Diabetes mellitus is a group of metabolic diseases characterized by hyperglycemia resulting from insulin resistance and/or insulin deficiency. Diabetes mellitus leads to a wide range of complications and, when poorly controlled, contributes to significant morbidity and mortality (see also Complications of Diabetes Mellitus chapter).

Type 1 diabetes, which accounts for approximately 5-10% of all cases, is caused by the autoimmune destruction of the insulin-producing beta-cells of the pancreas resulting in absolute insulin deficiency and hyperglycemia.[1] People with type 1 diabetes may also develop other autoimmune conditions (e.g., Hashimoto’s thyroiditis, Graves’ disease, celiac disease, rheumatoid arthritis, and lupus).[2] Most cases present in childhood with acute symptoms of polyuria and polydipsia. Approximately one-third present with life-threatening diabetic ketoacidosis.[3] Type 1 diabetes requires insulin treatment, along with therapeutic lifestyle interventions that are aimed at minimizing insulin doses and protecting against vascular complications.

Type 2 diabetes, which accounts for more than 90% of diabetes cases, is characterized by insulin resistance and relative, rather than absolute, insulin deficiency.[1] Complications include cardiovascular disease, nephropathy, neuropathy, retinopathy, peripheral vascular disease, and cerebrovascular disease, as well as damage to musculoskeletal, hepatic, and digestive systems, and impaired cognitive function and mental health.[4]

Latent autoimmune diabetes in adults includes a heterogenous group of conditions that is phenotypically similar to type 2 diabetes, but patients have autoantibodies of the type commonly observed in type 1 diabetes. Diagnostic criteria include age 30 years or older, no insulin treatment for 6 months after diagnosis, and presence of autoantibodies to glutamic acid decarboxylase, islet cells, tyrosine phosphatase (IA-2α and IA-2β), or insulin.

Patients with maturity-onset diabetes of youth typically present before 25 years of age, have only impaired insulin secretion, and have a monogenetic defect that leads to an autosomal dominant inheritance pattern.[5]

Diabetes begins with the accumulation of lipid particles in muscle and liver cells, coming largely from the diet. This accumulation of intramyocellular lipid and hepatocellular lipid causes insulin resistance in which muscle and liver cells are less responsive to insulin’s action and less able to remove glucose from the bloodstream.[6] Recent studies suggest that lipid accumulation in the pancreas may further contribute to insufficient insulin secretion.[7]

While type 2 diabetes is commonly and mistakenly understood by laypeople to be the result of sugar consumption, this assumption is not supported by current evidence.[8] Type 2 diabetes starts as insulin resistance resulting from intramyocellular and hepatocellular lipid accumulation. This process can occur in normal-weight individuals, but about 60-80% of patients are overweight. Prevalence in children is increasing rapidly due to poor dietary habits and increasing weight problems.[9]

People with type 2 diabetes may present with classic symptoms of hyperglycemia or may be asymptomatic or have only mild symptoms, such as intermittent blurred vision or nocturia. Diabetes in these patients may be diagnosed during a routine examination. Type 2 diabetes is often accompanied by hypertension and lipid abnormalities, conditions that, like hyperglycemia, are amenable to dietary interventions.

Prediabetes. Type 2 diabetes may be preceded by prediabetes, in which fasting blood glucose or 2-hour postprandial oral glucose tolerance test (OGTT) values are above the normal range but not yet at the level of diabetes (see Diagnosis below). Approximately one-third of US adults have prediabetes.[10] They are at increased risk of type 2 diabetes and cardiovascular disease, and often also have hypertension, dyslipidemia, and obesity.[1] The rate of progression from prediabetes to type 2 diabetes varies with the population studied and the criteria used to define prediabetes.[11] In the Diabetes Prevention Program Outcomes Study, 11% of the control group progressed to diabetes within 2.8 years.[12] Prediabetes and its associated conditions are highly amenable to lifestyle interventions.[13]

Gestational diabetes mellitus is impaired glucose tolerance during pregnancy. In 2019, its prevalence in the US was estimated at 22%.[14] Many cases are actually previously undiagnosed type 2 diabetes, as women of childbearing age are not routinely screened for diabetes.[1] Treatment with dietary modification and/or medication (usually insulin) is essential to prevent short- and long-term risks to mother and fetus. Although blood glucose levels often normalize postpartum, women who had gestational diabetes have a high risk of developing type 2 diabetes. This risk may be reduced with diet and lifestyle changes.
Other forms of diabetes can result from other disease states or treatments and include the following:

• Monogenic diabetes syndromes (such as neonatal diabetes and maturity-onset diabetes of the young [MODY], referenced above)
• Diseases of the exocrine pancreas (such as cystic fibrosis)
• Diabetes caused by pancreatectomy (e.g., related to pancreatic cancer)
• Drug- or chemical-induced diabetes (such as with glucocorticoid use, in treatment of HIV/AIDS, or after organ transplant)[1]

Risk Factors

Genetic factors play an important role in type 1 diabetes. A close family history of type 1 diabetes significantly increases the risk of type 1 diabetes in subsequent generations, as multiple genes are associated with its development.[15] Genetic predisposition alone, however, does not explain the rapid rate of increase of type 1 diabetes over the past 50 years. This is especially true in individuals over the age of 50, who are less likely to have any genetic risk.[16] Though not well understood, it has recently been suggested that an interplay between genes that influence mucosal immunity, along with alterations of the intestinal microbiota and increased intestinal permeability (leaky gut), may play roles in the pathogenesis of type 1 diabetes.

