Type 2 Diabetes Mellitus Treatment & Management

Updated: May 01, 2023
  • Author: Romesh Khardori, MD, PhD, FACP; Chief Editor: George T Griffing, MD  more...
  • Print

Approach Considerations

The goals in caring for patients with diabetes mellitus are to eliminate symptoms and to prevent, or at least slow, the development of complications. Microvascular (ie, eye and kidney disease) risk reduction is accomplished through control of glycemia and blood pressure; macrovascular (ie, coronary, cerebrovascular, peripheral vascular) risk reduction, through control of lipids and hypertension, smoking cessation, and aspirin therapy; and metabolic and neurologic risk reduction, through control of glycemia.

New abridged recommendations for primary care providers

The American Diabetes Association has released condensed recommendations for Standards of Medical Care in Diabetes: Abridged for Primary Care Providers, highlighting recommendations most relevant to primary care. The abridged version focusses particularly on the following aspects:

  • Prediabetes

  • Self-management education

  • Nutrition

  • Physical activity

  • Smoking cessation

  • Psychosocial care

  • Immunizations

  • Glycemic treatment

  • Therapeutic targets

  • Diagnosis and treatment of vascular complications

  • Intensification of insulin therapy in type 2 diabetes

The recommendations can be accessed at American Diabetes Association DiabetesPro Professional Resources Online, Clinical Practice Recommendations – 2015. [176]

Type 2 diabetes care is best provided by a multidisciplinary team of health professionals with expertise in diabetes, working in collaboration with the patient and family. [2] Management includes the following:

  • Appropriate goal setting

  • Dietary and exercise modifications

  • Medications

  • Appropriate self-monitoring of blood glucose (SMBG)

  • Regular monitoring for complications

  • Laboratory assessment

Ideally, blood glucose should be maintained at near-normal levels (preprandial levels of 90-130 mg/dL and hemoglobin A1C [HbA1c] levels < 7%). However, focus on glucose alone does not provide adequate treatment for patients with diabetes mellitus. Treatment involves multiple goals (ie, glycemia, lipids, blood pressure).

Aggressive glucose lowering may not be the best strategy in all patients. Individual risk stratification is highly recommended. In patients with advanced type 2 diabetes who are at high risk for cardiovascular disease, lowering HbA1c to 6% or lower may increase the risk of cardiovascular events. [177]

A study from the ACCORD Study Group found that setting the treatment target for HbA1c below 6% in high-risk patients resulted in reduced 5-year nonfatal myocardial infarctions. However, patients who did not achieve the treatment target experienced increased 5-year mortality. [178]

Review of blood glucose logs must be part of any diabetes management plan. Both iron and erythropoietin treatments commonly prescribed in patients with chronic kidney disease cause a significant decrease in HbA1c without affecting blood glucose levels. [179]

With each health-care system encounter, patients with diabetes should be educated about and encouraged to follow an appropriate treatment plan. Adherence to diet and exercise should continue to be stressed throughout treatment, because these lifestyle measures can have a large effect on the degree of diabetic control that patients can achieve.

A study by Morrison et al found that more frequent visits with a primary care provider (every 2 wk) led to markedly rapid reductions in serum glucose, HbA1c, and low-density lipoprotein (LDL) cholesterol levels. However, how such a strategy can work globally remains a challenge due to available resources and economic restrictions. [180]

The United Kingdom Prospective Diabetes Study

The care of patients with type 2 diabetes mellitus has been profoundly shaped by the results of the United Kingdom Prospective Diabetes Study (UKPDS). This landmark study confirmed the importance of glycemic control in reducing the risk for microvascular complications and refuted previous data suggesting that treatment with sulfonylureas or insulin increased the risk of macrovascular disease. Major findings of the UKPDS are displayed in the images below.

Major findings from the primary glucose study in t Major findings from the primary glucose study in the United Kingdom Prospective Diabetes Study (UKPDS).
Results from metformin substudy in the United King Results from metformin substudy in the United Kingdom Prospective Diabetes Study (UKPDS).
Findings from the blood pressure substudy in the U Findings from the blood pressure substudy in the United Kingdom Prospective Diabetes Study (UKPDS).

Significant implications of the UKPDS findings include the following:

  • Microvascular complications (predominantly indicated by the need for laser photocoagulation of retinal lesions) are reduced by 25% when mean HbA1c is 7%, compared with 7.9%

  • A continuous relationship exists between glycemia and microvascular complications, with a 35% reduction in risk for each 1% decrement in HbA1c; a glycemic threshold (above the upper limit of normal for HbA1c) below which risk for microvascular disease is eliminated does not appear to exist

  • Glycemic control has minimal effect on macrovascular disease risk; excess macrovascular risk appears to be related to conventional risk factors such as dyslipidemia and hypertension

  • Sulfonylureas and insulin therapy do not increase macrovascular disease risk [96]

  • Metformin reduces macrovascular risk in patients who are obese [181]

  • Vigorous blood pressure control reduces microvascular and macrovascular events; beta blockers and angiotensin-converting enzyme (ACE) inhibitors appear to be equally effective in this regard

American College of Physicians guidance statement

According to a 2018 guidance statement by the American College of Physicians (ACP), “Clinicians should aim to achieve an HbA1c level between 7% and 8% in most patients with type 2 diabetes.” The ACP said that this higher target is aimed at helping patients benefit from glycemic control while avoiding the adverse effects—associated with low blood sugar, medication burden, and costs—of stricter targets. The ACP stated that evidence does not indicate that medication therapy to reduce the HbA1c level to 7% or less results in reduced mortality or in decreased macrovascular complications, such as heart attack or stroke, compared with a reduction to about 8%. [182, 183]

However, experts from the American Diabetes Association (ADA) and the American Association of Clinical Endocrinologists (AACE) have expressed skepticism about the higher target, noting that the guidance statement does not take into account the cardiovascular disease benefits of newer drugs, which themselves frequently reduce HbA1c levels. In response, a coauthor of the ACP statement observed that other guidelines have also not specifically accounted for these newer medications in their recommended HbA1c levels and that research on such drugs has primarily been in patients either with cardiovascular disease or at high risk of developing it. [182]

Joint consensus statement on remission

In a 2021 joint consensus statement from the American Diabetes Association, the Endocrine Society, the European Association for the Study of Diabetes, and Diabetes UK, the term “remission,” as it applies to type 2 diabetes, is defined as the presence of an HbA1c level below 6.5% (< 48 mmol/mol) at least 3 months after glucose-lowering pharmacotherapy has been halted. This applies whether the remission has been achieved by way of lifestyle, bariatric surgery, or other means. [184, 185]


Pharmacologic Therapy

Early initiation of pharmacologic therapy is associated with improved glycemic control and reduced long-term complications in type 2 diabetes. Drug classes used for the treatment of type 2 diabetes include the following:

  • Biguanides

  • Sulfonylureas

  • Meglitinide derivatives

  • Alpha-glucosidase inhibitors

  • Thiazolidinediones (TZDs)

  • Glucagonlike peptide–1 (GLP-1) agonists 

  • GLP-1 and glucose-dependent insulinotropic polypeptide (GIP) agonists 

  • Dipeptidyl peptidase IV (DPP-4) inhibitors

  • Selective sodium-glucose transporter–2 (SGLT-2) inhibitors

  • Nonsteroidal mineralocorticoid receptor (MR) antagonists 

  • Insulins

  • Amylinomimetics

  • Bile acid sequestrants

  • Dopamine agonists

A literature review by Alfayez et al indicated that GLP-1 agonists, DPP-4 inhibitors, and SGLT-2 inhibitors put patients with type 2 diabetes at no additional cardiovascular risk. Moreover, patients taking GLP-1 agonists actually showed significant reduction in such risk, while those on SGLT-2 inhibitors demonstrated a significant decrease in hospitalization for heart failure events. [186]

A position statement from Primary Care Diabetes Europe (PCDE) provides a clinical decision–making model that primary care clinicians can use in the pharmacologic management of patients with type 2 diabetes, including when addressing cardiovascular disease and its risk.


Metformin is the only biguanide in clinical use. Another biguanide, phenformin, was taken off the market in the United States in the 1970s because of its risk of causing lactic acidosis and associated mortality (rate of approximately 50%). Metformin has proved effective and safe. [187] A nested case-control analysis found that, as with other oral antidiabetic drugs, lactic acidosis during metformin use is very rare and is associated with concurrent comorbidity. [188]

Metformin lowers basal and postprandial plasma glucose levels. Its mechanisms of action differ from those of other classes of oral antidiabetic agents; metformin works by decreasing hepatic gluconeogenesis. It also decreases intestinal absorption of glucose and improves insulin sensitivity by increasing peripheral glucose uptake and utilization.

A study by Sun et al suggested that metformin also exercises glucose control by reducing levels of Bacteroides fragilis in the gut. The research indicated that reduction of the microbe, which has bile salt hydrolase activity, leads to an increase in the bile acid glycoursodeoxycholic acid (GUDCA), which in turn inhibits signaling of the intestinal farnesoid X receptor (FXR), a nuclear receptor that plays a role in regulating hepatic bile acid. [189, 190]  Unlike oral sulfonylureas, metformin rarely causes hypoglycemia.

Patients on metformin have shown significant improvements in hemoglobin A1c and their lipid profile, especially when baseline values are abnormally elevated. In addition, metformin is the only oral diabetes drug that reliably facilitates modest weight loss. In the UKPDS, it was found to be successful at reducing macrovascular disease endpoints in obese patients. [191] The results with concomitant sulfonylureas in a heterogeneous population were conflicting, [192] but overall, this drug probably improves macrovascular risk.

In January 2017, the American College of Physicians (ACP) released a guideline update recommending the use of metformin as a first-line treatment for type 2 diabetes. The update also recommended consideration of the addition of a drug from one of the following classes—sulfonylureas, thiazolidinediones, dipeptidyl peptidase IV (DPP-4) inhibitors, or selective sodium-glucose transporter–2 (SGLT-2) inhibitors—to metformin when a second oral therapy is thought to be needed to aid glycemic control. However, the second recommendation was graded as "weak," with the evidence of moderate quality, by the ACP. [193, 194]

A study by Vashisht et al that examined data from more than 246.5 million patients found that when used along with metformin therapy for type 2 diabetes, treatment with sulfonylureas, DPP-4 inhibitors, or thiazolidinediones was equally effective in reducing the HbA1c level to 7% or below that of total hemoglobin. However, compared with DPP-4 inhibitors, there was a slightly increased risk of myocardial infarction and eye disorders associated with sulfonylureas. [195, 196]

Kooy et al found improvements in body weight, glycemic control, and insulin requirements when metformin was added to insulin in patients with type 2 diabetes mellitus. No improvement of an aggregate of microvascular and macrovascular morbidity and mortality was observed; however, reduced risk of macrovascular disease was evident after a follow-up period of 4.3 years. These results support continuing metformin treatment after the introduction of insulin in patients with type 2 diabetes mellitus. [197]

Pradhan et al did not find an association between improvement of glycemic control with metformin or insulin and reduction of inflammatory biomarker levels in patients with recent-onset type 2 diabetes. [198] Patients were randomized to 1 of 4 groups: placebo, placebo plus insulin glargine, metformin only, and metformin and insulin glargine. No difference in levels of the inflammatory biomarker high-sensitivity C-reactive protein was shown between study participants who received insulin or metformin and those who did not.

A retrospective, nationwide cohort study found that metformin is associated with a low risk of mortality in patients who have diabetes and experience heart failure compared with treatment that includes a sulfonylurea or insulin. [199] Roussel et al studied the expanded use of metformin in groups of patients with diabetes previously considered high risk for possible drug-related adverse outcome and found a decrease in mortality in these patients. [200]

A study by Gross et al found no difference in benefit between drug classes in patients already on metformin and sulfonylurea. The patient's clinical circumstances must guide selection. [201]

In a meta-analysis of 20 publications comprising 13,008 cancer patients with concurrent type 2 diabetes, Yin et al found that patients treated with metformin had better overall and cancer-specific survival than those treated with other types of glucose-lowering agents. [121, 122] These improvements were observed across cancer subtypes and geographic locations.

Risk reduction was significant among patients with prostate, pancreatic, breast, colorectal and other cancers, but not for lung cancer. [122] However, it remains unclear whether metformin can modulate clinical outcomes in cancer patients with diabetes.


Sulfonylureas (eg, glyburide, glipizide, glimepiride) are insulin secretagogues that stimulate insulin release from pancreatic beta cells and probably have the greatest efficacy for glycemic lowering of any of the oral agents. However, that effect is only short-term and quickly dissipates. Sulfonylureas may also enhance peripheral sensitivity to insulin secondary to an increase in insulin receptors or to changes in the events following insulin-receptor binding.

