients do, however, require immunosuppression.12,13 Some of the unsolved problems include: determining the ideal implantation site, the large number of islet cells needed for a transplant (from 200 000 to 400 000), and the need to use antirejection (immunosuppressive) agents that have numerous adverse effects. For example, cyclosporine can have deleterious effects on the kidneys and on islet cell function and, in conjunction with high doses of glucocorticoids, can actually cause diabetes.
The ultimate goal of islet cell transplant research is to use this technique as a treatment for IDDM without generalized immunosuppression and early enough in the course of the disease to prevent or retard the development of diabetes complications. This would make it possible to perform islet cell grafts in young patients before the onset of complications. For this goal to be realized, there remains the obstacle of a very limited availability of human pancreatic tissue. Therefore, it is likely that progress will depend on the use of discordant islet xenografts, and methods will have to be discovered to implant islets without the need for immunosuppression.
Islet Transplantation With Tmmunoisolation
The most effective way to prevent immune rejection of a transplanted xenograft without the need for immunosuppressive drugs may be by immunoisolation of the transplanted islets14; this is the underlying principle of a biohybrid, artificial pancreas. Immunoisolation systems have been conceived in which transplanted tissue is separated by an artificial barrier from the immune system of the host. This approach has the potential to allow allogeneic transplantation without immunosuppression and to resolve the problem of human islet procurement by permitting use of islets isolated from animal pancreases.
The devices combine synthetic, selectively permeable membranes with living transplants and are, therefore, referred to as biohybrid artificial organs. Immunoisolation of islets in biohybrid devices offers a distinct advantage because the islets are isolated from the immune system of the host by a selectively permeable membrane. Substances of low molecular weight, such as nutrients, electrolytes, oxygen, and bioactive secretory products, are exchanged across the membrane, while immunocytes and other transplant rejection effector mechanisms are excluded.
Three major types of biohybrid pancreas devices have been studied: devices anastamosed to the vascular system as atrioventricular shunts,15 diffusion chambers,16 and microcapsules17 placed intraperitoneally, subcutaneously, or in other sites. Results in diabetic rodents and dogs indicate that biohybrid pancreas devices significantly improve glucose homeostasis and can function for more than a year. Before this technology can be used in human subjects, several technical problems must be overcome.
Since the discovery and large-scale production of insulin, diabetes mortality in children in developed countries has become uncommon. Instead, morbidity and mortality after 15 to 20 years of diabetes, caused by long-term micro- and macrovascular complications, have escalated. Visual impairment and blindness, kidney dysfunction and uremia, peripheral and autonomie neuropathy, and cardiovascular disease threaten all patients with IDDM.
The DCCT has proven definitively that nearnormal glycemie control reduces the occurrence and progression of both retinopathy and nephropathy by 30% to 70%.6 However, it is important to note that less than 5% of the patients treated intensively maintained hemoglobin AIc levels within the normal range despite the enormous efforts of expert diabetes treatment teams and the provision of free medical care and diabetes supplies. In the future, normoglycemia may be attainable with closed-loop insulin deliv' ery systems or islet transplantation. At present, alternative treatments (other than meticulous glycemie control) are needed to prevent or postpone diabetes complications because near-normal glycemie control cannot be achieved in most patients with IDDM.
In the last decade, research has provided valuable knowledge about the etiology and natural history of the late complications of diabetes. Early detection and treatment of diabetes complications is now standard clinical practice, and we can look forward, in the near future, to even more effective methods to prevent diabetes complications.