Additional contributors to type 1 diabetes risk include exposure to viruses associated with islet inflammation, birth weight greater than 4,500 g (9.9 lb), preeclampsia, oxidative stress from beta-cell exposure to nitric oxide (NO) and oxygen radicals (O2), and vitamin D deficiency.[17],[18],[19],[20] Some evidence also links exposure during infancy to cow’s milk proteins to type 1 diabetes risk.[21],[22],[23]

For type 2 diabetes, adults who are overweight (body mass index [BMI] ≥ 25 kg/m² or ≥ 23 kg/m² in Asian Americans) and have any of the following are at increased risk:[1]

• Regular consumption of meat or dairy products (see Nutrition Recommendations for more information)[24]
• Physical inactivity
• 1st-degree relative with diabetes
• High-risk race/ethnicity (e.g., Black, Latino, Native American, Asian American, Pacific Islander)
• Women who were diagnosed with gestational diabetes
• Hypertension ≥ 140/90 mm Hg or on therapy for hypertension
• HDL cholesterol level < 35 mg/dL (0.90 mmol/L) and/or a triglyceride level > 250 mg/dL (2.82 mmol/L)
• Women with polycystic ovary syndrome
• A1C ≥ 5.7% (39 mmol/mol), impaired glucose tolerance, or impaired fasting glucose on previous testing
• Other clinical conditions associated with insulin resistance (e.g., severe obesity, acanthosis nigricans)
• History of cardiovascular disease
• Smoking[25]

For gestational diabetes mellitus, risk factors include age ≥ 35 years, overweight, sedentary lifestyle, excessive gestational weight gain, excessive central body fat deposition, family history of diabetes, short stature (< 4 ft 9 in), excessive fetal growth, polyhydramnios, hypertension or preeclampsia in the current pregnancy, history of recurrent miscarriage, offspring malformation, fetal or neonatal death, macrosomia, gestational diabetes during prior pregnancies, and polycystic ovary syndrome.[26] Black, Asian, Native American, and Hispanic individuals have greater prevalence of gestational diabetes than non-Hispanic white individuals.[27]

Diagnosis

Endocrinopathies, such as Cushing’s disease, acromegaly, pheochromocytoma, and hyperthyroidism, may impair glucose tolerance and should be ruled out.

Prediabetes, Type 1 Diabetes, and Type 2 Diabetes

Diabetes and prediabetes are diagnosed based on laboratory values reflecting blood glucose control.[28] Normal values are fasting plasma glucose less than 100 mg/dL, a 2-hour OGTT less than 140 mg/dL, and an A1C of 5.6% or less.

Prediabetes or categories for increased risk for diabetes are:

• A1C 5.7-6.4%
• Impaired fasting glucose (IFG): fasting plasma glucose 100 mg/dL to 125 mg/dL
• Impaired glucose tolerance (IGT): 2-hr plasma glucose in the 75-g OGTT 140 mg/dL to 199 mg/dL

Diabetes is diagnosed based on the following values:

• A1C ≥ 6.5%
• A fasting plasma glucose concentration ≥ 126 mg/dL
• A 75-g, 2-hour OGTT result of 200 mg/dL or greater (if no symptoms of hyperglycemia are present, repeat testing is needed to confirm results)
• A random plasma glucose of 200 mg/dL or greater when hyperglycemia symptoms are present

In the absence of unequivocal hyperglycemia, a diagnosis of diabetes requires 2 abnormal results from the same sample or 2 separate test samples.[1]

Black individuals may have higher A1C levels than non-Hispanic white individuals despite similar fasting and postprandial blood glucose levels.[1] A1C levels may also be influenced by the presence of certain hemoglobinopathies and in conditions associated with increased red blood cell turnover, such as pregnancy and hemodialysis. To improve diagnostic specificity, the Veterans Administration/Department of Defense (VA/DoD) recommends that A1C values between 6.5-7.0% be confirmed with fasting plasma glucose levels.[29]

Type 1 diabetes often presents with ketoacidosis in children and adolescents, which is caused by partial or total insulin deficiency and normally requires hospital admission and intensive care. Beta-cell destruction can be slower in some individuals, however, resulting in a slower, progressive rise in glucose, which may worsen with stress.[1] When patients present with clear symptoms of hyperglycemia, a single random plasma glucose ≥ 200 mg/dL (11.1 mmol/L) is enough to diagnose the patient. The most common form of type 1 diabetes is immune-mediated and results from a cellular-mediated autoimmune destruction of the beta-cells of the pancreas. It can be identified by the presence of islet cell autoantibodies and antibodies to GAD (GAD65), insulin, the tyrosine phosphatases IA-2 and IA-2β, and zinc transporter 9 (ZnT8).[1] Idiopathic type 1 diabetes is less common, in which the patient presents with diabetic ketoacidosis but has no evidence of autoimmunity. Low plasma C-peptide levels can confirm that the pancreas is secreting little or no insulin.

In all asymptomatic adult patients, testing for type 2 diabetes should begin at age 45 and may be repeated after 3 years if all results are normal. Testing should be considered at any age for those adults who are overweight and have at least 1 of the risk factors noted above for development of diabetes, especially women who are considering pregnancy.

Gestational Diabetes

Screening for gestational diabetes is a routine part of prenatal examinations. Because the poor eating habits and obesity that are common in the US and many other countries put many women of childbearing years at risk for type 2 diabetes, it is important to test for undiagnosed type 2 diabetes at the first prenatal visit for those individuals with risk factors (see above) using standard diagnostic criteria.