Sulfonylureas are indicated for use as adjuncts to diet and exercise in adult patients with type 2 diabetes mellitus. They are generally well-tolerated, with hypoglycemia the most common side effect. The first-generation sulfonylureas are acetohexamide, chlorpropamide, tolazamide, and tolbutamide; the second-generation agents are glipizide, glyburide, and glimepiride. The structural characteristics of the second-generation sulfonylureas allow them to be given at lower doses and as once-daily regimens.

One study exonerated the sulfonylurea group of oral agents as the chief cause of cardiovascular death in diabetic patients admitted with acute myocardial infarction. However, even though sulfonylureas were safer in general, within the group, the use of glyburide was associated with highest mortality (7.5%) compared with other sulfonylureas, such as gliclazide and glimepiride (2.7%). [202] This raises an important concern about whether the use of glyburide should be avoided.

Meglitinide derivatives

Meglitinides (eg, repaglinide, nateglinide) are much shorter-acting insulin secretagogues than the sulfonylureas are, with preprandial dosing potentially achieving more physiologic insulin release and less risk for hypoglycemia. [203] Although meglitinides are considerably more expensive than sulfonylureas, they are similar in their glycemic clinical efficacy.

Meglitinides can be used as monotherapy; however, if adequate glycemic control is not achieved, then metformin or a thiazolidinedione may be added. Meglitinides may be used in patients who have allergy to sulfonylurea medications. They have a similar risk for inducing weight gain as sulfonylureas do but possibly carry less risk for hypoglycemia.

Alpha-glucosidase inhibitors

These agents delay sugar absorption and help to prevent postprandial glucose surges. Alpha-glucosidase inhibitors prolong the absorption of carbohydrates, but their induction of flatulence greatly limits their use. They should be titrated slowly to reduce gastrointestinal (GI) intolerance.


TZDs (eg, pioglitazone [Actos], rosiglitazone [Avandia]) act as insulin sensitizers; thus, they require the presence of insulin to work. They must be taken for 12-16 weeks to achieve maximal effect.

These agents are used as monotherapy or in combination with sulfonylurea, metformin, meglitinide, DPP-4 inhibitors, GLP-1 receptor agonists, or insulin. They are the only antidiabetic agents that have been shown to slow the progression of diabetes (particularly in early disease).

In the Canadian Normoglycemia Outcome and Evaluation (CANOE) trial, glycemic parameters and insulin sensitivity improved in patients taking rosiglitazone and metformin in year 1 but deteriorated in the years thereafter, as in the placebo arm. Beta-cell function remained relatively stable in both groups for the first 2 years but then deteriorated progressively in subsequent years. The investigators attributed the lower rate of incident diabetes in the rosiglitazone/metformin group to the early effect of treatment. [204]

In a study by DeFronzo et al, pioglitazone was found to reduce the progression to frank diabetes by 72% in patients with IGT. [205] However, the drug was associated with significant edema and weight gain.

In the Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication (DREAM) trial, rosiglitazone reduced the incidence of diabetes by 62%. It also improved the achievement of normoglycemia by 70% in patients with IFG and by 64% in patients with both IFG and IGT. [206]

A study by Phung et al investigated oral agents used for prevention of type 2 diabetes and found that TZDs resulted in a greater risk reduction than biguanides. Sulfonylureas and glinides had no benefit. [207]

TZDs generally decrease triglyceride levels and increase HDL cholesterol levels. They increase LDL cholesterol, but this increase may involve large, buoyant LDL, which may be less atherogenic.

Pioglitazone in patients unresponsive to combination therapy

Charpentier et al concluded that the early addition of pioglitazone in patients who are not responding to dual therapy is beneficial, decreasing HbA1c, as well as improving FPG levels and other surrogate markers. [208] In this study, patients (n=299) with type 2 diabetes mellitus uncontrolled by combination therapy with metformin and a sulfonylurea or a glinide were randomly assigned to receive add-on therapy with either pioglitazone 30 mg daily or a placebo.

Among patients with a baseline HbA1c level of less than 8.5%, 44.4% of patients in the pioglitazone group achieved an HbA1c level of less than 7% after 7 months, compared with only 4.9% of patients in the placebo group. In patients with a baseline HbA1c level of 8.5% or greater, 13% of those in the pioglitazone group achieved an HbA1c level of less than 7%, while no patients in the placebo group saw the same reduction. [208]

Adverse effects

While TZDs have many desirable effects on inflammation and the vasculature, edema (including macular edema) and weight gain may be problematic adverse effects, especially when TZDs are administered with insulin or insulin secretagogues. [209] These effects may induce or worsen heart failure in patients with left ventricular compromise and occasionally in patients with normal left ventricular function. TZDs have not been tested in patients with New York Heart Association class III or IV heart failure.

Fluid retention from TZDs has been considered resistant to treatment with loop diuretics, because of upregulation of renal epithelial sodium channels. However, a randomized, double-blind, placebo-controlled, crossover study by Rennings et al found that response to the loop diuretics furosemide and amiloride were preserved in rosiglitazone-treated subjects with insulin resistance. [210]

The use of pioglitazone for more than 2 years is weakly associated with an increased bladder cancer risk, with the highest risk among patients who took pioglitazone the longest and at the highest cumulative doses. [211, 212, 213] Constant surveillance and vigilance is needed. Ninety-five percent of these cases were detected in early stage. The US Food and Drug Administration (FDA) currently recommends not prescribing pioglitazone for patients with active bladder cancer and using it with caution in patients with a history of bladder cancer.

A meta-analysis indicated that in women with type 2 diabetes, long-term (ie, 1 y or longer) use of TZDs doubles the risk of fracture. [214] Although in this study, TZDs were not found to have significantly increased fracture risk among men with type 2 diabetes, risk of fracture in males has since been reported.

Rosiglitazone restrictions

In response to data suggesting an elevated risk of myocardial infarction in patients treated with rosiglitazone, the FDA has restricted access to this drug. [215] The use of rosiglitazone is limited to patients already being successfully treated with this agent and to patients whose blood sugar cannot be controlled with other antidiabetic medicines and who do not wish to use pioglitazone, the only other TZD currently available.

Health-care providers and patients must be enrolled in the Avandia-Rosiglitazone Medicines Access Program in order to prescribe and receive rosiglitazone. Patients who are enrolled in the access program receive their medicine by mail order through certified pharmacies that participate in the program.

Glucagonlike peptide–1 agonists

GLP-1 agonists (ie, exenatide, liraglutide, albiglutide, dulaglutide) mimic the endogenous incretin GLP-1; they stimulate glucose-dependent insulin release, reduce glucagon, and slow gastric emptying. The use of a GLP-1 in addition to metformin and/or a sulfonylurea may result in modest weight loss. Animal data suggest that these drugs prevent beta-cell apoptosis and may in time restore beta-cell mass. The latter property, if proven in humans, would have tremendous therapeutic potential.


A comparison by Bunck et al of 1 year's therapy with either exenatide or insulin glargine in metformin-treated patients with type 2 diabetes found that exenatide provided significantly greater improvement in beta-cell function. Reduction in HbA1c was similar with the 2 drugs. Beta-cell function and glycemic control returned to pretreatment values following discontinuation of exenatide or insulin glargine, suggesting that long-term treatment is required to maintain the beneficial effects of these drugs. [216]

The addition of exenatide in patients receiving insulin glargine as basal insulin helps to improve glycemic control without the risk of increased hypoglycemia or weight gain. This benefit, however, is accompanied by a significant increase in adverse events such as nausea, diarrhea, vomiting, and headache. [217]

Exenatide has greater ease of titration (only 2 possible doses, with most patients progressing to the higher dose) than does insulin. Although the original product requires twice-daily injections, a long-acting exenatide formulation that is given once weekly (Bydureon) has been developed and has been found to provide significantly greater improvement in glycemic control than does the twice-daily formulation. [218] Once-weekly exenatide injections result in improvements in glycemic control and body weight regardless of age, gender, race, duration of diabetes or BMI. [219] Bydureon was approved by the FDA in January 2012.

In the DURATION-5 (Diabetes Therapy Utilization: Researching Changes in A1C, Weight and Other Factors Through Intervention With Exenatide Once Weekly) study, the exenatide once-weekly formulation provided significantly greater improvement in HbA1c and FPG levels than did the twice-daily preparation. Additionally, less nausea was observed with the once-weekly exenatide formulation. [220]

For patients with type 2 diabetes inadequately controlled with metformin, the injectable agent exenatide was found, in one clinical trial, to be more effective than insulin detemir. [221, 222] A clinical trial involving 216 patients with A1c baseline levels >7.1% despite treatment with metformin found that once-daily injections of exenatide resulted in a significantly greater number of patients achieving target A1c than treatment with detemir. At 26 weeks, 44.1% of the exenatide group had achieved an A1c of 7% or less compared to 11.4% of the detemir group.


Indicated as an adjunct to diet and exercise, liraglutide therapy is aimed at improving glycemic control in adults with type 2 diabetes. In addition, results from the LEADER (Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results) clinical trial led to liraglutide’s approval for risk reduction of major cardiovascular events (cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke) in adults with type 2 diabetes mellitus and established cardiovascular disease. With a median follow-up of 3.8 years, risk for the trial’s composite endpoint of cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke was 13.0% for patients treated with liraglutide, compared with 14.9% for those given placebo therapy. [223]

In June 2019, liraglutide was approved by the FDA for use in children aged 10 years or older with type 2 diabetes. Approval was based on the ELLIPSE clinical trial, in which patients aged 10 years to less than 17 years received up to 1.8 mg/day of subcutaneous liraglutide or placebo. After 26 weeks, the mean HbA1c level was 0.64% lower in the liraglutide patients and 0.42% higher in the placebo patients. [224]


The glucagonlike peptide-1 (GLP-1) receptor agonist albiglutide (Tanzeum) was approved by the FDA in April 2014 as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. [225, 226] GLP-1 agonists augment glucose-dependent insulin secretion. Approval of albiglutide was based on a series of individual phase III trials (Harmony 1-8) that included approximately 5,000 individuals.

In an open-label 32-week study in 805 patients with type 2 diabetes inadequately controlled with oral drugs, Pratley and colleagues found that reductions in HbA1c with once-weekly albiglutide injections were clinically meaningful but less than those seen with daily liraglutide injections (0.78% vs 0.99%, respectively). Patients who received albiglutide had fewer gastrointestinal events than those who received liraglutide (35.9% vs 49.9%) but had more injection-site reactions (12.9% vs 5.4%) and less weight loss (0.64 vs 2.19 kg). [227, 228]

The dosage of albiglutide in the study was 30 mg once weekly titrated to 50 mg at week 6. The dosage of liraglutide was 0.6 mg once daily titrated to 1.2 mg at week 1 and 1.8 mg at week 2. [227, 228]


Dulaglutide (Trulicity) was approved by the FDA in September 2014 as adjunctive therapy to diet and exercise to improve glycemic control in type 2 diabetes mellitus. [229] It is administered as a once-weekly subcutaneous injection. [229, 230] Approval was based on six clinical trials (AWARD studies) involving a total of 3342 patients who received dulaglutide as monotherapy or as part of combination therapy. [229] Dulaglutide was noninferior to daily liraglutide in one study and superior to the oral dipeptidyl peptidase–4 (DPP-4) inhibitor sitagliptin in another. Adverse effects included nausea, diarrhea, vomiting, abdominal pain, and decreased appetite. [229]

AWARD-1 compared dulaglutide weekly doses of 0.75 mg or 1.5 mg compared with exenatide injectable solution BID. The mean A1C reductions were dulaglutide, 1.5 mg, 1.5%; 0.75 mg, 1.3%; exenatide solution, 1.0%; placebo, 0.5%. [230]

In September 2020, the FDA approved higher doses (ie, 3 mg or 4.5 mg weekly) of dulaglutide for patients requiring additional glycemic control. This approval was based on results from the AWARD-11 clinical trial. [231]  

AWARD-3 compared dulaglutide with insulin glargine titrated to target. Mean A1C reductions were dulaglutide 1.5 mg, 1.1-1.6%; 0.75 mg, 0.8-1.6%; and insulin glargine 0.6-1.4%. Dulaglutide was shown to be noninferior as monotherapy compared with metformin in the AWARD-3 trial. Mean A1C reductions were dulaglutide 1.5 mg, 0.8%; dulaglutide 0.75 mg, 0.7%; compared with metformin 0.6%. [232]

AWARD-5 compared dulaglutide with sitagliptin in patients taking metformin. At the 52-week primary endpoint, mean A1C reductions were dulaglutide 1.5 mg, 1.1%; 0.75 mg, 0.9%; compared with sitagliptin 0.4%. [233]

In February 2020, the FDA approved dulaglutide for major adverse cardiovascular (CV) event reduction (with regard to CV death, nonfatal myocardial infarction, and nonfatal stroke) in adults with type 2 diabetes mellitus in whom established CV disease or multiple CV risk factors are present. Approval stemmed from the REWIND (Researching cardiovascular Events with a Weekly INcretin in Diabetes) study, a multicenter, randomized, double-blind, placebo-controlled trial that found that the incidence of primary composite outcomes (nonfatal myocardial infarction, nonfatal stroke, or CV-related death) was 2.4 per 100 person-years in the dulaglutide group, versus 2.7 per 100 person-years in the placebo group. [234]

Dulaglutide is not recommended for use as first-line pharmacologic treatment for type 2 diabetes, and it is contraindicated in patients with personal or family history of medullary thyroid carcinoma or in those with multiple endocrine neoplasia syndrome type 2. [229] The label will include a boxed warning that thyroid C-cell tumors have been observed in animal studies. Required postmarketing studies will include studies in children, a medullary thyroid carcinoma case registry, and a cardiovascular outcomes study in high-risk patients. [229]


Lixisenatide (Adlyxin) was approved by the FDA in July 2016 as adjunctive therapy to diet and exercise to improve glycemic control in type 2 diabetes mellitus. It is administered by subcutaneous injection once daily within 1 hour before the first meal of the day. The starting dose is 10 mcg/day SC for 14 days and is then increased on day 15 to 20 mcg once daily.

