Author + information
- Received April 7, 2008
- Revision received November 4, 2008
- Accepted November 6, 2008
- Published online February 3, 2009.
- Edward S. Horton, MD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Edward S. Horton, Joslin Diabetes Center, One Joslin Place, Boston, Massachusetts 02215
Type 2 diabetes mellitus is a progressive disease characterized by early impairment of beta-cell function and ultimately loss of beta-cell mass. Hence, a single daily injection of a long-acting insulin is commonly initiated after intensification of oral antihyperglycemic therapy. Hemoglobin A1C should be measured every 3 months and therapy adjusted if the target is not met. As beta-cell function continues to decline, it is often necessary to add exogenous bolus insulin therapy, using short-acting insulin analogs or regular insulin. Alternatively, the use of pre-mixed insulin preparations, combining both long-acting and short-acting insulins, may be used.
Diabetes is highly prevalent in patients with cardiovascular (CV) disease, and its optimal management is critical to reduce adverse outcomes. The pathogenesis of type 2 diabetes mellitus (T2DM) is complex and involves progressive impairment of pancreatic beta-cell function with loss of normal patterns of glucose-stimulated insulin secretion, excess glucagon secretion, and peripheral insulin resistance. The result of these abnormalities is hyperglycemia, both in the fasting and the post-prandial states (1–3). Successful treatment of hyperglycemia in T2DM requires an understanding of the factors that regulate fasting blood glucose concentrations and the excessive increase in glucose that occurs after meals.
There are now several classes of oral antihyperglycemic drugs (OADs), as well as some injectable peptide hormones, that target different defects in glucose regulation and are effective either as monotherapy or in combination with other agents that have complementary mechanisms of action (Table 1).There are also a large number of insulin preparations available, including human insulin and insulin analogs, ranging from long-acting basal insulins that provide stable plasma insulin concentrations for up to 24 h or longer, as well as short-acting insulins that are used for bolus therapy to control post-prandial hyperglycemia. Various pre-mixed insulin preparations that combine both long-acting and short-acting insulins are also available for use (Table 2).Achieving optimal glycemic control in people with T2DM is challenging for CV specialists and requires the effective use of both OADs and insulin replacement therapy to achieve target hemoglobin A1C (A1C) levels as close to normal as possible (≤6%) and at a minimum ≤7% without excessive risk of hypoglycemia or other unacceptable side effects. An updated consensus statement on the management of hyperglycemia in T2DM, recently published by the American Diabetes Association and the European Association for the Study of Diabetes, provides an algorithm that can serve as a general guideline (4).
Causes of Hyperglycemia in T2DM
In the fasting state, plasma glucose concentration is normally maintained at 70 to 100 mg/dl by a balance between glucose uptake in both insulin-dependent (e.g., skeletal muscle and adipose tissue) and noninsulin-dependent (e.g., central nervous system and blood elements) tissues and glucose production by the liver through a combination of glycogenolysis and gluconeogenesis. Hepatic glucose production is regulated primarily by insulin, glucagon, and the availability of substrates for gluconeogenesis; although during situations of stress, the counter-regulatory hormones epinephrine, norepinephrine, cortisol, and growth hormone also play a role to increase hepatic glucose production. In T2DM, fasting hyperglycemia results primarily from excessive hepatic glucose production in the context of insulin resistance in both the liver and the peripheral tissues, decreased insulin secretion, and excess glucagon secretion. Agents that reduce insulin resistance (particularly in the liver), increase plasma insulin concentrations, or decrease glucagon secretion can effectively reduce fasting hyperglycemia.