Screening typically occurs between 24-28 weeks of gestation in women who are at low risk and not previously known to have diabetes. Diagnosis can be accomplished by either a 1-step or a 2-step approach:[1]

1. The 1-step strategy is conducted at 24-28 weeks gestation with a fasting 75-g OGTT, measuring fasting, 1-hour, and 2-hour plasma glucose levels. Diagnosis is made when plasma glucose levels exceed the following:

1. Fasting: 92 mg/dL (5.1 mmol/L)
2. 1-hour: 180 mg/dL (10.0 mmol/L)
3. 2-hour: 153 mg/dL (8.5 mmol/L)

2. The 2-step strategy involves a 1-hour challenge with a 50-g oral glucose load. A venous serum or plasma glucose greater than or equal to 130 mg/dL, 135 mg/dL, or 140 mg/dL is considered a positive screen, with greater sensitivity at the lower threshold and greater specificity at the higher threshold.[30] If a patient screens positive, she will then proceed to a 3-hour, 100-g OGTT after an 8-hour fast, which is diagnostic for gestational diabetes when glucose values are elevated at 2 different times. The cutoff values for elevated glucose are 95 mg/dL at fasting, 180 mg/dL at 1 hour, 155 mg/dL at 2 hours, and 140 mg/dL at 3 hours.

The 1-step approach is the newer of the 2 recommended screening strategies. It may significantly increase the number of women diagnosed with gestational diabetes, increasing medical costs and medicalization of pregnancies.[31] However, there is some evidence that there may be benefits to intervening at lower glucose levels by decreasing large-for-gestational-age births and preeclampsia.[1]

In pregnancy, mildly abnormal glucose levels can lead to fetal complications, which is why fasting serum glucose greater than 92 mg/dL is considered abnormal and treatment is started very early. Treatment should be considered in women with fasting glucose even below the diagnostic values for diabetes due to the increased risk of adverse outcomes, which can occur even with a normal OGTT.[32] Women with gestational diabetes should be screened 6-12 weeks postpartum using the OGTT with the normal diagnostic criteria. Lifelong screening should be conducted every 3 years, and the nutritional interventions described below should be implemented to reduce the risk of type 2 diabetes.

Treatment

Nutrition, exercise, and other lifestyle interventions are essential parts of treatment for all forms of diabetes. As described below, there is particular value to a low-fat vegan diet favoring low-glycemic-index foods. For individuals with type 1 diabetes, such interventions can reduce insulin requirements, reduce the risk of complications, and minimize the effect of the disease on the patient’s life. Insulin treatment will nevertheless be necessary, regardless of lifestyle measures.

For type 2 diabetes, conventional interventions have used lifestyle and medication to slow the progression of diabetes and its complications. Currently, only roughly 1 in 4 people with diabetes reaches all ideal glycemic, blood pressure, cholesterol, and smoking cessation targets.[33] A new, more aggressive focus on diabetes remission aims to improve all risk factors associated with diabetes and its complications to improve patient quality of life and reduce morbidity and mortality.

For all types of diabetes, Diabetes Self-Management Education and Support provides patients and family members with knowledge, decision-making, and skills needed for diabetes self-care. [34]Treatment also includes referrals to health care professionals as appropriate (podiatrists, ophthalmologists, registered dietitians, dentists, and mental health professionals). American Diabetes Association (ADA) guidelines also provide recommendations on vaccinations (e.g., for influenza, pneumococcal pneumonia, and hepatitis B).

A1C and Blood Glucose Targets

Hemoglobin A1C is an indicator of blood glucose during the previous 2 to 3 months. Maintaining an A1C below 7% has long been recommended as a primary goal of care for patients with either type 1 or type 2 diabetes.[35],[36],[37],[38] However, evidence from large randomized trials shows that, for people with type 2 diabetes, tight control achieved with medication does not reduce mortality or prevent severe macro- and microvascular complications.[39],[40],[41],[42],[43] Moreover, pharmacologic efforts at tight control can lead to episodes of hypoglycemia. Concerns about lack of efficacy and risks to patients with such management strategies prompted the American College of Physicians to recommend a target A1C range of 7-8% for most patients with type 2 diabetes when medications are used.[44] The ADA suggests a target of < 7.5% for adults aged 65 and older who are otherwise healthy, and a target of < 8.0-8.5% for those with cognitive impairments or in poor health.[45]

When using lifestyle interventions alone (without medications), there is no specified lower limit for A1C based on any safety consideration.

Self-monitoring of blood glucose is critical for people with type 1 diabetes and people with type 2 diabetes on insulin. For patients who do not require insulin, self-monitoring may be useful for titrating medications that can cause hypoglycemia, during concurrent illness, or when modifying diet and exercise routines.[46]

Pharmaceutical Treatment for Type 1 Diabetes

Type 1 diabetes should be treated with continuous insulin infusion via a subcutaneous insulin pump, or multiple insulin injections per day of both basal and prandial insulin.[47] The choice of agent(s), frequency of administration, and delivery device(s) are tailored to the patient’s needs.

Insulin is available in several forms that differ in duration of action:

• Long-acting insulins provide basal coverage. Insulin glargine is usually administered daily; insulin detemir may be given once or twice a day.
• NPH is also a basal insulin but has an intermediate length of action and is often administered twice a day.
• Short- (regular) or rapid-acting insulin analogues can be administered in boluses before meals to curb postprandial blood glucose elevations and to correct premeal elevations.
• Intermediate insulin is available premixed with short- or rapid-acting insulin when 2 insulin types are required.
• Premixed insulins are primarily used in patients with type 2 diabetes.

Amylin is a beta-cell hormone that is co-secreted with insulin. Pramlintide, a synthetic amylin analogue, is injected at mealtimes along with fast-acting insulin. Pramlintide reduces postprandial rises in blood glucose concentrations and suppresses appetite, which may lead to weight loss. Caution should be exercised, however, as severe hypoglycemia may develop.