The FDA approved lixisenatide based on results from the GetGoal worldwide clinical program and from the Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trials. [235, 236, 237]

The drug's safety and efficacy was assessed by the GetGoal program, which included 13 clinical trials of adults with type 2 diabetes mellitus (n >5000). The primary efficacy endpoint, HbA1c reduction, was achieved by all of the GetGoal studies. [235, 236]

The ELIXA study demonstrated that in patients with type 2 diabetes who had had a recent acute coronary syndrome, cardiovascular adverse events did not increase in those taking lixisenatide compared with patients taking placebo. The study included 6068 adults with type 2 diabetes, 39% of whom had a recent non–ST-segment-elevation myocardial infarction, 43% of whom had ST-segment-elevation myocardial infarction, and 17% of whom had unstable angina. At a median of 25 months, the investigators found that cardiovascular death, myocardial infarction, stroke, or hospitalization for unstable angina had occurred in 13.4% of the lixisenatide patients and 13.2% of the control patients. [237]

A study to assess efficacy and safety of lixisenatide monotherapy in type 2 diabetes found a once-daily dose of the drug improved glycemic control. Once-daily monotherapy significantly lowered postprandial glucose and was well tolerated by patients with type 2 diabetes. [238]


In December 2017, the FDA approved once-weekly semaglutide SC (Ozempic), a GLP-1 receptor agonist, as a glycemic control–improvement agent in adults with type 2 diabetes. It is administered as a subcutaneous injection once weekly. Meant as an adjunct to diet and exercise, semaglutide was approved following eight phase 3a studies (the SUSTAIN trials). 

Semaglutide SC gained FDA approval in January 2020 for risk reduction of major adverse cardiovascular events (MACE) in adults with type 2 diabetes and heart disease. Results from the 2-year randomized study SUSTAIN 6 (Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes) led to approval for the new indication. In the trial, 3297 adults with type 2 diabetes and established cardiovascular disease (CVD) received either injectable semaglutide or placebo, in addition to stand-of-care therapy. [239, 240]

Risk for the primary composite outcome—first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke—for patients treated with semaglutide SC was a significant 26% below that for placebo patients. However, cardiovascular risk reduction primarily resulted from declines in nonfatal stroke (39%) and nonfatal myocardial infarction (26%), with cardiovascular death rates being similar between the semaglutide and placebo patients. [239, 240, 241]

In September 2019, the FDA approved the first oral GLP-1 receptor agonist, a form of semaglutide available under the brand name Rybelsus. It is administered as a once-daily oral tablet. Approval of the oral tablet was based on results from the phase 3 PIONEER trials (n=9543). The trials included head-to-head studies of oral semaglutide compared with sitagliptin (DP4 inhibitor), empagliflozin (sodium-glucose cotransporter-2 [SGLT2] inhibitor), and liraglutide 1.8 mg (GLP-1 agonist). A1c reduction was found with oral semaglutide, as well as, via a secondary endpoint, body weight reduction. [242, 243, 244]

In January 2020, the FDA updated the clinical studies section of Rybelsus’s prescribing information, with results added from the randomized, placebo-controlled study PIONEER 6 (Trial Investigating the Cardiovascular Safety of Oral Semaglutide in Subjects With Type 2 Diabetes), which were released in June 2019. The report found oral semaglutide to be associated with a nonsignificant 21% reduction in three-component MACE. The information added to Rybelsus’s label related to cardiovascular safety, not benefit. [245, 240]  

Dual glucagonlike peptide-1 and glucose-dependent insulinotropic polypeptide agonist 


A dual agonist of glucagon-like peptide-1 (GLP-1) receptors and glucose-dependent insulinotropic polypeptide (GIP), tirzepatide was approved by the FDA in May 2022 for improvement of glycemic control in adults with type 2 diabetes mellitus, serving as an adjunct to lifestyle modifications. Insulin secretion is stimulated, glucagon secretion is suppressed, and gastric emptying is delayed by GLP-1 agonists, while, depending on conditions of hypoglycemia or hyperglycemia, GIP agonists may raise or lower glucagon levels, respectively. However, as a dual agonist, tirzepatide may more greatly impact glucose lowering and weight control than will a selective GLP-1 agonist.

In a phase-3 trial comparing tirzepatide with semaglutide in patients with type 2 diabetes, Frias et al found that, in terms of lowering HbA1c, tirzepatide at all doses was noninferior and superior to semaglutide. With regard to body weight reduction, tirzepatide was again determined to be superior to semaglutide. [246]   

Dipeptidyl peptidase IV inhibitors

DPP-4 inhibitors (eg, sitagliptin, saxagliptin, linagliptin) are a class of drugs that prolong the action of incretin hormones. DPP-4 degrades numerous biologically active peptides, including the endogenous incretins GLP-1 and glucose-dependent insulinotropic polypeptide (GIP). DPP-4 inhibitors can be used as a monotherapy or in combination with metformin or a TZD. They are given once daily and are weight neutral.

A study comparing the efficacy and safety of monotherapy with sitagliptin or metformin in treatment-naive patients with type 2 diabetes found no statistical differences between the 2 drugs in terms of decreases in HbA1c and fasting glucose levels. The 1050 participants in the study had baseline HbA1c levels of 6.5-9% and received sitagliptin (100 mg qd) or metformin (1000 mg bid) for 24 weeks. [247]

In this study, the incidence of adverse GI effects was lower with sitagliptin than with metformin (11.6% vs 20.7%). Specifically, diarrhea (3.6% vs 10.9%) and nausea (1.1% vs 3.1%) were significantly less common with sitagliptin. [247]

A study by Vilsboll et al in patients receiving stable-dose insulin therapy (with or without concomitant metformin) found that the addition of sitagliptin produced a greater reduction in FPG (by 15 mg/dL [0.8 mmol/L]) and 2-hour postprandial glucose (by 36.1 mg/dL [2 mmol/L]) than did placebo. Sitagliptin reduced HbA1c by 0.6%, while no reduction was seen with placebo. In addition, 13% of patients attained an HbA1c level of less than 7% with sitagliptin, compared with 5% with placebo. [248]

A study by Pérez-Monteverde et al found that a combination of sitagliptin and metformin was associated with improved glycemic control and less weight gain when compared with pioglitazone in the treatment of patients with type 2 diabetes mellitus. [249]

Adding linagliptin to treatment in patients with type 2 diabetes mellitus that has been inadequately controlled with a metformin and sulfonylurea combination improves glycemic control. Because it has predominantly nonrenal excretion and is a clinically nonrelevant substrate for cytochrome-450 isoenzymes, this drug possesses the benefits of having a low risk of drug-drug interaction and of being safe to use in patients with renal insufficiency. [250]

Upper respiratory tract infections have been increasingly reported among users of DPP-4 inhibitors compared with users of other antidiabetic drugs. [251] However, further research is needed to evaluate the scope and underlying mechanisms of this phenomenon. On the other hand, a meta-analysis suggested that treatment with DPP-4 inhibitors could reduce the risk of bone fractures. [252]

A case-control, retrospective, observational study by Solerte et al suggested that in patients with type 2 diabetes mellitus who are hospitalized with COVID-19, glucose lowering with sitagliptin rather than treatment with insulin reduces the likelihood of 30-day death by 77%. The investigators also indicated that the rate at which intensive care or mechanical ventilation is needed in these patients is significantly reduced as well. [253, 254]

Selective sodium-glucose transporter-2 inhibitors

Canagliflozin is the first SGLT-2 inhibitor approved in the United States. [255, 256] SGLT-2 inhibition lowers the renal glucose threshold (ie, the plasma glucose concentration that exceeds the maximum glucose reabsorption capacity of the kidney). Lowering the renal glucose threshold results in increased urinary glucose excretion. A second SGLT-2 inhibitor, dapagliflozin (Farxiga), was approved by the FDA in January 2014, [257, 258] and another, empagliflozin, approved in August, 2014. [259, 260]

Dosage adjustments are required for canagliflozin in patients who have renal impairment (ie, estimated glomerular filtration rate [eGFR] < 60 mL/min/1.73 m2). Dapagliflozin should not be used if eGFR is < 60 mL/min/1.73 m2. Also consider lowering the dose of insulin or insulin secretagogues to reduce the risk of hypoglycemia when coadministered with SGLT-2 inhibitors.

FDA approval of canagliflozin was based on global phase 3 clinical trials that included over 10,000 patients. [255, 256] In a trial evaluating canagliflozin monotherapy efficacy and safety in 584 adults with type 2 diabetes mellitus inadequately controlled with diet and exercise, treatment for 26 weeks with canagliflozin 100 or 300 mg daily resulted in a statistically significant improvement in HbA1C with both doses compared with placebo. [261]

Canagliflozin add-on combination therapy to metformin and/or sulfonylureas showed a reduction in fasting glucose and a greater proportion of patients achieving an HbA1C level less than 7%. [262] Add-on therapy to insulin and comparative data to thiazolidinediones and to dipeptidyl peptidase-IV inhibitors have also shown improved postprandial glucose levels and HbA1C levels. [262]

In October 2019, canagliflozin received FDA approval for the treatment of diabetic kidney disease (DKD) and, in patients with type 2 diabetes and DKD, reduction in the risk of hospitalization for heart failure. Approval stemmed from the outcome of the phase 3 CREDENCE study, which found that at median 2.62-year follow-up, the risk of renal failure and cardiovascular events was lower in patients with type 2 diabetes and kidney disease who were treated with canagliflozin than in patients on placebo. [263]

Dapagliflozin is indicated as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. [257, 258]  It can be employed as monotherapy, as initial therapy with metformin, or as an add-on to other oral glucose-lowering agents, including metformin, pioglitazone, glimepiride, sitagliptin, and insulin. [264, 265, 266, 267]  

In October 2019, dapagliflozin gained an indication to reduce hospitalization for heart failure in adults with type 2 diabetes and cardiovascular risk. FDA approval was based on the Dapagliflozin Effect on Cardiovascular Events–Thrombolysis in Myocardial Infarction 58 (DECLARE-TIMI 58) study, in which 17,160 patients were evaluated (median 4.2-year follow-up). The investigators found that in patients on dapagliflozin, the rate of cardiovascular death or hospitalization for heart failure was 4.9%, compared with 5.8% for patients on placebo. [268]  

Like dapagliflozin, empagliflozin is also approved as an adjunct to diet and exercise to improve glycemic control. The drug’s safety and effectiveness were evaluated in 7 clinical trials with 4480 patients with type 2 diabetes. The pivotal trials showed that empagliflozin improved hemoglobin A1c levels compared with placebo. [259, 260]  In late 2016, the FDA also approved empagliflozin for a new indiction, specifically, the prevention of cardiovascular disease–related death in adults with type 2 diabetes who also have cardiovascular disease. [269, 270]  The new approval was based on results from the (Empagliflozin) Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME), which included more than 7000 patients. [271]

Indicated as an adjunct to diet and exercise, ertugliflozin therapy is aimed at improving glycemic control in adults with type 2 diabetes. Ertugliflozin's approval stemmed from a series of nine phase-3 clinical trials in which statistically significant improvements were seen in HbA1c, fasting plasma glucose, body weight, and systolic and diastolic blood pressures, in adult patients with type 2 diabetes. [272, 273, 274]

Nonsteroidal mineralocorticoid receptor antagonists

In July 2021, the FDA approved finerenone (Kerendia) for inhibition of the effects in adults of chronic kidney disease associated with type 2 diabetes, including sustained estimated glomerular filtration rate (eGFR) decline, end-stage kidney disease, cardiovascular death, nonfatal myocardial infarction, and hospitalization for heart failure. It is the first nonsteroidal mineralocorticoid receptor (MR) antagonist to be approved for this purpose. Approval was based on the FIDELIO-DKD trial, a placebo-controlled study that involved over 5700 patients with type 2 diabetes to whom the maximum-tolerated dose of renin-angiotensin system inhibitor (RASI) was already being administered. However, until more data on finerenone is gathered, RASIs and SGLT-2 inhibitors will be the preferred agents for slowing chronic kidney disease in type 2 diabetes. [275, 276, 277]


Ultimately, many patients with type 2 diabetes mellitus become markedly insulinopenic. The only therapy that corrects this defect is insulin. Because most patients are insulin resistant, small changes in insulin dosage may make no difference in glycemia in some patients. Furthermore, because insulin resistance is variable from patient to patient, therapy must be individualized in each patient.