Diabetic retinopathy is the most frequent complication of IDDM and the leading cause of new blindness among Americans 20 to 74 years of age. After about 4 years, background nonprolrferative retinal changes develop, and after 20 years of IDDM, nearly all patients have some evidence of diabetic retinopathy. Vision-threatening proliferative diabetic retinopathy (PDR) usually occurs after 10 to 15 years. The growth of new blood vessels in front of the retina can cause preretinal or vitreous hemorrhages, and visual loss and blindness result from progressive vitreous hemorrhage, traction retinal detachment from scarring, or diabetic macular edema.18
The Wisconsin Epidemiologie Study of Diabetic Retinopathy showed that 67% of patients diagnosed before 30 years of age had PDR after 35 years of IDDM and 3.6% were legally blind.19 Laser photocoagulation, performed early in the course of PDR and before visual acuity is lost from diabetic macular edema, greatly reduces the risk of moderate and severe visual loss. The 5-year risk of severe visual loss is reduced to less than 5% by laser photocoagulation.18
Because PDR is usually asymptomatic, patients must be examined regularly to identify eyes at high risk so that laser treatment can be performed in a timely manner. To this end, screening strategies have been devised that are based on the natural history of the onset and progression of diabetic retinopathy. Proliferative changes that threaten vision typically do not appear within the first 5 years of IDDM or before puberty. Therefore, the ADA recommends annual screening for diabetic retinopathy beginning 5 years after the onset of diabetes and, in general, not before puberty. All patients old enough to understand, and the parents of young children with IDDM, should be informed about the possibility of retinopathy, made aware that it usually is an asymptomatic disease, and that effective therapy exists to prevent severe visual loss. The importance of annual eye examinations should be stressed for all teenage patients in the hope that they will continue to have routine follow-up eye examinations when they become independent.
The examination to detect diabetic retinopathy includes ophthalmoscopy through dilated pupils, fundus photography, and, rarely, fluoroscein angiography, and requires the expertise of an ophthalmologist. Examinations by general medical staff or even by diabetologists may fail to recognize proliferative retinopathy in 30% to 50% of cases. Nonmydriatic cameras are useful for large-scale screening. Preventive examinations and timely therapeutic interventions should substantially decrease severe visual loss in patients with diabetes.20
While diabetic retinopathy is currently an incurable but treatable disease, better strategies to identify, treat, and prevent diabetic eye disease are being sought. Researchers are seeking to understand the cause of altered retinal blood flow early in the course of diabetes with a view to developing pharmacologie methods to treat or prevent retinopathy. Links between activation of protein kinase C and decreased retinal blood flow have been discovered. The specific roles of vasoactive agents such as angiotensin II, histamine, and oxygen are under investigation.
Diabetic nephropathy is the most devastating late complication of IDDM and the most common cause of end-stage renal disease in the United States. The clinical syndrome afflicts one third of IDDM patients. Inexorable progression occurs after the appearance of overt proteinuria and culminates in either end-stage renal failure or death due to premature coronary artery disease. At the stage of overt proteinuria, interventions can postpone end-stage disease by a few years but cannot prevent it. To be effective, therefore, intervention must occur at an earlier stage.
It is now known that there is an earlier stage of incipient diabetic nephropathy, which is amenable to therapeutic interventions.21 This stage, known as microalbuminuria, is characterized by a subclinical (micro) increase in urinary albumin excretion. Normal urinary albumin excretion is less than 30 mg/day; indeed, urinary albumin excretion is usually less than 10 mg/day in most patients with well controlled IDDM of short duration. Overt proteinuria (Albustix positive) occurs when albumin excretion exceeds 300 mg/day. Urinary albumin excretion in the range of 30 to 300 mg/day or 20 to 200 µg/minute in a timed urine collection, or an albumin/creatinine ratio exceeding 20 µ-g/mg in a spot urine sample, is considered to be microalbuminuria.22 Spot testing avoids the inconvenience of obtaining a timed, quantitative urine collection. Repeated measurements, however, are required to avoid false-positives caused by nonspecific elevations in urinary albumin excretion that may occur with exercise. Microalbuminuria cannot be detected on a routine urinalysis, but reliable assays for measurement of micro amounts of albumin are now available in many specialized laboratories. Our practice at the Joslin Clinic is to examine patients annually after the onset of puberty for the presence of microalbum inuria.
Recent research has demonstrated that three interventions, near-normal glycemie control, blood pressure control, and reduced dietary protein consumption, can alter the course of diabetic nephropathy. Improved glycemie control does not affect overt proteinuria but does reduce the risk of progression from microalbuminuria to overt proteinuria and decreases albumin excretion.6 Because microalbuminuria may not be completely preventable despite improved glycemie control, additional measures to prevent nephropathy are needed.