After meal ingestion, insulin is normally secreted in a biphasic manner and glucagon secretion is suppressed. The early, rapid release of insulin (first phase secretion) and suppression of glucagon play a major role in regulating hepatic glucose metabolism; hepatic glucose production rapidly decreases and net hepatic glucose uptake occurs. In addition, insulin stimulates peripheral glucose uptake in skeletal muscle and adipose tissue. The post-prandial increase in insulin secretion and decrease in glucagon is potentiated by the intestinal release of the incretin hormones glucagon-like peptide-1 and glucose-dependent insulinotropic polypeptide, both of which increase glucose-dependent insulin secretion from pancreatic beta cells (the incretin effect) and suppress glucagon secretion from the alpha cells. Decreases in all of these responses occur in T2DM, resulting in excessive increases in plasma glucose after meals. Normally, post-prandial blood glucose concentration reaches a peak (usually ≤140 mg/dl) at about 45 to 60 min after food ingestion and returns to baseline levels by 2 to 4 h, depending on the composition of the meal, as well as the rates of gastric emptying, food digestion, and absorption. In T2DM, peak blood glucose levels typically occur 1 to 2 h after a meal, are much higher than normal, and are slow to return to baseline values. This results in sustained hyperglycemia, which is associated with increased oxidative stress, impaired endothelial function, and other deleterious effects on the vascular system. Effective treatment for post-prandial hyperglycemia is therefore a major target for treatment to reduce long-term risks and complications of hyperglycemia.
A1C is a measure of the glycation of the hemoglobin molecule over the lifespan of red cells and is expressed as a percentage of the total hemoglobin. In people with a normal red cell production rate and survival time, normal values are between 4% and 6%, using current methods and standards. The A1C provides an indication of average blood glucose concentrations during the preceding 2 to 3 months and is considered to be the reference standard for assessing overall glycemic control in people with diabetes. The A1C measurement does not distinguish between the contributions of fasting versus post-prandial hyperglycemia, although some general assumptions can be made. When fasting blood glucose levels are maintained close to normal, for example, post-prandial increases in glucose that occur throughout most of the day will have a relatively greater impact on A1C levels than when fasting glucose levels are also high. Therefore, treatment strategies to lower A1C to as close to normal as possible must address improving both fasting and post-prandial glucose control. Goals of therapy dictate that A1C should be as close to normal as possible without incurring an excessive risk of hypoglycemia, and in general to ≤7%, in individual patients. Recent estimates from National Health and Nutrition Examination Survey data indicate that, in the U.S., progress is being made in achieving these goals, but there is still a long way to go. The percentage of people with diagnosed diabetes with A1C ≤7% has increased from 37% in 1999 to 2000 to 56.8% in 2003 to 2004 (5) (Fig. 1).
Relationship Between Blood Glucose and CV Disease Risk
It is now well established that elevated A1C levels are closely related to both microvascular and macrovascular complications of diabetes (6–9). However, the relative contributions of fasting and post-prandial glucose concentrations to CV disease risk are less defined (10). The bulk of evidence supports the conclusion that post-prandial hyperglycemia is a more important determinant of CV disease risk than fasting glucose (11–16), which has resulted in increased emphasis on treating post-prandial hyperglycemia in T2DM management.
Two recent trials assessed the effects of intensive versus standard glucose targets on CV events. The ACCORD (Action to Control Cardiovascular Risk in Diabetes) study included 10,251 subjects (mean age 62 years) with T2DM and either confirmed CV disease or combinations of risk factors and/or conditions suggesting a high likelihood of CV disease (17). The A1C targets were <6% (intensive group) and 7% to 7.9% (standard group). To meet these goals, study investigators could use all currently available glucose-lowering drugs. The primary outcome was a composite of CV death, nonfatal myocardial infarction (MI), or stroke. At 1 year, median A1C levels were 6.4% and 7.5% in the intensive-therapy and standard-therapy groups, respectively, and these levels were maintained throughout a mean follow-up of 3.5 years. The primary outcome occurred in 6.9% and 7.2% of patients, respectively (hazard ratio [HR]: 0.90, 95% confidence interval [CI]: 0.78 to 1.04, p = 0.16). All-cause mortality was significantly higher in the intensive-therapy group, 5.0% versus 4.0% (HR: 1.22, 95% CI: 1.01 to 1.46, p = 0.04). In contrast, nonfatal MI was significantly lower in the intensive-therapy group, 3.6% versus 4.6% (HR: 0.76, 95% CI: 0.62 to 0.92, p = 0.004). There was no significant difference in the rate of stroke. Because of the increased all-cause mortality rates in the intensive-therapy group versus the standard-therapy group, the intensive regimen was discontinued and patients in this arm of the study were switched to standard glycemic therapy.
The ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified-Release Controlled Evaluation) trial included 11,140 subjects who had T2DM, were 55 years of age or older (mean age 66 years), and had either a history of CV disease or at least 1 other CV risk factor (18). The A1C goal in the intensive group was ≤6.5%, whereas that in the standard group was based on local guidelines. The glucose-lowering regimen used in the intensive arm was based on a modified-release formulation of the sulfonylurea gliclazide; participants in the other arm continued with their usual glucose-lowering regimens except gliclazide. The macrovascular primary outcome was a composite of CV death, MI, and stroke. The microvascular primary outcome was a composite of new/worsening nephropathy (development of macroalbuminuria, doubling of serum creatinine to a level of at least 2.26 mg/dl, need for renal replacement, or death caused by renal disease), retinopathy (development of proliferative retinopathy, macular edema, or diabetes-related blindness; or retinal photocoagulation therapy). After a median of 5 years, mean A1C values were 6.5% and 7.3% in the intensive-treatment and standard-treatment groups, respectively. The macrovascular primary outcome occurred in 10.0% and 10.6% of patients, respectively (HR: 0.94, 95% CI: 0.84 to 1.06, p = 0.32). The microvascular primary outcome occurred in 9.4% and 10.9% of patients, respectively (HR: 0.86, 95% CI: 0.77 to 0.97, p = 0.01). The combined primary outcomes occurred in 18.1% and 20.0% of patients, respectively (HR: 0.90, 95% CI: 0.82 to 0.98, p = 0.01). All-cause mortality and secondary CV outcomes did not differ significantly between the groups (Fig. 2).
The ACCORD and ADVANCE results indicate that in patients with diabetes who are at high risk for CV events it would be prudent to set a goal A1C of approximately 7%. The American Diabetes Association recommends a fasting glucose of 70 to 130 mg/dl and a post-prandial target of <180 mg/dl (19). Ongoing clinical outcomes trials may provide further clarification.
In the VADT (Veterans Affairs Diabetes Trial) study, 1,791 veterans with T2DM were randomized to intensive versus standard treatment (20). Among trial participants, approximately 40% had a prior CV event, whereas approximately 80% were hypertensive; most subjects were also overweight or obese. These subjects had higher glucose levels than participants in either ACCORD or ADVANCE, with a baseline A1C of 9.5%. A variety of drugs was used in both arms, with A1C levels reduced to 6.9% in the intensive arm and 8.4% in the standard-treatment group. Blood glucose levels had no impact on outcomes; both groups had fewer events than expected, which may have been a result of the general control of other risk factors. Interestingly, a substudy found that coronary artery calcification correlated with retinopathy. As with ADVANCE, there was no suggestion that more intensive antidiabetes therapy increased CV events.
Management of Hyperglycemia in T2DM
Fundamental to all treatment regimens for managing hyperglycemia in T2DM is the education of patients to self-manage their disease. The establishment of a lifestyle program that includes a healthy diet, physical activity, and weight loss of 5% to 10% of initial body weight, for those who are overweight or obese, is also critical. These strategies alone, if successfully implemented, can result in significant improvement in both fasting and post-prandial blood glucose levels and in A1C. However, many patients are not successful in achieving these goals and will require therapy with OADs or insulin to manage their glucose levels. In addition, lifestyle modifications require time to achieve results; current recommendations dictate beginning OADs immediately, usually metformin alone or in combination with another oral agent such as a sulfonylurea, thiazolidinedione (TZD), or dipeptidyl-peptidase-IV inhibitor as initial combination therapy. In some patients with severe hyperglycemia, insulin may also be required as initial therapy.