A study by de la Pena et al found that although the overall insulin exposure and effects of 500 U/mL of insulin were similar to those of 100 U/mL of insulin, peak concentration was significantly lower at 500 U/mL, and the effect after the peak was prolonged; areas under the curve were similar for both doses. This observation should help guide therapy. [278]

A range of insulin preparations, individual and premixed, is currently available. The Agency for Healthcare Research and Quality (AHRQ) has reviewed the use of premixed insulin analogues in patients with type 2 diabetes mellitus. Conclusions for which the strength of evidence was high are as follows [279] :

  • For lowering postprandial glucose, premixed insulin analogues are more effective than either long-acting insulin analogues alone or premixed neutral protamine Hagedorn (NPH)/regular human insulin 70/30

  • For lowering HbA1c, premixed insulin analogues are as effective as premixed NPH/regular human insulin 70/30 and more effective than long-acting insulin analogues

  • The frequency of hypoglycemia reported with premixed insulin analogues is similar to that with premixed human insulin and higher than that with oral antidiabetic agents

Long-acting insulins used in the United States include insulin glargine (Lantus, Toujeo) and insulin detemir (Levemir). Insulin glargine has no peak and produces a relatively stable level lasting more than 24 hours. In some cases, it can produce a stable basal serum insulin concentration with a single daily injection, though patients requiring lower doses typically are given twice-daily injections. Insulin detemir has a duration of action that may be substantially shorter than that of insulin glargine but longer than those of intermediate-acting insulins.

Toujeo 300 U/mL is a newer dosage strength and form of insulin glargine than Lantus 100 U/mL, having been approved by the US Food and Drug Administration (FDA) in February 2016. Compared with those of Lantus 100 U/mL, the pharmacokinetic and pharmacodynamic profiles of Toujeo are more stable and prolonged; the duration of action exceeds 24 hours. Clinical trials showed comparable glycemic control between Lantus and Toujeo, although the trials noted the need for higher daily basal insulin doses (ie, 12-17.5%) with Toujeo. The risk for nocturnal hypoglycemia was lower with Toujeo in insulin-experienced patients with type 2 diabetes, but this was not the case for insulin-naïve patients with type 1 DM or for patients with type 2 DM. [280]

With its March 2018 approval by the FDA, Toujeo Max SoloStar became the highest capacity long-acting insulin pen on the market. Toujeo Max necessitates fewer refills and, for some diabetes patients, fewer injections to deliver the required Toujeo dosage. [281]

A new ultralong-acting basal insulin, insulin degludec (Tresiba), which has a duration of action of up to beyond 42 hours, has been approved by the US Food and Drug Administration (FDA). This new basal insulin forms a soluble multihexamer after subcutaneous injection to provide a depot effect that is long lasting. It is indicated for diabetes mellitus types 1 and 2. A combination product of insulin degludec and the rapid-acting insulin aspart was also approved (Ryzodeg 70/30). Approval was based on results from the BEGIN trial [282, 283] that showed noninferiority to comparator productions. The cardiovascular outcomes trial (DEVOTE) comparing cardiovascular safety of insulin degludec to that of insulin glargine in patients with type 2 DM is ongoing. A combination product (Ryzodeg) was also approved that contains insulin degludec plus a rapid-acting insulin (insulin aspart).

A study by Zinman et al found that insulin degludec provides comparable glycemic control to insulin glargine without additional adverse effects. [284] A reduced dosing frequency may be possible because of its ultralong-action profile. Careful study is needed when making a decision regarding reduced dosing frequency.

A rapid-acting inhaled insulin powder (Afrezza) for types 1 and 2 diabetes mellitus was approved by the FDA in June 2014. Approval was based on a study involving over 3,000 patients over a 24-week period. In persons with type 1 diabetes, the inhaled insulin was found to be noninferior to standard injectable insulin when used in conjunction with basal insulin at reducing hemoglobin A1c. In persons with type 2 diabetes, the inhaled insulin was compared to placebo inhalation in combination with oral diabetic agents and showed a statistically significant lower hemoglobin A1c. [285]

The first inhaled insulin (Exubera) was approved by the FDA in January 2006 as a rapid-acting prandial insulin. It did not produce better glycemic control than did conventionally injected insulins, and it required a mildly cumbersome device and skill to deliver an accurate dose (up to a few minutes to deliver 1 dose) and pulmonary function monitoring due to concerns about lung toxicity over time. Exubera was withdrawn from the market in October 2007, not because of safety concerns but because too few patients were using the product for its continued sale to be economically feasible.

In September 2017, the FDA approved the rapid-acting insulin aspart Fiasp for the treatment of adults with diabetes. This human insulin analog is formulated with niacinamide, which aids in speeding the initial absorption of insulin. Dosing can occur at the beginning of a meal or within 20 minutes after the meal commences. In a study of adult patients with type 1 DM, Fiasp could be detected in the circulation about 2.5 minutes after it was administered. Maximum insulin levels occurred approximately 63 minutes after the drug’s administration. [286, 287]

Insulin and cancer

On July 1, 2009, the FDA issued an early communication regarding a possible increased risk of cancer in patients using insulin glargine (Lantus). [288] The FDA communication was based on 4 observational studies that evaluated large patient databases and found some association between insulin glargine (and other insulin products) and various types of cancer.

Further evaluation is warranted, however, before the link between insulin use and cancer is confirmed. The duration of these observational studies was shorter than that considered to be necessary to evaluate for drug-related cancers. Additionally, findings were inconsistent within and across the studies, and patient characteristics differed across treatment groups.

In a study by Suissa et al, insulin glargine use was not associated with an increased risk of breast cancer during the first 5 years of use. The risk tended to increase after 5 years, however, and significantly so for the women who had taken other forms of insulin before starting insulin glargine. [289]

A study by Johnson et al found the same incidence rate for all cancers in patients receiving insulin glargine as in patients not receiving the drug. Overall, no increase in breast cancer rates was associated with insulin glargine use, although patients who used only insulin glargine had a higher rate of cancer than did those who used another type of insulin. This finding was attributed to allocation bias and differences in baseline characteristics. [290]

A study by Steansdottir showed that different drug regimens used to accomplish intensified glycemic control did not alter the risk of cancer in patients with diabetes. [291] This study differs from previous studies, in which metformin use was associated with lower cancer risk.

The FDA states that patients should not stop taking insulin without consulting their physician. An ongoing review by the FDA will continue to update the medical community and consumers with additional information as it emerges. Statements from the ADA and the European Association for the Study of Diabetes called the findings conflicting and inconclusive and cautioned against overreaction.


Pramlintide acetate is an amylin analog that mimics the effects of endogenous amylin, which is secreted by pancreatic beta cells. This agent delays gastric emptying, decreases postprandial glucagon release, and modulates appetite. [292]

Bile acid sequestrants

Bile acid sequestrants were developed as lipid-lowering agents for the treatment of hypercholesterolemia but were subsequently found to have a glucose-lowering effect. The bile acid sequestrant colesevelam is FDA-approved as an adjunctive therapy to improve glycemic control. It has a favorable, but insignificant, impact on FPG and HbA1c levels. [293]

A study in patients with early type 2 diabetes who were receiving metformin found that the addition of colesevelam reduced HbA1c levels to a degree that was statistically significant but that may have been clinically irrelevant, as no data show that a 0.3% reduction of HbA1c produces a better outcome than a 0.2% reduction of HbA1c. Achievement of LDL cholesterol goals was also improved with the use of colesevelam, but it is not known whether that result correlates with significantly different outcomes in these patients. [294]

Colesevelam is a relatively safe addition to the menu of choices available to reduce LDL cholesterol in patients with prediabetes. It should be avoided in patients with hypertriglyceridemia (a rule that applies to bile acid sequestrants in general).

Dopamine agonists

In 2009, the FDA approved a quick-release formulation of bromocriptine mesylate (Cycloset) as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. Bromocriptine is a centrally acting dopamine D2 receptor agonist. When given in a single timed morning dose, it is thought to act on circadian neuronal activities within the hypothalamus to reset the abnormally elevated drive for increased plasma glucose, triglyceride, and free fatty acid levels in fasting and postprandial states in insulin-resistant patients. [295]

Quick-release bromocriptine may be considered for obese patients who do not tolerate other diabetes medications or who need only a minimal reduction in HbA1c to reach their glycemic goal. This agent has the benefits of not causing hypoglycemia and weight gain. In addition, a randomized trial of bromocriptine in 3095 patients found that cardiovascular events were less frequent in the treatment arm than in the placebo arm. [296]

Adverse events most commonly reported in clinical trials of bromocriptine included nausea, fatigue, vomiting, headache, and dizziness. These events were more likely to occur during initial titration of the drug and lasted a median of 14 days. Nausea and vomiting were not described as serious.

Bromocriptine can cause orthostatic hypotension and syncope, particularly on initiation of therapy and dose escalation. Caution is advised when treating patients who are receiving antihypertensive therapy; orthostatic vital signs should be evaluated at baseline and periodically thereafter.

Comparison of oral antidiabetic agents

In 2007, the AHRQ compared the effectiveness and safety of oral diabetes medications for adults with type 2 diabetes, with a 2011 update. [297, 298] The AHRQ found little evidence to support predictions as to whether a particular medication is more likely to be effective in a given patient subgroup or to cause adverse effects in a particular patient.

The AHRQ concluded that although the long-term benefits and harms of diabetes medications remain unclear, the evidence supports the use of metformin as a first-line agent. On average, monotherapy with many of the oral diabetes drugs reduces HbA1c levels by 1 percentage point (although metformin has been found to be more efficacious than the DPP-4 inhibitors), and 2-drug combination therapies reduce HbA1c about 1 percentage point more than do monotherapies.

Other AHRQ findings included the following:

  • Metformin decreased LDL cholesterol levels relative to pioglitazone, sulfonylureas, and DPP-4 inhibitors

  • Unfavorable effects on weight were greater with TZDs and sulfonylureas than with metformin (mean difference of +2.6 kg)

  • Risk of mild or moderate hypoglycemia was 4-fold higher with sulfonylureas than with metformin alone; this risk was more than 5-fold higher with sulfonylureas plus metformin than with a TZD plus metformin

  • Risk of heart failure was higher with TZDs than with sulfonylureas

  • Risk of bone fractures was higher with TZDs than with metformin

Diarrhea was more common with metformin than with glitazones.


Management of Glycemia

In 2013, the American Association of Clinical Endocrinologists (AACE) issued a comprehensive new type 2 diabetes treatment algorithm--the first to incorporate obesity, prediabetes, and cardiovascular risk factor management. [299, 300]

Obesity management was incorporated into the algorithm because it is now clear that weight loss also reduces blood glucose. The authors suggest that obesity management can be considered first-line treatment for people with prediabetes. The prediabetes section of the algorithm considers cardiovascular risk factors and the options of antihyperglycemic or antiobesity therapy, though without making a recommendation regarding which form of treatment is better.

As in the AACE's earlier glycemic-control algorithm, the level of treatment depends on the initial hemoglobin A1c (HbA1c). (Lifestyle modification, including weight loss, is a component of all treatments.) Whereas the earlier algorithm recommended an HbA1c of 6.5% or lower as the goal for most patients, the current algorithm refines this advice, recommending an HbA1c of 6.5% or lower for healthy patients without concurrent illness and at low risk for hypoglycemia but individualized target HbA1c values greater than 6.5% for patients with concurrent illness and those who are at risk for hypoglycemia.


Metformin is the preferred initial agent for monotherapy and is a standard part of combination treatments. Advantages of metformin include the following:

  • Efficacy

  • Absence of weight gain or hypoglycemia

  • Generally low level of side effects

  • High level of patient acceptance

  • Relatively low cost

The dose of metformin is titrated over 1-2 months to at least 2000 mg daily, administered in divided doses (during or after meals to reduce gastrointestinal [GI] side effects). Exercise increases metformin levels and interferes with its glucose-lowering effect. [301]

Metformin may also decrease the risk of dementia associated with type 2 diabetes. In a 2013 observational study of 14,891 patients aged 55 years and older with type 2 diabetes, treatment with metformin signlificantly lowered the risk of developing dementia. [302] Only patients who initiated therapy with a single drug (metformin, sulfonylureas [SU], thiazolidinediones [TZDs], or insulin) during the study period were included.