Glucagon injection is an antidote to severe hypoglycemia for an unconscious patient. Patients who are at risk and live with someone who can be trained to use glucagon should be provided with instructions and a prescription.

Pharmaceutical Treatment of Type 2 Diabetes

Dietary changes, along with exercise, are the first-line treatments for type 2 diabetes and are described in the Nutrition Considerations section below. Nonetheless, many patients will also require medications. When the decision is made to treat hyperglycemia, the Diabetes Medication Choice Decision Aid can be used to help decide on an agent that best fits the patient’s values and preferences, keeping in mind that there is little meaningful difference between medications in reduction of cardiovascular and all-cause mortality, while there are significant differences in cost, side effects, effect on weight, and contraindications.[48],[49]

Oral Agents

Biguanides. Metformin is recommended as the first-line agent, as it provides effective glycemic control, does not promote weight gain, is usually well tolerated, and is affordable for many patients. It decreases hepatic gluconeogenesis and increases insulin sensitivity. Metformin modestly reduces the risk of cardiovascular events, irrespective of its impact on glycemia, and may also reduce mortality because of an anticancer effect.[50] It is contraindicated in heart failure, renal insufficiency, liver disease, excessive alcohol intake, serious infection and illness, and other disease processes. Gastrointestinal disturbances are common but usually abate over time. Chronic metformin use results in vitamin B12 deficiency in 16-30% of patients, symptoms of which may be misdiagnosed as diabetic neuropathy. Chronic use may also result in anemia.[51] Vitamin B12 monitoring and supplementation are therefore essential.

None of the agents listed below, when used to achieve tight glycemic control, have shown an absolute benefit for the majority of study participants on patient-centered outcomes, such as prevention of kidney failure, loss of eyesight, amputation, or mortality.[42]

Sulfonylureas. These medications work by stimulating the pancreatic beta-cells to release insulin. Common examples include glipizide, glyburide, and glimepiride. Glipizide is shorter acting and may be preferable for geriatric patients or those with renal or hepatic insufficiency. Sulfonylureas may cause hypoglycemia and weight gain, as well as dose-related gastrointestinal effects such as nausea, diarrhea, and constipation. They often cease to be effective within a few years because their use accelerates beta-cell failure compared with other medications. Their use is therefore often discontinued in an attempt to preserve beta-cell function.

Thiazolidinediones. Pioglitazone and rosiglitazone increase insulin sensitivity in muscle, fat, and the liver, decrease glucose production, and may also increase insulin secretion. These medications are generally well tolerated by patients and do not cause hypoglycemia as monotherapy. However, they are contraindicated in heart failure, and pioglitazone and rosiglitazone may cause or exacerbate congestive heart failure. Rosiglitazone use is associated with an increased risk of myocardial ischemic events such as angina or myocardial infarction. These elevations in cardiovascular risks are of particular concern, given that one of the principal goals of diabetes treatment is to attempt to reduce cardiovascular risk.
Weight gain, fluid retention, and hepatic injury are additional risks, especially when combined with insulin, and liver enzymes should be routinely monitored.

Pioglitazone should not be used in patients with active bladder cancer, and its use may raise the risk of developing bladder cancer. Thiazolidinediones may also cause macular edema and increased incidence of fractures.

Meglitinides. Nateglinide and repaglinide stimulate insulin secretion but are shorter acting than sulfonylureas and must be taken before meals. Dose reduction is required for hepatic or renal impairment. They may cause hypoglycemia, weight gain, and flu-like symptoms.

Alpha-glucosidase inhibitors. Acarbose and miglitol inhibit the conversion of carbohydrates to monosaccharides, slow the absorption of glucose, and lower postprandial glucose values. They need to be taken at the beginning of a meal. Flatulence, abdominal cramping, and diarrhea are common side effects and may limit compliance.

DPP-IV inhibitors. Alogliptin, saxagliptin, sitagliptin, and vildagliptin inhibit the enzyme that degrades endogenous incretin hormones. This results in increased glucose-dependent insulin secretion, decreased glucagon secretion, and delayed gastric emptying. Sitagliptin and saxagliptin must be given at a lower dose for those with renal insufficiency. These agents are expensive and are most effective for A1C reduction in drug-naive patients.

SGLT-2 inhibitors. Canagliflozin, dapagliflozin, and empagliflozin decrease the reabsorption of glucose in the kidneys, increasing glucose in the urine, which can result in urinary tract and genital mycotic infections. These agents may also cause hypotension and should be used cautiously with antihypertensive medications, especially diuretics, angiotensin-converting enzyme inhibitors, and angiotensin II receptor blockers. These agents increase urination. They may also increase LDL cholesterol concentrations and, when used in combination with insulin or insulin secretagogues, may cause hypoglycemia. Renal and liver function and potassium levels should be monitored. Canagliflozin may reduce bone mineral density and increase fracture risk.[52] These agents are expensive.

Noninsulin Injectables

GLP-1 analogues. These injectable synthetics are used to increase mealtime insulin secretion in type 2 diabetics, usually as an add-on to metformin or an insulin secretagogue. They vary from twice daily taken 30-60 minutes before meals (exenatide, liraglutide, and lixisenatide), to once daily (liraglutide), and once a week (albiglutide, dulaglutide, and exenatide extended-release). They are less likely to cause hypoglycemia when compared with insulin or insulin secretagogues and may promote satiety, leading to decreased appetite and modest weight loss. Nausea, vomiting, and diarrhea are common side effects and can lead to renal impairment. Reduced doses should be used in patients with renal impairment. These agents are expensive and carry a risk of thyroid tumors and pancreatitis, which may be fatal.

