Obviously, NIDDM is not a benign condition. Other articles in this supplement review the high incidence of morbidity and mortality accompanying NIDDM. Although it has been recognized that those persons with the highest levels of hyperglycemia are at the greatest risk for developing complications, evidence confirming the benefits of intensive glycemic management has only recently become available. Both the Stockholm Diabetes Intervention Study [3] and the Diabetes Control and Complications Trial (DCCT) [4] were long-term studies designed to evaluate whether intensive insulin therapy with near normalization of glycemia influenced the microvascular and neurologic complications of IDDM. Both studies have provided compelling evidence that near normalization of blood glucose levels can delay the development and progression of retinopathy, nephropathy, and neuropathy in patients with IDDM.

A major question derived from these studies is whether the conclusions about the benefits of glycemic management are applicable to persons with NIDDM as well. Substantial evidence reviewed elsewhere in this supplement indicates that the severity and duration of hyperglycemia is a critical factor in the pathogenesis of microvascular complications in both forms of diabetes. Thus, it seems warranted to expect that patients with NIDDM would also benefit from improved glycemic control and to advise that management strategies should be instituted to achieve the best possible glycemic control.

This review focuses on how glucose control and insulin resistance interact and, specifically, how each influences the application of intensive glycemic management in NIDDM. How poor glucose control affects the severity of insulin resistance and how insulin resistance influences the achievement of good glucose control are considered. Most comments in this article refer primarily to obese patients with NIDDM. although they may also be applicable to nonobese patients with NIDDM. The management and influence of hypertension and hyperlipidemia and clotting abnormalities that commonly accompany NIDDM have recently been reviewed and are not considered in this article [5]. However, several treatment modalities used to improve metabolic control in NIDDM, including diet, exercise, and various pharmacologic therapies, are commented on briefly before discussing insulin use in the intensive management of hyperglycemia in NIDDM.

Pathophysiology of NIDDM

Before proceeding to discuss the treatment approaches used for intensive management of hyperglycemia in NIDDM, understanding the pathophysiology of hyperglycemia in NIDDM is essential. Three basic abnormalities that characterize NIDDM contribute to the development of hyperglycemia [6] Figure 1, including excessive glucose production by the liver, impaired insulin secretion, and peripheral insulin resistance primarily occurring in liver, adipose, and muscle tissues. Both obese and nonobese patients with NIDDM have the same underlying pathophysiology, but their expression and contribution to the development of hyperglycemia may differ.

View larger version (35K): [in this window] [in a new window]   Figure 1. Mechanisms of hyperglycemia in NIDDM. The X on each arrow indicates the presence of a defect that contributes to elevated glucose levels (Reproduced from Henry RR. Prospects for primary prevention of type II or non–insulin-dependent diabetes mellitus. Diabetes Reviews International. 1994; 3:2-5, with permission.).

In obese patients with NIDDM, severe insulin resistance in the liver and peripheral tissues predominates. Although the pancreas may produce a large quantity of insulin, it is insufficient to overcome the severe insulin resistance, and hyperglycemia ensues. In contrast, nonobese patients with NIDDM tend to have milder degrees of insulin resistance with hypoinsulinemia caused by deficient insulin secretion as the predominant abnormality. The extent or severity of each of these abnormalities also varies within individuals and accounts for the wide range of fasting and postprandial hyperglycemic levels that occur. Understanding these pathophysiologic variants of NIDDM is important because the strategies used to implement intensive management can be influenced by the different patterns of clinical expression. Insulin therapy can be effective for both forms of NIDDM, but the amount of exogenous insulin needed to achieve glycemic control in obese patients with NIDDM is usually large to overcome the severe insulin resistance.