During 5 years of followup, dementia was diagnosed in 1487 (9.9%) patients. [302] Compared with patients starting SU, those starting metformin had about a 20% reduced risk for dementia. Compared with patients starting TZD, those starting metformin had a 23% lower risk. [302]

Conversely, starting SU treatment (compared with metformin) was associated with a 24% increased risk for dementia; starting TZD treatment was associated with an 18% increased risk; and starting insulin treatment was associated with a 28% increased risk. [302]

Dual-drug therapy

If the patient fails to safely achieve or sustain glycemic goals within 2-3 months, another medication should be added. The choice should be guided by patient characteristics (eg, a DPP-4 inhibitor if both postprandial and fasting glucose levels are elevated; a GLP-1 agonist if postprandial glucose levels are strongly elevated; a TZD if the patient has metabolic syndrome and/or nonalcoholic fatty liver disease). [303]

Failure of initial therapy usually should result in addition of another class of drug rather than substitution. Reserve the use of substitution for cases in which patients experience intolerance to a drug because of adverse effects.

Considerable debate exists regarding which second agent to add to (or use initially in conjunction with) metformin. An outline of the therapeutic approach generally used by the author is presented in the first 2 images below. An idealized scheme for glucose and insulin patterns is presented in the third image below. The author finds that keeping such an idealized scheme in mind is helpful when treating and educating patients, even if the patient is trying to replicate it with less intensive insulin therapy.

Treatment of type 2 diabetes mellitus. Treatment of type 2 diabetes mellitus.
Simplified scheme for using insulin in treating pa Simplified scheme for using insulin in treating patients with type 2 diabetes mellitus.
Simplified scheme of idealized blood glucose value Simplified scheme of idealized blood glucose values and multiple dose insulin therapy in type 2 diabetes mellitus.

Because TZDs not infrequently cause weight gain and edema, the author usually reserves these agents for patients who cannot use metformin, as a result of intolerance or contraindications. Exceptions to this practice may include patients of relatively normal weight who have marked insulin resistance, such as patients of Asian heritage.

Before adding a second agent for a patient who is taking an insulin secretagogue, the clinician should warn the patient about the possibility that the second agent will induce hypoglycemia. If hypoglycemia occurs, the dose of the insulin secretagogue, not the newly added agent, should be reduced.

Triple-drug therapy

If 2 drugs prove unsuccessful after 2-3 months, the next step is triple therapy. The third drug may be an oral agent from a third class of antidiabetic drugs, basal insulin (typically at bedtime), or the injectable drug exenatide. The expense and adverse effect profile of TZDs make their use in an oral triple therapy approach less desirable.

The addition of exenatide to 1 or 2 oral agents (eg, metformin and/or a sulfonylurea) is attractive because of its simplicity (ie, only 2 possible doses of exenatide, with easy titration compared with insulin); although expensive, it avoids hypoglycemia. If basal insulin is used, the insulin dose is titrated to the fasting glucose concentration, which the patient can measure at home.

Glucose values

Some patients need reduction of their oral antidiabetic agent to prevent daytime hypoglycemia as the bedtime insulin is initiated or increased and the fasting glucose concentration decreases. If a GLP-1 agonist is used, the author monitors fasting and postprandial sugars, expecting a marked flattening of the postprandial rise in glucose concentrations.

Measurement of glucose patterns in patients with type 2 diabetes, particularly those who have central obesity and hepatic steatosis, often reveals that the highest preprandial glucose level of the day is before breakfast (because of disordered hepatic glucose production overnight), with a "stair-step" decrease during the day (after the usual postmeal rise). These higher-than-desired morning glucose values do not necessarily dictate abandonment of the current therapeutic regimen, provided that the HbA1c level is at target.

For patients trying to achieve near euglycemia, premeal glucose values of 80-120 mg/dL are the goal, with the patient going to sleep at night with a value at least 100 mg/dL. In patients with less stringent glycemic goals (eg, because of advanced age, advanced complications, or severe concomitant disease), preprandial glucose values of 100-140 mg/dL are desired. Because of the limitations of therapies, essentially no patient is able to achieve these goals all the time if, in fact, insulin is needed to treat their disease.

For patients who primarily have fasting hyperglycemia, basal insulin is the easiest way to correct this abnormality. Basal insulin is typically scheduled at bedtime but can be given at suppertime if that is more convenient for the patient.

The goal of a combined daytime oral agent plus once-a-day insulin is to lower the fasting glucose level to 100 mg/dL by titrating the insulin. When this target is achieved, the oral agents can be effective in maintaining preprandial and postprandial blood glucose levels throughout the day. If a regimen combining oral agents and insulin fails to lower glucose levels into the normal range, patients should be switched to a daily multiple-injection schedule with a premeal rapid-acting insulin and a longer-acting basal insulin.

Insulin regimens

A necessary condition for twice-daily insulin to succeed is a regimented lifestyle, with mealtimes regularly spaced and insulin injections taken at essentially the same time every day, including weekends and holidays. Lack of regularity in the schedule is self-defeating for this approach to therapy.

The author limits the use of premixed insulin to patients who may have trouble mixing their insulins. The author prefers premixes containing regular insulin if the premix is administered to maintain better midday coverage. Premixes with rapid-acting medications can be used if the midday meal is small. A systematic review found that glycemic control with premixed insulin analogues (ie, mixtures of rapid-acting and intermediate-acting insulin analogues) is similar to that with premixed human insulin. [304]

Multiple daily dosing

Conventional multiple daily dosing of insulin gives the patient the greatest flexibility. In this approach, long-acting insulin (eg, glargine, detemir) is generally given once daily as the basal insulin, and rapid-acting insulin (eg, aspart, glulisine, lispro) is administered just before each meal.

The basal component can be administered at any time of day as long as it is given at the same time each day. However, interpreting glucose patterns is probably easiest if the basal insulin is administered at or near bedtime. The basal insulin can then be titrated to the morning sugar, and the bolus premeal insulin can be titrated to the next premeal sugar and, in some cases, a postprandial (2 h) value.

All insulin injections should preferably be administered in the abdomen, although they can also be given in the thigh, hip, or buttock regions. Adiposity blunts the pharmacodynamics of the basal insulins NPH, glargine, and, especially, detemir. [305]

Insulin dosing can be safely reduced in patients with renal insufficiency without compromising glycemic control. [306] Dosing based solely on weight is not advisable in these patients, who have reduced lean body mass and water retention.

Continuous subcutaneous insulin infusion

The American Association of Clinical Endocrinologists and American College of Endocrinology released a consensus statement on insulin pump management: [307]

  • Based on currently available data, continuous subcutaneous insulin infusion (CSII) is justified for basal-bolus insulin therapy in patients with type 1 diabetes mellitus.

  • Only providers whose practice can assume full responsibility for a comprehensive pump management program should offer this technology.

  • The ideal CSII candidate is a patient with type 1 diabetes mellitus or intensively management insulin-dependent type 2 diabetes mellitus who is currently performing 4 or more insulin injections and 4 or more self-monitored blood glucose measurements daily; is motivated to achieve optima blood glucose control; is willing and able to carry out the tasks that are required to use this complex and time-consuming therapy safely and effectively; and is willing to maintain frequent contact with their health care team.

  • Adult patients

    • At CSII initiation, the patient should have daily contact with the pump trainer. a return visit with the endocrinologist/diabetologist/advanced practice nurse is advised within 3-7 days after CSII initiation.

    • Educational consults should be scheduled weekly or biweekly at first, then periodically as needed.

    • Specialist follow-up visits should be scheduled at least monthly until the pump regimen is stabilized, then at least once every 3 mo.

  • Pediatric patients

    • CSII is indicated for pediatric patients with elevated hemoglobin A1C (HbA1C) levels on injection therapy; frequent, severe hypoglycemia; widely fluctuating glucose levels; a treatment regimen that compromises lifestyle; and microvascular complications and/or risk factors for macrovascular complications.

    • Ideal pediatric candidates are those with motivated families who are committed to monitoring blood glucose 4 or more times per day and have a working understanding of basic diabetes management.

    • Patient age and duration of diabetes should not be factors in determining the transition from injections to CSII.

Intensified basal-bolus regimen

An intensified basal-bolus regimen of insulin glargine and insulin glulisine provides better glycemic control than does a standard, premixed insulin regimen, in patients with long-standing, insulin-treated type 2 diabetes mellitus, according to a study by Fritsche et al. In this open-label, randomized, multinational trial, an intensified insulin regimen combining insulin glargine (once daily) with premeal insulin glulisine (basal-bolus group; n=153) was compared with twice-daily conventional therapy with premixed insulin (n=157).

The mean decrease from baseline HbA1c was -1.31% for the basal-bolus group, versus -0.80% for the premix patients, with more patients in the basal-bolus group attaining HbA1c of 7% or less. Moreover, significantly lower blood glucose levels were observed in the basal-bolus group than in the premix group. [308]

Postprandial glycemic control

Glycemic control is a function not only of fasting and preprandial glucose values but also of postprandial glycemic excursions. Emphasis on postprandial glucose measurements has been fueled to some degree by the availability of short-acting insulin secretagogues, very-short-acting insulin, and alpha-glucosidase inhibitors, all of which target postprandial glycemia.

While postprandial glucose levels are a better predictor of macrovascular disease risk early in the course of loss of glucose tolerance, it remains to be seen whether targeting after-meal glucose excursions has a greater effect on the risk of complications than do more conventional strategies. A study by Siegelaar et al seriously questions the notion that targeting postprandial glucose variability favorably affects cardiovascular outcomes in patients after myocardial infarction. [309] Clearly, more studies are needed.

Intuitively, one would assume that therapies that normalize preprandial and postprandial glycemia (or that come close to normalizing them) would be optimal. Whether such a strategy can be achieved without untoward adverse effects and with further reductions in microvascular and macrovascular disease risk (compared with regimens used in the UKPDS) using newly available therapies is open to question. Practically speaking, most patients are fully occupied trying to handle conventional glucose monitoring and insulin dose adjustment.

Eating a high-protein prebreakfast snack, such as one with soy yogurt, is a simple way to achieve better postbreakfast glycemic control, according to a study by Chen et al; this study confirms a phenomenon observed in healthy humans nearly a century ago (Staub, 1921). [310]

Glycemic monitoring

Decisions about glycemic management are generally made on the basis of HbA1c measurements and the results of self-monitoring of blood glucose (SMBG). HbA1c is measured at least twice yearly in patients with stable glycemic control who are meeting treatment goals and quarterly in patients whose therapy has changed or who are not meeting treatment goals. [2]

If a total glycated hemoglobin (GHb) measurement is used, the number is 1-2% higher. However, the laboratory should provide a correlation of GHb values with HbA1c values. [3, 96, 170]

Glycemic targets

A guideline from the American College of Physicians (ACP) recommends that an HbA1c target of less than 7% is appropriate for many patients. [311] Some organizations (eg, the American Association of Clinical Endocrinologists, [159] the International Diabetes Federation) recommend a glycemic target of less than 6.5% for HbA1c, although this is a general target that always has to be individualized according to patient characteristics and health conditions

The ACP advises, however, that an HbA1c of 7% may not be an appropriate target for all patients. [311] Goals should be tailored to the individual patient and should take the following considerations into account:

  • The patient's preferences

  • Risk for complications from diabetes

  • Comorbidity

  • Life expectancy

In a meta-analysis of 13 studies, intensive glucose lowering had no significant effect on all-cause mortality or cardiovascular deaths. A reduction in nonfatal myocardial infarction and microalbuminuria was noted. However, patients experienced a 2-fold increased risk of hypoglycemia. [312]

Risks of and considerations in intensive treatment

Risk for hypoglycemia is almost always the limiting factor in achieving the lowest possible HbA1c that does not cause undue harm. Unfortunately, some practitioners and their patients pursue a particular HbA1c value despite uncertain benefit or unacceptable risk, with significant risk for side effects.

Factors that can produce an unfavorable risk-benefit ratio for intensive blood glucose lowering include advanced age, other major systemic disease, and advanced microvascular and neuropathic complications. For example, in an elderly patient, risk considerations may include the possibility of falling and breaking a hip during a hypoglycemic episode. (Such considerations are addressed in Type 2 diabetes mellitus in older people: a brief statement of key principles of modern day management including the assessment of frailty. A national collaborative stakeholder initiative, from Diabetes UK.)

In elderly patients who have a life expectancy of less than 5 years or in any patient with a terminal disease, tight control may be unnecessary. Patients with cardiovascular or cerebrovascular disease may also need higher preprandial blood glucose targets (eg, 100-150 mg/dL) to prevent severe hypoglycemia.