The next major recommendation directed toward CV specialists is to avoid clinical inertia in advancing therapy to achieve the A1C goal. A1C should be measured every 3 months and therapy adjusted if targets are not met. Common guidelines dictate the following: metformin doses should be gradually titrated upward to achieve maximal effectiveness (usually 2,000 to 2,500 mg/day) while minimizing gastrointestinal side effects and a second oral agent or insulin should be added if needed. The most commonly used and least expensive oral combination is metformin plus a sulfonylurea; however, hypoglycemia caused by the long-acting effects of sulfonylureas on insulin secretion may be a limiting factor. Shorter-acting insulin secretagogues such as nateglinide or repaglinide carry less risk of hypoglycemia and may be useful in this situation. Recently, sitagliptin, a dipeptidyl-peptidase-IV inhibitor, has been shown to be as effective as the sulfonylurea glipizide when added to metformin, and shows a significantly lower risk of hypoglycemia (21).
Another commonly used combination is metformin plus a TZD, which decreases both fasting and post-prandial glucose by reducing hepatic glucose production and improving peripheral insulin sensitivity. The major disadvantages of TZDs are the associated weight gain and fluid retention that can result in heart failure in some patients (22). Recent concern about a possible increased risk of MI with rosiglitazone (23), but not with pioglitazone (24), has prompted caution in using rosiglitazone in patients at high risk for CV disease (4). However, an interim analysis of another rosiglitazone trial showed no statistically significant effects on MI, although there seemed to be an increased risk of congestive heart failure (25). In the ADOPT (A Diabetes Progression Trial) study, rosiglitazone was associated with higher rates of upper and lower limb fractures compared with metformin or glyburide (26). This effect was more pronounced in women than in men. In the PERISCOPE (Pioglitazone Effect on Regression of Intravascular Sonographic Coronary Obstruction Prospective Evaluation) study, pioglitazone was associated with a higher rate of fractures compared with glimepiride (27).
Alpha-glucosidase inhibitors, such as acarbose and miglitol, may also be used to improve post-prandial hyperglycemia and reduce A1C levels in T2DM. By slowing the rate of digestion and absorption of complex carbohydrates in the intestine, alpha-glucosidase inhibitors blunt post-prandial increases in blood glucose. However, many patients find the gastrointestinal side effects associated with this drug class to be a significant deterrent, despite the option of gradual titration to minimize gastrointestinal problems; these agents, therefore, are not used extensively in the U.S. They can be very effective, however, in people who traditionally eat a diet high in complex carbohydrates.
Other effective approaches to improve glycemic control, with the added benefit of achieving significant weight loss in T2DM, include treatment with exenatide or pramlintide, both given by injection 2 to 3 times daily. Exenatide is a long-acting glucagon-like peptide-1 mimetic that increases glucose-dependent insulin secretion and restores suppression of glucagon secretion in response to meals. It also delays gastric emptying, increases satiety, and results in weight loss in the majority of patients. Both fasting and post-prandial glucose levels are improved with exenatide, which exerts its major effect post-prandially. Some cases of pancreatitis associated with exenatide treatment, including acute hemorrhagic pancreatitis, have been reported (28,29). On the basis of review of these cases, the U.S. Food and Drug Administration has asked the manufacturer to strengthen the labeling of acute pancreatitis in the product label. Pramlintide also decreases post-prandial hyperglycemia, mainly by decreasing the rate of gastric emptying, and is associated with weight loss in most patients.