Patients with diabetic nephropathy commonly have higher blood pressures than patients with normal rates of abumin excretion. Antihypertensive therapy, in general, significantly slows the rate of decline of kidney function in patients with overt proteinuria. Angiotension-converting enzyme (ACE) inhibitors have a specific renal protective effect, independent of their effect on blood pressure, in patients with diabetic nephropathy. Recently, a randomized trial in 409 patients with IDDM and overt proteinuria prospectively compared the ACE inhibitor captopril with placebo.23 Independent of its antihypertensive properties, captopril treatment reduced the risk of a doubling of serum creatinine concentration by approximately 50% and was associated with a 50% reduction in the risk of the combined end points of death, dialysis, and transplantation. A meta-analysis of 100 smaller clinical trials of patients with either overt proteinuria or microalbuminuria showed that ACE inhibitors reduced urinary protein excretion more than other antihypertensive drugs at comparable systemic blood pressures.24
The beneficial effect of ACE inhibitors is established for patients with overt proteinuria and appears to be promising in normotensive diabetic patients with microalbuminuria. At the present time, it would be inappropriate to prescribe these drugs as prophylaxis against nephropathy in normotensive patients with IDDM who do not have microalbuminuria. Their safety profile is generally favorable, but ACE inhibitors may cause fetal damage during the second and third trimesters of pregnancy. Angiotensionconverting enzyme inhibition should be discontinued as soon as pregnancy is discovered - ideally, within the first trimester. Until reliable methods are available to identify those patients destined to develop diabetic nephropathy, the use of ACE inhibitors should be considered only in patients with proteinuria, microalbuminuria, or hypertension.
A diet low in protein has been shown to slow the decline of renal function in patients with overt nephropathy. There are sparse data about the effects of reduced dietary protein in patients with microalbuminuria. Nevertheless, at the present time, we recommend a "prudent protein diet," individualized to meet the protein requirements for physical growth and development.
It is now known that genetic factors are important in the development of diabetic nephropathy.25 Investigators are attempting to identify the specific gene or genes that contribute to the development of diabetic nephropathy. Soon, therefore, it may be possible to determine a persons genetic risk for nephropathy at the time of diagnosis of IDDM. Institution of primary prevention, for example, with an ACE inhibitor, before the appearance of microalbuminuria might be warranted in individuals at high risk. In addition, strenuous efforts to maintain meticulous glycemie control from the time of diagnosis would be especially important for such patients.
OTHER APPROACHES TO PREVENTION AND TREATMENT OF DIABETES COMPLICATIONS
Chronic hyperglycemia is a mediator of late complications; however, the exact pathogenesis of diabetesspecific microvascular disease remains unknown. In addition to its hemodynamic effects, hyperglycemia activates the polyol pathway through the enzyme aldose reducise26 and by a nonenzymatic process causes irreversible glycation (glycosylation) of macromolecules.27 Activation of the polyol pathway results in the accumulation of the nondiffusible sugar alcohol, sorbitol, while nonenzymatic glycation produces modified macromolecules called advanced glycosylation end-products (AGEs). Whereas ACE inhibition appears to favorably modify the hemodynamic consequences of hyperglycemia, these other pathways also offer potential opportunities for therapeutic interventions.
Investigators are searching for nontoxic inhibitors of the enzyme aldose reductase and for an agent to block transcription of the aldose reductase gene. In addition, AGE formation may be inhibited by aminoguanidine, a drug that reacts with byproducts of glucose, which would otherwise lead to the formation of AGEs. Human and animal trials of aldose reductase inhibitors and of aminoguanidine are currently underway.
1. American Diabetes Association. Position statement. Prevention of type 1 diabetes mellitus. Diabas. 1990;39:1 151-1 152.
2. Keller RJ, Eisenbartfi GS, Jackson RA. Insulin prophylaxis in individuals at high risk of type 1 dtabetes. LanciL 1 993 ;341:92 7-928.
3. Bingley PJ. Bonifacio E, Gale EAM. Can VK really predici IDDM; Diabetes. 1993;42:213-220.
4. Pickup JC. Sampling and sensing blood glucosi. Lanca. 1993;342:1068.
5. Robinson MR. Eaton RP, Haaland DM. et al. Noninvasive glucose monitoring in diabetic patients: a preliminary evaluation. Ciin Cfiem. 1992;38;1618-1622.
6. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes im the development and progression of long-tenn complications in insulin-dependent diabetes meli itus. N EnglJ MaJ. 1993;329:977-986.