Goals of Intensive Therapy in NIDDM

Once the results of the Stockholm [3] and DCCT [4] studies became available, there was widespread acceptance that normoglycemia should be the goal of therapy for IDDM. This enthusiasm for normoglycemia has also been transferred to the management of NIDDM, as is evident from the recent recommendations of the American Diabetes Association [7]. In their previous advocated therapeutic objectives [8], a fasting or preprandial glucose level of 7.8 mmol/L (140 mg/dL) and a glycosylated hemoglobin level of 8% were deemed acceptable. The new goals for glycemic control are a fasting (preprandial) glucose level of less than 6.7 mmol/L (< 120 mg/dL) and a glycosylated hemoglobin level of less than 7% (normal range, 4% to 6%). The scientific rationale for these goals in NIDDM is the reasonable assumption that benefits comparable to those shown for IDDM may result. However, as reviewed by Colwell in this supplement (see "The Feasibility of Intensive Insulin Management in NIDDM: Implications of the Veterans Affairs Cooperative Study on Glycemic Control and Complications in NIDDM"), the pathophysiologic differences between the two types of diabetes usually require that much larger doses of exogenous insulin be used to reach these glycemic targets in NIDDM as compared with IDDM, especially in obese patients. As discussed later, the administration of these larger doses may cause adverse effects.

Glucose Toxicity

Chronic hyperglycemia, in addition to being a marker of uncontrolled diabetes, is now recognized as having deleterious effects on insulin secretion and action and, as such, is a self-perpetuating abnormality ("glucotoxicity"). Thus, in addition to the potential for reducing the long-term complications of diabetes, reducing hyperglycemia may also be expected to decrease the insulin resistance and the reduced insulin secretion underlying the disease.

Therapy of Hyperglycemia

Weight Loss

Weight loss is one of the most effective forms of therapy for obese patients with NIDDM, particularly when it is combined with an exercise program as part of a multidisciplinary team approach. Even modest amounts of weight loss reduce day-long glycemia in most obese patients with NIDDM by reducing elevated hepatic glucose output, improving insulin secretion, and increasing peripheral insulin sensitivity [9]. Two effects of diet and weight loss therapy need to be distinguished. After the start of hypocaloric diet therapy for NIDDM, a prompt and often dramatic decrease in the serum glucose level may occur, primarily from reduced hepatic glucose output. With more prolonged caloric restriction and weight loss, both lean and fat mass are reduced, and tissue insulin sensitivity is enhanced. Despite the impressive and often dramatic benefits from weight loss, the ability of persons to maintain lower weight and metabolic benefits for prolonged periods has been limited. Weight loss and prevention of weight gain should continue to be encouraged as part of a comprehensive treatment program, but it is doubtful whether this form of therapy alone is sufficient to achieve or maintain the earlier stated glycemic goals in most obese patients with NIDDM.

Oral Antidiabetic Agents

Until recently, only the sulfonylurea class of glucose-lowering medications was available for oral treatment of hyperglycemia of NIDDM not controlled by diet, exercise therapy, or both. These agents tend to be most efficacious in the early course of diabetes when insulin secretory reserve is present and relative insulin deficiency is not the predominant abnormality. In keeping with this observation, sulfonylureas work primarily by enhancing basal and stimulated insulin secretion, which reduces hepatic glucose output and facilitates peripheral glucose disposal [10]. Prolonged use of sulfonylureas also results in improved insulin action, which is believed to be an effect secondary to improved glycemia and reduced glucotoxicity. Most persons are able to achieve some improvement in glycemic control with sulfonylureas, but achieving and maintaining long-term near-normal glycemia is uncommon with this medication alone.

More recently, metformin has been introduced to the American market. This agent has been available in Europe for decades, and much is known about its mechanism of action [11]. In brief, metformin reduces hepatic glucose production and increases peripheral glucose utilization and can be used either alone or combined with a sulfonylurea since their mechanisms of action differ [12]. Metformin has been used in the United Kingdom Prospective Diabetes Study [13], and the long-term results of its use are summarized by Turner in this supplement (see "United Kingdom Prospective Diabetes Study 17: A 9-Year Update of a Randomized, Controlled Trial on the Effect of Improved Metabolic Control on Complications in NIDDM"). Briefly, the results indicate that metformin, either alone or combined with sulfonylurea, is associated with less weight gain than either sulfonylurea or insulin alone. Thus, it appears to be a particularly good choice for the treatment of obese patients with NIDDM. Nevertheless, with the therapeutic choices available today, the attainment of normoglycemia in NIDDM over long periods will eventually require inclusion of some form of insulin therapy for most patients, either alone or combined with oral agents.