For patients older than 65 years, a recent consensus statement from the American Diabetes Association and the American Geriatrics Society recommends adjusting treatment goals for glycemia, blood pressure, and dyslipidemia according to life expectancy and the presence of comorbidities. The statement suggests 3 broad groupings [313, 314] :

  • Healthy: Patients with few coexisting chronic conditions and intact cognitive and functional status

  • Complex/intermediate: Patients with multiple coexisting chronic illnesses or 2 or more impairments in activities of daily living (ADL) or mild to moderate cognitive impairment

  • Very complex/poor health: Patients in long-term care or with end-stage chronic illnesses or moderate to severe cognitive impairment or with 2 or more ADL dependencies

Corresponding HbA1c targets might be less than 7.5%, less than 8%, and less than 8.5%, respectively, for the 3 groups above.

Additionally, patients with alcoholism or other serious substance abuse problems and patients with severe, uncontrolled mental illness may be unable to effectively participate in the care of their diabetes. Consequently, they are at high risk for severe hypoglycemic reactions if near-normal glucose levels are targeted.

Finally, patients with hypoglycemia unawareness (ie, lack adrenergic warning signs of hypoglycemia) or those with recurrent episodes of severe hypoglycemia (ie, hypoglycemia requiring treatment by another person) should also have high target levels, at least temporarily. Fortunately, patients with type 2 diabetes mellitus (unlike those with long-standing type 1 disease) usually maintain adequate hypoglycemia awareness. This greatly facilitates hypoglycemic therapy (ie, insulin secretagogues, insulin) in patients with type 2 diabetes.

Self-monitoring of blood glucose

Daily SMBG is important for patients treated with insulin or insulin secretagogues to monitor for and prevent hypoglycemia, as well as to optimize the treatment regimen. The optimal frequency of SMBG for patients with type 2 diabetes is unresolved, but it should be sufficient to facilitate reaching glucose goals.

The author often utilizes no or minimal SMBG in patients using lifestyle changes alone or agents that do not cause hypoglycemia (eg, metformin, TZDs, glucosidase inhibitors). Patients using multiple insulin injections should use SMBG at least 3 times a day. [2]

A task force from the Endocrine Society evaluated the following potential uses for continuous glucose monitoring:

  • Real-time, continuous glucose monitoring in adults in hospital settings

  • Real-time outpatient monitoring in children and adolescents

  • Real-time outpatient monitoring in adults

The Task Force developed recommendations regarding benefits in maintaining target levels of glycemia and limiting the risk of hypoglycemia. [315]

Although the use of continuous glucose monitors (CGMs) has primarily been aimed at persons with type 1 diabetes, the devices have come to increasingly be employed in type 2 diabetes. [316] For example, the CGM device sugarBEAT (Nemaura Medical; Loughborough, UK) was developed specifically for individuals with type 2 diabetes, being produced for such patients who are not at high risk for hypoglycemia; it can also be utilized in persons with prediabetes. The sugarBEAT CGM uses an adhesive patch and sensor, drawing glucose molecules from the interstitial fluid just beneath the skin’s surface for measurement. A Bluetooth connection is employed to transmit data every 5 minutes to a smartphone app. The device is worn for 14 hours at a time during the day and for just 2-4 days monthly (in contrast to CGMs for patients with type 2 diabetes, which are utilized every day). Although sugarBEAT is noninvasive, a once-daily fingerstick is still needed for calibration. In May 2019, it received a CE (Conformité Européenne) Mark in Europe for use as a class IIb medical device. Although still awaiting FDA approval in the United States, permission has in the meantime been granted for this CGM to be marketed as a “wellness” device, with sugarBEAT in this capacity producing retroactive reports for the physician and patient rather than real-time values. [317]


Dietary Modifications

For most patients, the best diet is one consisting of the foods that they are currently eating. Attempts to calibrate a precise macronutrient composition of the diet to control diabetes, while time-honored, are generally not supported by the research. Caloric restriction is of first importance. After that, individual preference is reasonable.

Modest restriction of saturated fats and simple sugars is also reasonable. However, some patients have remarkable short-term success with high-fat, low-carbohydrate diets of various sorts. Therefore, the author always stresses weight management in general and is flexible regarding the precise diet that the patient consumes. Also, the practitioner should advocate a diet composed of foods that are within the financial reach and cultural milieu of the patient. For example, patients who participate in Ramadan may be at higher risk of acute diabetic complications. Although these patients do not eat during the annual observance, they should be encouraged to actively monitor their glucose, alter the dosage and timing of their medication, and seek dietary counseling and patient education to counteract any complications. [318]

Weight loss

Modest weight losses of 5-10% have been associated with significant improvements in cardiovascular disease risk factors (ie, decreased HbA1c levels, reduced blood pressure, increase in HDL cholesterol, decreased plasma triglycerides) in patients with type 2 diabetes mellitus. Risk factor reduction was even greater with losses of 10-15% of body weight. [319, 320]

A study by Lazo et al attested to the benefits of lifestyle intervention, which aimed at a minimum weight loss of 7%, on hepatic steatosis in patients with type 2 diabetes. [321] Since there is no known treatment for nonalcoholic fatty liver disease, a weight loss strategy may help to prevent progression to serious liver damage.

One-year results from the open-label, randomized Diabetes Remission Clinical Trial (DiRECT) demonstrated a type 2 diabetes remission rate of 46% in participants who underwent intervention with a very low–calorie liquid diet. The regimen included gradual reintroduction of food after 3-5 months and ongoing weight-loss maintenance support. The remission rate reached 73% in just those patients in the intervention group who at 12 months had maintained a weight loss of at least 10 kg. [322, 323]

Mediterranean-style diet

Esposito et al reported greater benefit from a low-carbohydrate, Mediterranean-style diet than from a low-fat diet in patients with newly diagnosed type 2 diabetes mellitus. [324] In a single-center, randomized trial, 215 overweight patients with newly diagnosed type 2 diabetes mellitus who had never been treated with antihyperglycemic drugs and whose HbA1c levels were less than 11% were assigned to either a Mediterranean-style diet (< 50% of daily calories from carbohydrates) or a low-fat diet (< 30% of daily calories from fat).

After 4 years, participants assigned to the Mediterranean-style diet had lost more weight and had demonstrated more improvement in some measures of glycemic control and coronary risk than had participants consuming the low-fat diet; 44% of patients in the Mediterranean-style diet group required antihyperglycemic drug therapy, compared with 70% of those in the low-fat diet group.

High-protein versus high-carbohydrate diet

A study by Larsen et al concluded that the long-term therapeutic effect of a high-protein diet is not superior to that of a high-carbohydrate diet in the treatment of type 2 diabetes mellitus. In this 12-month trial, 99 overweight or obese diabetic patients followed a low-fat diet (30% total energy) that was either high in protein (30% total energy) or high in carbohydrate (55% total energy); both groups benefited equally. [325]

It should also be noted that already-attenuated glucose disposal is not worsened by postprandial circulating amino acid concentration. Therefore, recommendations to restrict dietary proteins in patients with type 2 diabetes seem unwarranted. [326]


In the Cardiovascular Health Study, phospholipid trans -palmitoleate levels were found to be associated with lower metabolic risk. [327] Trans -palmitoleate is principally derived from naturally occurring dairy and other ruminant trans -fats. Circulating trans -palmitoleate is associated with lower insulin resistance, incidence of diabetes, and atherogenic dyslipidemia. Potential health benefits, therefore, need to be explored.

Advanced glycation end products

Food-derived, pro-oxidant, advanced glycation end products may contribute to insulin resistance in clinical type 2 diabetes mellitus and may suppress protective mechanisms. Advanced glycation end-product restriction may preserve native defenses and insulin sensitivity by maintaining a lower basal oxidative state. [328]

Other considerations

Oral ginseng (or ginsenoside) does not improve pancreatic beta-cell function. Routine use is not recommended. [329]

Pasta enriched with biologically active isoflavone aglycons improves endothelial function in patients with type 2 diabetes mellitus and favorably affects cardiovascular disease risk markers. [330]

In patents with type 2 diabetes mellitus, impaired fasting glucose or impaired glucose tolerance at high risk for cardiovascular disease, addition of n-3 fatty acids does not reduce risk of cardiovascular events, including death from cardiovascular causes. [331]


Activity Modifications

Most patients with type 2 diabetes mellitus can benefit from increased activity. Aerobic exercise improves insulin sensitivity and may improve glycemia markedly in some patients.

Structured exercise training of more than 150 minutes per week is associated with greater HbA1c reduction; however, physical activity helps lower HbA1c only when combined with dietary modifications. [332]

The patient should choose an activity that she or he is likely to continue. Walking is accessible to most patients in terms of time and financial expenditure.

A previously sedentary patient should start activities slowly. Older patients, patients with long-standing disease, patients with multiple risk factors, and patients with previous evidence of atherosclerotic disease should have a cardiovascular evaluation, probably including an imaging study, prior to beginning a significant exercise regimen.

Balducci et al showed that a supervised, facility-based exercise training program, when added to standard treatments for type 2 diabetes mellitus, yields better results than does simply counseling patients to exercise. [333]

A randomized, controlled trial by Church et al emphasized the need to incorporate both aerobic and resistance training to achieve better lowering of HbA1c levels. [334] Aerobic exercise alone or in combination with resistance training improves glycemic control, circulating triglycerides, systolic blood pressure, and waist circumference. [335] The impact of resistance exercise alone, however, remains unclear.

Loimaala et al found that long-term endurance and strength training resulted in improved metabolic control of diabetes mellitus and significant cardiovascular risk reduction, compared with standard treatment. However, exercise training did not improve conduit arterial elasticity. [336]

In a 3-month trial, Hegde et al found that yoga can be effective in reducing oxidative metabolic stress in patients with type 2 diabetes mellitus. However, yoga did not impact waist-to-hip ratio, blood pressure, vitamin E, or superoxide dismutase. [337]


Bariatric Surgery

In morbidly obese patients, bariatric surgery has been shown to improve diabetes control and, in some situations, normalize glucose tolerance. It is certainly a reasonable alternative in carefully selected patients if an experienced team (providing appropriate preoperative evaluation, as well as technical surgical expertise) is available.

In 2011, the International Diabetes Federation Taskforce on Epidemiology and Prevention of Diabetes released a position statement on bariatric surgery. The task force recommended bariatric surgery as an appropriate treatment for people with type 2 diabetes mellitus and obesity who have been unable to achieve recommended treatment targets using medical therapies, particularly if other major comorbidities exist. [338, 339]

According to guidelines released in 2016 by the 2nd Diabetes Surgery Summit (DSS-II), an international consensus conference, bariatric surgery should be considered even for type 2 diabetes patients with mild, class 1 obesity (BMI 30.0-34.9 kg/m2) if their hyperglycemia is inadequately controlled with optimal treatment. In addition, the guidelines state that bariatric surgery should be a “recommended option” for type 2 diabetes patients with class 3 obesity (BMI 40 kg/m2 or above) no matter what level of glycemic control has been achieved. The guidelines also say that of the different forms of bariatric surgery, Roux-en-Y gastric bypass seems to have the best risk/benefit profile for the majority of patients with type 2 diabetes. [340, 341]

Kashyap and colleagues demonstrated that bariatric surgery improved glycemic control in patients with type 2 diabetes. [342] The study compared the metabolic effects of 2 types of bariatric surgery (Roux-en-Y gastric bypass or sleeve gastrectomy) combined with intensive medical therapy with intensive medical therapy alone in 60 patients with uncontrolled type 2 diabetes and moderate obesity. At 24-month follow-up, glycemic control improved in all 3 groups. Body fat reduction was similar in the 2 surgery groups, with patients in the gastric bypass group showing a greater absolute reduction in truncal fat. [342] Insulin sensitivity increased significantly only in the gastric bypass group, and pancreatic β-cell function increased significantly more in these patients compared with those in the other 2 groups. [342]

A retrospective study of 252 patients with type 2 diabetes mellitus who underwent bariatric surgery found that 44% of the patients were able to do without insulin treatment by median 7-year follow-up. The patients had a mean preoperative BMI of 46 kg/m2, with those who underwent the Roux-en-Y procedure showing a 1.4 percentage-point reduction in HbA1c at 7 years, and those who underwent sleeve gastrectomy showing a 1.6 percentage-point reduction. [343]


Laboratory Monitoring

Because diabetes mellitus is a multisystem disease, focusing solely on blood glucose is inadequate. The image below lists appropriate laboratory parameters in the global assessment of patients with type 2 diabetes mellitus. Obviously, patients with abnormalities need more frequent monitoring to guide therapeutic interventions. Drug-specific monitoring is also necessary (eg, serum creatinine and vitamin B12 in patients taking metformin, serum transaminases for patients taking a TZD).

Laboratory monitoring guidelines for patients with Laboratory monitoring guidelines for patients with type 2 diabetes mellitus.

Monitoring for Diabetic Complications

The ADA recommends initiation of complications monitoring at the time of diagnosis of diabetes mellitus. [2] This regimen should include yearly dilated eye examinations, annual microalbumin checks, and foot examinations at each visit.