Amylin analogue. Pramlintide (described above) may also be used in type 2 diabetes.

Insulin

The insulin formulations described above for type 1 diabetes are appropriate for use in type 2 diabetes when a patient is symptomatic, unable or unwilling to make lifestyle modifications, unable to take metformin or already on 2 g of metformin daily, and able to safely inject insulin, monitor blood glucose levels, and recognize and treat hypoglycemia. There is no evidence of benefit for starting insulin therapy early in the course of type 2 diabetes. In several trials (ACCORD, VADT, and NICE-SUGAR), type 2 diabetes patients who were prescribed insulin to achieve an A1C ≤ 7% were eventually found to have an increased incidence of severe hypoglycemia and cardiovascular mortality. Apart from its use for symptom control (ideally for short periods), there is no benefit to insulin use in people with type 2 diabetes, and it may actually reduce quality of life.[50]

When insulin is required in type 2 diabetes, the best regimen is the one that (1) results in adequate blood glucose control with a dosing schedule that is acceptable to the patient, and (2) causes the fewest episodes of hypoglycemia.

Clinicians should avoid the common practice of overtreatment of type 2 diabetes with medications and should de-intensify insulin and other pharmaceutical interventions through dose reduction or discontinuation.[45] In some cases, insulin and other agents can be withdrawn, with blood glucose levels maintained through lifestyle interventions alone.

A caveat to this approach applies to patients with presumed type 2 diabetes and autoantibodies, who are less likely to respond to the oral agents. Such patients may require insulin therapy (see type 1 diabetes) and are at increased risk of ketoacidosis.

Treatment of Gestational Diabetes

Dietary interventions are discussed in detail under Nutritional Considerations below. When adequate glycemic control is not attained through nutrition modification and physical activity, insulin should be prescribed. Oral antihyperglycemics are not approved for the treatment of gestational diabetes by the FDA; however, the ADA supports the use of oral therapy (metformin and/or glyburide) during pregnancy when the patient is unable to take insulin due to cost, complexity, cultural influences, or other considerations.[53] Still, these medications should not be offered as first-line therapy because long-term safety for offspring has not been demonstrated, and efficacy for glycemic control has not been found to be adequate.[53]

Role of Physical Activity

A sedentary lifestyle is associated with increased risk for impaired glucose tolerance and diabetes.[54] Exercise and diet-exercise programs that produce weight loss significantly reduce the risk for type 2 diabetes.[55],[56],[57],[58]

Exercise alone has little or no effect on body weight.[59] However, in persons with established diabetes, exercise reduces blood glucose and plasma lipid concentrations and improves insulin sensitivity.[60] Exercise also reduces cardiovascular complications of diabetes, including high blood pressure, left ventricular diastolic dysfunction, arterial stiffness, systemic inflammation, and left ventricular mass.[61]

For prevention of type 2 diabetes and for improved glycemic control and other health benefits for type 1 and type 2 diabetes, the ADA recommends at least 150 minutes per week of moderate to vigorous aerobic exercise, divided among at least 3 days per week, with at most 2 consecutive days of rest. For patients with a BMI > 30, longer sessions are recommended to achieve 200-300 minutes per week.[62] All adults with diabetes should avoid being sedentary for longer than 30 minutes at a time, and engaging in flexibility and balance training is recommended 2-3 times per week for older adults with diabetes. If not contraindicated, resistance training should be performed 2-3 sessions per week.[34],[63]

Women who either have or are at risk for gestational diabetes can also benefit from exercise. Independent of BMI, women who regularly engage in moderate exercise (e.g., brisk walking) are at reduced risk of gestational diabetes.[64] In women with gestational diabetes, exercise helps control blood glucose without the use of insulin.[65]

Because exercise reduces blood glucose, medications (especially insulin acting during the time of the activity) may need to be adjusted or carbohydrate intake increased on exercise days. A form of treatment for low blood glucose should be available for those on medications that can cause hypoglycemia who are engaging in strenuous exercise. Patients with type 1 diabetes have a variable response to exercise, and care needs to be taken in recommending the type and duration of exercise.[34]

It should be noted that exercise does not take the place of diet changes.

Nutritional Considerations

Diet and prevention of type 1 diabetes. Compared with a stable and relatively low incidence of type 1 diabetes in the first half of the 20th century, incidence rose in the latter half of the century and is now increasing by 3% per year, a rate faster than can be explained by genetic factors.[66],[67] Possible explanations include changes in infant diets, particularly the lack of breastfeeding and increased exposure to cow’s milk, the role of hygiene, differences in gut microbiota and gut permeability, and a high maternal BMI during pregnancy.[68] The following considerations have emerged as potentially important candidates for preventive strategies, although none has been definitively established:

Breastfeeding. In a pooled analysis of 43 observational studies involving nearly 10,000 patients, exclusive breastfeeding for greater than 2 weeks and more than 3 months was associated with a 25% and 13% lower risk for type 1 diabetes, respectively.[69] Four systematic reviews implemented by the US Departments of Agriculture and Health and Human Services examined the relationship between breastfeeding and risk for type 1 diabetes. They found limited evidence that never (versus ever) breastfeeding babies is associated with a higher risk, moderate evidence that breastfeeding babies for shorter (versus longer) periods is associated with a higher risk, and limited evidence that exclusive breastfeeding is associated with higher risk for type 1 diabetes.[70] Longer duration of breastfeeding may reduce risk for type 1 diabetes, presumably by prolonging avoidance of exposure to cow’s milk proteins or other antigens, increasing protection against infections, enhancing the infant’s immune responses, and increasing beta-cell proliferation.[71] Children who were never breastfed had a 2-fold increased risk for developing type 1 diabetes compared with those who were breastfed, according to 2 large cohorts from Scandinavia.[72]