Insulin Therapy

Only a few studies have evaluated the effects of intensive insulin therapy to normalize serum glucose levels in NIDDM. A summary of the recently completed Veterans Affairs Cooperative Study on Glycemic Control and Complications in NIDDM is presented by Colwell in this supplement (see "The Feasibility of Intensive Insulin Management in NIDDM: Implications of the Veterans Affairs Cooperative Study on Glycemic Control and Complications in NIDDM"). The studies reviewed for the current article were chosen to provide a brief overview of the different methods of insulin delivery commonly used for intensive insulin management of NIDDM, including the possible benefits and complications. The effects of intensive insulin therapy and glycemic control on measures of insulin action and resistance are emphasized.

Continuous Subcutaneous Insulin Infusion

Garvey and colleagues [14] studied 14 patients with uncontrolled NIDDM and used continuous subcutaneous insulin infusion therapy to normalize serum glucose levels. At the study outset and before treatment, these patients were mildly overweight (relative weight, 1.19 ± 0.06), with a mean age of 50 years, a fasting glucose level of 15.9 ± 0.9 mmol/L (286 ± 17 mg/dL), and a glycosylated hemoglobin level of 13.0% ± 0.7% (normal range, 6.3% to 8.2%). The study group comprised patients with new-onset diabetes and those who had failed treatment with diet, oral agent, and insulin regimens. After 3 weeks of continuous subcutaneous insulin infusion therapy, mean fasting glucose levels fell to less than 5.6 mmol/L (100 mg/dL), and glycosylated hemoglobin levels fell to within the normal range at 8.1%.

Figure 2 shows that the basal hepatic glucose output, which was initially increased, fell more than 40% to within the normal range after continuous subcutaneous insulin infusion therapy. The close correlation between basal hepatic glucose output and fasting plasma glucose before and after treatment suggests that insulin therapy lowered the fasting glucose level primarily by lowering hepatic glucose output. Peripheral insulin action was evaluated by a series of hyperinsulinemic euglycemic clamps, which showed that the initial severe insulin resistance was markedly improved after intensive insulin therapy. As shown in Figure 3, peripheral glucose disposal improved on average approximately 70% to 80% with treatment but still remained 30% to 40% lower than normal in control subjects. Achieving this level of glycemic control required a mean exogenous insulin dose of more than 100 U/d, which resulted in the induction of marked hyperinsulinemia with serum insulin levels ranging from 69 ± 13 to 142 ± 21 µ U/mL. No mention was made of increases in body weight, but the evaluation period may have been too short to detect appreciable changes. This study was among the first to show that normoglycemia could be achieved and maintained in NIDDM over the short term but required the administration of large doses of insulin with the development of hyperinsulinemia. The data presented below will show it is likely that the large doses of insulin and not the method of administration resulted in these improvements.

View larger version (27K): [in this window] [in a new window]   Figure 2. Mean ± SE rates of basal hepatic glucose output in 14 subjects with NIDDM before and after intensive insulin therapy and in age-matched control subjects. (Reproduced from reference 13 with permission.).

View larger version (26K): [in this window] [in a new window]   Figure 3. Mean in vivo dose-response curves for the control subjects () and subjects with NIDDM before and after (\#9679;) intensive insulin therapy. (Reproduced from reference 13 with permission.).

Conventional Subcutaneous Insulin Therapy

A longer trial of exogenous insulin therapy in patients with NIDDM has recently been reported by Henry and colleagues [15]. These investigators used an intensively managed and monitored program of conventional subcutaneous neutral protamine Hagedorn (NPH) and regular insulin given before breakfast and supper for 6 months to achieve near-normal glycemic control in 14 patients with NIDDM. The patients in the study were obese (weight, 93.5 ± 5.7 kg; body mass index, 31.4 ± 1.9 kg/m²), and all had had secondary failure to oral sulfonylureas. Their mean fasting glucose level before treatment and off all medication was 15.7 ± 0.7 mmol/L (283 ± 13 mg/dL), and their glycosylated hemoglobin level was high at 7.7% (normal range, 3.6% to 4.9%). Although fasting and day-long endogenous hyperinsulinemia existed before treatment, these insulin levels were insufficient to overcome the severe insulin resistance that was present. Throughout the study, patients were seen every 1 to 2 weeks by a physician and metabolic nutritionist and hospitalized at monthly intervals for a 24-hour glycemic profile and fasting metabolic parameters on a standard diet.