A study by Cigolle et al found that middle-aged and older adults with diabetes have an increased risk for the development of geriatric conditions (eg, cognitive, vision, and hearing impairments; falls). [344] These conditions substantially contribute to morbidity and functional impairment. The authors concluded that adults with diabetes should be monitored for the development of geriatric conditions at a younger age than was previously considered.

The risk for early development of Parkinson disease is 36% higher in patients with diabetes mellitus. [345] However, a systematic review from Cereda et al found no conclusive evidence of this association. [346]

A high overall risk for pancreatic neoplasm is noted in individuals with diabetes mellitus, particularly in those aged 45-65 years. [347]

The incidence of complications widely vary among the Asian subgroups, suggesting the need for an ethnic stratified nuanced approach in evaluation and surveillance. [348] One size does not fit all.


Management of Hypertension

Blood pressure goals

The role of hypertension in increasing microvascular and macrovascular risk in patients with diabetes mellitus has been confirmed in the UKPDS and Hypertension Optimal Treatment (HOT) trials. [349, 350]

The ADA’s Standards of Care in Diabetes—2023 advised that the target blood pressure for people with diabetes be below 130/80 mm Hg, if it can be safely achieved. (This is below the 140 mm Hg systolic target that the ADA had advised in 2012.) [351, 352, 353]

In patients with greater than 1 g/day proteinuria and renal insufficiency, a more aggressive therapeutic goal (ie, 125/75 mm Hg) has been advocated. According to the Veterans Affairs Diabetes Trial, however, a diastolic blood pressure of less than 70 mm Hg increases the risk of cardiovascular disease in patients with diabetes. [354]

The ADA recommendations endorse blood pressure measurement “at every routine clinical care visit” and state that patients whose blood pressure is found to be elevated (at or above 130/80 mm Hg) should undergo multiple readings to confirm hypertension, including blood pressure assessment on a separate day. [355, 356, 351, 352, 353]

Cardiovascular risk and hypertension medications

In a safety review, the FDA found no clear evidence of increased cardiovascular risk with the hypertension drug olmesartan in diabetic patients. [357] This review addressed data from the Randomized Olmesartan and Diabetes Microalbuminuria Prevention (ROADMAP) study, which found an increased risk of cardiovascular mortality in patients taking olmesartan, as well as data from an epidemiologic study of Medicare patients, which suggested a similarly increased risk.

Because of discrepant survival results in diabetics and nondiabetics in the Medicare study, the FDA concluded that the evidence for the increased cardiovascular risk was not conclusive and did not support recommending that olmesartan not be used in patients with diabetes. [357, 358]

Pharmacologic therapy

While angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARB), diuretics, beta blockers, and calcium channel blockers are all considered acceptable initial therapy, the author prefers inhibitors of the renin-angiotensin system (ie, ACE inhibitors, ARBs) because of their proven renal protection effects in patients with diabetes. Many patients require multiple agents. Diuretics or calcium channel blockers frequently are useful as second and third agents.

The ALTITUIDE Trial investigating the impact of direct renin inhibitor with aliskiren on cardio-renal outcomes in patients with diabetes mellitus when used as an adjunct to angiotensin converting enzyme inhibitor or angiotensin receptor blocker therapy failed to show any benefit. On the contrary, it might be harmful due to an increased risk for hyperkalemia and hypotension despite marked reduction in proteinuria. [359]

A study by Hermado et al showed that treatment with antihypertensive medications taken at bedtime provides better ambulatory blood pressure control, as well as significant reduction in cardiovascular morbidity and mortality when compared with taking medications upon waking. [360] The study was conducted in patients taking 1 or more antihypertensive drugs and had a median follow-up of 5.4 years.


Management of Dyslipidemia


Dyslipidemia is common in patients with type 2 diabetes mellitus and often takes the form of high triglyceride and low HDL cholesterol levels. Trials have shown that the use of statins is effective for primary and secondary prevention of coronary heart disease (CHD) events in patients with diabetes.

In the ADA’s Standards of Care in Diabetes—2023, the target for LDL cholesterol was set at under 70 mg/dL for persons with diabetes aged 40-75 years who have a higher cardiovascular disease risk, such as individuals with at least one atherosclerotic cardiovascular disease risk factor. [352, 353]

In persons with diabetes who have established atherosclerotic cardiovascular disease, an LDL cholesterol level of under 55 mg/dL was advised. [352, 353]

See Management of Coronary Heart Disease for guidelines on statin use in persons with diabetes.


Fibrates may reduce CHD events in patients with isolated low HDL cholesterol. Whether therapy aimed more at triglyceride reduction and HDL cholesterol elevation (ie, fibrates, niacin) is effective in CHD-event reduction in primary prevention remains to be determined. [361]

Beta blockers

Vasoconstricting beta blockers are known to reduce HDL cholesterol levels and increase triglyceride, LDL cholesterol, and total cholesterol levels. The vasodilating beta blocker carvedilol (mixed alpha1, beta1, and beta2 blocker) has not been associated with the aforementioned effects.

In a randomized, double-blind trial in patients with type 2 diabetes mellitus receiving renin-angiotensin blockers, the addition of carvedilol for blood pressure control resulted in a significant decrease in triglyceride, total cholesterol, and non ̶ HDL cholesterol levels. Patients given metoprolol (a vasoconstricting beta blocker) were significantly more likely to be started on statin therapy or, if already on statin therapy, to require an increase in the dose, than were patients taking carvedilol. [362]


Management of Coronary Heart Disease

There is contradictory epidemiologic evidence as to whether diabetes is in fact a CHD risk equivalent. For the present, however, that is the position adopted by most groups, such as the National Cholesterol Education Program (NCEP) and the ADA. [361]

Although the risk for CHD is 2-4 times greater in patients with diabetes than it is in individuals without diabetes, control of conventional risk factors is probably more important in event reduction than is glycemic control. Control of hypertension, aspirin therapy, and lowering of LDL cholesterol levels are vitally important in reducing CHD risk.


The ADA recommends that patients with diabetes who are at high risk for cardiovascular events receive primary preventive therapy with low-dose, enteric-coated aspirin. For patients with aspirin hypersensitivity or intolerance, clopidogrel is recommended. [363]

However, a randomized, controlled trial from Japan found that using low-dose aspirin as primary prevention did not reduce the risk of cardiovascular events in patients with type 2 diabetes. [364] These investigators subsequently reported that low-dose aspirin therapy reduces cardiovascular risk only in patients with a glomerular filtration rate (GFR) of 60-89 mL/min; low-dose aspirin had no beneficial impact if the GFR was above 90 mL/min or below 60 mL/min. [365]

A study by Okada et al reported that low-dose aspirin therapy (81-100 mg) in patients with diabetes who are taking insulin or oral hypoglycemic agents does not reduce atherosclerotic events. [366] This is yet another argument against using low-dose aspirin for primary prevention of cardiovascular disease in patients with moderate or severe diabetes.


The Scandinavian Simvastatin Survival Study (4S) showed a 42% reduction in CHD events in diabetic patients with simvastatin therapy (mean dose 27 mg daily, with LDL reduction approximately 35%). Participants in 4S had known CHD and very high LDL cholesterol levels. [367]

A smaller reduction was seen in the Heart Protection Study (HPS) in patients with CHD or other vascular disease and diabetes. [368] Patients in the HPS treatment arm received simvastatin 40 mg daily. Lesser degrees of risk reduction have been shown in other secondary prevention studies in patients treated with pravastatin with mild to moderate LDL cholesterol elevation at baseline.

Atorvastatin, 10 mg daily, did not reduce CHD risk among diabetic patients with hypertension and no previous CHD who were enrolled in the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT). [369] In contrast, the Collaborative Atorvastatin Diabetes Study (CARDS) showed a significant reduction in CHD risk in patients with type 2 diabetes mellitus and 1 other risk factor when treated with atorvastatin 10 mg daily. [370]

Some studies have suggested that statin therapy may be associated with an increased risk of developing diabetes. In a pooled analysis of data from five statin trials, intensive-dose statin therapy was associated with increased risk of new-onset diabetes compared with moderate dose statins. [371]

A study by Ahmadizar et al of subjects over age 45 years who had no diabetes at baseline reported that compared with individuals who have never used statins, the risk of incident type 2 diabetes development in persons who have ever taken statins is 38% greater, with the likelihood being particularly high in persons with impaired glucose homeostasis and in individuals who are overweight/obese. However, analyses stratified at baseline for gender and body mass index (BMI) indicated that statin use was not significantly associated with type 2 diabetes in women or in persons with a normal body mass index (BMI). [372, 373]

The American Diabetes Association (ADA) provided recommendations on the use of statins in patients with diabetes to align with those of the American College of Cardiology and the American Heart Association. [374]

  • The ADA recommends statin use for nearly everyone with diabetes.

  • The ADA guidelines divide diabetes patients by 3 age groups:

    • Younger than 40 years: No statins for those with no cardiovascular disease (CVD) risk factors other than diabetes; moderate intensity or high-intensity statin doses for those with additional CVD risk factors (baseline LDL cholesterol 100 or greater, high blood pressure, smoking, and overweight/obesity); and high-intensity statin doses for those with overt CVD (including previous cardiovascular events or acute coronary syndrome).

    • Age 40-75 years: Moderate-intensity statins for those with no additional risk factors, and high-intensity statins for those with either CVD risk factors or overt CVD.

    • Older than 75 years: Moderate-intensity statins for those with CVD risk factors; and high-intensity statins for those with overt CVD.

  • Lipid monitoring for adherence is recommended as needed, and annual monitoring is advised for patients younger than 40 years who have not yet started on statins.

  • There is a new BMI cut point of 23 kg/m2 (instead of 25 kg/m2) for screening Asian Americans for prediabetes and diabetes, based on evidence that Asian populations are at increased risk at lower BMIs relative to the general population.

  • The premeal glucose target of 70-130 mg/dL was changed to 80-130 mg/dL to better reflect new data that compared average glucose levels with HbA1c targets.

  • The goal for diastolic blood pressure was raised to 90 mm Hg from 80 mm Hg to better reflect data from randomized clinical trials. (This followed ADA's 2013 shift from a systolic target of 130 mm Hg to 140 mm Hg.) In the ADA's Standards of Care in Diabetes—2023, however, these numbers were again reduced, with the the target blood pressure for people with diabetes advised to be below 130/80 mmHg, if it can be safely achieved. [352, 353]

  • With regard to physical activity, the document now advises limiting the time spent sitting to no longer than 90 min.

  • The ADA does not support e-cigarettes as alternatives to smoking or to facilitate smoking cessation.

  • Immunization against pneumococcal disease is recommended.

  • A new HbA1c target of less than 7.5% for children is now recommended.

HDL cholesterol therapy

The benefits of raising HDL cholesterol levels in patients with type 2 diabetes remains uncertain. Some of the statin trials suggest that statin therapy eliminates some of the excess risk from low HDL cholesterol levels in patients with LDL cholesterol elevation at baseline.

The Veterans Administration HDL Intervention Trial (VA-HIT) showed an approximately 22% reduction in CHD events in patients with diabetes and known CHD when HDL cholesterol levels were increased by approximately 6% by gemfibrozil. [375] This was a population with low LDL cholesterol levels, however, so whether these same benefits would accrue in patients with elevated LDL cholesterol who are treated with a statin before their low HDL cholesterol is addressed is unclear.

Triglyceride therapy

An elevated triglyceride level is a common abnormality in type 2 diabetes mellitus. However, whether therapy to reduce triglycerides helps to reduce CHD events has not been determined from clinical end-point trials.


The Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) study, which was conducted in 2368 patients with type 2 diabetes mellitus and heart disease, showed no significant difference in the rates of death and major cardiovascular events between patients undergoing prompt revascularization and those undergoing medical therapy with insulin or insulin-sensitizing drugs. [376] These data emphasize the need to customize therapy to the patient’s circumstances and therapeutic goals.


Management of Ophthalmologic Complications

Patients with established retinopathy should see an ophthalmologist at least once every 6-12 months, as necessary. Three-year retinal screening may be feasible for patients with mild diabetes and no retinopathy. [377]

Early background retinopathy may reverse with improved glycemic control. More advanced retinopathy does not regress with improved glycemia and may worsen, although rarely, with short-term marked improvements in glycemia. Hypertension control is of paramount importance in these latter patients. Results of the randomized, placebo-controlled DIRECT-Protect 2 trial suggested that treatment with the ARB candesartan may improve mild to moderate retinopathy in patients with type 2 diabetes. [378]

Macular edema has been reported in a proportion of patients who experience fluid retention as a side effect of TZDs. [379] Resolution typically follows cessation of the TZD, although diuretics have been prescribed in such cases.

Laser photocoagulation has markedly improved the ability of ophthalmologists to preserve sight in patients with diabetes and proliferative retinopathy or macular edema. Laser therapy is effective in decreasing macular edema and preserving vision but is less effective in restoring lost vision.