Because T2DM is a progressive disease characterized by early impairment of beta-cell function and ultimately loss of beta-cell mass, insulin replacement therapy is required for many patients. Most OADs lose effectiveness over time, requiring frequent monitoring of A1C levels and adjustment of the treatment regimen to achieve or maintain adequate glycemic control. In the ADOPT study, treatment with the sulfonylurea glyburide showed a less durable effect than treatment with metformin, and both were less effective in maintaining long-term glucose control than the TZD rosiglitazone, suggesting that beta-cell function was better preserved with rosiglitazone (26). Maintenance of glycemic control with insulin, by targeting both fasting and post-prandial hyperglycemia, may also be effective in preserving beta-cell function and mass by reducing glucose toxicity and oxidative stress, as both of these effects have been implicated in the process of beta-cell destruction (2).
Insulin Therapy in T2DM
The most common first step in the initiation of insulin therapy is to use a single daily injection of a long-acting insulin analog such as glargine or detemir insulin. These modified insulins are absorbed slowly and result in nearly constant basal plasma concentrations, which in most patients last for 24 h or longer. Long-acting insulins differ from the intermediate-acting neutral protamine Hagedorn insulin, which has a shorter duration of action, requires at least 2 injections daily, and tends to show less reproducible pharmacokinetics in some patients. Although the single insulin injection is most commonly given in the evening, it may be given at any time during the day; dose adjustments are based on the fasting glucose concentration, usually targeting a value of 100 to 120 mg/dl. Adjustments are made every 2 to 3 days, until the desired target fasting glucose is achieved without significant hypoglycemia at other times.
As beta-cell function declines, it often becomes necessary to add exogenous bolus insulin therapy, using rapid-acting insulin analogs or regular human insulin, to control post-prandial hyperglycemia. Rapid-acting analogs such as lispro, aspart, or glulisine insulin are generally preferable to regular human insulin because they are more rapidly absorbed and their action more closely mimics the normal physiological insulin response to meals. The dosage is adjusted based on both the pre-meal blood glucose concentration and the estimated carbohydrate content of the meal, using a pre-determined correction factor for treating elevated glucose levels and an insulin–carbohydrate ratio to match the insulin dose to the carbohydrate load. These values are determined on an individual basis for each patient and are designed to replicate as closely as possible the normal patterns of insulin secretion throughout the day. Excellent reviews are available that discuss the various insulin preparations and their uses (30).
An alternative approach to basal–bolus therapy is the use of pre-mixed insulin preparations that combine both long-acting and short-acting insulins in various combinations, including 50:50, 75:25, and 70:30 mixtures of insulin analogs or human insulin. These preparations are usually administered 2 to 3 times daily before meals to provide a more convenient regimen for patients than the 4 daily injections required by basal–bolus therapy. In many patients, this approach is adequate, although a recent study found that a regimen using 3 daily injections of pre-mixed insulins was less effective in lowering A1C than a basal–bolus regimen involving 4 injections daily (31).
An increasing number of people with T2DM are now using insulin pumps to administer continuous subcutaneous insulin infusions of rapid-acting insulin analogs to simulate normal insulin secretory patterns throughout the entire 24-h period. With these pumps, multiple basal rates may be programmed to adjust for changing insulin requirements at different times of the day, and different patterns of bolus infusions may be chosen to match individual needs. Insulin pumps, combined with continuous glucose monitoring systems, have made it possible to fine tune blood glucose regulation to achieve excellent A1C levels with lower risk of hypoglycemia.
With all forms of insulin therapy, the major risk is hypoglycemia; careful monitoring of both pre-prandial and post-prandial blood glucose concentrations and appropriate insulin dosage adjustments are required. As A1C approaches normal values, the risk of hypoglycemia increases and may become a limiting factor to achieving the desired target. Factors such as changes in diet, physical exercise, emotional stress, or concurrent illness may require adjustments in insulin dosage. Additionally, patients with altered gastric emptying caused by gastroparesis or with hypoglycemia unawareness may have particular difficulty in adjusting insulin dosages to maintain adequate glucose control.