Biochemical measurements obtained during the 24-hour profile studies at baseline and at monthly intervals throughout the 6-month study are shown in Table 1. Near-normal glycemia was achieved within 1 month, and few additional glycemic benefits occurred thereafter. By the end of the study, the glycosylated hemoglobin level was just above the upper limit of normal. Interestingly, despite this aggressive management, minimal biochemical or symptomatic hypoglycemia and no instances of severe hypoglycemia requiring assistance occurred during the study. This relative lack of hypoglycemia is likely the result of severe insulin resistance present in both liver and muscle.

As in the study by Garvey and coworkers [14], this level of glycemic control required a mean total insulin dose of approximately 100 U/d with serum concentrations in the range of 400 to 500 pmol/L (65 to 87 µ U/mL). Despite efforts to reduce caloric intake to maintain stable body weight, a mean weight gain of 8.7 kg occurred. No relation was found between the amount of weight gained and pretreatment glycosuria. Thus, weight gain could not be attributed to the decrease in loss of calories in the urine with improved glycemic control. Important correlations were found, however, among the total exogenous insulin dose, mean serum insulin levels, and total weight gain Figure 4, suggesting that these parameters may be related. Thus, these results indicate that near-normal glycemia can be achieved in obese patients who have NIDDM with intensive insulin therapy over an extended period. Whereas hypoglycemia was not problematic in these studies, hyperinsulinemia and weight gain were.

View larger version (19K): [in this window] [in a new window]   Figure 4. Correlation of total weight gain with total insulin dose (A) and mean serum insulin level (B) at study completion. For panel A, r = 0.62, P < 0.02; for panel B, r = 0.67, P < 0.01. (Reproduced from reference 14 with permission.).

Changes in basal hepatic glucose output and peripheral glucose uptake were also evaluated before and after the 6 months of intensive treatment and are shown in Figure 5. As in the study by Garvey and colleagues [14], the basal hepatic glucose output fell 44%, and a strong correlation was again present between this output and fasting plasma glucose levels before and after treatment. In contrast, however, peripheral glucose uptake determined during a hyperinsulinemic euglycemic clamp increased only 17%, possibly because of the large associated weight gain and prolonged hyperinsulinemia that was induced. This study also showed another important feature of insulin therapy in NIDDM. Peripheral glucose uptake was stimulated only minimally during the glucose clamp, although exogenous insulin levels were nearly 10-fold greater than mean peak postprandial levels achieved by these patients. Thus, the primary mechanism by which intensive insulin therapy achieves normoglycemia in NIDDM is suppression of hepatic glucose output rather than stimulation of peripheral glucose uptake. From these studies, hyperinsulinemia and possibly weight gain appear to be inevitable consequences of intensive therapy with conventional subcutaneous insulin administration.

View larger version (21K): [in this window] [in a new window]   Figure 5. Mean levels of basal hepatic glucose output (A) and peripheral glucose uptake (B) before and after 6 months of intensive insulin therapy in 14 subjects with NIDDM. (Reproduced from reference 14 with permission.).

Combination Studies

In an effort to define the optimal insulin treatment program for NIDDM, Yki-Jarvinen and coworkers [16] compared the effects of four insulin regimens used to achieve glycemic goals close to those advocated by the American Diabetes Association. They studied a total of 153 patients who had failed to achieve optimal glycemic control on maximal doses of glipizide or glyburide, either alone or in combination with metformin. The patients were mildly obese with a body mass index of 27 to 29 kg/m² and had poor glycemic control with mean fasting blood glucose levels of approximately 12.5 mmol/L (225 mg/dL). They were randomly assigned to receive one of five regimens that included four different insulin treatment arms and a control group on oral hypoglycemic agents alone for 3 months. The four insulin arms included two groups treated with the usual oral hypoglycemic agent plus either NPH insulin given before breakfast (the morning-NPH group) or NPH given at 2100 hours (the evening-NPH group) and two groups in which oral hypoglycemic agents were discontinued and either NPH and regular insulin (in a ratio of 70 units to 30 units, respectively) were given before breakfast and dinner (the two-injection group) or NPH insulin was given at 2100 hours with regular insulin before breakfast, lunch, and dinner (the multiple-injection group).