Diabetes can affect the lens, vitreous, and retina, causing visual symptoms that may prompt the patient to seek emergency care. Visual blurring may develop acutely as the lens changes shape with marked changes in blood glucose concentrations. This effect, which is caused by osmotic fluxes of water into and out of the lens, usually occurs as hyperglycemia increases, but it also may be seen when high glucose levels are lowered rapidly. In either case, recovery to baseline visual acuity can take up to a month, and some patients are almost completely unable to read small print or do close work during this period.

Patients with diabetes also tend to develop senile cataracts sooner than persons without diabetes. Development of senile cataracts is not related to the degree of glycemic control, however.


Management of Diabetic Neuropathy

Peripheral neuropathy is the most common complication observed in patients with type 2 diabetes in outpatient clinics. Patients may have paresthesias, numbness, or pain. The feet are involved more often than the hands.

Improved glycemic control early may alleviate some of the symptoms, although sometimes symptoms actually worsen with lowering of blood glucose levels. Later symptomatic therapy largely is empirical and may include the following:

  • Low-dose tricyclic antidepressants

  • Duloxetine

  • Anticonvulsants (eg, phenytoin, gabapentin, carbamazepine)

  • Topical capsaicin

  • Various pain medications, including nonsteroidal anti-inflammatory drugs (NSAIDs)

Protection of the feet by applying lubricating agents (but not between the toes) and wearing appropriate footwear (shoes and socks or stockings) is important. Daily inspection of the feet after bathing is mandatory. In patients with advanced neuropathy, water temperature must be checked by a companion or with a thermometer. Soaking the feet generally is not recommended and may be harmful.

Gastroparesis is usually less of a problem in patients with type 2 diabetes mellitus than in those with type 1. Improved glycemic control, discontinuation of medications that slow gastric motility, and the use of metoclopramide may be helpful. Metoclopramide use preferably should be limited to a few days at a time, as long-term use has been linked to tardive dyskinesia. [380]

Autonomic neuropathy may manifest as orthostatic hypotension. Such patients may require volume expanders or adrenergic agents. Patients with cystopathy may benefit from cholinergic agents.

Acute-onset mononeuropathies in diabetes include acute cranial mononeuropathies, mononeuropathy multiplex, focal lesions of the brachial or lumbosacral plexus, and radiculopathies. It is important to consider nondiabetic causes for cranial nerve palsies, including intracranial tumors, aneurysms, and brainstem stroke. [381]

For more information see Diabetic Neuropathy and Diabetic Lumbosacral Plexopathy.


Management of Infections

Diabetes predisposes patients to a number of infectious diseases, including the following:

  • Malignant otitis externa

  • Rhinocerebral mucormycosis

  • Bacteriuria

  • Pyuria

  • Cystitis

  • Upper urinary tract infection

  • Intrarenal bacterial infection

  • Skin and soft tissue infections

  • Osteomyelitis

For more information, see Infections in Patients with Diabetes Mellitus.


Management of Intercurrent Medical Illness

Patients with intercurrent illness become more insulin resistant because of the effects of increased counterregulatory (ie, anti-insulin) hormones. Therefore, despite decreased nutritional intake, glycemia may worsen.

Patients on oral agents may need transient therapy with insulin to achieve adequate glycemic control. In patients who require insulin, scheduled doses of insulin, as opposed to sliding scale insulin, are far more effective in achieving glycemic control. [382, 383]

Metformin is a special case. If patients taking metformin have any illness that leads to dehydration or hypoperfusion, the drug should be temporarily discontinued because of a possible increased risk of lactic acidosis.


Management of Critical Illness

Standard practice in intensively ill patients has been to provide tight glycemic control through intensive insulin therapy. Research evidence, however, has called this practice into question.

A meta-analysis found that in critically ill adult patients, tight glucose control is associated with an increased risk of hypoglycemia but not with significantly reduced hospital mortality. [384] A large, international, randomized trial among adults treated in an intensive care unit (ICU) found that intensive glucose control (target, 81-108 mg/dL) resulted in higher mortality than did a blood glucose target of 180 mg/dL or less. [385]

However, large, single-center studies using more accurate glucose measurements have shown a benefit to intensive glycemic control in critical illness. [386] This remains an area of important ongoing research.

Results of the Diabetes and Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) trial suggested improved outcomes in patients with type 2 diabetes with acute myocardial infarction or stroke who receive constant IV insulin during the acute phase of the event to maintain blood glucose values of approximately 100-150 mg/dL. [387] However, these results were not confirmed in the follow-up trial, DIGAMI-2. [388]

A post-hoc analysis of the DIGAMI-2 study revealed that glucose-lowering drugs impact prognosis differently. Insulin may be associated with increased risk of nonfatal cardiac events, whereas metformin seems to be protective against risk of death. [389]


Pharmacologic Considerations in Surgery

Surgical patients may experience worsening of glycemia for reasons similar to those listed above for intercurrent medical illness. Patients on oral agents may need transient therapy with insulin to maintain blood glucose at approximately 100-180 mg/dL.

In patients who require insulin, scheduled doses of insulin (eg, glargine once daily plus glulisine before meals, as opposed to sliding-scale insulin, are far more effective in controlling glucose. Intensive glucose control in surgical ICU patients appears to reduce the risk of septicemia, but as with other critically ill patients, this may come at the cost of increased risk of hypoglycemia. [384]

A standardized protocol can be effective in transitioning patients who have diabetes and acute coronary syndrome to subcutaneous insulin once oral feeding has resumed. This is based on insulin requirement during the previous 12 hours. Half of the amount is given as basal insulin, and the remainder is given as prandial insulin. [390]

For patients who can eat soon after surgery, the time-honored approach of administering half of the usual morning dose of neutral protamine Hagedorn (NPH) insulin with 5% dextrose in the IV infusion is acceptable, with resumption of scheduled insulin (perhaps at reduced doses) within the first 1-2 days. With the availability of newer basal insulins (ie, glargine, detemir), options have expanded. A full dose of basal insulin can be given, and rapid-acting insulin can be administered when meals are consumed.

Patients receiving basal insulin can often receive their usual dose if they are given IV glucose during surgery, with appropriate intraoperative and postoperative monitoring of glucose. Oral antidiabetic agents can be restarted when the patient is stable and eating.

Insulin secretagogues should be used with caution in the hospital, since food intake may be interrupted by diagnostic tests and procedures. Metformin may have to be started at a lower dose and gradually titrated to full dose due to GI side effects. Since TZDs have such a long biologic effect, their omission in the hospital is usually inconsequential. The role of incretins in the hospital has not yet been defined.

For patients who require more prolonged periods without oral nutrition and for major surgery, such as coronary artery bypass grafting and major abdominal surgery, constant infusion IV insulin is preferred. Discontinue metformin temporarily after any major surgery until the patient is clearly hemodynamically stable and normal renal function is documented. Discontinuing metformin for at least 48 hours in this situation until proof of normal renal function is established is the current standard.


Prevention of Type 2 Diabetes Mellitus

Guidelines from the American College of Clinical Endocrinologists for the prevention of type 2 diabetes mellitus in patients at risk recommend the following measures:

  • Weight reduction

  • Proper nutrition

  • Regular physical activity

  • Cardiovascular risk factor reduction

  • Aggressive treatment of hypertension and dyslipidemia

Lifestyle improvement

The Diabetes Prevention Program (DPP) trial has shown that modest lifestyle changes (eg, 4-5% sustained weight reduction for approximately 3 y) reduce the risk for diabetes in patients at high risk by 58%. [391] Eight health-care facilities participated in an instructive study of group-based lifestyle intervention that should help other agencies/states emulate strategies used to affect positive lifestyle changes for the prevention of diabetes. [392]

In an 11-year, population-based cohort study of over 200,000 men and women without evidence of diabetes, heart disease, or cancer at baseline, good lifestyle decisions in combination significantly reduced the risk of developing diabetes. For each additional positive lifestyle factor (eg, with regard to diet, physical activity, or smoking) in the low-risk group, the odds for diabetes were 31% lower [393]

Yeh et al found that although cigarette smokers are at increased risk for type 2 diabetes, smoking cessation leads to higher short-term risk. [394] In this prospective cohort study in 10,892 middle-aged, nondiabetic adults, 1254 persons developed type 2 diabetes during 9 years of follow up.

The adjusted hazard ratio of incident diabetes among persons in the highest tertile of pack-years was 1.42, compared with persons who had never smoked. However, in the first 3 years after quitting smoking, the hazard ratio was 1.73; the risk then gradually decreased, disappearing completely at 12 years. Yeh et al recommended that smoking cessation in smokers at risk for diabetes be coupled with strategies for prevention and early detection of diabetes.

A significant inverse correlation has been found between the risk of diabetes and the intake of magnesium, which plays an important role in insulin action and glucose homeostasis. In a meta-analysis, the summary relative risk of type 2 diabetes for every 100 mg/day increment in magnesium intake was 0.86. [395]

Interest in the impact of phylloquinone intake on glucose tolerance and insulin sensitivity has a long history. A 2012 report suggests a beneficial role for phylloquinone in diabetes prevention in elderly subjects with high cardiovascular risk. However, caution is advised in patients who are concurrently being treated with anticoagulant drugs such as warfarin. [396]

Pharmacologic prevention

Drugs from several classes have been studied in the prevention of diabetes. However, the FDA has not approved any drug for the treatment of prediabetes or the prevention of type 2 diabetes. [397]


The ADA recommends that, in addition to lifestyle counseling, metformin be considered in selected patients with prediabetes. [2] ADA criteria for preventive metformin therapy are as follows:

  • Obesity

  • Age younger than 60 years

  • Both impaired fasting glucose (IFG) and impaired glucose tolerance (IGT)

  • Other risk factors (eg, HbA1C >6%, hypertension, low HDL cholesterol, elevated triglycerides, or a family history of diabetes in a first-degree relative)

In the DPP, metformin 1700 mg daily was about half as effective as lifestyle intervention in reducing risk among subjects with elevated fasting and postload plasma glucose concentrations. [391] Over an average follow-up period of 2.8 years, the incidence of diabetes was 11.0, 7.8, and 4.8 cases per 100 person-years in the placebo, metformin, and lifestyle groups, respectively.


Analysis of available data from the DPP suggests that troglitazone was effective in preventing diabetes. This effect was also seen in the Troglitazone in Prevention of Diabetes (TRIPOD) study of Hispanic women with a history of gestational diabetes. After troglitazone was withdrawn from the market because of hepatotoxicity, the continuation of TRIPOD in the Pioglitazone in the Prevention of Diabetes Study demonstrated slowed progression of subclinical atherosclerosis with glitazone treatment. [398]

In the Diabetes Reduction Assessment With Ramipril and Rosiglitazone Medication (DREAM) trial, investigators concluded that rosiglitazone at 8 mg daily reduces the incidence of type 2 diabetes mellitus in patients with IFG and/or IGT. At the end of this prospective, multicenter study, composite outcome of diabetes or death from any cause was 11.6% in the rosiglitazone group versus 26% in the placebo group. [206] Ramipril did not produce significant reduction in the same composite outcome. [399]


Acarbose (100 mg three times a day) was shown in the Study to Prevent Non-Insulin Dependent Diabetes Mellitus (STOP-NIDDM) to reduce diabetes rates by approximately 25% in patients at high risk for the development of type 2 diabetes. [400] This 6-year, international, multicenter, double-blind, placebo-controlled, randomized investigation included 1,368 subjects with IGT.


Stroke Prevention in Diabetes

The 2010 American Heart Association/American Stroke Association (AHA/ASA) guidelines for the primary prevention of stroke include the following recommendations for patients with diabetes:

  • Regular blood pressure screening

  • Physical activity; 30 minutes or more of moderate-intensity activity on a daily basis

  • A low-sodium, high-potassium diet to reduce blood pressure; a diet emphasizing consumption of fruits, vegetables, and low-fat dairy products (eg, the Dietary Approaches to Stop Hypertension [DASH] diet) may lower stroke risk

  • A blood pressure goal of less than 130/80 mm Hg

  • Drug therapy with ACE inhibitors or ARBs

  • Statin therapy, especially in patients with other risk factors; monotherapy with fibrates may also be considered to lower stroke risk

The AHA/ASA guidelines note that the benefit of taking aspirin for the reduction of stroke risk has not been fully demonstrated in diabetic patients.



Primary care providers can care for patients with type 2 diabetes mellitus adequately. The multiple facets of disease treatment (eg, nutrition, exercise, smoking cessation, medications, complications monitoring) and data management (eg, glucose levels, blood pressure, lipids, complications monitoring) must be continually addressed.

Inability to achieve adequate glycemic (or blood pressure or lipid) control usually should be a clear indication to consult a diabetes specialist. When a patient has developed advanced complications, a diabetes specialist cannot be expected to be able to lessen the burden of these complications.