Many recent technological advances, including the use of insulin pens rather than traditional syringes and needles for insulin administration, have made it easier for patients to use insulin therapy to manage their diabetes. The newer models of insulin pens are convenient, easy to use, and provide more accurate dosing than conventional syringes and needles. The major disadvantage is that the pens do not allow mixing of insulin for a single injection. However, this is not a practical problem because pre-mixed insulins are available and current basal–bolus regimens do not require mixing insulin preparations.
Finally, continuous glucose monitoring systems, which record interstitial glucose concentrations for up to 7 days, are excellent tools for determining fluctuations in glucose throughout the day, providing information needed to make appropriate adjustments in insulin therapy.
Hyperglycemia, both in the fasting and the post-prandial states, is a major risk factor for microvascular and macrovascular damage in individuals with T2DM. Effective treatment programs should be implemented early in the course of the disease and adjusted at regular intervals to maintain target A1C levels as close to normal as possible, ≤7% at a minimum, without causing an excessive risk of hypoglycemia in individual patients.
Because of the progressive course of T2DM, primarily attributable to deterioration of pancreatic beta-cell function and ultimately loss of beta-cell mass, treatment with OADs usually becomes less effective over time, necessitating insulin replacement therapy in many patients.
Initially, replacement of physiological basal insulin concentrations with a once-daily injection of a long-acting modified insulin analog, in addition to continued use of OADs, is sufficient to provide adequate glycemic control. However, many patients will eventually require multiple daily injections of insulin, using a combination of long- and short-acting insulins in a basal–bolus approach to control both fasting and post-prandial glucose levels, or the use of pre-mixed insulin preparations administered 2 to 3 times daily before meals. An alternative is the use of an insulin pump to provide continuous subcutaneous insulin infusions.
A wide variety of insulin preparations are now available that make it easier to simulate the normal insulin secretory pattern. Improved insulin delivery systems, including insulin pens and programmable insulin pumps, make it easier for patients with T2DM to initiate and maintain insulin therapy to better regulate blood glucose control.
Dr. Horton has received grant or research support, consulting agreements, or honoraria from Amylin Pharmaceuticals, Inc., Eli Lilly and Company, Abbott Laboratories, Merck & Co., Inc., Medtronic, Inc., Novartis AG, Novo Nordisk, Hoffmann-La Roche, Inc., Takeda Pharmaceuticals North America, and GlaxoSmithKline.
- Abbreviations and Acronyms
- hemoglobin A1C
- confidence interval
- hazard ratio
- myocardial infarction
- oral antihyperglycemic drug
- type 2 diabetes mellitus
- Received April 7, 2008.
- Revision received November 4, 2008.
- Accepted November 6, 2008.
- American College of Cardiology Foundation
- Butler A.E.,
- Janson J.,
- Bonner-Weir S.,
- et al.
- Steppel J.H.,
- Horton E.S.
- Ford E.S.,
- Li C.,
- Little R.R.,
- Mokdad A.H.
- Stratton I.M.,
- Adlae A.I.,
- Neil H.A.,
- et al.
- Kuusisto J.,
- Mykkänen L.,
- Pyörälä K.,
- Laakso M.
- Klein R.
- Middelbeek R.J.,
- Horton E.S.
- Donahue R.P.,
- Abbott R.D.,
- Reed D.M.,
- Yano K.
- Cavalot F.,
- Petrelli A.,
- Traversa M.,
- et al.
- American Diabetes Association
- Abraira C.,
- Duckworth W.C.,
- VADT Investigators
- Nauk M.A.,
- Meininger G.,
- Sheng D.,
- Terranella L.,
- Stein P.P.,
- Sitagliptin Study 024 Group
- Nesto R.W.,
- Bell D.,
- Bonow R.O.,
- et al.,
- American Heart Association; American Diabetes Association
- Home P.D.,
- Pocock S.J.,
- Beck-Nielsen H.,
- et al.,
- RECORD Study Group
- Denker P.S.,
- Dimarco P.E.
- Rosenstock J.,
- Ahmann A.J.,
- Colon G.,
- et al.