Avoidance of early introduction of cow’s milk. In a classic 1992 study of 142 children with type 1 diabetes, every child had antibodies to cow’s milk proteins, and these antibodies were found to be capable of attacking the pancreatic beta-cells.[73]

Greater intake of cow’s milk protein was associated with increased risk of islet autoimmunity, a precursor to type 1 diabetes, in children with a low/moderate risk genotype.[21] Cow’s milk protein formula and enterovirus infection before the age of 3 months were found to be associated with islet autoimmunity.[22] Incidence of beta-cell autoantibodies was reduced by age 3 in children at genetic risk of type 1 diabetes who were randomized to a bovine-insulin-free formula, though additional studies are needed to determine if this is protective against development of type 1 diabetes.[23]

Human studies show that cow’s milk intake in childhood is associated with an increased risk for development of islet cell autoantibodies and type 1 diabetes.[21]

Although the contribution of cow’s milk to type 1 diabetes remains to be proven, the American Academy of Pediatrics concluded that avoiding early exposure to cow’s milk may reduce the risk.[74]

An interventional pilot study suggested that substituting a hydrolyzed casein formula in the place of cow’s milk formula may decrease the risk of beta-cell autoimmunity over 4 years.[75] Later results from this study, however, contradicted this finding.[76]

Avoiding early introduction of gluten-containing foods. A high degree of concordance between type 1 diabetes and celiac disease has been observed.[77] In epidemiologic studies, supplementing infant diets with gluten-containing foods before 3 months of age or later than 6 months is associated with increased risk for developing islet cell autoantibodies and type 1 diabetes.[78] In some children, both early (before 3 months) and late (after 7 months) introduction of cereals was associated with increased risk of islet autoimmunity, suggesting that there may be a window of exposure to cereals, outside which initial exposure increases islet autoimmunity risk in genetically susceptible children.[79],[80]

Researchers are seeking ways to prevent autoimmune attacks of pancreatic beta-cells, with the goal of reducing the risk for developing insulin-dependent diabetes. So far, evidence suggests that the following dietary factors may protect against the development of type 1 diabetes: breastfeeding rather than exposure to dairy-based formulas, probiotic supplementation, and vitamin D supplementation during both pregnancy and the infant’s first year of life.[81],[82] All of these actions modulate the production of proinflammatory cytokines that are known to be involved in the pathogenesis of type 1 diabetes.[83]

Diet and prevention of type 2 diabetes. Diet and lifestyle significantly affect the likelihood that type 2 diabetes will manifest and influence its course after diagnosis.

As noted above, type 2 diabetes begins as insulin resistance, resulting from the accumulation of fat particles in muscle (intramyocellular lipid) and liver (hepatocellular lipid) cells. To the extent this fat buildup is avoided, diabetes risk remains low.

Individuals in Asia and Africa who follow traditional diets low in animal fat and high in complex carbohydrates and who remain physically active have a far lower incidence of diabetes than those who follow a Western diet and activity pattern.[84] A number of reviews have concluded that plant-based dietary patterns rich in fruits, vegetables, legumes, and nuts are the most effective approaches to the prevention and management of type 2 diabetes.[85],[86],[87],[88] Vegetarians also have a lower prevalence of diabetes than their omnivorous counterparts.[24],[89],[90] In a cohort of 60,903 participants in the Adventist Health Study-2, the vegan group had the lowest prevalence of type 2 diabetes: 2.9% compared with 7.6% in the nonvegetarian group.[91]

More than 6,000 adults ages 50 and older from the Third National Health and Nutrition Examination Study (NHANES III) were followed for 18 years. When categorized for animal protein intake, those in the high-protein-intake group had a 73-fold increased risk of diabetes mortality, compared with those in the low-protein-intake group; a moderate intake was associated with a 23-fold increase, while vegetable protein was not associated with an increase in risk.[92] A systematic review and meta-analysis of randomized controlled trials revealed that replacing animal protein with plant protein improves glycemic control in individuals with diabetes.[93] A meta-analysis including more than 50,000 nondiabetic white individuals found that, regardless of genetic risk of diabetes, meat consumption was associated with higher fasting glucose and insulin concentrations.[94]

Several possible mechanisms link meat consumption to diabetes risk, apart from the large amount of fat in meat that contributes to the buildup of intramyocellular and hepatocellular lipid.[24] Higher meat consumption is associated with an increase in body weight, as meat is typically higher in calories, compared with plant-based foods.[91],[95],[96] Meat is the second-leading source of saturated fat (after dairy products). Intake of saturated fat reduces both insulin secretion and insulin receptor activity and is associated with impaired glucose tolerance, insulin resistance, gestational diabetes, and type 2 diabetes.[97],[98] Meat intake is also associated with inflammation, a greater deposition of fat in visceral tissue, and an unhelpful accumulation of stored iron.[99] A high intake of heme iron (found in meat) is associated with a significantly increased risk for type 2 diabetes.[100]

Nitrates in processed meats have also been associated with diabetes.[99] Advanced glycation end products found in meat products cooked at high temperatures are highly oxidative metabolites that damage beta-cells, reduce insulin sensitivity, and contribute to the development of microvascular complications as well as cardiovascular disease.[99] L-carnitine in meat is metabolized to trimethylamine by the microbiota and then further metabolized to trimethylamine N-oxide (TMAO) in the liver. Serum TMAO significantly increases the risk of cardiovascular disease.[99]