7. Fischer U, Ertle R.Abel P, et al. Assessment of subcutaneous glucose concentration· validation of the wick technique as a reference for implanted electrochemical sensors in normal and diabetic: dogs. Diabeiologia. 1957:30:940-945,
8. Pickup JC. In vivo glucose monitoring: sense and sensorhilirv. Diabetes Care. 1993:16:535-539.
9. Sibley RK, Sutherland DER, Coen FC, Michael AF. Recurrent diabetes mellitus in the pancreas iso- and allugrafr. A light and electron microscopic and immunohistochemical analysis of four cases. Lefe Imiesi. 1985;53:132-140.
10. Lacy PE. Status of islet cell transplantation. Diofvics Ritieuu. 1993:1:76-92.
11. Ricordi C, Lacy PE, Fmke EH, Olack BJ, Scharp DW. Automated method fot isolation of human pancreatic islets. Diabetes. Ì988;37:41 3-420.
12. Wamock GL, Kneteman NM, Ryan E, Seelis RE. Rabinovitch A, Rajotte RV. Normoglycemia after transplantation of freshly isolated and crytipreserved pancreatic islets in type I (insu I in -dependent) diabetes mellitus. DiofceioJogia. 1991;34:55-58.
13. Scharp DW. Lacy PE, Santiago JV, et al. Results of our first nine intraportal islet allografts In type 1, insulin-dependent diabetic patients. Transplantation. 1991 ;51:76-85.
14. Lanza RP, Sullivan SJ, Chick WL. Islet transplantai ion with immunoisolatiun. Diabetes. 1992:41:1503-1510.
15. Sullivan SJ1 Malti T, Borland KM, et al. Biohybrid artificial pancreas: long-term implantation studies in diabetic, pancrcateciomked dogs. Science. 1991:252:7 18-721.
16. Lanza RP, Butler DH. Borland KM, et al. Successful xenonansplantaiion of a diffusion -based biohybrid artificial pancreas: a study using canine, bovine, and poicine islers. Trompiani Pmc. 1992:24:669-671.
17. Lim F. Sun AM. Micioencapsulated islers as bioaitificial endocrine pancreas. Science. 1980:210:908-910.
18. Aiello LM. Cavallerano JD. Ocular complications of diabetes mellitus. In: Kahn CR, Weir GC, eds. Joslm's Diatelo Mettimi. 13th ed. Philadelphia. Pa: Lea & Febiger; 1994:771-793.
19. Klein R, Klein BEK. Moss SE. Oavis MD, DeMets DL The Wisconsin Epidemiologie Study of Diahetic Reiinopathy, II: prevalence and risk of diabetic rctinopathy when age at diagnosis is less than 30 years. Arch Opthalmol. 1984:102:520-526.
20. Owens DR. Dolben J. Young S, et al. Screening fm diabetic retinopaihy- Diotel Med 1991 :8( symposium ):S4-S10.
21. Mogensen CE. Prediction of clinical diabetic nephropalhy in IDDM patients: alternatives tu microalbumi nuria. Diatelo. 1990;39:761-767.
22. Herman W, Hawthorne V, Hamman R, el al. Consensus Statement. Am J Kidney Du. 1989:13:2-6.
23. Lewis EJ, Hunsicker LG1 Bain RP, Rohde RD for the Collaborative Study Group. The effect uf angiotensin-convcrtmg-cnzYme inhiriitkm on diabetic nephropathy, N Engl J Mid. 1993; 329:1456- 1462.
24. Kasiske BU Kalit RSN, Ma JZ, Liao M. Keane WE Effect of ani ihypertensive therapy on the kidnrf in palients with diahetes. a meta-regreuKTi analysis. Ann iruem Atei 1993:118:129-138.
25. Krolewski AS, Doria A, Magre J- Warran JH. Housman D. Molecular genetic approaches to the identification of genes in the development of nephropathy in insulin-dependent diabetes mellitus. J Am Soc Nephrol. 1992;3:S9-S17.
26. Frank RN. The aldose reductase controversy Diabeta. 1994:43:169-172.
27. Brownlee M. Glycanon products and the pathogenesis of diabetic complications. Diabetes Care 1992;15:1835-1843.