The changes in glycemic control in the various treatment arms are shown in Figure 6. The mean diurnal blood glucose and glycosylated hemoglobin levels were similar in all groups during the 6-week run-in period and improved substantially and to similar near-normal levels in the four insulin treatment arms. Although mean diurnal blood glucose changes were different in the four insulin treatment arms at various times of day, the overall changes in day-long glycemia were similar. In the four insulin treatment groups combined, the mean diurnal blood glucose concentration after 3 months was substantially less at 8.9 ± 0.2 mmol/L (160 ± 4 mg/dL) compared with 11.3 ± 0.6 mmol/L (204 ± 11 mg/dL) in the control group. This equivalent improvement in glycemia in the different treatment arms required approximately twice as much exogenous insulin in the two arms taking no oral hypoglycemic agents (43 ± 2 and 45 ± 3 U/d) as in the two taking oral agents (19 ± 1 and 20 ± 2 U/d). Mean diurnal free insulin levels increased in all four groups, but the increment in the evening-NPH group was lower than the other three insulin-treatment groups by 50% to 65%. Weight gain was also substantially less in the evening-NPH group as compared with the morning-NPH, two-injection, and multiple-injection groups (1.2 ± 0.5 kg as compared with 2.2 ± 0.5 kg, 1.8 ± 0.5 kg, and 2.9 ± 0.5 kg, respectively). The study also found in the combined insulin treatment groups an important relation between change in glycosylated hemoglobin levels and body weight and change in the mean diurnal serum free insulin levels, which suggests a causal relation.

View larger version (31K): [in this window] [in a new window]   Figure 6. Mean diurnal blood glucose level (top) and glycosylated hemoglobin level (bottom) in the control group (+); morning-NPH group ; evening-NPH group (\#9679;); two-injection group ; and multiple-injection group (\#9632;) before and after 12 weeks of treatment. The normal reference range of glycosylated hemoglobin level is 4% to 6%. (Reproduced from reference 15 with permission.).

In each of the insulin treatment studies described above, levels of glycemia approached those recommended by the American Diabetes Association; to achieve this improved glycemia, increased exogenous insulin was required and associated with concomitant increased peripheral hyperinsulinemia and weight gain. In the study by Yki-Jarvinen and coworkers [16], the adverse effects were ameliorated by the addition of sulfonylureas. More recent reports have also suggested that similar improvements can be made using metformin and insulin combined [17].

Hypoglycemia was uncommon in all studies of NIDDM in contrast to studies of IDDM. Thus, in both NIDDM and IDDM, weight gain is concomitant with improvement of glycemic control. However, in NIDDM, weight gain seems to be greater and was associated with decreased peripheral action of insulin. We are currently faced with the dilemma of the long-term benefits of improved glycemia in NIDDM being offset by the potential adverse effects of excessive circulating insulin levels and weight gain. The evidence associating peripheral hyperinsulinemia with accelerated atherosclerosis is reviewed in this supplement by Stern (see "Do NIDDM and Cardiovascular Disease Share Common Antecedents?").

Every effort should be made to achieve the best possible glycemic control using the lowest dose of insulin. Continued emphasis on diet and exercise and combined insulin and oral hypoglycemic agent therapy is required to attain this goal.

Newer therapeutic agents designed to directly reduce insulin resistance are now being studied [18]. One of these, troglitazone, in addition to lowering glucose levels also reduces blood pressure and lipid levels [19].

Conclusion

Top Conclusion Author & Article Info References

Intensive management of NIDDM to normalize glucose levels is feasible but may require large doses of exogenous insulin to offset the presence of insulin resistance, which can be severe when obesity also exists. After a period of intensive insulin management with improved glycemic control, basal hepatic glucose production rates are reduced to those seen in nondiabetic patients, and peripheral insulin action is improved. Hyperinsulinemia and weight gain frequently accompany intensive insulin management, but their severity may be attenuated by combining insulin therapy with maximal doses of oral hypoglycemic agents, particularly when insulin is given in the evening. These studies suggest that, if intensive insulin management is implemented in NIDDM, special attention should be given to achieving the best glycemic control with the lowest dose of insulin. Possible future strategies that may assist in attaining this goal include the use of novel insulin formulations and delivery systems as well as therapies designed to specifically reduce insulin resistance, such as troglitazone.