Diets richer in plant-based foods—fruits, vegetables, whole grains, and beans—result in lower postprandial glucose and are associated with a significantly lower risk for type 2 diabetes.[101],[102],[103] High-fiber diets often contain micronutrients important in glucose tolerance, including magnesium and vitamin K. A study of more than 127,000 men and women found that consuming the highest amount of magnesium was associated with a roughly 35% lower risk for development of type 2 diabetes compared with those consuming the least amount.[104],[105],[106]

The Diabetes Prevention Program demonstrated that dietary changes designed to reduce body weight, combined with regular exercise, can significantly reduce the risk for type 2 diabetes. Risk reduction was 58% in the group that combined diet and exercise compared with placebo. Those on drug (metformin) treatment had a 31% reduction. In individuals aged 60 and older, the risk reduction was 71% with diet and exercise.[12]

Recent reviews have concluded that, as in type 2 diabetes, diets to prevent and manage gestational diabetes should be high in complex carbohydrates, fiber, fruits, vegetables, and whole grains.[107],[108]

Nutritional Management

The optimal dietary approach to type 1 diabetes is to implement a high-fiber, plant-based diet that is low in fat. A decrease in dietary fat may be linked to lower insulin needs as fat from food can diminish insulin action and intensify the liver’s release of glucose into the blood. The improved insulin sensitivity and decreased insulin need may persist for hours after eating a relatively lower-fat meal.[109],[110]

A high-fiber diet results in lower insulin requirements and improved management of blood glucose and lipids.[111] A 2020 publication of 2 case studies showed that individuals with type 1 diabetes who adopted plant-based diets rich in whole carbohydrates—including fruits, vegetables, whole grains, and legumes—were able to decrease insulin doses and decrease or maintain A1C levels.[112] The ability of plant-based diets to reduce cardiovascular risk factors is likely to be important in type 1 diabetes as well.

The dietary approach to type 2 diabetes has become considerably more aggressive in recent years, aiming not simply to improve body weight and glycemic control, but to reduce and even eliminate medication use and, when possible, achieve remission. The ADA defines diabetes remission as the maintenance of euglycemia (complete remission) or prediabetes level of glycemia (partial remission) with no diabetes medication for at least 1 year.[113]

According to the American College of Lifestyle Medicine, remission of type 2 diabetes should be the clinical treatment goal and is facilitated with a plant-based diet that focuses on fruits, vegetables, whole grains, legumes, and nuts and seeds, while eliminating animal products, added fats, and sugar.[114] Regular exercise is also a key treatment component.

In a consensus statement, the American Association of Clinical Endocrinologists and the American College of Endocrinology encourage a plant-based dietary pattern for weight attainment and maintenance for patients with type 2 diabetes.[115]

The ADA has taken a more moderate approach, suggesting that nutrition therapy should assist in achieving glycemic targets, improving body weight, and reducing cardiac risk factors.[113] The ADA does not endorse one eating pattern over another, instead recommending an individualized approach to medical nutrition therapy. The ADA recommends carbohydrate counting for people with type 1 diabetes and those with type 2 diabetes on flexible insulin dosing. For those on fixed insulin dosing, a consistent intake of carbohydrate is recommended. The ADA does not recommend any specific macronutrient distribution or eating pattern but does encourage consuming carbohydrate foods such as whole grains, vegetables, fruits, legumes, and dairy products. It emphasizes foods high in fiber and low in glycemic load, and discourages the consumption of foods and beverages high in sugar, specifically sugar-sweetened beverages. In individuals with a BMI ≥ 25, the ADA recommends reducing caloric consumption by 500-750 kcal/day (or 1,200-1,500 kcal for women and 1,500-1,800 kcal/day for men) to achieve a weight loss of 5% to ≤ 7%. The ADA also recommends that adults who consume alcohol do so in moderation, limiting alcohol to ≤ 1 drink/day for women and ≤ 2 drinks/day for men, and limiting sodium consumption to < 2,300 mg/day.[116]

Evidence from observational studies and clinical trials suggests that the optimal regimen for type 2 diabetes is a low-fat, vegan diet favoring minimally processed foods (sometimes referred to as a whole food, plant-based diet), along with moderate exercise.[114]This intervention has several simultaneous benefits for individuals with type 2 diabetes:

Improved glycemic control. A low-fat, plant-based diet reduces intramyocellular and hepatocellular lipid and increases insulin sensitivity, and may even improve beta-cell function.[117],[118] In a randomized clinical trial, in participants whose exercise and medications remained unchanged, a low-fat, plant-based diet reduced A1C by 1.2 absolute percentage points, compared with a reduction of 0.4 points achieved with a conventional diet.[119]

Weight loss. In addition, because low-fat, plant-based diets are both low in fat and high in fiber, they typically cause covert reductions in energy density and energy intake, which are not fully compensated for by increased food intake.[120],[121],[122] As a result, low-fat, vegan diets are associated with significant weight loss and reduction in fat mass and visceral fat. This is an important effect given that increased body fat, especially visceral fat, is associated with insulin resistance.[123],[124],[125] Randomized trials with individuals who are overweight have shown that low-fat vegan diets cause weight loss even in the absence of intentional limitations of energy intake.[126]

Improved plasma lipid concentrations. Because grains, legumes, vegetables, and fruits are low in fat, devoid of cholesterol, and rich in soluble fiber, plant-based diets are highly effective for reducing LDL cholesterol. Vegan diets typically cause plasma cholesterol levels to fall sharply, and the lipid-lowering effect of a plant-based diet is augmented by the inclusion of soluble fiber (e.g., oats, barley, or beans), almonds, soy protein, and sterol-containing margarines.[127],[128] It should be noted that, while most plant oils are low in saturated fat, coconut and palm oils are exceptions and will raise plasma cholesterol concentrations. The same is true for hydrogenated oils, which should be avoided.

Improved blood pressure. Vegan diets consistently improve blood pressure.[129] This occurs partly because plants are naturally low in sodium (which raises blood pressure) and rich in potassium (which lowers it). However, because plants are very low in saturated fat, they also reduce blood viscosity, improving artery wall compliance and blood flow, with the net effect of reducing blood pressure. Further, as vegan diets reduce body weight, blood pressure continues to gradually fall.

Carbohydrate type may influence glucose control. A review of 5 studies of individuals with type 1 or type 2 diabetes showed that low-glycemic-index diets significantly reduced A1C concentrations.[130] A systematic review and meta-analysis of 29 randomized clinical trials in people with type 1 or type 2 diabetes found out that diets low in glycemic index resulted in significant improvements in glycemic control, blood lipids, adiposity, blood pressure, and inflammation beyond concurrent treatment with medications or insulin.[131] Diets that provide carbohydrates in unrefined form or as low-glycemic-index foods are also more effective than low-carbohydrate diets in managing gestational diabetes.[107]

Furthermore, diets richer in fiber tend to produce lower postprandial blood glucose concentrations compared with fiber-depleted diets, and high-fiber diets have been shown to improve glycemic control in individuals with type 2 diabetes.[132]

The acceptability of vegan diets in clinical studies is similar to that of seemingly more-moderate therapeutic diets.[133],[134],[135],[136] Because vegan diets are based on the elimination of certain foods, they require no specified limits on portions, calories, or carbohydrates, and may be simpler to understand than regimens that limit quantities of certain foods without proscribing any. While individuals vary in their adherence to therapeutic diets, studies suggest that the more far-reaching the diet changes that are recommended by clinicians, the more changes patients actually make.[137]

Regarding alcohol, moderate intake, defined as no more than 1 drink per day for adult women and no more than 2 drinks per day for adult men, has not been shown to impair glycemic control, though excessive alcohol intake may cause hyperglycemia and weight gain.[34] In patients on insulin or insulin secretagogues, there is a risk of hypoglycemia, or delayed hypoglycemia due to impairment of gluconeogenesis in the liver during alcohol metabolism.[138] Additional drug interactions with excess alcohol intake include risk of lactic acidosis with metformin, and ketosis with SGLT2 inhibitors.[139],[140] People with diabetes should be counseled on these potential concerns.

Many patients with diabetes incorrectly assume that they should avoid carbohydrates in order to control blood sugar; however, low-carbohydrate diets often cause persistent elevations of LDL cholesterol, which are of particular concern given that cardiovascular disease is the principal risk of type 2 diabetes.[141],[142],[143] A systematic review and meta-analysis found that individuals most strictly adhering to carbohydrate-restricted diets had a roughly 30% greater risk for all-cause mortality compared with those who were least adherent.[144]

Vitamin B12

Vitamin B12 supplementation has an important role in several situations. First, a plant-based (vegan) diet is the diet of choice for prevention and management of all forms of diabetes. Because plant products do not contain vitamin B12 except when fortified with it, people who follow a mostly or exclusively plant-based diet must supplement with vitamin B12.[145] Also, because metformin interferes with B12 absorption, people taking metformin are at risk of deficiency.[146],[147] Vitamin B12 deficiency is also a concern for any adult over the age of 50 because of decreased absorption from reduced production of gastric acid and intrinsic factor with normal aging.[148]

In each of these situations, vitamin B12 supplementation is essential. The recommended dietary allowance of vitamin B12 for adults is 2.4 µg per day. All commercial supplements contain higher doses than this. A supplement of 100-200 µg can be taken daily, or a dose of 500-1,000 µg can be taken several times per week.[149],[150] Those who take metformin, proton pump inhibitors, and certain other medications may be at risk for vitamin B12 deficiency. Checking the B12 status of the patient with diabetes at least once a year may be warranted.[151]

Several studies have suggested that cinnamon may improve glycemic control and reduce the risk of type 2 diabetes.[152]

High-dose vitamin C intake may reduce the accuracy of blood glucose meters.[153]

Orders

A low-fat, vegan diet, favoring minimally processed, low-glycemic-index foods.

Nutrition consultation to support patients in implementing a healthful diet.

Pregnant patients should work closely with a dietitian to ensure adequate nutrient intake.

Vitamin B12 supplementation 100 µg per day or 500 to 1,000 µg taken 3 to 4 times per week. Vitamin B12 levels should be regularly monitored in patients taking metformin.

Patients on medications that cause hypoglycemia should be referred to a certified diabetes care and education specialist for instruction in preventing and treating hypoglycemia.

See also Basic Diet Orders chapter.

Exercise prescription, individualized.

Smoking cessation.

Refer to ophthalmologist, podiatrist, dentist, and other specialists as indicated.

What to Tell the Family

The risk of developing type 2 and gestational diabetes can be reduced through healthful dietary habits and regular physical activity (see Basic Diet Orders chapter). These measures are important even in individuals who are not overweight. In established type 1, type 2, and gestational diabetes, the risk of diabetes complications and the diseases and conditions that often accompany diabetes, such as high blood pressure, heart disease, and obesity, is reduced through healthful dietary habits, particularly a low-fat, plant-based (vegan) dietary pattern emphasizing low-glycemic-index foods.

Exercise, both aerobic and resistance training, is a helpful component of diabetes management. Smoking cessation can greatly reduce the risk of heart disease, peripheral vascular disease, and other complications of diabetes. For some, self-monitoring of blood glucose and use of appropriate oral medication and/or insulin are also recommended. Family members can help prevent and facilitate management of diabetes by following similar dietary and lifestyle recommendations and being aware of how to recognize and treat low blood sugar.

References